This is a beautiful looking site but is decidedly designed for Japanese readers. I will leave cultural commentary aside. While the romanized names are large, they are incredibly awkward to read being rendered vertically with rotation. I would recommend rendering the romanji vertically without rotation or using a Western left-to-right rendering. While I have some rudimentary Japanese reading skills, lack of furigana above the kanji color names further limits the readability.
> This is a beautiful looking site but is decidedly designed for Japanese readers.
Well, yeah, it's a site in Japanese with specifically Japanese subject matter, why is that a problem?
> I will leave cultural commentary aside. While the romanized names are large, they are incredibly awkward to read being rendered vertically with rotation
IMO, vertically-with-rotation is fine, probably even ideal, for reading the romanized text given the rest of the design, which presumably is appropriate for the main target audience.
The only real readability problem I see is that the text color doesn't change for contrast with the background color.
> I would recommend rendering the romanji vertically without rotation
Please don't. Consistently rotated context is easier to read as words than vertical words of horizontal letters.
> or using a Western left-to-right rendering
That's done for the Romanized text where space permits (e.g., for the selected color in the right-side banner.) But unless you are suggesting that the site should be designed first and foremost for Westerners, I don't see any more of that as reasonable.
> IMO, vertically-with-rotation is fine, probably even ideal, for reading the romanized text given the rest of the design, which presumably is appropriate for the main target audience.
I'm soso with the rotated text but the font choice just compounds the readability issue. the lines are so thin that, when rotated, they just become overly antialiased.
That being said, if you look around you'll basically _never_ see characters from horizontal languages on ANY public billboards that are rotated. They're always vertically oriented letters. While it may work for you personally, graphic designers learned a long time ago that this doesn't work for the general populace.
> people in the English speaking world forget that not everything is optimized for them.
people everywhere forget that not only are there people who process things differently than them, but that on the web we have the technology to accommodate most of those needs.
In this case the whole thing could be rotated one way for folks with their browsers set to a language that is traditionally vertical and another way for folks where their browsers are left-to-right languages, and rotated and flipped for folks with right-to-left languages.
It is a terrible limitation of Latin characters that it's hard to read vertically. However the hard coded as images labels is inexcusable. Zooming in causes the images to look upscaled and terrible.
I think it might actually be more readable (as someone exclusively versed in languages that can be read left-to-right) if the letters were oriented horizontally,
L
I
K
E
S
O
. Otherwise it's a bit difficult to recognize their shapes.
I see Latin characters embedded in vertical CJK text a lot and have encountered both the method the website uses and the one you proposed. In my opinion your method is harder to read because most people read entire clusters of letters at a time, and if the letters were arranged vertically you then have to slowly read it letter by letter instead. In contrast (for me at least) it's very easy to read 90° rotated text by simply tilting my head. I think the 90° rotation method is also the official way of embedding latin text in vertical CJK as well.
I think that’s actually a great point, which I failed to really realize because the words in this case were novel so I was reading them one character at a time, so I actually responded with “how do I make this unfamiliar thing easier for me to read right now”. I would expect being able to recognize rotated Latin script well enough with a bit of practice as well.
Japanese is traditionally written top to bottom, right to left. Vertical typesetting (mainstream in print media but rarely seen on the Web, even though CSS has decent support of TB-RL CJK typesetting now) follows this rule, rotating any words from LR-TB writing systems 90 degrees clockwise.
The idea may seem unconventional at first approach, but this page’s layout—impractical as it is—is like a breath of fresh air to my eyes.
in places with romanized lettering it was long ago determined that this is much more difficult for readers to process. It may be a breath of fresh air for you but for most readers it's difficult to mentally process.
However, I find that the approach described by saagarjha has the objective issue that letters in one word are more difficult to visually connect to each other, and would require a monospace font specially designed for vertical writing.
It's a long list of Japanese color names with accompanying RGB values. Some are well known in Japan (sakura, nadeshiko), the vast majority are obscure.
They also use an off-the-wall half-Hepburn/half-wapuro romanization scheme that's the virtual equivalent of nails on chalkboard if you care about this sort of thing (CYOHSYUN 長春 etc).
On the plus side, kunrei-shiki romanisation, which is the official government style and has been standardised as ISO 3602, is in practice ignored even by that Japanese government.
In terms of nails on blackboards, I ask that you bathe for a moment in the exquisite splendour of the romanisation that is "Mt Huzi", Japan's highest peak.
This is not a well designed website. The names of the colors aren't even searchable (because it's not actual text), and unless Japanese people read roman letters differently than we do, letters rotated 90 degrees is awful for readability and should never be used. Text color is also never adapted to background color, pick "Sakura" for example and the white foreground text is completely unreadable.
