CUTCODEDOWN
Minimalist Semantic Markup

Welcome Guest
Please Login or Register

If you have registered but not recieved your activation e-mail in a reasonable amount of time, or have issues with using the registration form, please use our Contact Form for assistance. Include both your username and the e-mail you tried to register with.

Poll

Has HTML 5 been bad enough to once again create a fork?

It sucks, of course it's time
It worked for the WhatWG!
Things are bad, but not that bad.
There's something wrong with HTML 5?

Author Topic: Loss of confidence in the W3C... is it time for a new HTML?  (Read 2571 times)

John_Betong

  • Full Member
  • ***
  • Posts: 218
  • Karma: +24/-1
    • The Fastest Joke Site On The Web
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #15 on: 5 Nov 2019, 10:52:12 am »
  • syntax more lenient than HTML_STRICT

Unambiguous communication of any kind, especially with computers, REQUIRES precision defined by HIGH standards.


True and rather than prevent script from running the browsers are remarkably considerate and render atrocious HTML and CSS scripts.


W3.org provides excellent validation tools and also supply validated script CSS which can be copied and pasted, HTML validated script is also available and highlights errors and warnings making it easy to rectify the problematic script.


Nothing better than a computer to fix computer errors :)
Retired in the City of Angels where the weather suits my clothes

Jason Knight

  • Administrator
  • Hero Member
  • *****
  • Posts: 1049
  • Karma: +188/-1
    • CutCodeDown -- Minimalist Semantic Markup
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #16 on: 5 Nov 2019, 10:58:51 am »
Running this forum site through the CSS w3.org validation service not only highlights all errors and warnings it even shows validated source code which can be simply copied and pasted.
Which then has to be sliced up into every single .template.php file, hoping that it doesnt' break the decade and a half old CSS (which it will), or the scripting for nothing (which it also will)... much less the hundreds of different pages that would have to be run past it resulting in making even more work....

Than starting over from scratch and doing it right from the start.

Seriously, download SMF and have a look at Themes/default/BoardIndex.template.php, and tell me how useful that "valid" code download is for helping with skinning issues.

Hence why this template is a placeholder until I have the time to do it right; or someone I trust is willing to step up and help... though I need the common.template.php done first before I'd even think about bringing someone else aboard, just so they can see the methodology.

When it's vomiting up 94k of markup to deliver 32k of plaintext on a forums -- likely not even 48k's flipping job -- the validation service's opinion of code that was written to 4 tranny standards really means nothing. ESPECIALLY since HTML 5 validation pretty much means even LESS than 4 tranny did... scary as that is.

It doesn't check for good practices, just for a set of rules that have lost ALL meaning since they undid all of the progress 4 Strict gave us.

The tools USED to be excellent -- they still are in terms of how it works / what it is trying to do... but the ALLEGED specification that is HTML 5 has pissed on its usefulness from so on-high you'd think the almighty just got back from a kegger. When the "specification" is incompetent trash, any tools based on it will be the same.

Again, see 4 Tranny... which is basically what 5 is.

Before HTML 5 came along, I'd be in 100% agreement with you. Now? It can go take a flying leap. We're right back to where we were in 1997.
« Last Edit: 5 Nov 2019, 11:17:24 am by Jason Knight »
We are all, we are all, we are all FRIENDS! For today we're all brothers, tonight we're all friends. Our moment of peace in a war that never ends.

Jason Knight

  • Administrator
  • Hero Member
  • *****
  • Posts: 1049
  • Karma: +188/-1
    • CutCodeDown -- Minimalist Semantic Markup
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #17 on: 5 Nov 2019, 11:08:53 am »
Actually...
I find that HTML5 is easier to write web pages and has some good tools for checking.

Do you have an example of something you've written in it? What I've seen so far of your HTML and CSS -- with your little URI checker -- doesn't inspire confidence in your statements. See how you got duped into the "presentational classes" lie where instead of repeating yourself in the CSS where it can be cached, you repeated yourself in the markup where it is likely NOT to be cached, and missed sub-page caching opportunities.

I'd like to see something you've done "correctly" with HTML 5 in your opinion. I have the feeling your over-reliance on tools -- like the validation services -- may have led you to utterly incorrect false conclusions.

Warning, if you do share, I'd likely do one of my 20 minute rewrites to illustrate the point.

I just get the feeling you and I have different definitions of the word "easy".
We are all, we are all, we are all FRIENDS! For today we're all brothers, tonight we're all friends. Our moment of peace in a war that never ends.

gleepower

  • Junior Member
  • *
  • Posts: 22
  • Karma: +1/-0
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #18 on: 12 Nov 2019, 03:45:59 pm »
One the one hand there is alot to be improved in the HTML standard

1. The complexity

I feel the HTML standard is very complex at this stage. There are lots of ways of doing the same thing which makes it harder for browser writers and more error prone and also harder for website writers to consume.

