231–240 of 249 posts
Earlier quoted context omitted.
The font size can change due to zoom, which is a must, because 16px is too small text and unreadable.
cmd+"+" has always worked regardless of what unit is used, though?
Earlier quoted context omitted.
GNU docs have good design in accordance with W3C best practices: rendering is user agent's job, and if it doesn't do the job per your preference, you change the user agent to the one that does or submit an RFE to your user agent vendor to support your preference.
> GNU docs have good design Who's side are you on here, stop proving fctorial's point :) > rendering is user agent's job GNU uses more CSS than the joke website in question for an uglier result; they're putting in more effort to make something that's less readable. I don't see how anyone can call that good design. It would take literally 2 CSS declarations to improve their website's design for >90% of the people look…
I'm on my side and against people who push their preferences on me.
>It would take literally 2 CSS declarations to improve their website's design for >90% of the people looking at their page, at no cost or inconvenience to any other user. Good user-interface design incorporates sensible defaults.
The sensible default is to follow user preferences. Those 90% people do it this way and GNU docs rightfully don't question it. User preferences can't be reasonably implemented on the server side, because different people have different preferences, so it's undecidable whose preferences to implement.
>I don't see any W3C best practices anywhere that argue column widths should be avoided
W3C best practice is to let user agent render text as appropriate.
http://motherfuckingwebsite.com/ touches on this too:
>You dumbass. You thought you needed media queries to be responsive, but no. Responsive means that it responds to whatever motherfucking screensize it's viewed on. This site doesn't care if you're on an iMac or a motherfucking Tamagotchi.
W3C often refers to unusual user agents to illustrate their point of client side rendering.
From " http://bettermotherfuckingwebsite.com/ " which is one up the chain: > A little less contrast WHY. WHY DO PEOPLE DO THIS. To give this person credit they have at least chosen some sensible values, rather than gone with something that was just about enough on an expensive graphic designer high-contrast monitor and then not tested it on anything else. Leaving the rest of us squinting to try and read text or manua…
https://contrastrebellion.com/ More high-contrast options exist then Black foreground on a White background. I'm not disagreeing with your point, I loathe grey on grey sites. Being visually impaired nothing makes me rage-quit your site quicker if my DarkReader extension can't unfuck a site. I'm posting https://contrastrebellion.com/ so all the replies and followup conversations can see it too.
Earlier quoted context omitted.
On a good monitor, the color contrast between #ffffff and #000000 is far beyond the contrast possible in nature. No reflective material can reach those colors or that contrast. That color contrast is perfect for a game, a movie or an image. But you don't want the feeling of something as bright as looking into the fucking sun mixed with the darkness of the deepest starless night. That's just painful, both in light mod…
> On a good monitor, the color contrast between #ffffff and #000000 is far beyond the contrast possible in nature. No, it is not. > No reflective material can reach those colors or that contrast. Even if that was true, nature includes light sources as well as reflective materials.
Emissive light sources in nature (except for rare occurrences like lightning, certain jellyfish and cherenkov radiation) are all black body radiation.
That's very different from the colors and contrast ratios possible by modern OLED monitors. The best example would be the blue LEDs used in modern OLED panels, which when they were introduced had a strange unnerving blue glow because that color just doesn't exist in nature.
Additionally, the contrast ratios possible by modern OLED panels are also far beyond anything in nature, as in nature anything that is bright, be it reflective or emmissive, will as result also light up its surroundings and reduce contrast naturally.
Earlier quoted context omitted.
This one has a dark theme though.
I have stopped waiting for website authors to figure out how to make a dark theme. I just have my user agent do it with https://darkreader.org/ Like ad blockers, sometimes you have to adjust it, but it usually Just Works. The main innovation from this is that my display has much improved contrast, because I can set the brightness for text on a black background, instead of setting the brightness for an entire screen o…
And for $4.99 Safari user. 30% paid to Apple and 70% for developer to recoup their $99/year cost.
Earlier quoted context omitted.
> GNU docs have good design Who's side are you on here, stop proving fctorial's point :) > rendering is user agent's job GNU uses more CSS than the joke website in question for an uglier result; they're putting in more effort to make something that's less readable. I don't see how anyone can call that good design. It would take literally 2 CSS declarations to improve their website's design for >90% of the people look…
>Who's side are you on here, stop proving fctorial's point :) I'm on my side and against people who push their preferences on me. >It would take literally 2 CSS declarations to improve their website's design for >90% of the people looking at their page, at no cost or inconvenience to any other user. Good user-interface design incorporates sensible defaults. The sensible default is to follow user preferences. Those 90…
> The sensible default is to follow user preferences.
