UI Design for Developers

Wanna know basics of UI Design. I basically freeze thinking about how my website should look and I am not skilled enough to just pickup a random design and start coding that , would love if I could just get some recommendations on where to start? Sorry for the frog

>>31782 I designed my site to work with the widest array of browsers possible. Mothra, abaco, w3m, netsurf, etc. Some styling is applied for "modern" browsers but it's minimal and really only exists to make dense text followed by dense pre elements easier to parse. Some responsive css too but it's super tiny.

>>32274 How difficult was ensuring compatibility with that many browsers? The only one I've heard of that you listed was netsurf and I think I've heard somebody mention w3m, but are you manually testing compatibility or using something else? I'd imagine with such a low userbase for most of those that there isn't just a different version of CanIUse.

>>31782 Emulate the look of 2000s websites. Here are some pointers: >https://developer.mozilla.org/en-US/docs/Learn/CSS/Building_blocks/The_box_model >https://makefrontendshitagain.party >http://catb.org/~esr/html-hell.html >http://bettermotherfuckingwebsite.com & https://thebestmotherfucking.website >>32275 >How difficult was ensuring compatibility with that many browsers? Don't use JS (or make sure your website works without JabbaShit). That's the number one most important thing. Also, avoid modern HTML5 and CSS3. You should manually test your site also with Lynx or Links (terminal browsers) and perhaps NetSurf to make sure everything works.

>>32275 >>32289 This anon is correct. No JS, minimal CSS nonsense, standards compliant HTML with no nonsense. I tested in the minimal browsers first, then wrote a minimal responsive stylesheet that is entirely optional and doesn't do much in the way of changing formatting aside from auto dark theme and improving readability on mobile. Manual testing is really simple if you're running a test server. You could probably even write a script or makefile rule to automatically open all the browsers. POSIX philosophy is: >write once, run everywhere Applying this idea to webshit, any website worth building is: >write once, read everywhere

>>32289 2000s websites look like dogshit, and the only reason you like them is because you grew up with them. >>32290 >POSIX philosophy is: >>write once, run everywhere No it's not.

I think POSIX philosophy is more like "write once, compile everywhere". >>32294 The new websites aren't better. For example, so-called "hamburger menus" suck, except for phones, and most websites should work without JavaScript.

My website is very simple, but there are tricks to ensure it works well in older WWW browsers when I do something fancy. As an example, the prime index of my website has a footer that uses CSS to stick to the bottom of the page, but I want someone using, say, Lynx to see it without scrolling to the bottom where it may be ignored unintentionally; I place the footer element at the top of the page to achieve this. Similarly, I use elements such as center, i, and b to style pages over CSS, because any WWW browser understands those elements, but not all understand CSS. Other little details are ensuring every fancy SVG has a PNG alternation and like things. For my programs, I often design the interfaces to be hard to misuse and, if feasible, base them around repeating a more primitive operation to create the entire display. The basic idea also extends to the more abstract meaning of interface. >>32290 The POSIX philosophy isn't followed, never worked, and isn't a philosophy. Although, pretending to be the originator of an idea is certainly in character for it. In brief, it's utter nonsense. Don't drink the soykaf.

>>32296 Objectively wrong. Le old web wasn't just pure html and starfield backgrounds like the modern day gayporwave meme version of the past. Back in the 2000s, most sites were designed around Internet Explorer, a proprietary browser owned by an evil company that brazenly and deliberately ignored standards. Lots of sites were also dependent on insecure Java applets and proprietary Flash, both of which were these massive behemoths that installed entire goddamn virtual machines on your computer just to display chat widgets in web pages. And then there's fuarrrking silverlight, which I'm not even going to touch. Don't get me wrong, newer websites suck too. But it's a better kind of suck.

>>32301 >a better kind of suck Given the practical choice between a completely inscrutable blob, and obfuscated code that runs inside a Free, auditable jail, yeah, I guess I'd take the latter, if that's what you mean. Arguably, therefore, the current plague is byte-for-byte less evil than the previous ones, but I do think it's more prevalent than them. I would say, and we probably just remember things differently, that a much greater proportion of sites now depend on javascript (as in their main content being unreachable without it) than ever did on those other technologies. Surely, this is in part encouraged by the technology being built into major browsers? But in fairness, it's probably not javascript's fault, mostly - the fact is that the web itself has changed, and most sites today are built not to share information, but exploit their visitors for profit.

The Design of Everyday Things http://libgen.rs/book/index.php?md5=9E4AF5615DB9CD4C1C8D0579B98EE549 Steve Krug - Don't Make Me Think http://libgen.rs/book/index.php?md5=A35D2795743E57BB795ADB7F14BE991E Refactoring UI http://libgen.rs/book/index.php?md5=C0008A5B7285DD078C6AFC3F5881B06D https://m.youtube.com/channel/UCVyRiMvfUNMA1UPlDPzG5Ow https://uxdesign.cc/ https://www.nngroup.com/topic/study-guide/

>>32309 >if that's what you mean. You hear me loud and clear. The Web is like kicking dead whales down the beach balls right now, but it could've and indeed did suck way more than it does.