2. The inaccessibility of the standard

The HTML standard is absolutely impenetrable if you are not very technical. I think it's very important for standards to not only be technically correct (this is not a sufficient condition for a good standard, but it is a necessary one) but to be clear and easy to understand. This makes it more democratic as we can have more implementations. I feel at this stage it's realistically impossible for one guy to write an HTML + CSS user agent at this stage.

3. The verbosity of the syntax

The xml syntax is very wordy. This is in stark contrast to formats like markdown which are more pleasant to write in and I feel easier to understand. Even more importantly, I think markdown emphesises the document/semantic nature of markup rather than thinking of elements to be chucked around by javascript.

Given all that though, I'm not sure how to cleanly get HTML attributes into a syntax as clean as markdowns. In latex and markdown, a newline is just assumed to start a paragraph. How would you attach a class to that?




However, given all that. HTML just has too much traction and inertia around it to make a move from it. Anyone can make a markup better than HTML, but making one good enough to make up for the vast amount of manhours to go into porting existing websites, rewriting user agents, education, existing tooling. Backwards compatibility isn't pleasant, but it's the reality we live in.


Perhaps an markdown-esque lighter HTML with a strict subset of the tags which could still have existing CSS applied to it? Or a just a validator which defines a much stricter subset of HTML5. This would get you around backwards compatibility.
The beaucracy is expanding, to meet the needs of the expanding beaucracy

Jason Knight

  • Administrator
  • Hero Member
  • *****
  • Posts: 1049
  • Karma: +188/-1
    • CutCodeDown -- Minimalist Semantic Markup
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #19 on: 12 Nov 2019, 07:12:02 pm »
The xml syntax is very wordy.
SGML, not XML.... at least if talking proper HTML and not XHTML. Kissing cousins, but not the same thing.

... and you say that like it's a bad thing. It's hardly "wordy" when the words describe what things ARE, or SHOULD BE in a professionally written document... or even any document written by anyone older than the age of 12 since this was late elementary to middle-school English back in the late '70's / early '80's; Which I can only assume is now doctorate level classwork at college given the state of education right now.

This is in stark contrast to formats like markdown which are more pleasant to write in and I feel easier to understand.
See, this is what I wanted to hear from people, even though I heartily disagree. For me, markdown is painfully aggravatingly cryptic -- Aaron and I had it out no-end over his little "creation" where all those easily forgotten use of characters made it painfully aggravating to work with, particularly with any content where you might need/want to nest meanings.

But that shows how differently we all think about things. I have the exact opposite reaction to markdown that you do. I find it aggravatingly cryptic and impossible to remember.

Even more importantly, I think markdown emphesises the document/semantic nature of markup
How is that so?  I see little to no semantics in it, since it appears to be almost entirely about writing content as presentation, choosing the cryptic symbols ENTIRELY because of how you want things to look -- the polar opposite of semantic markup!

The whole point of HTML is to say what things are, or would be in a professionally written document for grammatical reasons. Markdown seems to exist to create visual style and jack-else.

I'm wondering how you came to that conclusion about markdown... possibly because you've never seen/been told how HTML is supposed to work in terms of leveraging semantics?

In latex and markdown, a newline is just assumed to start a paragraph.
Actually they're not assumed to mean anything, and that's what's wrong with them. The only effect they have is presentational. You use newlines for headings and lists, does that also make them GRAMMATICAL paragraphs? Of course not. That's why from the point of view of what HTML is for, LaTex and Markdown are rinky crippled toys that have more in common with the abortive train wreck disaster that was HTML 3.2, than any proper semantic markup practices.

For all intents and purposes what they do is more akin to the deprecated <center> or <font> tags, and attributes like border or bgcolor than any proper 21st century HTML.

Of course with 90%+ of the people out there still vomiting up HTML 3.2 and masquerading it as HTML 4 Tranny or slapping 5 lip-service around it, one can hardly be surprised the notion and good practice of "separation of presentation from content" goes unused by people not qualified to write a single blasted line of HTML. Aka those who use front-end frameworks, those who use classes to say what they want things to look like, or those who choose their HTML tags based on what they want things to look like. ALL broken choices based on false assumptions.

Which is how we get people who make websites that start with H5 and have anchors just thrown into NAV any-old-way in complete ignorance of what HTML is, what it is for, or how to use it. (again, see bootcrap developers)

I think that's what frustrates me the most about all this junk. HTML is easy, people make it hard seemingly on purpose and why? The best answer we can get is "I don't want to change" or "wah wah, eye dunz wunna lurns". NOT exactly inspiring.