> W3C best practice is to let user agent render text as appropriate.
But nothing about including CSS goes against this. CSS is clientside rendering, it has nothing to do with forcing preferences on anyone. Your browser has the ability to override provided styles and either modify or ignore CSS. If it doesn't then I suggest, in your words, "changing the user agent to the one that does, or submiting an RFE to your user agent vendor."
This is why we use CSS. So that we can separate style from content and lay out a page with nice defaults that can still be overridden by the user.
And not only is CSS customizable and overridable, because it's completely separate from content it is also optional. Any of your users can choose not to download your CSS file and they'll get served the exact same HTML content that they're getting now. So when I say that there's no cost or inconvenience for users who don't like your styles, I literally mean that. There is no cost to them because they don't have to download the file. They can disable all 3rd-party CSS in their user agent if they don't want it, and then they don't even need to waste time making an extra network request.
Imagine taking this attitude with other applications as well. Should we be upset that Emacs ships with default keybindings? Is that as an undecidable problem? When I install a MELPA package, should I get mad that the author is pushing their default config on me? There's no downside to having optional defaults that users can partially override or fully ignore at their own convenience.
> W3C best practice is to let user agent render text as appropriate.
If this is genuinely GNU's attitude, then why does this page include style resets?
You're heavily misreading the W3C's position here. Letting the user agent render text as appropriate does not contradict providing CSS, and I've never heard of the W3C advising authors to avoid CSS.
first: there is no such concept of "page" in HTML. a web page is a document without specified dimensions by the original design of the web, not a "page" in a sense like pages in press-related applications (pdf, word processors, A4). so if you force any width to a DOM box, it'll be either too big or too small for someone (except if you know your target audience uniformely watches fullscreen browser on a 1900×1200px display - or i don't know whatever is the most popular screen resolution nowadays), and you can not reliably force min/max width in px without knowing what are the font sizes in the box, which is often unknown because the font renderer does not disclose it to the DOM renderer, let alone to the css author: remember it must be the user's free choice to set preferred font size for paragraph/headline/emphasizes, because you neither can uniformize all display and viewport size, user's eyesight, nor is it sensible to design your site for all possible viewport when the layout can be more reliably computed by the presentation application when the user's preferences are available.
and secondly: can not users just resize the browser window to prevent long lines? it's one click-and-drag move.
my point is: content authors should only decide on semantic design, not details (not specific color or size, or texture) ; and the reader should fill the specifics and present the document to the user according to his preferences.
Earlier quoted context omitted.
>Who's side are you on here, stop proving fctorial's point :) I'm on my side and against people who push their preferences on me. >It would take literally 2 CSS declarations to improve their website's design for >90% of the people looking at their page, at no cost or inconvenience to any other user. Good user-interface design incorporates sensible defaults. The sensible default is to follow user preferences. Those 90…
> and against people who push their preferences on me. > The sensible default is to follow user preferences. > W3C best practice is to let user agent render text as appropriate. But nothing about including CSS goes against this. CSS is clientside rendering, it has nothing to do with forcing preferences on anyone. Your browser has the ability to override provided styles and either modify or ignore CSS. If it doesn't t…
>There's no downside to having optional defaults that users can partially override or fully ignore at their own convenience.
That's what GNU docs do, they use defaults: i.e. they follow user preferences.
Earlier quoted context omitted.
Very true, however better has chosen #444 whereas #333 seems to work a bit more consistently. One thing you should always consider during the requirements phase of a web application is whether your app is likely to be used projected on a conference room wall. Things like contrast and fonts make huge differences in such hostile settings, and available screen real estate is also considerably limited.
Great point yeah. I also think it should be easier to tell browsers to make text high contrast, design be damned. Fundamentally it's an accessibility issue.
When people are already getting short with each other before you even get to the actual reason for the meeting, you know it's going to be a long day.
But corporate wanted our fonts to match their other product. I don't know if it was my imagination or not, but it felt like that act broke a couple people on the team, myself included.
