Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Which idea irritates you because it has not been taken far enough?
24 points by amichail on Nov 1, 2009 | hide | past | favorite | 69 comments
As an example, I am irritated by evolution not being taken far enough on the net.

Sure, you can view open source as evolution in action, but much of that relies on human intelligence rather than a more direct evolutionary process (e.g., like genetic programming).

One can imagine some sort of evolutionary process that is midway between genetic programming and human intelligence.

See: http://en.wikipedia.org/wiki/Human-based_computation




I think all of the test & measure stuff we talk about so much in the web context -- A/B testing, analytics, etc -- would revolutionize the world if it were applied with one tenth the dedication by human institutions which are not teeny little web startups.

This isn't to say that we have the patent on The Scientific Method. However, there are darn few organizations which seem to have it written into their DNA. Toyota does, famously. Imagine what would happen if your typical big-city school system did. ("First order of business: we had 37 A/B tests running in first grade classrooms last week. 31 of them did not lead to substantial results. Let's hear from our lead statistician about one promising innovation, which improved vocabulary retention among our lowest performing quartile of black students by 15%...")


Multivariate testing is an extremely limited tool. It's great when you're doing everything else right and not much better than rearranging the deck chairs on the titanic in ever other case.

I think it's dangerous (but very tempting) to try to apply it too generally. One passionate visionary thinking outside the box can do more than 1,000 statisticians ever could.


"... what would happen if your typical big-city school system did."

Education science never made it into US K-12 education. We Americans have a very fluffy concept of what education's about. We worry more about socialization than education (Neil Postman was right).

For a looooong time there have been US educational research journals with data far ahead of current practice, but many US educators never get much chance to see them. And I've never heard of a US (K-12) school that's doing anything empirical (beyond, oh, statistics anyway).


COLLECTIVE CONSCIOUSNESS!

I sort of feel like the internet is as close as we get to it now, with Twitter and other microblogging services bringing us a step closer. But really, I want full-on instant telepathic communication with anyone. Cell phones don't cut it. Gone would be the days of humming a tune in your head, whose name you do not know, only to realize you cannot emulate those noises with your mouth. Good luck finding out who plays that song. If everyone could read everyone else's thoughts, you'd just hum the tune in your head and the name would come to you right away.

I don't think we're good enough with gen. prog. right now to do anything useful or safe with it. GM crops seems to have negative health and environment effects. When we can open up notepad and write the ooDNA code for my new pet mini-liger, then we could probably start programming our food.

Also, I kind of think we need a new form of math. While maths currently describe our universe well, I think it is overly complex and non-intuitive. This leads me to believe that we could create a simpler language, one that builds off itself -- like a programming language which is built from AND, NOT, OR -- which is naturally recursive like the universe. These days you have to build an entire framework to encompass your latest quantum theory.


How is math not intuitive? Where is it needlessly complex? How is math not already "naturally recursive"? (What does that mean?) How is the universe naturally recursive? What does it mean to "build math off itself"? Are we not doing that already? I think you are complaining about math and then proposing that we build math back up exactly as it is now.


Math isn't NOT intuitive as far as math itself goes. It happens to be a very nice language. I tend to think that relativity and quantum mechanics do not mesh well because of the language we use to describe them. While standard maths is NP complete, it does not seem to inherently describe the universe without getting all crazy complex. I encourage you to try and comprehend even some of the simplest M-Theory equations. (not saying I do, but if math was really the language of the universe m-theory equations would be quite simple).

You know how in science and math people describe certain things as 'elegant' ? For example, E=mc^2 is very elegant. If math were elegant, you could get from here (higgs bosons) to there (atoms) with a simple equation which has been recursed x times to produce a mathematical model of an atom.

Math isn't bad, it's quite useful in fact. It's just not the natural language of the universe like everyone thinks it is.


There is a good possibility though that the universe is inherently complex in certain aspects for our brains. Human brains are evolved to survive in the part of the physical world observable by human organs, not to comprehend the essential forms of the universe.