It's like the LAZY "I don't want to have to rewrite" -- fine, stick with your outdated crap. That's no excuse for NEW websites and why the people who spent most of the past two decades vomiting up "transitional" markup are so damned annoying. Just like the people who use frameworks to slide-around the validation whilst still using the same broken bloated ignorant methodologies that were the core of HTML 3.2 and 4 tranny, undoing 20+ years of progress! Now they get to slap HTML 5's doctype atop the same incompetent trash and back-slap each-other over how "modern" they are?

As a great man once said whilst defending the Ardennes: NUTS!
« Last Edit: 13 Nov 2019, 01:35:29 am by Jason Knight »
We are all, we are all, we are all FRIENDS! For today we're all brothers, tonight we're all friends. Our moment of peace in a war that never ends.

gleepower

  • Junior Member
  • *
  • Posts: 22
  • Karma: +1/-0
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #20 on: 13 Nov 2019, 04:02:11 pm »
I definitely feel HTML could be less wordy. Maybe removing the requirement to close off tags and using whitespace instead? HTML is also a massive pain to parse, since SGML is actually really darn difficult if I remember correctly. Again I'm not sure exactly how this would look but it has to still keep the semantic elements of HTML intact. I really think it would be great if non technical people such as lawyers could write documents in HTML, but I feel the syntax scares them off. Less setup cruft would be nice as well -> I think html 5 nailed making the head smaller.

Maybe some thing like:

[paragraph] (
blah blah bla hcontent
   [heading1] ( this is a heading)
   [object] src:blsdfsdfsdf ()
)

The need for a pointless closing tag annoys me.
« Last Edit: 13 Nov 2019, 04:06:06 pm by gleepower »
The beaucracy is expanding, to meet the needs of the expanding beaucracy

Jason Knight

  • Administrator
  • Hero Member
  • *****
  • Posts: 1049
  • Karma: +188/-1
    • CutCodeDown -- Minimalist Semantic Markup
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #21 on: 13 Nov 2019, 05:46:32 pm »
I definitely feel HTML could be less wordy. Maybe removing the requirement to close off tags and using whitespace instead?
Relying on whitespace is always a disaster. Having one misplaced tab or space completely borking your logic is just begging for things to go wrong.

No matter what the python fans say. I should like Python, but to me it's probably one of the hardest languages to work with alongside Ruby... and that's coming from someone who can hand assemble RCA 1802 machine language and has code floating through space.

But to keep it fair, I think that C syntax is aggravatingly cryptic because is uses {} instead of begin/end. I like seeing words that say what things are or do... probably because the first thing I "really" learned after assembly was Pascal.. .and then Modula... because few if any people working with microcomputers in the late '70's and most of the '80's gave a flying purple fish about C. hell, C was the language we made fun of the "big iron dinosaurs" for still using.

In that way closing tags are a blessing because they say WHAT is being closed. It's also why I add even more words when closing in the form of comments before DIV to say the same.

Code: [Select]
<div id="content">
  Your content here
<!-- #content --></div>

Is simple and plainly obvious as to what's going on. Certainly far more understandable and clear than what you presented:

Code: [Select]
[paragraph] (
blah blah bla hcontent
   [heading1] ( this is a heading)
   [object] src:blsdfsdfsdf ()
)

Where you have symbols in more complex code that could mean damned near anything... much less what the blazes makes a depth one heading (site title) and an object part of a grammatical paragraph? :D

Admittedly I see that all the time in code, paragraphs around single images, paragraphs around tables or lists, paragraphs around headings, paragraphs around sentence fragments.  All basically saying the developers should have failed 5th grade English class for not having learned what a paragraph is.

HTML is also a massive pain to parse, since SGML is actually really darn difficult if I remember correctly.
Compared to most formats not really so. You look for the <, see if it's followed by a / or not, extract everything until the next >, remove any spaces around = just in case someone goes full Pakled with how they write HTML, split by the remaining spaces, done.  Maybe check against the list of "empty tags" to auto-close those.

It's not a rocket science format. Parsing SGML is the easy part. It would be even more so if we stopped with the "do as you will" nonsense of not stopping dead in your tracks on error, something that only existed in HTML as a result of it being sent over telnet at the time and dropped characters being a reality of darpanet communication at the time.

I think html 5 nailed making the head smaller.
One of the few things in HTML 5 I praise. Honestly, I think it's about as small as it needs to be, it's just people keep derping in endless pointless META now that not one legitimate UA makes use of, or isn't redundant to other meta. See that whole "opengraph" BS where the only useful property for 99.99% of websites is og:image.

The need for a pointless closing tag annoys me.
... and yet you have one, a painfully cryptic and aggravatingly meaningless close parenthesis. Something where any competent developer would likely add a comment after it to say what's being closed anyways, so why not force the closure for clarity?

