Hacker News new | past | comments | ask | show | jobs | submit login
Always Bet on Text (2014) (graydon2.dreamwidth.org)
570 points by asyrafql on Feb 17, 2021 | hide | past | favorite | 338 comments



As usual, context matters. If you're trying to convey complex thoughts about abstract matters, then language is the way to do it. If you are trying to convey deep emotional states, then a photo or video is probably better. You cannot convert the 1st amendment into photos, and you cannot convert the Mona Lisa smile into words.

Just try to use words to describe either of these photos and see how they fall short:

https://img.buzzfeed.com/buzzfeed-static/static/2015-10/19/1...

https://img.buzzfeed.com/buzzfeed-static/static/2015-10/19/1...

Between recorded voice and written text, I personally prefer written language. I like the ability to consume at my own pace, go back and re-read tricky bits, and easily search or quote. Plus it's much easier to use TTS to convert text to speech than the other way around (at the moment). Maybe someday technology will remove this boundary but currently that's my stance.


Strong disagree on photos. I find photos to be an excellent way to convey emotions that didn't exist. Also very common to have multiple people conclude very different things from a photo. Very misleading.

I'll also add that well written prose has moved me (and many others) emotionally in more powerful ways than any photo ever has.

Your Great Depression era photo conveys emotions only because of the context around it - context I know primarily because of text. If I showed this photo to a relative of mine half way across the world, they would merely see an ordinary woman with two sleepy kids.


>Also very common to have multiple people conclude very different things from a photo.

But that's true for text, too. The US just had a very major politicized event where two sides had two wildly different conclusions over what one man's words meant. And there's a grand legal tradition of taking one specific document and interpreting what its words mean and applying it to a plethora of situations. There's plenty of people that come to wildly different conclusions over what certain amendments do or do not mean.

Hell, look at literary analysis! You can give people a book like 1984, and some people will say in all confidence that it is explicitly anti-socialism, when the author said the literal opposite. And that's something with a supposedly clear ideological message, to say nothing of works with less concrete themes.


All three of your examples concern people with a vested interest in arriving at a certain conclusion established long before they ever actually look at the text in question. No communication medium will solve the problem of humans trying to bend that communication to fit their pre-existing opinions.

For neutral observers, I would definitely argue that text is far more concrete than photos. There's a whole host of reasons for this in my opinion, but one that hasn't been mentioned in this thread is that humans are used to seeing images in three dimensions, whereas photos are typically two-dimensional. By removing an entire axis of data you're introducing far more room for ambiguity than could have existed previously.


The ideological message in 1984 is anything but clear when you compare it to what the message would look like had George Orwell merely written a nonfiction treatise, without resorting to any literary devices.

If you know nothing of Orwell or the historical importance of 1984 and you read the book, it certainly wouldn't be clear that Orwell was trying to make a point, let alone what that point is. Some people would guess it does, others would say "It's just a story." Orwell intentionally obfuscated his point in choosing the mode of delivery (fiction). Indeed, this intentional obfuscation/ambiguity by fiction writers is what keeps literature professors employed.

The difference is that you can read a work like 1984 and at least know it is fiction, and that the message, if any, is hidden. Whereas you cannot look at a photo and easily tell if it's documenting what you think it is, no matter how much the photographer tries. At best you can see the shapes, and to a much lesser extent, things like colors/contrast. What the photographer was trying to convey is much more open to interpretation than 1984 is.


I disagree on obfuscation, and I think Orwell would too.

Orwell probably knew that if he wrote a dry academic tome, no one would read it, and that would obscure the message by default.


Sorry, when I said "clear ideological point" I meant "clear that some ideological point was being made", not that it was clear what the message was. I thought I made that clear by immediately pointing out that people disagree on what the point of the story is.

The fact that I'm even having to make this clarification is kind of the point of my argument. And that you can even argue that it wouldn't be clear that 1984 is trying to make a point is evidence that text is not inherently clear.

Also, I don't know how you can claim that it's easier to tell what text is documenting than a picture. It's much easier to lie with text.


> And that you can even argue that it wouldn't be clear that 1984 is trying to make a point is evidence that text is not inherently clear.

No, it's evidence that George Orwell was obfuscating his point. When you make a point through fiction, that is obfuscation. To make my point clear, had you made a faithful version of 1984 into a movie, people who know nothing of the book would still not be clear about whether it has a point or not. It has nothing to do with it being in text form.

> Also, I don't know how you can claim that it's easier to tell what text is documenting than a picture. It's much easier to lie with text.

The two are not contradictory. It's easy to tell what is being documented. Whether it is true is a different story.

The crux of my discussion isn't about what happens when the author/photographer is intentionally trying to mislead, but when they are not trying to mislead. If an author writes a nonfiction treatise it leaves a lot less room for ambiguity. When I read PG's essays, I don't have discussions with others on what he really means. A photo, without context, can inspire several emotions, all unrelated to reality, and this is true no matter how hard the photographer tries. Case in point, when I was in school, my teacher showed the class a photo he had taken from a newspaper - but without any context. He asked the students to try to write a news article where the photo would make sense. Obviously, the responses were widely different, and none was close to the reality. In some stories, the subjects in the photos were victims. In others, they were the villians. The reality turned out to be their being heroes.

Even when you provide accurate context to a photo, there are no shortage of instances where once you see the video of the event and/or hear from the subject, you find out that while the context was correct, the photo is showing a mere moment that is not reflecting what most people who see the photo are thinking. Take a video of someone having a long conversation, where they laugh often. It's not hard to capture a frame where it looks like the person is in pain.


You’re comparing incredibly arbitrary things. Is a photo comparable to a tweet, sentence, paragraph or novel? Is a video equatable to an essay?

You’ve convinced me that it’s all incredibly relative and complex, very much impossible to set up an equation with which to even compare it, and there’s no clear cut answer - and further, I don’t see why one side even needs to win, or why anyone cares beyond some personal attachment or game or identity pride or maybe just a childhood love of Wittgenstein.

Are symbols text? All of math could be represented as symbol. Are spoken words text? What about animal calls? The intro to Wall-E conveys so much, did I need text to understand that? I’m persuaded by a seeing a video of beautiful village in its natural state, or nature. Li Ziqi is maybe the most compelling video on all of YouTube, and it has no audio or text whatsoever.


John Steinbeck arguably did that for your first image: https://en.m.wikipedia.org/wiki/The_Grapes_of_Wrath

It's not really apples to apples of course. Steinbeck describes much more, but also takes much more time to consume. The point though is that it's possible to convey very complex emotional states in literature or poetry


For anyone who isn't familiar with the subject of that photo:

https://en.wikipedia.org/wiki/Florence_Owens_Thompson


I mean, if your best argument is that one of the greatest authors of a generation needed an entire book to capture the emotion of a single picture, that argument doesn’t seem very strong to me :)


Sure, and the author recognizes that. FTA: "Pictures may be worth a thousand words, when there's a picture to match what you're trying to say."


Abstract thoughts are conveyed with a class of language, this has not worked well in its predominance historically ( only math survived) similarly how music is communicated but not accepted as a universal language.


The question is whether these advantages are because text is a better communication medium, or because of the limitations of our technology?

Take using a computer to type a word versus draw a picture: Effectively everyone who uses a computer can type a word, but I'd bet less than 10% could draw a circle.

But put a piece of paper in front of someone and effectively everyone can draw a circle.

This points to there being a limitation in the technology for working with other forms of media, not the effectiveness of the communication medium itself.

This isn't to say that text isn't also a better communication medium, but it is to say, until the technology has improved for communicating with other media, it's difficult to compare without basing the decision on the limitations of the technology.

In other words, most of the perceived advantages of text are really advantages of text being easier to represent digitally (or generally reproduced, e.g., printing press), not advantages of text as a communication medium itself.


It is not plausible that, but for the limitations of technology, you could substitute a picture for the argument presented in your post.


Walter Ong said something like (paraphrasing): "If the phrase 'a picture is worth a thousand words' is true, then why does it need to be a phrase at all?"


Cause that phrase is just 7 words. A picture would have 993 other unnecessary words.


Never mind the 993, show us a picture that conveys those seven -- and any 993 other unnecessary ones you want. If the seven were true, that should be easy. Heck, I won't hold it against you if there are as many as 1,042 extra words, or as few as 763.

But I betcha you can't find one that clearly says what seven little words of text do. Which goes to show that those seven little words are a lie.


I don't disagree, I think some ideas are better expressed as text others by other forms of media. E.g., that's why we use graphs, diagrams, storyboards, flowcharts, etc... in addition to text to communicate.

For the record, I came up with this thought, about computers and software being optimized for text, by realizing I learned the best by reading information presented as a combination of text and media (e.g., a textbook with diagrams), but when I communicate with a computer, I always just use text, why? Well, making media with a computer is a huge PITA, I'm assuming because you program the computer itself with text, so its entire interface seems optimized for that. (This is particularly interesting looking in contrast to smartphones and tablets, which are entirely not optimized for text.)


I take your point, as I often sketch things myself to help me think, but only pencil-on-paper. I would guess there are tools for sketching with a stylus on a tablet, but I have never felt a pressing-enough need to push me into checking them out.


Ah come on, this is not accurate at all.

We have tons of tech now: audio, video, charting, etc.

Circles are pretty easy in powerpoint or any such software.

The tech is obviously here already.

But still, why do people prefer sending chat texts instead of calling or video calling? Surely not because of tech.


> But still, why do people prefer sending chat texts instead of calling or video calling? Surely not because of tech

It turns out that people who are highly fluent in a language with a compact alphabet like English prefer sending text chats. My Chinese colleagues and friends routinely send many more voice chat messages than text because it is much more efficient than typing mandarin.

Also in India, I've observed that people fluent in vernacular but not in written English primarily send voice chat messages - again because the vernacular text input is very inefficient

So interestingly does my daughter who can type English text easily but as she always grew up in a world her parents and her friends parents had smartphones, she finds it a lot more comfortable to send voice chat messages.

The other benefit of chat is asynchronocity. You're not forcing the other party to do a context switch and signalling that they can get to ot.


No, GP is right, for a simple reason: look at your input devices. A keyboard and mouse/touchpad combo is most effective at interacting with text. The story would be entirely different if the main interface was e.g. a stylus. A instant messaging service that sends hand written scribbles instead of text follows much more naturally from that.


This is why fax machines persist in Japan.


Some of us are on a desktop computer, all of us are on a mobile phone.

You do realize how cumbersome that mobile keyboard is right. With a click of a button, you can record audio and/or video.

But still, text messages rule them all.

Who sends texts on desktops anyway? A small minority.


No. A lot of people record voice messages, but they need to be in an environment providing enough privacy. But messengers emphasize text heavily in their UIs - again because of keyboards. Keyboard less touch devices are too novel an innovation compared to when instant messaging was cast into its final form.


> but they need to be in an environment providing enough privacy.

Yet another argument in favor of text.


Just an argument against audio and video. With a stylus instead of a keyboard you can still scribble amd draw naturally. That was my original example.


Wikipedia's List of codecs page (https://en.wikipedia.org/wiki/List_of_codecs) is a good example of the explosion of complexity that arises when representing media digitally.

To this day, this complexity is still mostly offloaded to users in the form of export settings, and supported playback formats. One basic requirement of the technology reaching the point good media support is if a user never encounters codecs, in either export settings or playback problems.


My kids have no idea what a codec is. Still they record and play videos and audio messages all the time.

I don't know in which world or time you are living, it's been a long time since I touch anything related to codecs.


This is a fair point, there's a lot you can get away without worrying about the complexity today, and that amount is certainly steadily growing (which is great!).

But particularly on the creation side, if you say, try and combine a vector created in one program with a bitmap created in another, you'll likely encounter some issues (e.g., color profiles), or taking an animation created in one program and overlaying it over a video in another, you'll often need to know your codecs. Or exporting a video optimized for social media, e.g., videos shared on Instagram and Twitter often have heavy compression artifacts.

I'm really looking for seamless creation and sharing freely combining media from different programs, with the same ease as cut, pasting, and editing text. (As well as interfaces that are as easy to use as manipulating pen and paper with your hands.)


> Why do people prefer sending chat texts instead of calling or video calling?

This isn't universally true. I (unfortunately, for me, since I hate it personally) know plenty of people who prefer voice or video calls. I even have people who will reply with voice recordings to my text messages.

Many people prefer voice or video.

I prefer text because of two main reasons:

1. Its asynchronous. I can reply as needed while not being disturbed if I happen to be in the middle of something.

2. Its not real time. That is, I can take my time to form my response. I can proof read, edit, clearly form my thoughts etc. In a voice/video recording, editing is difficult, in a live setting, I'm under more time pressure to finish my sentences rather than thinking about them more.

I don't prefer text because of technology reasons, except those that make text more asynchronous and easier to proof read and edit.

But many people I know or have interacted with hate text and prefer voice or video.


There's a lot of selectivity happening on this post. Nobody here has worked in an office where the management prefer video calls instead of Slack messaging? I've seen no shortage of people on HN argue that this kind of text-based communication is inferior to video calls or screen shares when solving tech problems.

It seems pretty obvious to me that text, video, audio, and images are good for different things, and that diversity is reflected in the ecosystem. There isn't a universal trend towards or away from text, not really. People are making fewer voice calls, but they're also consuming more content on Youtube instead of on blogs. They're also moving a lot of their social life onto platforms like TikTok and Instagram, which are primarily video/image.

I don't see a universal trend here in either direction. I think people are cherry-picking a couple of isolated trends or points and making really broad, generalized statements that aren't warranted.

The whole idea of picking a "best" medium here is silly anyway. It's childish, in the same way that people used to argue about whether books, movies, or games were the 'best' artistic medium. Different mediums are good for different things.


Actually I would argue that people send texts despite it often being highly inefficient. It often takes an infuriating amount of text messages to agree on something that would only take a 2min call. It's just we're so used to sending chats.


Chat has other advantages. Lets say you're trying to agree on a restaurant for dinner:

- If I'm busy right now I don't need to stop what I'm doing to answer the call, or you don't need to try again later.

- Need 2 minutes to look up your schedule and suggest a time or find a restaurant? That's fine, I can do something else until you reply and you don't feel like you need to rush the process.

- 6 hours later I forgot which restaurant we agreed on? It's ok, just open the chat, it's right there.

- The chat is searchable, if 2 months later I don't remember the restaurant name I can probably find in a few seconds by searching some related keywords like "food", "restaurant", "dinner", etc.


This depends on the people involved. I find most of my programmer friends prefer text in most circumstances, but most people in sales, marketing, accounting, HR and management that I've worked with tend to prefer voice, video, face to face. As do people in other industries (eg my mother loves voice and video and while she does use text a lot, mostly to talk to me, often finds it tedious).


This is why my default is text until a call becomes obviously better. "Where do you want to go for dinner?" ten messages of hemming and hawing, "Call me when you're free". A call forces an interruption on the person, or a game of phone tag. Texting can be answered when each person is able, and can be used to coordinate the call.


Don't tell Gimp that it's easy to draw a circle...


In my case because of the asynchronous nature of text versus vide or voice calls. An effective compromise is voice notes in modern messenger systems, which preserves asynchrony while also allowing the expressiveness of speech.


There's an entire dialect of English that has evolved just to handle expressiveness in text formats. There's a standard and reproducible grammar, and most people (who do or have spent any time on the internet in the last 15 years) use it a second language. Messenger services allow the flexibility of asynchronous communication with the added benefit of synchronous communication when it's important (that's why it's preferred often to email, for example).

Voice notes, like email, are fully async and require actually listening to someone talk, which changes the communication media that you're working in.

Messaging has risen to the prevalence it has because of the features it brings to the table, not in spite of them.


Imagine you have to communicate something precise to me and I am stuck in a meeting:

- you send me a picture, it may be ambiguous;

- you leave me a voice call, I have to go out of the meeting room to listen to it

- you write a message, it just works. If there is info to use, I will easily copy your message in any application (Excel, Word, SAP...).

Also, text can be stored easily, so I will keep an archive of your text, probably far less an archive of your voice message or diagram, as they are very heavy.


For an example of shiny new technology being better than text, have a look at Kevin Simler's interactive essay

http://35.161.88.15/interactive/going-critical/

There is something about playing with the sliders and watching that plague die out and flare up, that makes the point with a vividness beyond words.


I think he's trying to express a point about the gradient is the whole human-communication-technology system than precise properties of any part of it. And I say "express" rather than "make a point" because I don't think he expects to convince anyone. It's summed-up life experience, not a proof.


If someone can draw a circle on a piece of paper they can do it on a computer. The only difference is that we start you practicing doing it on paper in preschool.


A circle doesn't usually convey very many ideas, though. Words can.


Circles are building blocks of pictures, which can convey ideas, just like words are a building blocks of sentences.


A circle has a fixed level of abstraction, and representing a complex thought requires many circles. Importantly, in a complex image each individual circle doesn't convey much information, it's the composition that encodes any information.

Words have a near arbitrary level of abstraction, so that single words can convey immense collections of ideas, so much so that new words can be defined that represent the ideas that would otherwise take many words to express.


  ◯ ◯

   ◠


> text is the most powerful, useful, effective communication technology ever, period.

On the other hand, many (technical/mathematical) concepts are more effectively explained using diagrams/images.

E.g. the very visual approach taken in Mathologer videos [1] makes difficult, esoteric, mathematical ideas accessible to a wider audience.

[1] https://www.youtube.com/watch?v=iJ8pnCO0nTY


Yes. See also Florence Nightingale using rose diagrams to illustrate preventable excess deaths to infection during war.

https://www.florence-nightingale.co.uk/coxcomb-diagram-1858/


Also where does text end and diagrams/images start?

E.g where does musical notation stand? For an untrained eye it is just a bunch of lines, dots and squiggles but others will read it like text and much more efficiently so than if it were written words.

Similarly the example about the Wikipedia text of human rights being supposedly impossible to convey via image. But what does the text convey to you if you don’t speak English?


> Also where does text end and diagrams/images start?

Text is the written form of language. And language is defined as a sequence of symbols. Thus, text is a subset of diagrams. Arbitrary diagrams can be of any topology (e.g. DAGs or even have cycles), however text always is of sequential topology (or a hierarchy as there is an isomorphic mapping to sequences).

Basically, if there is a canonical way of reading it out loud, then it has an explicit sequence and is thus text.

By that rule it follows that: Musical notation is text as it is strictly sequential (in time) and maybe hierarchical if there are multiple tracks. Most mathematical notations are text, as they are also sequential, even matrices are only a hierarchy of sequences which in turn can be serialized. Chemical / molecular diagrams are not text, they can contain cycles. Same goes for electronic and logic circuit schematics or rail way diagrams found in public transportation.

> Similarly the example about the Wikipedia text of human rights being supposedly impossible to convey via image

Schematics and diagrams are only use full if / because they are subject to a certain defined visual vocabulary. So if you had a visual vocabulary in that domain then you could express them as a diagram.


Also, what does it convey if translated into (say) Egyptian hieroglyphs, if you're an ancient Egyptian who can read hieroglyphs...


Hieroglyph writers have to resort to tricks like punning to express ideas that are not simple pictures. There's a reason why written language became more abstract.


On the other, other hand. I find text far superior for creating and editing technical diagrams and images. Give my graphviz over any GUI editor to do the same.

In this case the diagrams are good for communication but terrible for expression, I think this distinction is being overlooked by many here.


Text is not the oldest - nor easiest to use -, text is a heavy abstraction born somewhat recently in human history.

The oldest and easiest is the visual in 3D, then comes the visual in 2D.

Those exist since we have eyes.

Our brain have dedicated and sizeable infrastructure for that. Children can communicate in 3D (gestures, posture, expressions, objects) and in 2D drawings before learning textural communication with great effort.

Text is more regular and reliable in certain contexts (not always, sometimes a pictogram or others are better), when the circumstances are proper for that.

Text has its uses, just like all the other forms, not being paramount, not at all!

(I'd also argue about that we could read old texts. Sometimes yes, but sometimes we cannot read present ones neither if the cultural and knowledge background is inadequate. Which is just aggravated by the ages)


As someone who worked long in the visual field (Film, VFX, Graphics, Art) and has a ton of experience with text based expression (studied philosophy, programming) in my eyes picture based communication certainly is powerful.

However reading pictures is much more subjective than most non-arts-educated might tend to realise. We had a weekly class where we would discuss scenes and pictures and what it evoked in people, what they "read" in it. The one big takeaway I got from this, is just how profoundly different a room full of people can see very clear pictures. This subjectiveness is even worse when you look at gestures and body language of actors – what one student saw as strong and self sufficient, the other might see as forceful and destructive etc.

Of course we also have codified visual languages (traffic signs, warning labels, ...), but they only will work for low complexity info ("Warning slippy surface", but not: "Watch out the last step of that stairway was built to high and might cause you to fall"). The low information density of such symbols is great if you want simple messages to be understandable by a big number of persons very quickly and with little cognitive overhead.

Text can shine when symbols don't suffice, when pictures are to vague, when gestures are unclear. Text is easy to create, modify and copy and I love it for that. My freelance time as a graphic designer convinced me that many people think they can communicate visually, while very few actually can. A lot of people already have issues with getting their message (be it text or spoken word) understood in the way they meant it at the side of the recipient, with visual language this is likely worse.


The thing is, "text" and "pictures" are somewhat vague super-categories that aren't very well defined in a lot of comments on this thread. It's ironic that a discussion about the precision of text is noisy because of of the imprecision of text.

Are emojies text ? the blog post found a 4000-byte Twitter logo and cited this as evidence of the efficiency of text, but they forgot that there is a vast store of similary-sized photos that cost nothing more than 3 or 4 bytes and sometimes less, comparable to the very same letters they are using to write the post (and conveying much more than a single useless-on-its-own sound). By any definition that depend on the fact that text is stored as contiguous bytes, emojies ARE text.

Are 2D tables text ? they are not linear (not as we see them) yet one table with entries arranged appropriately could be much more percise and expressive than a truck-load of prose paragraphs. And an xy plot is much more expressive than both. 2D tables is not that much more expensive in storage than text though (a bunch of html tags surrounding the text is one representation), and an xy plot could be derived from a table by computation. So with a little bit of computation you gain the precision and expressiveness of a visual plot with the storage overhead of a bunch of numbers. This example (simplistically) mirrors current work in ML on compression of photos and videos. Imagine if computers are advanced enough that a picture is no more storage than an emojii, all arguments. about "efficiency" of text that appeal to its economic representation in computer memory become moot then.

The point I'm making is: "text" vs. "pictures" is not fine-grained enough of a distinction. The real distinction is [ "numbers" vs. "graphs" vs. "letters" vs. "tables" vs. ...), its a big multiway competition between various forms of symbols and symbol-organization schemas humans invented for a variety of purposes and arguments for or against any of them shouldn't appeal to how they are represented or communicated (text could be drawn on photos or arranged on powerpoint slides, images could be compressed and redrawn to a high degree of fidelity). And there is a lot of paradigms "in-between" that piggy-backs on one of the two representation but aren't either of them (plots, tables, emojies). Tools and technological limitations are also to blame for a lot of pictures' shortcomings, the ability to copy-paste being one of them.

A lot of discussions here would benefit alot if all parties state beforehand exactly what are their definition of text and what their argument appeals to in that definition.


The reason human knowledge develops so fast is the existence of text. If we rely on visual in 3D we will be the same as other animals. Apparently now we are behaving badly since we are consuming videos and pictures rather than text. I believe text is superior because of its abstraction, when you read, you are creating alongside the author.


Re: history.

I agree on the power of text as a technology, and the role it played. China & the Roman Republic/Empire are good examples of this.

But.... I think it's worth remembering that we lost something as we gained something. We tend to severely underestimate oral "technologies." Scholarship existed before writing. History, geography, etc. Text spent centuries or millennia as a peripheral media. It was mostly used for accounting in Mesopotamia & the Levant for thousands if years. Sometimes for religious, magical or political reasons. It wasn't a major medium for philosophy, storytelling, history or such until much later... So writing didn't really play much of a "knowledge accumulation" role until pretty late in the game.

I suspect that it developed so slowly because oral traditions were hard to beat. They had their own advantages. A song was an efficient way to learn history.

An important, if subtle, fact is that mediums are not just for communication. They're modes of thought. Text and speech will yield different ideas. Mathematics are a huge example. Ways of conveying mathematical concepts (eg negative numbers) enables us to conceive of mathematical concepts. If you write an essay, the ideas/conclusions you will have will be different. Even the difference between a scroll and a codex (book) can make a big difference. That difference is evident if you compare the modern practice of Judaism (scroll tradition) to Islam and Christianity (book traditions).

Socrates/Plato give us a nominally dividing line between the oral and written approaches. Socrates may have even been illiterate, but either way, his main medium was oral. In fact, most Greek philosophy came from the "mostly oral" period. This is why Plato, Aristotle, (Diogenes?) and others of that generation become so important. They're the link. They wrote down ideas created by oralists. This is how they could be accessed by macedonians, Romans and such.

I wonder if the charming, curious style we associate with the likes of Socrates or Diogenes is inherent to oralism. Compare them to later, literary philosophers... The literalists are far more grim. Senecca comes to mind. Even Aristotle. He's not as grim as roman philosophers, but he is a lot more serious. The oralists were playful... and greek/roman philosophy (imo) declines as writing overtakes oral traditions.

Socrates' thoughts "On the Forgetfulness that Comes with Writing" are recorded (tellingly) by Plato. It's not Plato's best piece, but very relevant to our times. If you memorise instead of using text, all your knowledge is inside your head. On paper, ideas are lifeless. Living ideas inside your head interact with each other, refine, create new ideas. When we convey them to one another, we can ask questions, read expressions, etc.

We don't just have books, we have the internet and pocket computers to access it. Socrates' point applies even more now.

One relevant example relevant to our times is "shades of uncertainty." Say you read an article about the economy, GDP growth, unemployment & such. A lot of that information is uncertain, either inherently or at this point in time. There may be dissident positions. That's usually lost in text but not in conversation.

I definitely think a short conversation about this year's economic data is more informative and deep than an article by that same economist.


> We tend to severely underestimate oral "technologies." Scholarship existed before writing.

This comes up a lot in Bible scholarship.

Many modern people assume the words of Jesus, for example, were lost to a game of telephone before being written down years later in the form we have today.

But conveying the teachers words with true fidelity to what he said was very important to a rabbi's students, and they had a culture and techniques to make sure they did this with high reliability.


I agree on the principle, but not the example :)

Year 0 in Judea (and Rome, Damascus...) was a pretty literate period and literacy was (the story of jesus confirms) already a religious requirement... bar mitzva or an ancestor of that custom. Bar mitzva translates roughly to "eligible to uphold commandments." You need to read for that, or so the custom implies.

I'm sure that most of the religious/rabbinical tradition was still oral, and that rabbis did a lot of oral teaching. But, I think high fidelity oral "technology" was already heavily diluted, especially in judaic culture. Scriptural worship starts very early. It's hard to say when exactly, but it had to have happened while hebrew/canaanite was still spoken in the region. Aramaic had overtaken hebrew circa 400-500 BC. From that point, high fidelity transmission was done with writing.

Also, the multicultural/multilingual/multiregional context makes it unlikely that a high fidelity oral tradition existed in early christianity. IMO, the new testament was almost certainly compiled from earlier written sources... and oral telephone. I mean, the new testament isn't even in the same language as the sermons.

The same can be said about mishnah/talmud... the jewish contemporary to the new testament. Traditionally, it is seen as a compilation of jewish oral traditions, received at mount sinai and maintained with fidelity for two thousand years. Realistically, there were earlier written versions of (eg) Rabbi Hillel's teachings available to the scribes who compiled the Mishnah.

The traditional reasoning for writing the "oral torah" (resulting in talmud/mishna) was that oral traditions were dying, and that writing was necessary for fidelity. Multiregionalism, multilingualism and such were to blame... and the christians would have had even more of those problems. Fewer, more dispersed. No institutions. No common language. No old traditions. It's possible that there was a tradition of reciting Jesus' sermons, but that would be kind of culturally out of place. I think it's pretty unlikely. If there was, I think the new testament would have been compiled in aramaic.

Also... there are quite a few convergences between new testament stories and other (broadly termed) rabbinical accounts from the period. John the Baptist has his own religion, for example, and in their books you get some of the same stories, but with John in the Jesus role. I suspect there were many others, but have no modern adherents.

Judaism of that period aggressively trimmed out any new or recent "revelatory" writings. New prophets, new conversations directly with god. That's what many of the "apocrypha" are. From then on, religious scripture needed to be wisdom received from oral traditions, old sages and stuff. No revelation.

These, to me, strongly suggests late 2nd temple judaism was no better at oral tradition than us. That said, you can have conceptual fidelity without having word-by-word fidelity. When Jesus paraphrased Hillel, he was reaching across 400 years of oral (probably/mostly) tradition. Christians doing the same thing 200 years later probably had that level of fidelity.


So true


That's true. But the oldest isn't necessarily the best; civilization happened at roughly the same time text started becoming widespread, and that doesn't seem like a coincidence.

It's also interesting that 3D is so hard to translate into a usable context in computing. VR has felt like a second-class citizen compared to a mouse and keyboard in terms of usability. Not that the two are mutually exclusive – it was nice having as many giant monitors as you wanted in a 3D space.

What's needed is a 3D web browser, with websites connected with portals that you can walk through. I think gather.town is surprisingly close to that. http://gather.town/


Please also see primitive.io which is a VR browser for GitHub.


> visual in 3D

I cannot see in visual 3D. If anything I can see in strictly bigger than 2D. (a flat 'screen' with a tinge of depth perception, which I fear I don't make much use of when I spend most of my time in front of a flat 2D computer screen)


Text, in the end, is not more than a 2D agreed representation. It's drawing + rules.


Agreed, with a counterpoint:

Text is efficient at transmitting data. If I want to describe a concept or an event, text is king.

However, media is more efficient at transmitting sentiment. It will take you far more than 4000 bytes of text to transmit the feeling or emotion an icon can convey, when used well. This is why we've (as a species) started using emojis, and why media leans to emotion and sentiment while text leans to data.

This is an exaggeration, life is almost always a grey area - but I hope you get my point.


Emojis evolved from simple text like ":)" or ":/". And for the new set of emojis, half the time I am not sure what the sentiment is being expressed!


Add to that that, depending on the emoji font, the emoji will look completely different[0]. So as the sender you have no guarantee that the recipient of your message will actually interpret your emoji as you intended. It really is beyond me why the Unicode consortium thought that putting emojis in a character set would be a good idea.

Sure, having a code point in your character set that represents a bird[1] makes sense but I really hate that font designers now have control over the way I get to express my emotions and how others perceive them.

[0]: Sure, ":)" and ":/" also depend on the font but much less.

[1]: https://news.ycombinator.com/item?id=26164409


I'm reminded of a story I heard where someone responded to news of a death in the family with what they took to be a crying face, but what the recipients took to be a crying laughing face.


Text can have problems with people guessing what things mean from the situation they first see them (where the meaning is not entirely unambiguous).

I've seen a similar story of someone thinking "lol" meant "lots of love" and used it to end messages regarding the death of a family member.


The thing is, with emojis the differences introduced by font designers can be very subtle. In the GP's case, the laughing emoji seemed so much out of place and violated social protocol so heavily that I assume the confusion was probably cleared up later on.

But what about cases where a font designer changes, say, an emoji face with a light, content smile to a shy smile? This has the chance to undermine the entire emotional content of a message, without sender and recipient ever noticing.

For a more concrete example, have a look at https://emojipedia.org/beaming-face-with-smiling-eyes/ . Is it just me who thinks that the glyphs by Apple and WhatsApp show a beaming face with a somewhat sly touch whereas most others simply display happiness?


The Softbank one looks like someone grinding their teeth in frustration to me.


> why the Unicode consortium thought that putting emojis in a character set would be a good idea.

That puts blame on the wrong organisation. The consortium is to the most part only standardising existing character repertoires. Emojis were popularised by a mobile phone equipment manufacturer as a marketing stunt. http://enwp.org/Emoji#History


Emoji are modern hieroglyphs. It's just an additional means of conveying meaning.

> font designers now have control over the way I get to express my emotions

This could have been a major issue, but media platforms recognized it and have provided their own pictograms for years, mostly compatible with each other.


They have a valid point, however. What is now widely considered to be a "water gun" emoji started out as an actual gun, but Apple decided at one point that they were going to be "progressive" and changed it to a water gun. To me, this upends the entire idea of emoji having a stable definition.


Words don't have stable definitions either. My point was that there is nothing special about emojis any more than there is about Chinese characters or Egyptian hieroglyphs.


Sure, but in the case of words, both sender and recipient at least agree about the content of the message, even if their interpretation is different.

In the case of emojis, both the sender's and the recipient's software agree on the message content but the actual persons no longer do because the content is suddenly being displayed very differently.


It's fascinating to me that they did this with the pistol, but :knife:, :bomb:, :dagger:, :crossed_swords:, :bow_and_arrow:, et al all look as threatening as ever.


But those weapons don't have the same highly political negative associations as guns do. Knives are tools more than weapons; bombs can kill a lot of people (see the 2013 Boston Marathon attack or IEDs in Afghanistan), but their typical stylization is also very cartoonish and there isn't an NRA for bombs; swords are archaic and mostly shown as medieval or pirate weapons; and bows and arrows as weapons are also archaic, and they're only used today for hunting.


> It really is beyond me why the Unicode consortium thought that putting emojis in a character set would be a good idea.

iirc some japanese character sets already had them (introduced by telecoms for text messaging?), so originally they were included for compatibility [i.e. to fulfill Unicode's goal of being able to encode everything]. then people discovered the emoji keyboard on iphones and it got popular


They did it for compatibility with Japanese encodings.


Additionally, before emojis, we were saying things like :shrug: in IMs and we conveyed the same thing. It was a specific variation of text, which I still imagine emojis to be. I don't think the shrug emoji is more efficient at conveying sentiment than :shrug:


Even back when I started, around the ascendency of the WWW when Usenet was still ABigThing™ (though after the start of the Eternal September, I'm not quite that long in the tooth!) there was already confusion with the simple text smilies.

Ignoring the arguments of whether to include the nose (the original form was :-) which I much prefer) on the basic examples, people introduced new combinations faster than I cared to pay attention to so knowing what they meant was not always easy.

It ended up that there were whole dictionary like lists of them in some Usenet groups' FAQ documents.

As soon as you have more than a few of any symbolic representation (text smilies, emoji, gifs/memes[†], ...), it becomes dynamic grammar in its own right and away from the core few it is a mess of people not understanding what you mean either because they don't get your reference or you have used a reference incorrectly (or, if incorrectly is the wrong term, in a manner differing significantly from its common use).

[†] it is less of an issue with meme images/animations as they usually have a text portion making them far less ambiguous, but the issue is still there overall


That’s part of the fun of emojis. It’s also a counterpoint to the original post. There is a lot of information in a pictorial emoji, so much so that it can be bent and reimagined to many contexts.


> media is more efficient at transmitting sentiment

I don't agree with that. A photo doesn't contain much information about the emotional state of the photographer, unlike a few lines in a diary. My travel diaries are much richer in sentiment than my travel photos.

To me, emojis are a form of data compression. Common concepts are compressed into symbols: happy, sad, car, eggplant with 3 drops. Emojis are macros. That works as long as you exchange common concepts ("I feel sick"), but it falls flat if you need to venture beyond that ("My back is itchy"). You couldn't write a country's constitution with emojis. At least you shouldn't.

Likewise, a picture only shows what's visible. Travel photos don't capture the smell, the temperature, or how you feel after staying up all night with a sick stomach. A few words in a diary will.


“Text is efficient at transmitting data. If I want to describe a concept or an event, text is king.”

Really? Why are video games so visual, in that case? To update you on the situation your character is in, inform you where the enemies are, give you feedback on your current health and objectives... most video games use a rich graphical system of colors and symbols overlaid over a high resolution image of the game world...

If text is more efficient at communicating ‘events’, why aren’t there ‘action text adventures’?


Because games are optimized for immersion and being roughly analogous to reality, not for efficiency of describing events. That would be the purpose of a news article. Pictures are included, but text is the focus.


There are action text adventure games, even massively multiplayer ones.

I probably have blown more time on text MUDs than all other computer gaming combined, but even back when I used to play things like counterstrike I spent an inordinate amount of time using text to chat with team mates or in the console binding and triggering macros.


Hot take: emojis are a type of text, for this very reason. The entire unicode standard is what humanity has settled on as "text".


Could you clarify what you mean with media?

According to my understanding a medium is a mean of communication, thus including text. OTOH it seams from your message that these are two complementary objects. Is my understanding correct?


I think the post conflates several quite unrelated concepts under the label text. Also what does he mean by text and information? For example he mentioned the "optical telegraph", which is a semaphore system which used a system of messages which AFAIK were not alphabetic text.

Another example if we compare logographic language systems (e.g. Chinese characters, hieroglyphs ...) to alphabetic systems. I guess we can say both are text, but logographic systems are close to pictures as well (pictographic systems even more so) and they can convey information in much less characters, at the cost that one needs to know a much larger "alphabet".

Similarly if the information we are trying to transmit is actually an image, transmitting the actual image is certainly much less information than transmitting a description of the image. That is also why the comparison with the bird and twitter image falls flat. The image conveys a specific image, not the generic image of a bird. To describe that specific would need many more characters than 4. Similarly if I have a large set of numbers, it's much more efficient to store and transmit in binary format not as text.


>..."optical telegraph", which is a semaphore system which used a system of messages which AFAIK were not alphabetic text.

The pulses of light in an optical fiber are not alphabetic text either, but we are not talking about baseband here.


Yes but that's unrelated. We can clearly say that digital (binary) representation of information has won for encoding and the information content is well understood since Shannon. However this is not really connected to the current discussion, because binary is clearly is not meant for direct human "consumption".

I was restricting myself to discuss abstractions that are directly used by humans. The messages send in the optical telegraph are conveying things more complex than letters which I would say is what we associate with text. Now you might say that those messages are text, but then we might as well say language is the most flexible and efficient way of communicating, and that's probably correct, but also completely meaningless in this context. That's the issue with this posts, it's either so general that is essentially meaningless, or so specific that it's clearly wrong in general.


I wouldn’t have thought of this without your comment but i don’t know if it’s related. Media, including text, is sort of a cognitive baseband.


I agree that the author has mislabeled "text".

What are characters if not pictures? What is text if not a sequence of pictures? What's a picture if not a depiction of a concept?

In a bag-of-words type of model, I see no difference between a word and a painting, no difference between the body of work of van Gogh and the body of work of Kafka. They both use multiple sets of concepts to compose new sets of concepts.

Pictures and text are both programming languages of concepts.


The medium of your reply contradicts the content.


That's a beautiful sentence, especially in this here context, but it flew over my head. Would you care to expand?


He alludes to the fact that the text you wrote as a comment is transmitted as "information" in the form of bytes.

But to formulate a counterpoint to this argument and support your original comment: The bytes we send via cables to transmit text are at a basic level just that: bytes. Ones and zeros. What makes these bytes understandable information is the fact that we have all agreed that a certain sequence of bits is assigned to a certain picture: A text symbol. This agreement has been a major source of pains and problems throughout the history of computing, think all these different text encodings and the problems that arise when you try to open a file with a different encoding than the one which it was saved with: You see a gibberish of symbols that do not make sense.

Which is why you have a point here: The pc does not care what information it transmits, it is only text because we have "taught" the pc to display a specific symbol (picture) when a certain sequence of bits is encountered.


If you expand the definitions of text and pictures so they encompass all the same objects, of course you find that their expressiveness is equivalent. At that point, however, your logic is no longer connected to the team world. But since you asked, the special ingredient in text is syntax.


Agree. With code in VR it is possible to combine 3D and text for a greater result. The 3D space can be used for architectural abstraction that can be spatially memorized, and the text can be used for detail.

Self-promoting here: my project primitive.io makes it possible to collaboratively review code in VR. We use our tool daily to explain code faster and with greater information retention.


Why are people so afraid of moving forward?

Imagine saying “always bet on punch cards”. Look at any whiteboard in any company or school. It’s filled with text and diagrams. Then watch how someone interacts with a PowerPoint or whiteboard. There is animation and dimensional extension. The way people think and organize and communicate thoughts is multifaceted and multidimensional. It only makes sense that we should be able to work and program in the same way.


I don't think people are afraid of moving forward. Plenty of alternatives have been proposed, and text remains as effective as ever.

The problem with punch cards is that they weren't effective: they were what technology permitted back then. But technology permits many other things besides text nowadays, and text still remains an amazing and effective piece of "technology", in the broadest sense of the term.

Are we communicating, right now, via animation or fancy PowerPoint slides?


I don’t think hybrid or non-textual methods have really been given their due or worked on so much, simply because the only-text sentiment is so pervasive.

I’m not proposing punch cards as effective. It was just a simple comparative example, which I thought was clear and that the rest of my post clarified what I actually meant.

> Are we communicating, right now, via animation or fancy PowerPoint slides?

It’s ironic you bring this up, because I don’t find this method particularly effective.


> I don’t think hybrid or non-textual methods have really been given their due or worked on so much,

The problem is that they really have. Visual thinkers in academia have been aggressively pursuing this in CS PhDs and research since the 80s or even 70s.

The results haven't been great. Visual programming works well for some limited cases. Rational Rose, which produces code from UML diagrams, came out in 1994 and has been in continuous development since, has not taken over the world.

The big problem is high information density. With text you can go through it in a linear fashion, even if it's very long. Diagrams tend to hit a breaking point where they're too complex to take in. There's no clear place to start. Then suddenly they are worthless.


It's academia and some important professions (lawyers, accountants) that are primarily audio-textual in their perception and those often occupy the most important decision-making positions, forcing their mode of operation on everybody else, even if 80% of people have a different dominant perception mode.


The first comment is interesting:

"Send me a link to a news story that turns out to be a video, or an audio file, and I’ll close it unconsumed: I haven’t got that kind of time. Send me a transcript: I’ll finish reading in half the time it would take me to passively sit there while it played, and I’ll more clearly remember it."

The author is ok with sitting and staring at the article because it's faster. What he misses is that it forces you to actually be in a position / context where you can read from the screen.

Personally I have been giving a lot of though about balancing how I consume content. Text vs audio. Walking / exercising or whatever instead of reading.

I don't believe we should optimise on "time to consume" but rather "healthiness of consume".


>Personally I have been giving a lot of though about balancing how I consume content.

Depending on the content I think text can still be a pretty good bet. Using a screen reader or assistive features on devices can allow you to have an article written in text read aloud to you. I've seen lots of setups too where people will save articles and have them processed to create audio files that they listen to on their runs or whatever.


I agree. Text-to-speech falls into my category of audio, as it frees your eyes.

Actually I've wrote about couple days ago: https://audiobased.app/articles/screen-time-and-headphones-t...

btw. I would be happy to learn what setups have you seen


The balanced information diet is interesting.

It's frustrating that although so much more content is publicly accessible for free today, I find on the whole that information density has decreased, probably due to advertising as monetization.

I prefer text because I can skim to the important parts and jump through links. Audio and Video need better methods for this.

If we could summarize information better, in text, audio, video, etc. I think the internet would be more useful and people would be able to communicate more effectively. We need publicly accessible channels of higher information density.


The gist of what he's saying is pretty sound, but he should have gone deeper. For a deeper argument on the importance of "text", the written word, exposition - I recommend reading Postman's Amusing Ourselves to Death.

People here are suggesting that images/video are superior, but there's the cliche, "a picture is worth a thousand words". Well, which thousand words? Do they convey the same thousand words to you as they do to me?

Words leave nothing to the imagination. This is one of the most important traits of good exposition. Its arguments/messages are out there, free of the primate dominance gestures, the biases, the emotional stimulants, waiting to be vetted for their logical soundness.


> Words leave nothing to the imagination.

Words leave plenty to the imagination too. Interpreting what authors might've meant in books is all we ever did in English classes in school. Interpreting what the founding fathers might've meant in the Constitution is also a fierce public debate.


This is the problem with these constant black and white debates on hacker news. They leave no room for subtly and you end up with people saying dumb shit like "Words leave nothing to the imagination."


I prefer text simply because I can read the same text faster than anyone could dictate it to me. Plus I can CTRL+F through text.


Text is free of dominance gestures and emotions ? Have you ever paid attention to the words a politician might say? or the lyrics of a rap song? there are words in text that convey more emotions and gestures than if the speaker was face-to-face spitting the words personally on you. Even more when you know the original speaker beforehand and your brain is animating the syntax with their face and voice.

It's just a tradeoff ladder. Pictures represent pure unmitigated creativity, photographs represent the world, text is pictures + grammar, mathematics is when you compress the grammar to a few dead-simple rules that yiu could teach a bunch of silicate rocks to understand. At each step in the ladder, you lose freedom and expressiveness broadness but you gain freedom and expressiveness in a much more Domain-Specific way. And off course there are imperfections and leakages everywhere.


You can teach someone to paint or take photos, with text. You can't teach someone to write with paintings, or with text-less video.


> Words leave nothing to the imagination.

If only. Differing interpretations of words are one of the dominant sources of political discord in the US. The things words leave to the imagination are killing people.


> Text is the most efficient communication technology. By orders of magnitude. This blog post is likely to take perhaps 5000 bytes of storage, and could compress down to maybe 2000; by comparison the following 20-pixel-square image of the silhouette of a tweeting bird takes 4000 bytes: <twitter logo>

Author is really excited about really old information technology. He makes some good points.


It's solid technology to be excited about.

For example. Graphical programming languages are nowhere. Closest I had was either VHDL or backed by copious amounts of XML like Xtend. And both of those have a textual Component as well.


> Graphical programming languages are nowhere.

They are super popular in game development. The most popular one is Unreal Engine's "Blueprint". Unity does not include visual scripting, but there are popular plugins on Asset Store, the biggest one is PlayMaker. Unity Technologies themselves plan to release their own, in-engine visual programming tool this year.

Apart from that, visual programming is commonly used for shader programming, video compositing and AI/animations scripting [1].

I am a programmer and like many of my peers I cannot stand those systems, but designers and artists love them. They are widely used everywhere from small indie games up to AAA releases, not only for simple logic but for modelling complex behaviours and flow controls too.

https://blueprintsfromhell.tumblr.com

[1] Example of AI graph form The Division (5:34 - 6:10): https://youtu.be/fZOZ2daE-lA?t=334


> I am a programmer and like many of my peers I cannot stand those systems

I am also a programmer but I love my graphical shader editor in Octane [0]! It's the perfect interface for defining a graphical data flow. For example, I can implement really complex spatial conditionals by creating a merge node controlled by a greyscale bitmap. E.g. to create a gold thread on a silk material, I can merge existing gold and silk shaders through a bitmap of a lace pattern, itself edited in Photoshop. Just by connecting some nodes.

[0] https://home.otoy.com/render/octane-render/


I agree, but it makes me sad that we've come to see XML as primarily tech for serialization of component models, service payloads, and configs; that is, all the things that XML should not be used for. When actually SGML from which it's derived is all about text editing and using minimal markup to create implicit textual hierarchy.


The funny thing about XML is that it was explicitly conceived to solve problems of representing, translating and transmitting component models and configuration.

The sad thing is that it was abused, horribly, in the early naughts when people tried to take it too far. On the one hand it was rushed out without the proper tooling and you had humans typing raw xml without the right amount of support from intelligent autocomplete coming from schema-aware editors. On the other, you had systems that tried to force XML into roles that are better suited for a programming language (I'm looking at you, ANT). And worst of all, it became the centerpiece of a bunch of tech that people really hated-- remember all that crazy WS-* shit, and "SOAP" (which could have stood for "Complex Object Access Protocol". All this wasn't the "fault" of XML which, IMHO, has rock-solid core concepts. XML got an undeserved bad rep from association to the failure of the things it was used with.

XML still survives, though. The tooling is good. If you type XAML or HTML in the right editors, it's serves the purpose competently and without drama. People aren't going crazy finding new XML based applications anymore. It's just doing it's job. I prefer pure XML for config files even now-- of course I never expect users to see it or type it. XML, at the end of the day, is intended to be manipulated by tools.

I wish I could say the same for yaml and to some extent json. I feel like these things are about to be abused like XML was.


> XML still survives, though. The tooling is good. If you type XAML or HTML in the right editors, it's serves the purpose competently and without drama.

HTML isn't XML, though XML was inspired by HTML and HTML has an XML serialization.

IIRC, some or all of the versions of XAML aren't, strictly, XML either, merely XML-derivedl0


> XML was inspired by HTML

No. HTML used to be "an application" of SGML. Meaning HTML brings a profile for SGML features (formerly, an SGML declaration determining things such as allowing tag inference and other minimization) plus a DTD grammar. Though HTML has also quirks for script/style elements, and URLs.

OTOH, XML, like SGML, is a markup meta-language and a proper subset of SGML (also with a fixed SGML declaration disallowing tag inference and almost all other minimization features). XML was introduced by W3C (the SGML "extended review board") as the markup meta-language for new vocabularies on the web going forward around 1997 to eventually replace HTML. While that hasn't happened, SVG and MathML have been specified using XML.

Details on (my pages) below.

[1]: http://sgmljs.net/blog/blog1701.html (the "TALK" slides)

[2]: https://www.youtube.com/watch?v=jy-b4jeJSas


> > XML was inspired by HTML

> No. HTML used to be "an application" of SGML.

The “No” is wrong. It's true that HTML used to be an SGML application, but that doesn't contradict that XML was inspired by HTML.

Specifically:

(1) First, HTML was an SGML application.

(2) Then the Web took over, making HTML a very popular SGML application.

(3) Then, inspired by the design of HTML, but seeking broader application without the complexity of unconstrained SGML, XML was created.


It's close enough. There are only a limited number differences that exist for good reasons.


Because graphical programming are just textual programming languages which are just overly complex (xml) and coming with attached visualization engine for its components. They are used only for limited domains such as composing some data flows or video node editors, etc. because visualization of anything more sophisticated would make one's head explode. You could just as well visualize components of AST of Python for very little benefit.

The only upside of graphical languages is that because of their limited capabilities, they focus on only one abstraction layer and that allows you to visualize that layer quite cleanly.

You can't program anything more sophisticated or step outside of the only abstracrion layer you are allowed to program in. Eg. you can't write your custom for loops by dragging and dropping boxes and you also cannot orchestrate dags or box diagrams using dragging other boxes around.


> Because graphical programming are just textual programming languages which are just overly complex (xml)

i don't think that's fair. i mean, on a computer, you have to serialize everything into sequences of bytes. you could also encode images into XML:

  <pixel r="255" g="80" b="110" />
  <pixel r="11" g="0" b="203" />
  ...
but i wouldn't say that makes them "textual".


Text based programming is inherently 1d, so isn't well suited for the highly parallelised computing we have today. 2d programming would make much more sense.

The issues with graphical languages arethings like being much harder diff/merge and version control generally, being dependent on IDEs, and severely limited by mouse usage. I don't see any of those issues as being impossible to overcome, but instead graphical programming is held back by always being implemented as a DSL with no attempt to meet the requirements of a general purpose language.


This, exactly.

I think there are three reasons why non-textual programming is constricted to a few domains, and is not widespread as general-purpose:

- As you point out, all the tooling around programming like editors, exchange formats, version control, debuggers etc. are only available for text, rendering everything else generally unusable.

- There are many failed attempts, with many still being textual in the end, only with over-the-top syntax highlighting, such as Scratch. These attempts have paved the way of non-textual programming with gravestones, making the field look like a graveyard, so nobody wants to get into it anymore. Scratch might still have its place in education as an gentle introduction to the concepts of programming, but no software developer is seriously going to use it for any big project.

- People are used to text and they will reject other approaches instinctively. You can read many statements in this discussion here like: "Graphical programming would make your head explode", which I think is similar to how people where afraid of using trains in the early days: "Driving to fast will kill you" and "Nobody can survive such speeds". It might take a whole new generation of programmers to accept anything non-textual as intuitive (which is why it could become popular with artists and designers I think).


> Graphical programming languages are nowhere

Not nowhere. Labview is/was pretty successful. But not generally successful, no.


Graphical programming languages are fairly popular for “scripting” and signal processing in various domains like visual effects, sound design, game dev, etc.

Sure you wouldn’t want to implement the underlying systems in them, but I do think they have their place and are often undervalued by developers.


If you think graphical programming languages are nowhere, that just means you do not have a very broad exposure to programming.

As was already pointed out, graphical languages are massive in game design. They are also used in audio processing, in scientific and engineering software through labView, and in various other places.

They are very popular in fields where we want to open up programming to more people than those who call themselves "programmers".


Most programming languages would be better characterized as tree structures than text, being only obscured by unclear (but no less rigid) syntax.


So Lisp? You still type it in as text.


Nearly every language will be processed as an AST, not as text.

Text is just a serialisation format for the AST. For some reason we still insist on editing the serialisation format instead of the actual data representation.


The Latin script originated in the 7th century BC. Is that old? Maybe. But not so old compared to how long our species has existed.


Writing itself is a little older, but not much. But writing is one of the first information technologies (that we have records of ;).


It really depends on what you define as text. Does it involve special notations such as math? Can it involve diagrams?

If all you mean is prose, as writing sentences in whatever native language you speak, then I must respectfully disagree with the OP.

Yes, text is marvelous. I'm an author of multiple books and have a passion for the written word, but there are things for which there are better ways to convey information.

A simple example is electronics. A circuit diagram is not text. It is a graphical representation using a standard notation. It is much easier to understand than spending paragraphs describing which component should connect to which other component.

Unless you decide that "text" should include such special representations. Then the question becomes: where do you draw the line between what you consider text and what is no longer text. In an isometric exploded view of some mechanical device text? Are architecture plans text? Because all those representations are better than the written word to convey their meaning...


I would draw the line like this. Text is anything you can represent using Unicode.

Source code is text. Emoji are text. Some simple math expressions are text, but more complicated ones involving specialist layout rules are not. Architecture plans, no way.


Unicode is not quite the right place for the line, I think. Cypro-Minoan is text (but not currently part of unicode). ASCII art is not text.

Linearity is part of it. Source code and emoji and simple math expressions can be seen as a linear stream of "characters" (defined loosely), even on a fundamental level. An architectural plan could be transliterated into a linear form, but that would be a transformation.

Of course then you need to explain why audio doesn't count as text. Discreteness? Ease of faithful reproduction?


Unicode answer is not so great because you can express compiled program in Unicode which obvious is non-text. Math expressions are always text because they always fit TeX or LaTeX.

Architecture is a beautiful example which provokes thinking. From first view it is a non-text, but an architect Antoni Gaudí has developed an attitude with hanging models which made his architecture plans to share some abilities of text even not being written. You may tell me I am crazy but his hanging models is a WYSIWYG model of something from real world so I consider it a text despite it could not be simply copied. IDK maybe a modern architect software shares that attitude, for make these models be copied easy. So my opinion is that architecture plan may be text at least from a structure point of view.


That's a reasonable but exactly backwards way of looking at it. Indeed, Unicode captures a wide range of what we consider text, but text is not Unicode. There are languages with symbols that Unicode can't represent but they are still text -- to claim otherwise is to erase them.

Something closer to a definition of text is parts of their criteria for encoding symbols [1] but even that is imperfect.

[1] https://www.unicode.org/pending/symbol-guidelines.html


Even if you include things like circuit diagrams as text I don't agree that you should always bet on text. Personally I think you should always pick text as a default, unless you have good reasons to chose something else and use text to elaborate on it.

Couple of examples:

- For a musician to convey a musical idea, the arguably most straight forward and effective way to do so is just to play/sing/record it. Any other means like musical notation, textual description, midi etc. require a lot of additional knowledge and work on both ends of the communication and come with their own drawbacks.

- The Wikipedia text about human rights in the article is a great example when written text is better than an image. But say I want to describe what the grand canyon looks like. Personally, I don't have the writing skills to really do it justice. On the other hand my phone has a camera so I can snap a couple of pictures that will get the other person a much better idea of how it is. In this case text is useful for additional information not visible in the image, like why the color of the rocks is how they are.

- If I'm living together with someone and want to discuss how to pain the walls, a pantone color swatch is going to be much more useful than textual description of the colors. Text is useful here after the decision to pass on the color code to the contractor.

- I reckon for most people it easier to follow instructions on how to cut up a chicken when they are in form of a video or image slide show over just written words.

I don't think the problem is whether text is better or not, but that often the choice of medium is not based on what makes most sense for the use case, but other considerations like what will yield better conversion numbers.


Let me try. Text is anything human-readable which supports easy editing in WYSIWYG way. For example: XML, maybe JSON, non-obfuscated source code, LaTeX notation for math, vector images, chess notation, etc. Diagrams are not text but they are easy to be describen in plain English, much easier than memes or photos.

Electric circuit is a beautiful example between text and non-text. You know EC is drawing but any EC is made of hierarchy of two-port networks, so it is like system of equations written in unusual for mathematics way. You can add or erase any element or change any nominal and recalculate a new circuit instantly. If you have a friend working in an electronics, you can give him any valid EC and ask him what he gonna do with it. His answer will be reading, not just looking at.


Your definition appeals to tools and technological limitations, smalltalk had this notion that anything is editable, including the IDE itself and the (running!) programs. Your definition + a smalltalk IDE would imply that a running program is text. Furthermore, the abilities of the message reciever factor into this as well : web pages are editable, but somewhat non-trivially. So the combination of tooling and abilities could be used to argue that everything is text: for a graphics artist with even a paint instance running, a picture is the epitome of human readability and editabiliy, while chess sounds like a gibberish unicode file opened with a wrong encoding.

That doesn't mean your definition is useless, its an interesting take to draw the lines between media along editabiliy, but I wouldn't call that a definition of text. It would exclude and include all the wrong things.

One is really starting to believe Wittgenstein that all disagreements boil down to definitions.


I would like to add that text, as a thousand-years-old-technology, has always included illustration too. Even in recent/digital text, you have this bird that doesn't take 4k bytes: [edit: unfortunately hackernews doesn't display the unicode character U+1F426]

Text is very powerful, but it doesn't need to always work by itself (arguably works best together with other media)


> [edit: unfortunately hackernews doesn't display the unicode character U+1F426]

That got a good chuckle out of me. It actually shows how there's not just "text".


It is just text (using Unicode nomenclature). The makers of the HN software deliberately broke the functionality in an obnoxious fashion.


They weren't obnoxious, just subscribed to a more restricted interpretation of "text" than yourself, and to which many people seem to agree.

It's clearly a continuum from 3D objects, to pictures, to icons and ideographs, to highly abstract sings representing words and ultimately sounds. That fact that some designer of character sets decided to put the limit somewhere and include a bird icon and not an Obama icon does not invalidate other interpretations.

By the way, there is a thing called Emojicon where a character for 'Albert Einstein' was proposed as valid 'text'.


While I totally agree with the OP, the point to remember is that text is a really terse storage as long as you have the Unicode mapping already stored to transform the binary compressed form to actual pictures of text (vs abstract forms or numeric codepoints when simply stored as "textual" bytes).

This does not diminish the value and expressiveness of text, but it needs to be said that in 5000 years time we'll need both the Unicode specification and those 2-4000 bytes to decipher the author's post.

It's just a cost of digital media.

The bigger issue is how do we ensure digital media perseveres for so long.


I dont think you would. If english is still understood, decoding ascii/utf-8 is trivial. Just think of it as a ceaser shift cipher with an offset of 64. Very trivial to decode with frequency analysis if you know its english. Either you understand that 65 represents the abstract concept A, or you have no idea what the letter A is. Either way, a picture of the letter A is not going to be helpful.

Having english still understood after 5000 years seems much harder. But hey, latin is pretty old and we still understand that.


> in 5000 years time we'll need both the Unicode specification and those 2-4000 bytes to decipher the author's post.

Nitpick: In all likelihood, an English dictionary would be enough. Even if the Unicode spec is lost, the text can probably be deciphered by using frequency analysis plus the dictionary to associate codepoints with characters.


Sure, but the English dictionary is a bit longer than the Unicode spec :)

My point was that there is a lot of implied data that undermines this "compression" of information into "simply bytes" today: I am pointing out the implicit in OP's assumption.

Maybe I did not choose the right addendum, but a pretty comprehensive addendum is needed (in all likelihood, an "ancient English dictionary" is needed anyways).



Apples and oranges. The post states that the used Twitter logo PNG is 4000 bytes, while it's only 723. Try to express that logo in text, English language and in all its details with 723 characters, so you are able to reproduce a pixel perfect representation. Not even possible with 4000 characters anyway. Don't bother. Text will be ambiguous, unless you describe every single detail. If the target is to communicate "Twitter logo" then sure, just write that. But if you compare 723 bytes of "data" with just a few specific information pieces contained within that data ... sure, text might be better.

> text is the most powerful, useful, effective communication technology ever, period.

Nothing of this is actually true, period.

> I do not post to this blog with the intention of entertaining Hacker News Debate Club and I frequently disable comments or friend-lock posts in order to avoid this sort of nonsense. I'm not interested in further discussion.

Ah yes, the true seal of quality. (sarcasm)


Plain text is simple, elegant and has open-standards - it traces its origins back to telegraph codes that we have been using for hundreds of years - and still forms the basis of most of the internet now. So I am inclined to agree with the statement "text is the most powerful, useful, effective communication technology ever, period".


By that same argument/logic you might as well argue cave paintings are superior over text. It traces back probably over 50000 years and is still universal today and doesn't even require understanding of a specific language - and it still forms the basis of most of society now. (art, painting, photographs, graffiti, ...)


Your reply is in text. Would you care to link a set of paintings to express your argument instead?


That's the point: take an arbitrary criterion, end up with an arbitrary answer.


It’s also in ascii, is that superior to all other character sets, or an accident of history?


> It’s also in ascii

No it isn't:

> 10:48:55.380 document.location.href

> 10:48:55.384 "https://news.ycombinator.com/item?id=26166797"

> 10:48:57.723 document.characterSet

> 10:48:57.726 "UTF-8"


If it was really Unicode I could use an eye roll emoji to reply with a succinct image, instead of all these words to describe my reaction.


That's HN filtering your input, nothing to do with the actual encoding.

Plenty of non-ASCII symbols show up perfectly fine. For example, '£'.


Please elaborate how text is most powerful (in which aspect anyway? define powerful) and most effective (surely it isn't or text wouldn't be the prime example for stuff that's good to compress).


He would have to use words for that... Could you help him showing some image of "powerful" and "effective"?


強大的 and 有效的, respectively? (I can see some powerful person in the middle of the first image.)


Right after you provide an image that's as effective as the text of the First Amendment.


Think of all the non-text technologies involved right now for me being able to read your strawman.


Can you explain the strawman you believe the parent is presenting? What is the original argument, and what is the misrepresentation?


I am not the source of the original statement, but my own interpretation of "powerful" aligns with simplicity and flexibility. Plain text has only slightly more structure than a stream of bytes, meaning it retains a lot of simplicity and flexibility. Yes plain text can be inefficient and is overused (a proprietary unpublished wire-format does not need to use JSON). However, 50 years from now, the only data I feel comfortable knowing I'll be able to read is plain text (and possibly also JPEG and a few other well-specified and simple binary formats). Many binary formats are effectively defined by large complex and transient code bases that target particular tool chains and APIs. A new binary format needs to justify itself, less so plain text.


The Twitter logo in SVG format is only 687 bytes and can be as large as you like: https://upload.wikimedia.org/wikipedia/en/9/9f/Twitter_bird_...


My SVG is extremely rusty, but it doesn't look as if that

     xmlns:v="https://vecta.io/nano"
serves any purpose besides advertising.


It's probably not the only removable part, the icon can be 414 bytes: https://github.com/edent/SuperTinyIcons#how-small


SVG is XML, which I'd say counts as text.


I don't think it counts for the purposes of this blog post.

You can't translate the Wikipedia quote in the article into an SVG except by just rendering the bare text.

Checking programmatically that text is about "bird" or "Twitter" is very easy. Checking that an SVG is about "bird" or "Twitter" is very hard.

It's not even efficient. It conveys the same message as the word "Twitter" in a hundred times the size. And it wouldn't be less efficient if it were a binary format.

It's maybe more suitable for carving into granite than a binary format, but you'd be better off carving the logo itself.


> Checking programmatically that text is about "bird" or "Twitter" is very easy. Checking that an SVG is about "bird" or "Twitter" is very hard.

That depends on what you want:

If you know what twitter is, then yes, just saying "twitter" is more efficient than a whole SVG file describing the twitter logo.

But if you don't know what twitter is, nor what its logo looks like, SVG is a very efficient way to describe the logo exactly. If you can have it rendered to your screen that's optimal, but if not, you can take it as instructions to redraw the logo yourself and still get an accurate impression of what the twitter logo looks like.

I that sense, "twitter" and the SVG of the twitter logo are both text, but they don't have the same meaning.

As for the granite, it is indeed one situation where just drawing the plain SVG logo is more effective, but SVG would still be viable. Over telephone, on the other hand, you just can't transmit an image. You can only attempt to describe it, and the more detailed you get with that, the closer you come to just using a code very similar to SVG.


How far do you take that line of reasoning? Base64 is text?


00101010010100110101001... is text too, but I agree SVG is just machine description, human sees the rendered result in that context.


it's not consumed as text


>Nothing of this is actually true, period.

Could you kindly verify that by repeating your post using something other than what amounts to a text file?

If not maybe that's an even truer seal of irony.


I read from parent's point that what you are trying to communicate matters. I take from that, that text works sometimes, and sometimes something else is a better form of communication.


Yes, it does help when you can read between the lines . . .


> Could you kindly verify that by repeating your post using something other than what amounts to a text file?

Could you kindly present me your comment without using all these inferior non-text technologies involved on my side to read it off my screen? That's the irony, if anything.


That sounds like a you problem. This site is one of a dwindling number that work flawlessly with a text-only browser, so if you view it in Links or similar (as I am right now) you can have a pure text experience without any of those pesky non-text technologies getting in the way!


> the "text-only browser"-argument

That's a bit short-minded, don't you think? Or is this some kind of troll meme at this point? Anyway, I don't expect you to have some off-spec working TLS/TCP/IP/ARP/ETH implementation purely based on text, yet alone are connected to the internet via some text-only telegraphy station utilizing those things. Pretty sure your output device (read as: screen) has some sort of analog/digital signaling as well that's not ASCII encoded. There are sure some pesky non-text technologies involved ...


~I'm not sure where you're going with this or how seriously you're taking it~, but if I have to I will dig out a couple of DECwriters and we can continue this discussion as a pure ASCII text conversation. (Just as soon as I find a null modem cable long enough to reach wherever you are.)

Edit: upon some rereading, I get that there was maybe more sarcasm in gp comment than I perceived at the time.


Wasn't that why they called it hyper-text?


In the USA the USPS usually does not fail to deliver a text document, no superior nor inferior technology required.

And they still reach places having no other form of comunication.

If it was not a post card, you would need to open the envelope please.

If you would like to read that kind of text off your screen, you might just need to scan it yourself or find someone who can.


What does the display (or TTS) technology used to read it matter in this discussion?

Obviously you need ways to display text in the physical world, be it a screen or a piece of handwritten paper. It's still text.


> What does the display (or TTS) technology used to read it matter in this discussion?

> Obviously you need ways to display text in the physical world, be it a screen or a piece of handwritten paper. It's still text.

And what about transfer? People keep claiming text is superior for transferring information, yet the majority my network stack or anything involved with the transport isn't text based.

The premise of the whole "text is the best technology" argument is based on cherry picking and strawman arguments anyway.


>The premise of the whole "text is the best technology" argument is based on cherry picking

I agree with that so naturally cherry-pick the thing expected to be the most universal and the last one standing.

Aren't everyone else's network stack usually sending packets of text when it's a plain text file? Even if it is hypertext.


> The post states that the used Twitter logo PNG is 4000 bytes, while it's only 723.

The article was written in 2014. Worth remembering when nitpicking the text of an article.


Do PNG files reduce in size over time?


Not by themselves, but maybe they didn't use an optimiser before. Or used a far better optimiser now. Or tweaked the logo, and now it's more amenable.to optimisation.


The author says in the post that it's a 20x20 image. That's only 400 pixels, and 4 bytes per pixel (RGBA) would only make it 1600 bytes with no compression at all.


Then that's quite some impressive, uh, decompression.


I don't know, but if they do, they seem to have a half-life of about 2.431184 years. That's interesting news.


Compare a book (a few megs) with a picture of the same size. What do you think can communicate more info? Do you have a single counter-example?


The beauty of both mediums is in the compression.

Text is that it's highly compressible while also being highly compressed (and lossy) info already. To make text we have to filter our own ideas, and then to interpret text we have to add a bunch of info back from context.

Pictures have the same tricks up their sleeve. Bitmaps are great but inefficient. If we put it through a lossy compression, it's a lot better. If we make it semi procedurally with a vector file, it's even better. How far can we take a compressed procedural file though? Well.. .kkreiger was about 97 KB, and stores a full 3d shooter. For a meg, you could store 100 such full games. https://en.wikipedia.org/wiki/.kkrieger

Text isn't best in every way. With clever design we can use all the same principles that make text efficient, and make other types of data more efficient. It is a very good default though.


There are single pictures you couldn't describe with terabytes of text(to a human, that is). Imagine for example trying to describe the milky way and northern lights. Yeah you could convey their looks in words but nobody is going to end up with the correct visual no matter how hard you try. Now add in other things, like emotion on people's faces. Some things just aren't possible to describe in full.


you somehow managed to describe the pictures without showing them


well, svg is text. The jpg/png icon will pixelate when zoomed, but svg will retain its quality no matter how much you zoom in... :)


> svg is text.

If SVG counts as text, then text is just binary. The encoding is not important, the final result (an image you see on a screen, in the case of SVG) is.

Besides, try reading out the actual SVG source text to someone and see if they can tell you what the image you are describing is. I dunno about you, but I find thinking in terms of paths and strokes to be rather meaningless.

Or another example, do you consider Wavefront .obj files text? They're ascii files, but I don't consider a series of "v 0.123 0.234 0.345 1.0" as something I would ever read in text form and have even the slightest idea of what the final 3D model actually is. I have to view it visually. Its "technically" text, but its no more useful as text (for a human text reader) than if it were binary data. Encoding isn't what makes text what it is, so non-text things encoded as text aren't really text.


SVG is definitely human-readable. It's just drawing instructions, draw a line from there to there, then another one from there to there, etc. I can read it to someone over the phone with a pen and graph paper, and he can draw it exactly like it should be represented, with zero loss of quality. It's no different than someone looking at list of cash transactions. Numbers are text too.

Of course, to extract meaning from this text (what is it?) requires drawing it. But you don't have to, you can just name it. You could just say "I'm sending you the stylised image of a blue bird, with the instructions on how to draw it using the SVG text standard". Or just, more clearly even, the following string of characters: "twitter_logo.svg"


> SVG is definitely human-readable.

No. A path tag with a 500-600 element `d` attribute is not human readable. It is not human understandable or parsable until it's converted into a visual format.

Ask a blind person whether they think SVG charts are accessible on their own without any additional markup or alternate explanations or controls. Ask them if they can read SVG line charts by stepping through the path property. Figuring out how to communicate complicated graphs/charts with lots of data without relying on any visual medium is a complicated problem that we are still trying to solve.

> You could just say "I'm sending you the stylised image of a blue bird, with the instructions on how to draw it using the SVG text standard". Or just, more clearly even, the following string of characters: "twitter_logo.svg"

By that logic, a PNG is also text, because I can do the same thing. All images can have alt tags. Even Javascript Canvas can have fallback text for when it doesn't render. My Twitter png can be transmitted as "twitter_logo.png".

What do people mean when they say text is preferable if not that the textual representation is more useful than the image representation? If you didn't feel like the SVG format was adding any useful information in its rendered form, you wouldn't waste the time sending information to render it. You would just send the alt text.

> I can read it to someone over the phone with a pen and graph paper, and he can draw it exactly like it should be represented, with zero loss of quality

Right, and when that person draws it, they will have converted it from a textual format into a visual one. I can send you the individual pixels of a small raster image over the wire in a textual format and you can sit down on a piece of grid paper and ink in squares until you come up with the final image. At which point you will no longer be looking at a piece of text.

To argue that the fact that we can encode a format over a UTF-8 stream or physically replicate your computer's drawing algorithm makes a thing text seems to me to be a misreading of the original article's point. The original article is not complimentary of Twitter's SVG, it explicitly brings it up as an example of a non-text medium.

If you expand the definition of text to include basically any storeable information that can be theoretically encoded into UTF-8, then sure, I would agree that text in that system is tremendously powerful. But that's not a useful category to talk about, it's so broad that it becomes meaningless.


> It's just drawing instructions, draw a line from there to there

That doesn't make it human-readable. I challenge you to tell me what (even the basic shape) this SVG is without loading it in a viewer (taken from a real SVG, not something handcrafted to prove you wrong, but the first thing I found online while searching for public domain SVG files):

https://gist.github.com/danielytics/90446a2314c1ad87017233eb...

Good luck figuring out what that is, even with graph paper. And that's a single layer, single (even if a bit complex) shape, single colour SVG. They only get more incomprehensible from here.


Is math that you not understand, really text?

Is human comprehensibility a pre-requisive to be called text? What if there was a written language no one understood, would that be text? What if someone broke the code for it? Would it suddenly become text?


If SVG is text, then so is StarWars.mpg.b64.


Some SVG is very readable and easy to modify. I'm no designer but I've opened up SVG files and made modifications to get what I want (remove elements, change colors, etc).

Not all SVG authoring tools make SVG's easy to read/modify. Tools in the 90's used to the same thing with HTML. Are you going to argue HTML isn't text?


> Are you going to argue HTML isn't text?

Go to google.com and view source. That's text, right?


Compression is part of human communication also. Sometimes it's used to keep the communication short, sometime to hide the meaning behind what is said.

lol, brb, PFA, W.H.O., IMHO, IANAL, GPA, GTA.. word soup for a different audience..


So because google.com is hard to read, HTML is not text?


> svg is text

Actually I thought about talking about SVG, which would be the natural result of taking a text representation of shapes ... and then use abbreviations until you are at some expert level language (SVG for example), which would then break the argument again. You might as well use some binary representation then, because most people don't speak fluent SVG.


Most people don't speak fluid tok pisin either; does that make it less of a language?

Also, you'd be surprised by the amount of people that write SVG by hand, which I doubt you'll see with PNG anytime soon.


Different SVG engines may show different result, that may not acceptable for logo.


Can you elaborate?

Outside of bugs, I was under the impression that SVG was actually pretty well specified and that (except perhaps for pathological examples created intentionally) the rendering of an SVG was quite consistent.

What am I not aware of?


That's a biased comparison. Obviously an image is better at communicating an image. By setting the goalposts at "pixel perfect recreation" you've already selected a winner.


re:Try to express that logo in text

"Twitter"

carries the same weight


Tell me in 723 characters or less what this "Twitter" looks like so I can make an exact copy of it. Text is ambiguous.


Images are the best way to describe images. Many ideas are better expressed in text than anything else we have.


HN doesn't render ASCII or ANSI art correctly however both methods can serve as representation of Twitter in text and graphic less then <700 characters.

     ,~
    ('v)__
   (/ (``/
    \__>' hjw
     ^^
   Twitter
Bird kindly borrowed from: https://www.asciiart.eu/animals/birds-land


This isn't text, you're using ASCII symbols as pixels in a grid to represent a raster image. This is literally just a raster image on an unconventional canvas.

If you treat a uniwidth character grid of characters as pixels, you have not converted an image into text, you have encoded text into an image. It's no different than using extended characters and font colors in a fixed width terminal window to make terminal art.

The fact that you use UTF-8 to encode something does not mean it will magically work in a screen reader, or that it will retain any of the advantages of pure text. I can encode any image or video file in a UTF-8 format and send it to you. Down this path lies calling everything text.

But if your "text" stops being readable if I disable monospace, or it stops working if I change the font, or it stops working if I change line wrapping, or it stops being communicable as soon as I try to read it out loud -- then it's not pure text. Using a UTF-8 character as a pixel does not change the fact that you're drawing an image out of pixels.


Interesting hypothesis. Thanks.


That does not even remotely represent the Twitter logo, though. Shape and color are completely lost information, which are essentially the key properties that make the Twitter logo the Twitter logo.


Valid point.


I see text as read optimized, but write expensive and not well suited or efficient for lots of scenarios:

- wood working, absolutely horrible to work from text, which I've done.

- group ideation, as much as I love IM'ing, audio + diagrams personally feels like a more effective way of communication.

- troubleshooting, too many times I've had to text my parents how to troubleshoot their router only to end up calling them and slowly talking them through it.

These are super specific, but I'd still wager audio and video trump text in a significant number of general scenarios that require communication of some kind.

Anything that needs read optimization i.e. this is information that needs to be communicated over and over, text is better. But often that's not a requirement.


too many times I've had to text my parents how to troubleshoot their router only to end up calling them and slowly talking them through it.

This is not an inherent problem of text, but a problem of synchronous vs asynchronous communication. Round-trip latency of a phone call is much shorter than IM. But I'd say that speech and text still use the same medium (language), so it's not really an argument either way.


The article is historically wrong. The oldest known cave murals etc. are 44000 years old [1], while cuneiform, heiroglyphics, Sumerian, Mesopotamian etc. are only about 3000-4000 years old [2]. A strange beautiful example of complex information presented visually which was in use until recently, are the Polynesian navigation charts [3]. Our primary method of communcation is probably visual and/or auditory.

The prominence and seeming economy of text in digital communications is probably because data is presented and transmitted as a one dimensional sequence of bits.

[1] https://en.wikipedia.org/wiki/Cave_painting

[2] https://en.wikipedia.org/wiki/History_of_writing#Cuneiform_s...

[3] https://www.khanacademy.org/humanities/ap-art-history/pacifi...


How much exact meaning can you infer from the murals?

Text is using the visual channel to transmit what language can convey, and language can convey generally almost everything we try express (likely more accurately and succinctly than drawings, gestures or other sounds).


Side note: the author is Graydon Hoare, creator of the Rust programming language.


That's really interesting, so thanks!

However, I am not sure if that is supposed to affect my opinion of his take on text vs other media? (I try not to let prejudices like "he's a really smart person" affect me when discussing a topic that is very approachable to a wider population)


It hopefully doesn't affect your opinion on his take, but it hopefully gives some context related to his background. He's got a bunch of experience in very low-level very back-end systems, involving a lot of abstract reasoning. I would guess people used to a lot of abstract reasoning would tend to be partial to plain text and/or mathematical notation.


Cool, that's a very interesting observation, so thanks for elaborating!

As originally worded, it only sounded like "hey, this dude is famous" (for HN value of famous :-), so I would appreciate the commentary you just shared right off the bat! Thanks again for the clarification!


The author may have overstated his case, but I hope the designers of GUI applications are listening: most of the icons on your menu bars mean nothing to me, so your application is hard to use. Substitute words.


My biggest pet-peeve is the hamburger icon instead of the world 'Menu'. You are saving 10px... for what!?


In defense of the hamburger menu, it's singular, nearly universal and well understood, I can deal with that just fine. This is very different from rows of icons in something like word where there's a huge array of functionality hidden behind obscure hieroglyphs.


> For what?!

I can think of several things: - more breath space in the UI - easier recognizability - less information overload (four character glyphs vs three plain lines) - consistency with other applications

And it's not like the hamburger menu symbol is particularly obscure.


In 1995 you needed to save 10px, they're just not as advanced any more.


Yes, there's a reason why the command line and shells are still a thing. Text supports a wider range of expression than any other medium we've developed thus far (and not for lack of trying!). Even Steve Jobs in his never-ending quest for UX minimalism couldn't eradicate the keyboard from the iPhone, despite all their UI innovations.

Text is also why XML, JSON, and Markdown are a thing.

When you can accomplish your tasks within a constrained range of expressivity, you do that (gas & break pedals, steering wheel, door handles, etc). But when you need greater expressivity, you're probably going to need text.


And somewhere in that discussion, I hear a faint echo of "org-mode" being whispered : ) Jokes aside, I often notice, how I am more productive, when relying on plain text, than relying on "enterprise" wiki systems, which make it impossible to export to any useful further processable format (looking at you, Confluence). One of my favorite features is, that it is easy to put text under version control, so that diffs have a recognizable meaning.


That's more of a dig at Confluence than it is in favor of text only systems. I've heard tons of good things about org-mode but Notion is also a pretty good wiki tool despite not being fully plain text.


Would you be willing to share what type of additional export formats you're looking for, and for what purposes?


Text has the benefit of extensibility and loose datatyping - you can encode all sorts of stuff as text, and with a bit of care, extend the "protocol" (human or machine) in such a way that receivers (current or future) can decode it (with a high degree of precision), even if you hadn't previously agreed on the protocol.

You can of course do this with multimedia or any other encoding, but text is pretty great at on-the-fly extensibility.


This is an important debate! Only one form of communication can be declared "the most powerful, useful, effective communication technology ever!!"


Case in point: text messaging. Quoting the Spolsky ('Not Just Usability', 2004) speaking about "social user interfaces."

Many humans are less inhibited when they’re typing than when they are speaking face-to-face. Teenagers are less shy. With cellphone text messages, they’re more likely to ask each other out on dates. That genre of software was so successful socially that it’s radically improving millions of people’s love lives (or at least their social calendars). Even though text messaging has a ghastly user interface, it became extremely popular with the kids. The joke of it is that there’s a much better user interface built into every cellphone for human to human communication: this clever thing called “phone calls.”

It's not just dates. It's "How ru?" & "running 3m late" and such. This has advanced to where text messaging is now a distinct written dialect, unintelligible to someone from 1993. Meanwhile, voice messages and such are more peripheral... even though they now work through the same UIs and we all have earpieces in our ears anyway. Text is powerful.

That said, text is not always the most powerful media. Photos/selfies and such have become a major 1-to-1 communication medium too. I often find that a phone conversation way more efficient than an email chain.

I also think there are categories of writing that shouldn't be. "Number articles" where an article is describing a company's financial's, for example. A lot of newspapers try to describe a table in essay form. The table would be better. That is still text though, in the sense that this article uses the term.

Choosing the most powerful medium or submedium is crucially important.


> Many humans are less inhibited when they’re typing than when they are speaking face-to-face.

Spolsky was right about this when he wrote that passage in 2004.

We should take seriously the idea that the lesser inhibition of textual communication has been a drawback on net, at least over the last 17 years.


It has been a factor, certainly. Probably both good and bad. Spolsky was a pioneer in his concepts of "social user interfaces," and that really showed when he did stackoverflow.

Lowering inhibitions is a broad statement. You can get more detailed. Inhibition is multifaceted, lots of flavours. We can be less inhibited about sending a quick "I love you," less inhibited about arguing, being mean, etc.

Over the last 17 years, all this stuff has experienced a massive multiplier effect. Both the positives and the negatives are multiplied and as culture grows around the technology everything gets more complex.


>Text is the oldest and most stable communication technology.

I disagree. Drawing pictures is the oldest communication technology. Pictures evolved into text eventually. Characters in early texts frequently are just small pictures. Understanding pictures is easier than understanding text because of smaller cognitive load.

Text helps with 2 things: condensing information and manipulating abstractions that don't have an unambiguous visual representation (such as hope or price). But it comes at the price of needing to learn the alphabet and dictionary and apply them to mentally decode what is otherwise just a cryptic drawing.

Text may be more efficient in some cases, but say it's better universally is moot. For instance, texts are very bad for representing non-linear, concurrent workflows. Pictures are way more better in this case.

>Text is the most efficient communication technology.

It heavily depends on what you're going to communicate. For a blog post, text may be better. In other cases, a picture may be worth a thousand of words.

The article may have good points, but it's full of poor statements.


If we agree that a picture is worth 1000 words, then...

Given the current state of information technology, I agree that we are most efficient at processing text. However, that can change pretty quickly. Storage mechanisms similar to DNA can make the difference between text and multimedia irrelevant. It will happen because nature already does that.


Pictures are worth a 1000 words, but that comparison holds when a human with context is interpreting the picture.

Our eyes have the ability to input so much information which is what makes pictures valuable.

If you simply need to transmit small/simple data then text is the way to go.

I get what the author is trying to communicate, but it seems a bit arrogant


Storage hasnt really been the bottleneck for a long time. A terabyte stores a lot of pictures.


I am talking about efficient use of available space. A terabyte may store a lot of pictures, but if we can achieve the data density of DNA, we may be able to process (store, transmit, transform...) far more amount of data far more quickly. Then text vs multimedia will become irrelevant.


Photography is about 200 years old. Film (moving images) is about 100 years old. The ability to easy create video is about 20 years old.

Compare all of that to text, which is at least thousands of years old. And then compare text to speech, which is hundreds of thousands of years old. Had a way to record and replay audio been developed before writing, it’s likely that you’d be listening to this comment right now, not reading it.

I’d say we’re at the extreme beginning of a highly audiovisual age. A millennium or two from now, writing “dead” words might seem as ancient to our descendants as foot messengers appear to us: useful for particular purposes but mostly irrelevant.

On a related note, the concept of logocentrism seems relevant here:

https://en.m.wikipedia.org/wiki/Logocentrism


This article annoys me whenever it reappears because it just isn’t clear on what it means by ‘text’.

Sometimes it seems to mean ‘the English language’ or ‘language’ more broadly. Other times it seems to mean ‘strings of ASCII encoded Latin letters’. Sometimes it seems to mean ‘pictures of arrangements of letter like symbols’. In general it just amounts to ‘linear streams of data’.

Sure, if you define text that broadly, it covers a lot of things that are great.

But it’s a definition that’s so broad it defies its own terms. Text, defined that way, encompasses SVG files. Or even base64 encoded PNG files if you want. So that Twitter logo can be unambiguously shared through ‘text’ too. Look - here’s a tweet-sized version: https://twitter.com/bbcmicrobot/status/1237867433064464394?s...

But there’s a weird cultural bias built in to the assertion that all those things are ‘just text’. Sure, for someone who uses a US keyboard to type Latin alphabet characters left to right, base64 encoded binary, svg, or BBC BASIC, is ‘just text’. But that’s not exactly a universal perspective.

In the limit, this amounts to ‘always bet on data transmission and storage’.