E=MC^2 could be just an aspect of reality that happens to coincide with the way observable physical world works.

Maybe a different math for describing the universe written by an alien intelligence with different sensory organs could be simple for them, not still incomprehensible for us. And that alien intelligence is just a jackpot hit in the evolutionary possibility space (and the jackpot may well have never been hit).


E=MC^2 could be just an aspect of reality that happens to coincide with the way observable physical world works.

All aspects of reality coincide with the way the physical world works. Or, what does that sentence mean?

Maybe a different math for describing the universe written by an alien intelligence with different sensory organs could be simple for them, not still incomprehensible for us.

I think there is a strong argument against this. That some alien race is really good at visualizing certain kinds of N-dimensional systems of some weird kind of space is plausible, but to throw all of mathematics out, I think, is not. They would have a mere superset of the mathematics we have.

In particular, take computation. Are aliens never going to deal with strings of units of data? If they do, right there you have the notion of their length, and of natural numbers. You have notions like concatenating strings. And the homomorphism between the two. And we're off on the road to abstract algebra.

It's very unlikely for smart aliens not to develop the same study of discrete math and abstract algebra that we do, unless they were never to use discrete units of information, and were only capable of processing infinite amounts of information at a time. Would they then have no use for the idea of associativity? Of proofs that consist of a finite set of symbols in a finite alphabet? Of proofs that consist of a finite set of symbols in an uncountably infinite alphabet? With any of these they would on the road to having a superset of our mathematical knowledge, rather than something different. We don't know how such a being would be physically possible, in the first place.

I think it would be more correct, and less exciting, to predict that some aliens might be smarter than us in certain ways.


All aspects of reality coincide with the way the physical world works.

I think observable is the keyword here... we happened to be able to observe enough that we can conclude that E = mc^2

As for your second point... I don't think I agree. The fact of the matter is that as humans, we are essentially unable to reason about how beings with a different set of senses would think. With respect to cosmology, we basically react to how light moves around and that forms the basis of our science (our other senses don't really come into the picture). Its very difficult to think about an intelligence based on some other source of information and impossible to know what sort of data would be present (since we don't have access to the information).


Ha! I started reading your comment, and I said to myself: "hey, this sounds like human_v2." Nice to see you around :)

So, are you saying that the rote structure of nuanced math is too complicated for it to be accurate or meaningful? I'm sure you're familiar with reductionism (http://www.absoluteastronomy.com/topics/Reductionism). Just imagine that mechanical duck as math.

One possible reason that we have not yet found "that one fundamental" thing in math is that our universe itself probably isn't a perpetual machine on its own (or, what we can perceive as our universe). So, there is a good chance that it is physically impossible for us to get to where you want us to be with our understanding of math. We wont know until we reach the singularity. But even then....

Anyway, my criticism of math has been that it has been getting far too philosophical. But, I am an amateur.


it seams you want someone to create a language that you could use to create itself. where it could be modified on the fly. where the syntax used to describe data would also be interchangeable with the syntax of the code.

(in 'tongue 'cheek)

> t

http://en.wikipedia.org/wiki/Lisp_(programming_language)


About collective consciousness,something like the WebBot[1] project uses it fairly well.I hear it uses the collective consciousness to predict <dramatic sound>The FUTURE</dramatic sound>

[1]http://en.wikipedia.org/wiki/Web_Bot


Functional programming.

I really like it when my computer crashes because someone decided that checking whether or not "-5435" was a valid array index was "too time consuming". Wait, I hate that. I have 4 cores that are idle 99% of the time. Let the computer do some self-checks that will keep my data intact.

(Read the ext3 source code some time and tell yourself that you trust important data to it. The only reason it works at all is because everyone has already been burned by the most obvious bugs.)


?... Automatic array-bounds checking doesn't require functional programming.

Besides, functional programming is not The One True Paradigm To Rule Them All either. Dataflow, for example, is an example of another superb paradigm.