Though I swear, the majority of code out there would be simpler to digest if people would stop slopping endless attributes and other such junk into a single line and instead obeyed the "rule of 76". HTML is bad for that, CSS is even worse. In that way I often thing semi-colons should be removed from the CSS spec and instead INSIST on linefeeds being the divisor between properties.
« Last Edit: 13 Nov 2019, 09:41:08 pm by Jason Knight »
We are all, we are all, we are all FRIENDS! For today we're all brothers, tonight we're all friends. Our moment of peace in a war that never ends.

fgm

  • Jr. Member
  • **
  • Posts: 60
  • Karma: +5/-0
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #22 on: 14 Nov 2019, 05:27:57 pm »
Tags like header, nav, main, section, aside and footer have no place in HTML since I understand it as a tool to describe the content, not a powerpoint presentation. A header or footer is part of the presentation structure unlike <h1>...<h6>, <p>, <ul>,<li>,<em> that define what the content is.

As coothead said, I'm also an amateur, but I get the point on what Jason Knight defends, and it makes lots of sense. Absence of sanity in website design is what made me abandon years ago, and CutCodeDown motivated me to continue developing websites.

mrMarcup

  • Junior Member
  • *
  • Posts: 5
  • Karma: +1/-0
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #23 on: 23 Feb 2020, 06:09:56 pm »
Sorry to necro this thread, but I wanted to inquire about how everyone feels about the new HTML 5 input types: https://www.w3schools.com/tags/att_input_type.asp

Jason Knight

  • Administrator
  • Hero Member
  • *****
  • Posts: 1049
  • Karma: +188/-1
    • CutCodeDown -- Minimalist Semantic Markup
Re: Loss of confidence in the W3C... is it time for a new HTML?
« Reply #24 on: 23 Feb 2020, 07:11:16 pm »
Sorry to necro this thread,
If you're adding to the conversation, no need to apologize.

but I wanted to inquire about how everyone feels about the new HTML 5 input types:

As much as I HATE the new "structural tags" considering them to be pointless idiotic redundancies, there ARE good things in HTML 5.

In the case of the new type="" for INPUT those are a "well it's about {expletive omitted} time!"

This is functionality that should have been present in HTML 20 years ago, and basically lets us throw a lot of pointless bloated garbage JavaScript in the trash. I'm always down for having to write less code because functionality was added to the specification.

I think that's what I hate about the structural tags and halfwitted trash like ARIA. Tags like SECTION, NAV, ARTICLE, HEADER, FOOTER, and so forth being a pointless redundancy and seeming to exist JUST to sate the wants and desires of the folks who went from endless pointless tables to endless pointless DIV. Aria roles seeming to be for the people who want everything to be a DIV and refuse to learn how to use the existing tags. Both  just mean writing more code for ZERO legitimate improvement in functionality.

Whilst things like the removal of type="" on SCRIPT, STYLE, and "LINK rel="stylesheet", shorter DTD, the smaller charset META, and the new INPUT type are all a blessing and more than welcome if not overdue because they mean less work and less code for the same or even better functionality.

.. and there are new tags that make sense.

TIME for example is another "way overdue" tag since it does serve a purpose, and can even be leveraged with scripting assistance to ajust times locally without having the user have to set their timezone with user accounts. I was actually going to post a snippet in the appropriate forum section about doing that soon.

RUBY, RT, RP, and BDI add much needed non-Latin descended language support. Whilst I don't use them they fix what the far east has been hacking around for decades. HTML was never supposed to be a "English only club".

DETAILS and SUMMARY add functionality used on websites all the time of showing/hiding sections of content -- now we have tags that make semantic sense and provide that! Actual functionality! -- part of what I hate about the "structural" tags is that not one legitimate UA treats any of them (apart from perhaps MAIN) as anything more than a DIV.

PICTURE to allow multiple SOURCE is a great addition as well, particularly when used for newer image formats that older browsers might not know. Since unknown tags are supposed to be treated as inline-level and these are content-less... but much like AUDIO and VIDEO I wish that all three tags were never added and they just used OBJECT instead since that was SUPPOSED to be the intent of OBJECT.

Hence why the original plan before the WhatWG came along was for even the IMG tag to go away for being redundant to object.

But the functionality is real nice. The ability to go:

Code: [Select]
<picture>
  <source src="images/test.webp"  type="image/webp">
  <img src="images/test.png" alt="A test image">
</picture>

Is freaking awesome given how in supported browsers webp can save you 20-50% in file sizes compared to alpha transparent PNG, and around 10% over non-transparent JPEG. Well worth the extra markup.

Laugh is you can do that in your CSS too. Just wrap the webp declarations in

@supports (background-image:url("image.webp")) {}

you don't even need the image to be valid, it doesn't actually request the file.

Fun stuff.

So I'm not saying it's all bad, but it could still be better by just removing the pointless redundancies.
We are all, we are all, we are all FRIENDS! For today we're all brothers, tonight we're all friends. Our moment of peace in a war that never ends.

 

SMF spam blocked by CleanTalk

Advertisement