When Japanese text is written vertically, Roman letters (with the exception of short acronyms) are indeed written with a 90-degree clockwise rotation. If you were reading a novel in Japanese that contains a brief English quotation, you would turn the book 90-degrees counterclockwise and then read it normally as you would in English. Non-fiction or technical works where English quotations, code examples, or mathematical formulas would be common are not usually written vertically.
As true as that may be, it's silly to be doing that with a lot of text on a website, which is primarily viewed on a computer monitor (or phone, but most people have auto-rotation on their phones enabled all the time anyway). When every colour name has the English underneath it, it seems like this would be a prime use case for left-to-right, not top-to-bottom.
Japanese readers don’t literally rotate the book unless it’s a long quotation. For names or a quote of several words you just get used to reading rotated text.
I’m not arguing this is a good design. The parent post said “... and unless Japanese people read roman letters differently than we do...”, so I was pointing out that, in this context, they do.
Tategaki (vertical script) is a traditional style and, anecdotally, younger Japanese people find it harder to read than yokogaki (horizontal script). Using tategaki on websites at all is unusual.
> Tategaki (vertical script) is a traditional style and, anecdotally, younger Japanese people find it harder to read than yokogaki (horizontal script)
Hello from Japan. Searching in Japanese gives no result and I fail to comprehend how is this even possible, given that almost everything is written in tategaki. Where did you get this info?
The aesthetics would be ruined by left to right writing. You don't make a website about "traditional" colors and purposely use a Mincho font and not also use the traditional orientation. The color name underneath isn't "English" either, it's Japanese written with Latin characters. The convention is to rotate left to right scripts when using vertical typesetting, complaining about it is like complaining about an American site spelling "colour" as color.
>You don't make a website about "traditional" colors and purposely use a Mincho font and not also use the traditional orientation.
Actually, you do. It would be an easier experience for the target audience (which I assume isn't primarily Japanese speakers, otherwise they wouldn't be using romaji for the names). I have nothing against aesthetic purity, but it gets in the way of content here. They are not inextricable.
>The convention is to rotate left to right scripts when using vertical typesetting, complaining about it is like complaining about an American site spelling "colour" as color.
The convention is fine; adherence to the convention in a medium to which the convention is clearly not suited is what I'm taking issue with. It may be fine for printed media - that doesn't make it fine, in this quantity, for the web. A better comparison would be a website for, say, an Arabic colour palette and the romanization is written backwards.
I have no trouble learning Japanese by reading books and manga with tategaki - but to say that it is the style best suited for anything trying to be "aesthetically Japanese", no matter the content or medium, is what I disagree with.
> which I assume isn't primarily Japanese speakers, otherwise they wouldn't be using romaji for the names
Romaji gets used for stylistic purposes for plenty of things that aren't meant for non-Japanese. Surely as a manga reader you would notice this?
I don't see what about a computer screen makes rotating text impossible. People don't literally rotate the thing they're reading when it's done in print either. Text rotated 90° gets used posters (which you can't rotate) too. I've even seen it done in America.
> Japanese people read roman letters differently than we do, letters rotated 90 degrees is awful for readability and should never be used
Unless Japanese people read words differently than we do, letters rotated to face the same direction as the words they compose is exactly the right way to do things; consistent but rotated word shapes are more recognizable than words composed of letters that are oriented at right angles to the way the word runs.
It's still an unresolved "problem" in "fluorescent" (or painted) business signs in the US --Crate & Barrel, for example. You will see this is most business (Main) Streets.
Some rotate 90, maybe most don't, but there are those that do. Also, some "go (read) up" some "go down".
It's interesting that a lot of color names listed here are pointing to actual things. Just like in english we'd have the "orange" color backed by the average color of oranges.
I also think they would only be used in phrases with the "xxxx color" cosntruct, and not as we'd say "blue" or "green". For instance azuki is a popular grain, so you'd say a shirt is "azuki color" and not just "azuki" IMO. That makes is a bit of cheating, as then anything can be set as a color, for instance "UPS color" or "post box color" or "Xbox death ring color".
But I guess our very history with naming colors is very ad hoc anyway.
How is this cheating? When you go for a palette in English, you pick from colors such as salmon, coral, wheat, eggplant, eggshell, etc.
Meanwhile, the construct “xxx color” in Japanese is used for very ordinary colors like brown (tea color), gray (ash color), and yellow (uhhh… yellow color).
What I believe gp means by cheating is that a color name is not a new construct, but just the name of a colored object with 'color; appended. Whereas in English a color name becomes distinct from the object.
'Orange' is a color name, inspired by the fruit but now an independent name; the Japanese cheating way would be calling it 'orange color,' referring to the fruit - in this case orange is not a color name. In 2000 years even if oranges are extinct or evolved out of existence we'll still be using 'orange' as a color, but the Japanese will no longer call it 'orange color.'
> What I believe gp means by cheating is that a color name is not a new construct, but just the name of a colored object with 'color; appended. Whereas in English a color name becomes distinct from the object.
The claim that the Japanese terms are somehow “not independent” from the objects and would disappear if the objects disappeared smacks of armchair linguistics to me. I could equally claim that women have fewer teeth than men do, as Aristotle did. Apparently he never bothered to count them.
If you keep repeating the idea that the Japanese language is somehow “cheating” whereas the English version is fundamentally different and inventing whole new words that are now divorced from their original meaning, I would like to see at least a small amount of reasoning or research to support it. The claims just don’t make sense at face value.
There has been a ton of research into the linguistics of color names.
And… if you are just trying to explain gp’s position rather than argue for it yourself, please don’t do that.
To clarify a bit, I didn't mean cheating in a much negative way.
Using physical goods as reference to color is the basic evolution for everyone. As I understand it "blue" in english for instance was initially pointing to a concrete thing (lead?) as well, but the words went through many transitions isolating the world for the color from its original counterpart.
But I think in english the concepts still stay more separated, for instance "rose" as a color has parted a lot from the common perception of an actual rose's color. Or looking at Faber Castel's color chart for instance [1], pink declinations are all defined relative to "pink", in contrast on the article's chart we'll have declinations like Ume (plum), Sakura (cherry), Ichigo (strawberry).
> But I think in english the concepts still stay more separated, …
My question is, why do you think that? It sounds like an unfounded assertion to me. I hear you repeating claims but color names have been extensively studied in linguistics, and you have direct access to Japanese dictionaries and encyclopedias online, and you can type Japanese color names into Google image search, so it should be relatively easy to come up with even a small bit of evidence.
> …for instance "rose" as a color has parted a lot from the common perception of an actual rose's color.
If you look up “rose color” in Japanese Wikipedia, you’ll see that it specifically defines it as a shade between red and magenta, even though roses come in many different shades (yellow, orange, white, cream, etc).
The problem here is that I am not a Japanese speaker, but this seems to be evidence that the term “rose color” is a specific color, and not a term that means “the color of a rose”.
Unlike English, this color is not the color of a peach fruit. It is also not the color of a peach blossom, despite the fact that “peach-blossom color” is listed as a synonym. Instead, it is the color that fabric takes when dyed with peach blossoms, but the practice of dyeing cloth with peach blossoms has fallen out of usage. You will find a different set of colors if you do a Google image search for 桃花 (peach blossom) and 桃花色 (peach blossom color). (You may find some sexual material in your image search, so maybe don’t do the search in the office.)
So the term “peach color” in Japanese refers to a specific shade of pink which is divorced from its meaning.
When I was a young'n hacking up little websites and personal projects 10 years ago I would always refer to nipponcolors.com for color ideas. Glad to see it's still online
In addition to those, paint manufacturers have a name for every shade in their range, which might vary from manufacturer to manufacturer. Magnolia is well-known, but there are literally hundreds of others. We once painted a wall bon voyage. It was a light blue.
I don't think one can draw a rigorous distinction between "well known and well-defined colour term" and "marketing department's name for a shade in the company range", but some are clearly at the "colour term" end of that continuum, and many are clearly at the marketing end.
I could have included "cherry blossom" in the list of English colour terms (In Japanese: Sakura) but a quick google shown that it refers to a range of related shades in English.
I think the many comments griping about this site's layout/readability/accessibility are missing the fact that it was made in 2010 - a time when HTML5 was relatively new, many modern web features didn't exist yet, and Flash was very much the norm for artsy dynamic sites like this.
As such, judge it for what it is - an art site trying to push the bounds of HTML at the time. It clearly wasn't made to maximize readability, and folks complaining about the color names not being searchable are hugely missing the point.
It's the same concept, but one color for every day of the year. It's a lot more browsable/readable, though it's not searchable at all and has no romaji.
I've always found it interesting that the Japanese use the character 赤 generically for the colour red but Chinese speakers use the character 紅. When they both refer specifically to different shades and tones of red and speakers of both languages would know the specific meaning of both characters.
A lot of them are named after tea. Is or was tea really that influential in Japan, or is there just a strong correlation between tea and colors? I have been to a tea market just once and I can only guess that color would be a quick way to tell them apart since they all look similar.
Japanese text is commonly written either left-to-right or top-to-bottom. You'll see top-to-bottom most often in newsprint or manga.
After clicking on the little 'person' icon in the lower right of the control you're taken to the author's personal site I assume. The site seems to indicate that they're either located in Japan or natively Japanese.
In this case the Japanese text is correctly oriented, however the English text is rotated 90 degrees as a result. Since they author is probably aiming for a Japanese audience this doesn't come off as odd.