The general argument is against "it's too slow". People avoid FP because "it's too slow" in favor of languges like C that get you the wrong answer really quickly. (Functional programming is not "too slow", of course; consider the popularity of Ruby and realize that SBCL, OCaml, GHC, ... are often 50x faster. Of course, those are all 3x slower than C, so clearly should be avoided...)


Everything. I wish I was born at least 500 years in the future. Science is just beginning and I was born at the wrong damn time. I could be flying my own space ship and visiting HN over The Subspacenet. No single advance could ever satisfy me or make up for my disappointment at being born too soon.

(Not that I'm not appreciative for being alive at all. I'm fortunate in many ways, but that doesn't stop me from imagining what could have been.)


If we look back at human history and extrapolate that trend, 500 years on we could be worse off than what we are now in technology. Civilizations no matter how strong they look, have a trend of rising and falling. And when it falls, it's knowledge gradually gets lost.

40 years back, human where more capable of visiting the moon then they are now.

Fundamental discoveries in physics is almost stuck now for 60 years, since discoveries of quantum mechanics.

The only gain we are seeing is in medical science, and information technology (in Engineering, not Science).

That's what I feel.


No. Technology has progressed over thousands of years.

Of course the world could go to complete shit. If it doesn't though, we'll end up with technology that's unimaginably advanced. Whether that's in 500 years or 50,000 years makes no difference.


there's not enough going on with 10 Foot Interface advancements. Boxee and XBMC rock, but a lot is happening and about to happen in this space. What's possible today was not possible 18 months ago whether it be hardware OR software. Boxee already has 600k users and that's with their alpha. I'd love to see hayzap or a casual games developer launch something on their platform. With that said, seeing more sites have a "10 foot interface" alternative, just like they have a gui for mobile devices would be great.

http://en.wikipedia.org/wiki/10-foot_user_interface


Here are a couple that came to mind:

1) Constructive feedback -- I want more, now :)

2) Web UI technologies -- progress, but at a snail's pace

3) Web APIs -- bring down those "walls"

4) Internet-driven TV apps -- why not?


For no. 1 - are you referring to the "casual observer" problem ?


* The network is the computer.

I want processes that migrate with me. When I login at work, I should be able to grab my webbrowser from home and continue it executing on my work machine. Or perhaps on my cellphone (although the UI probably would be horrible). When I reboot my laptop, processes get migrated onto another machine I have permissions for (my desktop?), and come back when the reboot completes.

* Distributed Storage.

I have a string of PC's I have own and/or have (legitimate!) access to, but I have problems storing data. Eg my desktop at home has gigs of local storage, but my home directory is mounted off the file server which is running out of space. Why can't I easily merge those two file stores?

* Augmented Reality.

* Mesh based Free Space Optics.

To be able to set up a wireless AP that has line of sight to another AP, that then sets up a mesh using FSO. FSO is near interference free.

* FPGA's for accelerating computation and reducing power consumption.

* Location/Situational awareness in cellphones.

Guessing better if it should be a loud ringtone based on ambient noise? Guessing who I'm going to call given that I always call this person at this time at this location?

* Better internet file distribution.

Knowing/Guessing something about network topology and doing something smart about it w.r.t choosing who to download from. (TCP_INFO to get TCP's estimated RTT for a link? Measuring how long it takes an IP address to download from you/to you and remembering that for your current network, and biasing which peer/server you use in future based on this). Using Source Specific Multicast when available to get network multiplier effects.

* End to End Encryption of communications.

* Natural Language Parsing

We know that computers have problems with ambiguous sentences, but a search engine where you give it a natural language query, and it parses it, and then asks for which parsing is correct, or asks you to resubmit your query in a less ambiguous way. People seem quite happy to iteratively query a search engine to get an answer they want, having a NLP guess the wrong meaning is easily rectified with a more specific query. Wolfram Alpha tries this. It however doesn't seem to provide you any obvious way to improve your query, variations on a theme tend to result in the same parse.


Capitalism-driven legislation. Particularly the taxation of negative externalities.


Democracy's design...

The scarce use of technology to encourage and access crowd wisdom. (Rather than the cleverness invested in things like gerrymandering and misleading advertising techniques.)

This is a similar phenomenon to the pathological cleverness used in so much social network advertising... you know, bait and switch, the devil's-in-the-details approach to supposed fellow decision makers.

I'm an anthropologist not a software guy. But the solutions are fairly obvious which I try to address elsewhere. Fortunately the reception is getting better in some places.


Voice-command computing. As Scotty said, "A keyboard? How quaint!"

(NOT for offices!) I want to tell my wearable to look up facts, do math, remember everything I tell it to remember, and comb through ALL the news keeping my personal preferences in mind. And quick instructions on what I can cook with the food I've got on hand ... which it remembers.


Bandwidth as a constitutional right in addition to life, liberty, and pursuit of happiness.


Life, liberty, and the pursuit of happiness are not guaranteed anywhere in the Constitution. Those three are a rhetorical flourish in the Declaration of Independence, and do not create enforceable legal rights for Americans. The Constitution guarantees enforceable negative rights, for example, it forbids the government from killing (or otherwise punishing) you without due process of law.

The Constitution, by design, doesn't establish positive rights. An example of a positive right is "You have the right to an education."

Positive rights suffer from implementation issues, expanding government creep as pressure groups suffer rights-envy, and the generic problems associated with state-provided services.

An example of an implementation issue: the Japanese Constitution guarantees that "All people shall have the right to receive an equal education correspondent to their ability, as provided for by law." This has been interpreted as making it unconstitutional to remove students from mainstream classrooms, regardless of behavioral disability or a distressing propensity to stab teachers in the neck.

An example of rights-envy: Once you have one positive right, every pressure group under the sun will want their own mandate. The argument will sound something like this: "The constitution guarantees bandwidth, food to eat is a lot more important than bandwidth, the constitution should guarantee food, too." Other favorites for expansion are health care, education, and sexual services. (Oh, Europe, what a continent.)

The American political consensus is that positive rights are not desirable at the federal constitutional level.


It's not HN until someone corrects you with facts.

I admit I wasn't clear that L,L,PofH was guaranteed anywhere in the constitution.

But with my one-liner, the point I hoped to get across was that I do think that it should be extended to access to the web and its information, as it becomes more ubiquitous and integrated in our lives. I named it "bandwidth" mostly as a remark to how I wished I had more of it.


Nice comment; though, I think the OP was just comedically endorsing Net Neutrality, which, iirc, asserts a negative right.


Net Neutrality isn't a negative right. If you own a network, net neutrality takes away from your ability to use it as you see fit; if you want, for whatever reason, to dick around with your customers and block Craigslist, you should be able to do so -- just like a newspaper can refuse to run an ad, or even refuse to write about a story.


In my opinion, newspaper is not a right analogy. Newspapers provide content, ISPs don't. I can understand if, say, National Geographic puts some restriction on the online content but why should my ISP decides if I can access National Geographic or not. Telephone networks is the right analogy here - will you allow your telephone provider to control whom you can call?


will you allow your telephone provider to control whom you can call?

Not now I won't, because that would be changing the service I'm getting to something I didn't sign up for.

If I got an offer tomorrow for a "limited" phone service I would turn it down, but I wouldn't consider it immoral or bad or something that shouldn't exist.


Look, you share an appallingly common ignorance of American history and law. Okay, fine. By why in the name of all that defies ignorance and stupidity would you equate life and liberty with Bandwidth for god's sake? Seriously.

Bandwidth: 15/dollars a month.


Embarrassingly so, I'm afraid. It's been a while since I read up on my American history and law as any good citizen should, I suppose.

But all the same, I still stand by my assertion above later in the thread.


Geoengineering and practical solutions to climate issues - as opposed to the rampant, ineffective cult of "environmentalism." Smart minds are working on geoengineering solutions - they just need cash and government support.


Ending world hunger. Children Education. Peace.


You know why we will never have peace? Because it upsets the status quo. The status quo gives power to many people who have, at present, sufficient power to maintain their ability to stifle the opportunity for peace.

It's really simple really, just stop blowing things up. Stop killing people. If people kill your people, ignore them. Repair the damage and keep going. Eventually, they'll give up.


I'd like our government to do more to help rather than hinder folks like Carl Malamud and the others at http://public.resource.org


Regional conflicts. And the idea of countries, in the first place.


The idea of countries is not taken far enough?


And/or the idea of regional conflicts? :)

Actually, paraschopra probably means the opposite, that is, these ideas are taken too far... or at least, thats my read.


Yep, right. I think resolution of regional conflicts hasnt taken far. Plus solution to arbitrarily defined regions called countries hasnt taken too far


Well, its actually a fairly new idea.

http://en.wikipedia.org/wiki/Peace_of_Westphalia

In any event, whats your better solution? Not to say you don't have one, but I'm curious.


One vote per registered person for elections, alleviating the need for the Electoral College. And maybe even representative government?


I think that every election year when the local polling place is packed to the brim with people waiting to get registered by slow, old people. Register every citizen centrally and put voting kiosks (that could double as places to pay taxes, etc maybe?) in grocery stores, gas stations, etc.


Tesla's global wireless communication and energy.


VR goggles, that stuff. But not with the huge 90's style visors, obviously. Displays integrated into sunglasses, etc.


Probably one of the most promising ideas that has lost much of its meaning....We the People of the United States, in Order to form a more perfect Union, establish Justice, insure domestic Tranquility, provide for the common defence, promote the general Welfare, and secure the Blessings of Liberty to ourselves and our Posterity, do ordain and establish this Constitution for the United States of America.


Do you really believe that? Are you really so blind to all the progress since 1788?


Of course I see progress, but the question was "what idea hasn't been taken far enough."


Ok. “… has lost all its meaning …” just sounded to me like it was all downhill since 1788.


Of course I said " lost much of its meaning" not "all of its meaning".


Sorry for the misquote.


"insure" domestic "Tranquility"?


Great. More insurance scams to pay high monthlies for.


hey at least it's not "insure Foreign Tranquility"


Intelligence, truthfulness, purposeness and politeness in advertising.

I'm continually amazed at the sheer stupidity of it all.


Platform as a Service.

Federation of micro-blogging.


Linus views open source, or at least the kernel, as guided evolution


teledildonics


One of the inherent problems with tactile feedback over a distance is the speed of light. The last haptic device I used received feedback from a computer every ms. Too much latency kills the experience for long distance tactile communication. How far could you reasonably "reach out and touch" someone? I'd guess maybe a hundred or few hundred miles.


As a general rule, the human brain doesn't notice any delay of less that 100ms. Anything less than that is "instantaneous".

Light takes 1.6ms to travel 300 miles, or 3.2ms for a round-trip, which is well below that threshold. To get up to a 100ms round-trip, the two parties would have to be 0.05 light-seconds apart, or just shy of 15000km (9314 miles).

That's plenty to get you North America, Europe and East Asia, but unfortunately Australia is a bit far from North America or Europe.

Obviously, that's assuming the signals are transferred at the speed of light with no other sources of latency, which is an ideal. But still, it's a far cry from "a hundred or few hundred miles".


marginal utility. hell i'd be satisfied if people would just accept supply and demand.


Pocket touch screen devices.

I want cheap, low-functionality touch screen devices. Like, an iPhone, but with only a calculator, a dictionary/thesaurus, and some other useful but simple applications (text reader/editor, todo list, etc).

Or how about an open-development touch screen device? Where programmers can create their own applications for it easily?


Sounds like you want any Palm or Windows CE device made in the last decade. I am not kidding, you described 90% of what I do on the 2003 PDA I picked up for $15. (The remaining 10% is gmail, skype, and an NES emulator.) Did I mention it is easy to program for? I've got every language from Scheme to J on board.

Welcome to the past, your dreams may be found at any yard sale.


You want an ipod touch. Run linux on it.


so a cheap iphone or droidphone...


open source hardware


Individual Freedom!




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

Search: