3
10Dev
54d

Do any of you guys who do web development actually use semantic HTML or is it basically the modern xHTML?

I, for one, don't. Old habits die hard

Comments
  • 4
    if you ask me semantics ftw and xhtml should die in horrible death
  • 0
    div, span, ul, ol, li, a, img, h1, h2, form, label, input, button.

    i literally can't remember when's the last time i needed to use any other tag, so i literally don't even properly know what you're asking.

    maybe if a technology wants me to be interested in how to use it properly and what its philosophy is, that technology should first try not to be utter shit.
  • 1
    Yep, I use all of them including the more obscure ones like hgroup, figure and dt/dd/dl. HTML isn't hard at all.
  • 1
    @Midnight-shcode html (the tag), head, body, meta, title, link, script, header, main, footer, nav, p, dl, dt, dd, strong, em, i, b (i and b are not deprecated), abbr, pre, picture, source, img, svg, noscript... just to name a few others that I'm using.

    Also, there is no "semantic HTML". It's just "using HTML properly", as in "like it has always meant to be".
  • 1
    @Fast-Nop
    "html (the tag), head, body, meta, title, link, script"

    yeah, of course, but I don't count those since I use them so rarely that I always have to look up how they're written properly (the ones with attributes, that is)
  • 0
    @Fast-Nop
    also, I don't believe any layer of webtech has ever been used "like it's always been meant to be used" for longer than about... 2 days since its introduction. mostly because if it were used in that way, we would still be stuck in how web looked in 1990, since nobody working on that tech seems to have any ability to anticipate even a year into the future.

    otherwise the web ecosystem would look very differently.
  • 0
    Yeah, I do. I stopped using XHTML since HTML5 became a standard.
  • 1
    Nobody cares about XHTML. ^^
  • 0
    Semantic HTML is better though, My motto is to use less divs and spans
  • 1
    @Midnight-shcode We wouldn't be stuck at all. HTML has always been meant for semantic document structuring. It's just that most web devs don't understand that because they can't think further than what they see on the screen, and that's because they have never written any professional document (e.g. academic paper, thesis whatever) in their whole life.
  • 2
    Any webdev should have this bookmarked:

    https://developer.mozilla.org/en-US...

    Spend some time learning this and you can avoid spamming divs and spans everywhere, even outside the context of a straight written document.
  • 0
    @Fast-Nop
    1. academics design something that makes sense to them, related to revolutionary groundbreaking technology which, if one cared to stop and think for a second, one would realize it's gonna get ubiquitous sooner or later.

    2. tech gets ubiquitous.

    3. non-academics try to use it in practice, as opposed to academic environment.

    4. turns out it makes little sense for non-academics and needs to be abused to be useful in any other than academic environment.

    5. widespread populace is blamed, instead of academic intelligencia who should have known better in the first place.

    6. nope, I don't think I accept your argument.

    (7. but that just might be due to my current contract very effectively reminding me just HOW MUCH do I hate almost everything about the web stack)
  • 2
    @Midnight-shcode Let's take popular example where many web devs screw up: heading levels.

    Wrong usage: abusing them for the font size of their default appearance. That also ruins accessibility because you can't navigate by heading levels - which blind people do with their screen readers.

    Correct usage: using heading levels for structuring the document into chapters and subchapters, and styling the heading appearance with CSS.

    Why do people screw up? Because they don't know how to write a professional document, which in turn is why they don't understand HTML.
  • 0
    @Fast-Nop
    "Why do people screw up? Because they don't know how to write a professional document, which in turn is why they don't understand HTML."

    which might be caused by the fact that most elements have default visual style which requires one to actively write 5-15 lines of CSS to get rid of.

    if the markup was supposed to be structural, either don't give it aesthetic implications, or give me a single CSS line to get rid of them.

    body {
    no-default-styles: true;
    }

    ...actually, on second thought, that should have been the default since HTML+CSS inception. the default styles should have been an opt-in:

    body {
    default-styles: enable;
    }
  • 1
    @Midnight-shcode CSS is meant to style the appearance. Complaining that you need it for its very purpose doesn't make any sense.

    If a web dev is already overwhelmed by writing 5-15 lines of CSS, maybe he should reconsider his career choice - instead of writing broken websites.
  • 0
    @Fast-Nop
    ok, i'll try again and simpler:

    if css is meant to style the appearance and html is meant to express semantic structure, then html shouldn't also style the appearance.
  • 1
    @Midnight-shcode Disagree. Three reasons:
    1) There has to be some default styling because making everything invisible is not an option.
    2) If CSS fails to load for whatever reason, the default styling is required as fallback.
    3) HTML doesn't even specify to style the default appearance in the first place, it's the browsers that do this.

    And also: incompetent and/or lazy web devs are not a valid argument.
  • 1
    @Fast-Nop
    2.
    body {
    default-styles: disable;
    }

    when css loads, the defaulta don't interfere with anything because they're disabled. when it fails to load, it won't process the directive, so default styles will be applied.

    your arguments are probably reasonable, i'm just working on a project that freshly reminds me how much i hate the whole webstack, from its philosophy, through the implementation, up to all the layers of post-hoc patchvork upon patchwork to make it halfway kindasorta almost capable of doing what people need to do with it.

    and since i'm fresh back into the hate mode, i don't think there's any arguments, however (theoretically) reasonable that would convince me that webstack isn't just collection of botched piles of shit.
  • 0
    @Midnight-shcode I think the lowest layers are somewhat OK-ish, but where it really gets fucked up, that's with CSS frameworks like Bootstrap, Tailwind, Bulma, or any other. These stem from a gross misunderstanding what CSS is even about, so of course they suck.

    And yeah, your CSS should have a CSS reset, but that's just a little copy/paste boilerplate, basically.
  • 2
    @Midnight-shcode It kinda naturally is considering the origins.

    Tim Berners-Lee said in the mid 90's "hey here's how you can create markup for a document with forms and everything, and you can even transmit them through the internet and not rely on the postal service!"

    Fast forward 2020 and we're using html to create camera apps on a web page.

    Brendan Eich said in the the mid 90's: "Heres a quick weakly typed dynamic language I created for Netscape to compete with Java and its static typing system. We only really need a simple "script" language to create simple interaction on the web"

    Fast forward 2020 and we are plotting graphs, playing games and doing cryptography on the internet.

    I would say they are at least trying to keep up with how we expect the Internet to service our needs.
  • 0
    @theuser exactly my point.
    and yes, trying, instead of doing the right thing which would be scrapping all of it and making tools designed for the purposes we want to use them for.
  • 0
    @Midnight-shcode Every dev wants to scrap everything all the time, mistakenly thinking that redoing everything from ground up would lead to better results.

    That's why devs need managers to keep them from doing that.
  • 0
    @Fast-Nop
    "mistakenly thinking that redoing everything from ground up would lead to better results."

    sometimes they are right, though.
    and I think webstack is one of those cases.

    you could liken it to the invention of wheel at time when nobody had any idea you could use horses to pull stuff, which for the reasons of backwards compatibility migrates across centuries into being used as wings for planes.

    i think at a point like that, scrapping the lift-generating-wheels altogether and designing something with the purpose of lift-generation in mind from the start, is warranted.

    and by definition, it's almost impossible to not improve on the tech in a situation like that, regardless of however much you botch the new design.
Add Comment