A lot of the listed benefits of text are only realizable when the text is coupled with a specific ‘interpreter’ - be that an SVG renderer, a BBC micro, or a human who speaks English.

Doing stuff with text with computers is hard! Lexers, parsers and tokenizers are probably the most common sources of security bugs in history. And if the text is natural language, we still don’t have reliable computer tools for dealing with it - understanding or generating.

So I just guess I don’t really know what the point of this piece is. Data is all there is. Linear streams of data are often a thing. Because of the history of computing, western language character sets and conventions are often used to capture them in the same format as we use for written language.


I do agree text wins when it comes to expressiveness. However, that expressiveness comes at a cost, just look how difficult it is for beginners to grasp the initial concepts of programming.

I've seen some people pick up this medium very quickly, and others struggle for months with little progress. However, almost anyone can pickup Sketch or Illustrator for creating UI prototypes very quickly.

The expressiveness of text is not always a strength. It's very hard to build programming languages without text, but I strongly believe we still program too much when building UIs. Excel demonstrates that people can quickly pickup a minimal programming language for connecting data to UI, I think an Excel-Sketch hybrid is where the future lies for building applications in particular.


This discussion right here is all text. Imagine if it was pictures instead. Imagine representing any of the comments in form of a picture. Won't be one picture representing a 1000 words. May be the another truth is that "a word is worth a 1000 pictures".


Concerns about 3D multimedia spaces replacing text makes as much sense as worrying about parchment, paper, or rectangle screens replacing text. Virtualized spaces are, much like those others, merely a transmission medium for text (and other media.) As immersive computing arrives (you may not think it is, but I assure you, it is) we will be refactoring a lot of these existing rectangle screens and other contexts where text is projected via physical processes into virtualized ones.

This is why in my 3D virtual space for work, Jel, fully collaborative text panels (synchronized via OT) are the primary kind of element you create: https://jel.app.


What I took away from this is: using combinations of a limited set of symbols to convey meaning is durable (as society advances or retreats and technology changes). This makes sense because human thinking, for the most part, seems to be symbolic (i.e. using a smaller, simpler thing to represent a larger more complex thing). Text is just one such form.

Drawings, sculptures, JPG files, videos etc. are more precise and are better at conveying a specific object or event, but not necessarily its meaning.

If I remember correctly, Neal Stephenson's Anathem deals with some of these themes.


The article is making a weird point and the comments here are talking past each other.

Author should be conveying the unique aspects of Textual media but also it’s shortcomings. Comments here should be discussing the benefits and shortcomings of all kinds of media.

Instead both are discussing which one is better - Text or something else. A lot of it is mutually exclusive. The Tianamen square tank man image is powerful and impossible to encode in text with the same effect, and Nabokov’s prose is impossible to paint a picture of or how we struggle to describe tasting notes.


Text is using a series of efficient pictures to communicate. I liken it to radix-based numbers. I could write 1,749 tallies on a piece of paper, or I could write the base-10 number.

I have to learn decimal numbers: my intended audience needs to learn decimal numbers. But the resulting efficacy is a thousandfold. I think of letters and words like digits, but for communicating broader concepts.

So when we lack the symbols or combinations to express something, we need a unique picture to do so. Pictures are worth a thousand words, but only when a thousand words won’t suffice.


There’s a real ‘ceci n'est pas une pipe’ treachery-of-images philosophical trap here, of course.

If text’s so darn great, after all, why do you need to draw a picture of it before I can understand it?


> “Pictures may be worth a thousand words, when there's a picture to match what you're trying to say.”

The author is saying that because pictures cannot easily capture arbitrary sentences, text is better. But the same applies with the roles reversed: text cannot capture arbitrary pictures! Instead of saying that one is better than the other outright, let’s move to media (for reading, composing, and programming) where the two can be intermingled appropriately, with as little friction as possible. We certainly have the technology for it...


Yes, visual diagrams and shapes are sometimes more efficient. But the main point is that text is the lowest (and cheapest) form of encoding that is still human readable/writeable. Text is always there even when other data formats are not feasible. If you can adequately represent your ideas in text, you are giving it the highest chance of being seen and distributed by other humans. You don't need special tools or artistic skills to replicate the data, anyone with a pen & paper or keyboard can do it!


Text and more broadly natural language is actually a stunningly low-bandwidth form of communication. The reason it works so well despite being low-bandwidth is that we humans share an incredible amount of shared experience and common knowledge, and so the limited amount of information in text can refer to a much wider set of knowledge and assumptions and our brains are able to make inferences and use context in a remarkable way to fill in the blanks.

To take an analogy, think of writing text as analogous to compressing data using a Huffman code. Our ideas correspond to the initial uncompressed data, natural language text corresponds to compressions of those source data, while our brains correspond to the Huffman tree that tells you how to decompress. With our brains/context we can recover the initial ideas, just like with the Huffman tree we can recover the uncompressed data. Without the Huffman tree, the compressed data are gibberish.

On the one hand this means that text is really powerful as the author says; we can store an incredible amount of information in a small amount space, and can reconstruct the source idea the text represents efficiently (if with some amount of ambiguity and error).

On the other hand text is very far from universal. Anyone who sees the Twitter logo sees the exact same thing (interpreting it as a bird, of course, requires the prior knowledge of what a bird is and looks like). However, anyone who sees a piece of text not only needs to understand the language it's written in, but also all of the ideas that it refers to. That's why we still have many examples of languages and texts that are undecipherable: we've lost the context they were originally written in. Even Egyptian hieroglyphics were undecipherable until the 1800s when people used the Rosetta stone to provide context to decipher it.

Text has further issues as well, chief among them its ambiguity. Not only is it easy to under-specify things in text, but it's also possible for the same piece of text to mean different things at different times and places.

As a culture these issues may be strengths; poetry and literature derive strength from this openness to interpretation, and many would argue so does law where statutes written centuries ago can be adapted to our time. But from a purely data storage and transmission perspective, these are clear weaknesses.


I think the one weakness of text is context. In 2,000 years, if someone finds an inscription reading "Thanks Obama", what will they make of it? They'll might know that Obama was a president of a country. Without context it would be hard to tell if this was 1) genuine sentiment 2) snarky criticism or 3) ironic. Good luck with "Covefe" or whatever.

My point is that context matters a lot, and (I assume) that it can only be reconstructed through a lot of text.


I’m so glad this post is still around. I cite it a lot when discussing creative tools with people who think that visual is the only way to go.

Obviously the best computer tools offer both visual and textual ways of working. But if you have to pick one, bet on text.

I was about to add a caveat about pure visual tasks, like image composition, but advances like DALL-E are starting to put those tasks into question as well. If I were making a photo editor today, I would bet on text.


I love Emacs and plain text. But claiming that text is everything is ignoring the power of images. The brain can immediately grasp the meaning of an image, visuals are immediately understood, some of the most remarkable things human beings have created is art, that is visual imagery.

Just because our current technology is best able to handle and deal with text says nothing about the power of sound, images, sketches, drawings and handwritten notes.


He uses a mathematical formula as an example of text, with its symbols and whatnot. I don't count that as "text", to me, its closer to a UML diagram or flowchart than it is to the text I'm using to write this comment. If you just conflate "all things represented through glyphs and symbols", then I find that too broad to be useful.


This looks like a specialization of Tim Berners-Lee's Rule of Least Power[1]:

  > suggests choosing the least powerful [computer] language suitable for a given purpose
https://en.wikipedia.org/wiki/Rule_of_least_power


The author seems to be arguing that "most general" = "best" - because text is the most general communication format, it's the best.

This argument is pretty trivially false.

By this logic, the best programming language is assembly, because it's the most general. Actually, writing opcodes directly might be more general, in case your assembler doesn't have translations for some undocumented opcodes. The best editor would be a hex editor. The best web browser would be netcat piped into a hex editor, as well. The best OS would be no OS at all (because OSes impose restrictions in order to make programming easier and safer). The best application, for any kind of application, would be an interpreter that would allow you to create your own, however you wanted.

Engineering is necessarily a tradeoff between generality and efficiency. Technology is largely used to make things more efficient, and so our tools always impose some constraints on the problem/solution space in order to be more efficient than not using that tool.

A common design pattern is a specialized fast-path and slower (but more general) fallback.

In addition, as many others have stated here: "the right tool for the right job". Videos, audio, pictures, and interactive tools will always be more efficient for certain problems. If your sole concern is generality, then yes, by all means, use text. However, this will almost never happen; your design space will almost always necessitate a tradeoff of generality with efficiency - in which case, pure plain text is rarely the solution.


Assembly language is not the most general because and particular assembly code can be run on only a few particular architectures. It's the higher-level languages that are more general.


Alright, I'll add a caveat: "assembly language is the most general for a particular architecture".

> It's the higher-level languages that are more general.

This is still false. Rust, for instance, is higher-level than C, while being less general - the borrow-checker constrains you from being able to do certain things. Haskell is even higher-level and even more restrictive.

The most general "programming language" is a Turing machine. Not coincidentally, it's also the least useful, and nobody programs in it.

This reinforces my point that "more general = more useful" is false.


It's a good post, and he makes excellent points.

Like any "hard and fast" rule, however, it tends to go a bit pear-shaped, when viewed from certain contexts.

As someone that has made the mistake of designing a "pure" iconic interface, I can tell you that alternatives to text UI can be quite difficult to implement[0].

But a well-designed symbolic UX can be leaps and bounds more effective than text.

In some contexts.

Basically, YMMV.

The main issue with text, is that is assumes that:

1) Everybody can read, and

B) Everybody is on the same page.

In any given day, I notice written signs everywhere. But the really important stuff tends to be done symbolically.

Notably, caution/danger signs and other warnings.

Road signs are almost always text, but the same thing I just mentioned, applies to important cautionary road signs. You can assume that anyone driving can read (written road test), so why use icons?

That's because we can process symbols much more quickly and effectively than text. A well-designed icon can be instantly recognizable. Take, for example, the classic radiation or biohazard icons.

They still need "training" to properly interpret; but nothing like the level of education required to simply read (and understand) the word "BIOHAZARD."

    ⠀⠀⠀⠀⠀⠀⠀⠀⠀⣠⡖⠁⠀⠀⠀⠀⠀⠀⠈⢲⣄⠀⠀⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⠀⠀⠀⠀⣼⡏⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢹⣧⠀⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⠀⠀⠀⣸⣿⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠈⣿⣇⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⠀⠀⠀⣿⣿⡇⠀⢀⣀⣤⣤⣤⣤⣀⡀⠀⢸⣿⣿⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⠀⠀⠀⢻⣿⣿⣔⢿⡿⠟⠛⠛⠻⢿⡿⣢⣿⣿⡟⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⣀⣤⣶⣾⣿⣿⣿⣷⣤⣀⡀⢀⣀⣤⣾⣿⣿⣿⣷⣶⣤⡀⠀⠀⠀⠀
    ⠀⠀⢠⣾⣿⡿⠿⠿⠿⣿⣿⣿⣿⡿⠏⠻⢿⣿⣿⣿⣿⠿⠿⠿⢿⣿⣷⡀⠀⠀
    ⠀⢠⡿⠋⠁⠀⠀⢸⣿⡇⠉⠻⣿⠇⠀⠀⠸⣿⡿⠋⢰⣿⡇⠀⠀⠈⠙⢿⡄⠀
    ⠀⡿⠁⠀⠀⠀⠀⠘⣿⣷⡀⠀⠰⣿⣶⣶⣿⡎⠀⢀⣾⣿⠇⠀⠀⠀⠀⠈⢿⠀
    ⠀⡇⠀⠀⠀⠀⠀⠀⠹⣿⣷⣄⠀⣿⣿⣿⣿⠀⣠⣾⣿⠏⠀⠀⠀⠀⠀⠀⢸⠀
    ⠀⠁⠀⠀⠀⠀⠀⠀⠀⠈⠻⢿⢇⣿⣿⣿⣿⡸⣿⠟⠁⠀⠀⠀⠀⠀⠀⠀⠈⠀
    ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣼⣿⣿⣿⣿⣧⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠐⢤⣀⣀⢀⣀⣠⣴⣿⣿⠿⠋⠙⠿⣿⣿⣦⣄⣀⠀⠀⣀⡠⠂⠀⠀⠀
    ⠀⠀⠀⠀⠀⠈⠉⠛⠛⠛⠛⠉⠀⠀⠀⠀⠀⠈⠉⠛⠛⠛⠛⠋⠁⠀⠀⠀⠀⠀
And, of course, the classic skull tends to convey a message that even the uneducated can understand.

I have learned the hard way, not to get too creative, when presenting GUI. I've learned to use platform conventions, and ISO symbols[1], where possible; even if I am not that thrilled with the aesthetics.

[0] https://littlegreenviper.com/miscellany/the-road-most-travel...

[1] https://www.iso.org/obp/ui/


Author says text is the oldest and most stable communication technology but his example is some pictures on rock.


FYI: I collect bet on text goodies in the Awesome .TXT / Text page [1].

Yes, Text, Text, Text - The past, present and future of writing (and publishing).

[1] https://github.com/officetxt/awesome-txt


It is interesting to think about it in the context of coding.

There is a strong opinion (which I share) that text coding is much more efficient than 'graphical' alternatives due to text flexibility and nice features ( easy compare, universal medium...)


While agree with the basic premise of the author, I find it amusing that I found the black background and font selection difficult to read. When I switched to reader view, I was experiencing pure text, and it was better for me.


Context < Subtext < Plaintext

Cute, overly self congratulatory turns of word aside, this had me look up the etymology of text. I hadn’t put it together before, but woven is such a fitting root for the fabric of our thoughts made manifest.


So happy to be on a website w/o any pictures or videos or bright emojis. To tell the truth I have some problems of communication in messengers where everyone considers cool to send animated sticker instead of words.


Actually a single letter is "a picture". So Text is a collection of small pictures. There is no such thing as "Text"- we give a meaning. Otherwise it is bunch of weird arrows and curves - pictures.


One can keep playing this game.

There is no such thing as pictures. It’s just photons that bounce off or are emitted by certain surfaces that fall on our eyes and are then transmitted as electric impulses to our brains which trigger synapses....it’s all just stuff firing in our brains.

It’s pretty obvious what the author of the article means by picture and texts. I don’t think there’s anyone who is not trying to be deliberately obtuse who would have a hard time figuring out what they mean by pictures and text here.


Yes the game. I understand what you mean.

But my intention was question validity of series of claims the author making using full loaded abstract words - " Text is everything" " text is the most powerful, useful, effective communication" . Of course the author should expect to be challenged on meaning of words.

Let's say you looking at some painting painted by a person who wanted "paint" the emotion inside his brain. Not sure you can completely paint the emotions as picture. But author claims is text can completely express the painting or the emotion - that is not true. Can someone write words to describe me - how lemon taste? And I get same feeling on my tongue after reading "text" ?


> Actually a single letter is "a picture".

It certainly is not.

Think about character recognition software. The point is to take a picture, and map it to a discrete value. The pictures that correspond to "a", for example, have a lot of variation. But the point of writing is to remove the ambiguity inherent in drawing something, and map it to one of a small set of discrete values.

So even before binary encoding systems, or even the printing press, text was a technology for conveying information with less ambiguity than drawing.

One could argue speech is similar. There are a lot of variations of sound corresponding to a phoneme, but language reduces a continuous stream of sound to a discrete sequence of phonemes in our brain, which are then disambiguated into words and sentences conveying concepts.

The whole continuous -> discrete mapping underlies both spoken language and text.


A letter is not a picture, at least not in the way we generally talk about pictures – it can be graphically manipulated and re-imagined in almost limitless ways and degrees and still perform its function. For most things-that-we-call-pictures, if you even made a relatively small adjustment to it, it would be a different picture.


That really depends on your language system. What you say is true for alphabetic systems, but quite different for logographic (e.g. Chinese) or pictographic systems.


I don't think they're different enough to refute my point. You see Chinese characters in a great variety of typographic and handwritten styles, all able to be read by fluent people.


It is equally true of the Chinese character system.

They have more "letters", but the point is still to unambiguously map a drawing of a character to a specific, discrete value.


Alphabetic glyphs and pictures are not the same thing. Glyphs are abstractions. Pictures are literal (ironic).


Something about hammers and nails and using the right tools for the job


Indeed, there are situations where you cannot convey your message using text.

One of the things I’ve learn doing e-commerce is that if you believe you can just use more text to compensate for short comings in you UX you’ll be very disappointed.

We had a subscription product, you where informed that you’d be sign up for a monthly charge seven times during checkout and people still complained. They just saw two prices and click on the lowest. No amount of text will fix people who are basically on auto-pilot.


This sounds great if you are very literate and well-vocabularied.

Most people in the world are not. Or don't speak English at all, another problem with text that you don't find with images.


There is a saying "a picture is worth a thousand words". The brain will recognize a white flurry cat from an image faster than from a phrase "white flurry cat".


> But let's hit the random button on wikipedia and pick a sentence, see if you can draw a picture to convey it, mm?

To be needlessly pedantic, my computer drew the image that conveyed this to me.

Less pedantically text is a medium of exchange for language, and a lossy one just like spoken word. I think there's a lot of power in its flexibility due to that lossiness. It's also one of its subtle weaknesses - we can read text from 5,000 years ago, but there's going to be much debate over understanding the text because of how much context has been lost to time.


As someone who has produced UML diagrams for a living for some time, this really talks to my soul... Thanks for posting.


the phonetic alphabet is a brilliant invention. made by thousands of people over thousands of years.

the least studied 'character' of this alphabet is the most imporant, the most critical character in any phonetic alphabet is the blank space. withoutitphoneticwritingsdoesnotmakesasmuchsesnse.


Obligatory reference to the Unix Koan "Master Foo Discourses on the Graphical User Interface", after reading this post:

> Master Foo said nothing, but pointed at the moon. A nearby dog began to bark at the master's hand.

> “I don't understand you!” said the programmer.

http://www.catb.org/~esr/writings/unix-koans/gui-programmer....


When breathing, always bet on air..


The day will come when we can think an image and then send it. It might change things.


> "Human rights are moral principles or norms that describe certain standards of human behaviour, and are regularly protected as legal rights in national and international law."

image: https://imgur.com/a/8J6yeRe


That is a great image conveying this concept.

But if you started from the image, and asked people to generate a sentence from it, you would get a lot of variation from the quoted sentence.


"Kids playing"


4000 bytes? That tweeting bird picture really is worth a thousand words ;).


On a 32 bit machine of course.


Nitpick:

> It can be translated.

It needs to be translated. No such issue with photos.


It depends, memes, short clips are super effective too.


And yet I can find toilets easily in any country with a simple logo but have to hunt for them if they are spelled out.

Images are wonderful if you don't live in a bubble where everybody speaks the same language.


Dumb post. It's not a contest. Use whichever is appropriate.

I design systems using both diagrams and text sometimes laid out in understood, non-linear patterns. I also enjoy music.


How is neuralink not being discussed here?


What is that image of Gnus doing there?


Ludwigstein cried after reading this


A picture is worth, like, 100 words


Yet Instagram is flourishing.


(2014)


I think text is a powerful mean to express ideas, but pictures are also. And it comes down to how people think: in text or in pictures?

I don't know much about it, but there in a talk, Jordan Peterson addressed that there are people who can think in text, in pictures or even both. And that trait says a lot about ones character.

Interesting stuff


Gopher vs HTTP.

One was only text.


> assuming we treat speech/signing as natural phenomenon -- there are no human societies without it -- whereas textual capability has to be transmitted, taught, acquired

I’ve been watching videos of pets (dogs, cats) being “button trained”—being taught to use (primitive, non-syntactic) language by making associations between things/acts/emotional states and the pressing of one or another audio-playing buttons that has been placed on the floor.

It has very much driven home the point for me, that “spoken language” is actually not something inherent and instinctual to humans (or to any species); but rather is a technology. It’s just a technology that’s rather simple to learn, if you have the right underlying hardware acceleration (e.g. a cerebral cortex)—and, crucially, a teacher. For other intelligent mammals, apparently the teacher is the only component they’re missing!

Language is a technology that humans in particular find very intuitive—at least at a young age when our brains are malleable—but not one we inherently start with. We absorb it easily if we’re immersed in a society where everybody uses language from birth. But in situations where that’s not true (feral children, some very broken homes) we don’t.

In a world where every human being instantly had all entrained structure in their neocortex erased, such that we were “reduced” to being upright hairless apes with the capacity for language but no knowledge of it, I don’t think we’d just instantaneously come up with the idea of language and begin attempting to develop languages to communicate, the way modern people instantly try to develop a creole of the languages they do know, when stuck in a situation with people who share no common language with them.

The idea to associate concepts with specific mouth-noises—and to condition others to use those same mouth-noises for the same concepts, to facilitate transmission of thought—might randomly arise in a few people, but it’d need to catch on and spread from there, just like any other technology. It would either need to be observed and copied, or actively taught.

And I hypothesize that that is what happened (pre)historically: at some point, there were several memetic “waves” spreading increasingly-technologically-advanced (e.g. syntactic, expressive) forms of language across human populations with brains already structurally amenable to them. Of course, each wave would only be a struggle to the generation that pioneered it; the next generation, being immersed in that new, more-complex language form from birth, would find it similarly intuitive.

(This makes me wonder whether we’ve yet hit the “limits of linguistic expressiveness” for our current brain size, such that we’d need to unlimit e.g. average skull diameter at birth to let us get any fancier with language; or whether we’ve still got some, ah, “headroom” left.)


>“spoken language” is actually not something inherent and instinctual to humans

I agree, as you say until you reach the point where newborns are already surrounded by it.

>It’s just a technology that’s rather simple to learn, if you have the right underlying hardware acceleration (e.g. a cerebral cortex)—and, crucially, a teacher.

No teacher needed whatsoever quite often, but maybe so usually, so I can not say crucial.

>And I hypothesize that that is what happened (pre)historically: at some point, there were several memetic “waves” spreading increasingly-technologically-advanced (e.g. syntactic, expressive) forms of language across human populations with brains already structurally amenable to them. Of course, each wave would only be a struggle to the generation that pioneered it; the next generation, being immersed in that new, more-complex language form from birth, would find it similarly intuitive.

The evolution of more expressive vocalizations might have been required along these lines for us to reach where we always thought modern man was to begin with.


There is a well known fact in the field of psychometrics: human general intelligence g can be modeled as two relatively independent sub-factors: verbal intelligence and spatial intelligence. Other sub-factors, if any, are way more speculative.

As a human being, it is quite possible to have verbal intelligence higher than spatial intelligence, and indeed I know some people with verbal tilt and some other people with spatial tilt. These people tend to think & approach problems differently, while having different strengths and weaknesses. Naturally, similar people cluster together, and some professions (e.g. lawyers, journalists, writers, programmers) are more amenable to verbally tilted persons, while other professions (mechanical engineering, airplane piloting, architecture) are amenable to spatially tilted persons.

Looking around via this lens, discerning cognitive styles inherent in design of the human experience is enlightening. One can see that our physical and social, educational environments and governing institutions are designed with one cognitive style in mind at the expense of the other: and this privilege goes to verbal cognitive style.

Let me offer a different perspective: to a person with higher spatial and weaker verbal cognition, this environment looks physically simplistic, tasteless, sometimes outright boring, often suffocatingly so. Utilitarian safety & simplicity prevailing over beauty and shape-being, denying the inhabitants possibilities of space meaningful by itself. Letters, words, strings of words are everywhere, starting with high-school where the recent historical trend of increasing verbalization of curriculum continues, and on to adult life where verbally intensive professions pay more and command significantly more power (note how in the aforementioned occupation list the second one contains less status-worthy & more specialized occupations). Limitless possibilities of rendered worlds on megapixel screens collapse into a flat-designed abstract hellscape of recursively composed words and menus. The brain of the child - a pinnacle of neuroplasticity - adapts, as relentless march of critical developmental periods continues, unnecessary white matter pathways wither away, while economically useful ones are potentiated and strengthened for the forthcoming endless competition with similar human beings, similarly shaped. The best and brightest in this game of words become lawyers and career politicians, movers and shakers of our world; but are they truly our best, and do they truly imagine the referents of their symbols ? Are we led by seeing or blind ?

Much could be said about benefits of verbal thinking, endless composability (to some, vacuous, denying interesting constrained structure) of syntax & grammar & semantics, and rightly so. Yet one wonders, which avenues of thought, of being, both alone and together as a people, were not taken. How a civlization of prevailing gestalt could look like ? Confronted with this state of affairs, one wonders, if "Always betting on text", pedal-to-the-metal, more-of-the-same is really going to bring us somewhere at all ?

If you, dear reader, have some latent spatial/geometrical imagination which was pushed away by economically profitable word manipulation engines that grew through you, maybe you too wonder about this question.


> The best and brightest in this game of words become lawyers and career politicians, movers and shakers of our world; but are they truly our best, and do they truly imagine the referents of their symbols ?

What would a high spacial intelligence, low verbal intelligence individual thrive as a President, CEO, or other type of leader?

Those jobs revolve around effective communication, and I believe that will always favor high verbal intelligence individuals.


These are not jobs but positions of power; the assumption that success in securing them is indication of merit in the sense of producing best outcomes for society at large, or even this specific firm or department in isolation, amounts to just world fallacy borne out of common market delusions. In truth, communication-first leaders are only as good as the team under their command, that really understands the managed domain, is. And since this team is largely replenished by people of the same stock, vying for the same position of power, they are not very good.

What results from this process of distributed re-interpretation of vaguely-worded, likely not very coherent, buzzwordy policies is a strangely incoherent and characteristically apathetic environment. For an extreme example of this pattern, see a typical UN press release and its vanishingly small influence on the state of things.

If we ask a hypothetical about the possible outcome of, say, very high spatial intelligence nation-state leader, we enter wild speculation territory.

I could try: "An object of harmonious art, but in political, life-shaping domain. Some precisely specified spacetime volume of greatness you'd want to be a part of once you saw a glimpse of it, even if you cannot quite put it into words", i.e. "A place of participation in a great multi-scale-harmonious vision".


At the very top, real leaders can sometimes do very well using a spokesperson instead of direct communication.


Can someone give me a tl;dr?

(Just kidding. Fantastic article.)


> We can read texts from five thousand years ago,

What's interesting is pretty much all file formats are still readable (The media they are stored on is often not)

If you have the file it should be readable with a program you can find on Google.

On topic, No, although the article might be technically correct the biggest thing holding back many consumer products is lack of ability for non-text.

Signal had massive issues for years because it didn't have emoji. Now it's up with the rest. It's so much easier now you can copy a photo from Signal.

Like all these apps it needs work, like where's my real gun emoji. But they'll get there.


This feels like an arbitrary perspective.

Text has all the same properties as an image.

It’s a composite of elements of varying height, width, and meaning to the whole.

Some sentences can be longer, or shorter. One element can be overloaded with meaning more than another.

I’m not really sure if there’s a point here at all.


I like how the modern meme can capture an entire gestalt in 640x480 px. It's an extremely efficient means of communicating amongst people.


Text is actually made of a bunch of little pictures. Pictures: check and mate!


Completely fails to mention speech.

And from the claims it makes ("text is the oldest") it seems either forgot about it, or is not even aware such thing exists.


Speech is not a communication "technology"; unlike writing, it didn't have to be "discovered".


Except nobody really reads text anymore. If you want to deliver a powerful message pics and vids have the most bang for the byte.


That's a really surprising sentiment to me.

I despise videos when I want to learn something. Give me a nice text (+ pictures if a subject requires it) so I can follow at my own pace: slow down, skip or skim as needed.


For me it depends a lot on what I want to learn. For some things (e.g. programming) text is much better.

For other things (e.g. repair a part on a bike) text requires images, to really be able to quickly grasp what one has to do.

Finally, for learning e.g. dance moves I really want a movie (or a live instructor).

So like usual "it depends".


Yeah, it definitely depends for me too, but getting the nuances right from a video wouldn't work for me at all, even in a thing like dancing. Live instructor is an entirely different proposition (you can choose to focus on the bits that you care about, move around to get a better view, etc.). And I can stop them and ask for them to repeat, slow down or split their movement into more granular "part-steps".

Videos are simply not that for me. Perhaps with actual 3D videos and improved tooling we might get there (eg. ability to zoom in on an instructor's hips in "infinite" resolution if you are having trouble following that particular bit of their motion).

And I mean, you can sure learn from a video, but I just find it harder. Eg. I can go look at dance or basketball moves and try to reconstruct them, but it's usually a couple of core points that need to be taken in to achieve them that are much easier to grasp when broken into fundamentals.

However, videos where the gist of the work is on the student, vs on the instructor, are much easier to make, which I believe is part of their popularity!


I think you're vastly underestimating the number of bytes it takes for a pic or vid to deliver that bang.


Nobody? You're not reading Hacker News posts and comments? Nobody is purchasing books anymore? There's no more blog posts or news articles? Writers no longer have jobs? Is that why GRR Martin can't finish Winds of Winter?




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: