Web Design Philosophy: Pure HTML versus CSS

Some Shit

If you insist on your own semantic for "center," for instance for some kind of topographical language in which "center" stands generically for "middle location" as in "Building 3," feel free to create one. But do not type documents employing such as HTML; they are not and you are objectively making a mistake doing so. You do not get to determine HTML tags' meanings.

This is good, but let me know when they allow me to do four-dimensional tables normally. It's ridiculous to have to use

>1, 2a, b...

(I can already do this using XML namespaces, display:list-item, and content:, but I want respective elements to be standardized.)

Frankly, I have deprecated the web and will only make new pages for gopher from now on. That means in plain US-ASCII.
Oldschool text BBS are fine too (ANSI or ASCII).

Attached: Amiga BBS - AmiExpress _X-Lxx9UsopV9k.webm (456x360, 2.1M)

Do your formats of choice support standardized markup? If not, then it is a bad choice that hurts operability, such as automated parsing. Please use a standardized format and don't require users to parse linebreaks and indentations on their own.

Demo:

" rel="nofollow" target="_blank">w3.org/1999/xhtml'>t { border: .1em solid black; padding: .25em; margin: .25em; }table > t { display: table-row; }table > t > t { display: table-cell; }table > t > t > t { display: list-item; }table > t > t > t > t:not(:last-of-type)::after { content: ', '; }" rel="nofollow" target="_blank">w3.org/1999/xhtml'> 12 12 12 12 12 12 12 12

Style tags should be dropped since HTML should be purely semantic. should be dropped and should not be used for design - only tabular data.

If everything is done properly then you can just use your own stylesheet when viewing websites and have it look the way *you* want it. Also works for non standard outputs like TTS because the schema is semantic.

Alas this will never happen again because muh js.

You can use whatever markup you want with gopher, so long as it's ASCII. You could even use HTML but it would be ugly as fuck, because nobody wants to read raw HTML.
With BBS, you don't even need anything special because you can just save messages to QWK format.
en.wikipedia.org/wiki/QWK_(file_format)

Attached: cottonwoodbbs10.jpg (567x426, 87.43K)

> should be dropped
That's stupid. It's a span with center as default css when css is enabled. It's pretty useful otherwise people will make a to make up for the loss which is computationally more unefficient.
I would even propose a and and . The left so you can have it inside the others
Now they make useless shit like which doesn't do anything else then but is "defined after purpose instead of style" which means NOTHING just wasted complexity and less conformity.

That said your page should look nearly the same with CSS disabled and fully working without JS. Install uMatrix and check it.

Also don't ever use button. Just use instead.

You get weird looks these days if you use optical media, and you want people to go back to some 80s networking technology?

I'm moving this way too. nroff is the shit for ASCII styling.
Nice webm.

So what? People are all over retro game consoles, even old computers to some extent. So it's not that far-fetched. Let's face it: the modern world is full of shit, and people miss the simpler times, when computing was actually fun.

This is functionally equivalent to supporting nothing. My question was in hope of a preexistent format, either equipped with a semantic like HTML of merely syntactical like XML, which would facilitate parsing and extraction. If the protocols you mention support any text document at all, then people will probably ignore or misuse formatting, sending whatever at all to users/user agents. This is not a solution. The solution would be a general-purpose semantic format like HTML, only without what ended up being the bane of HTML, namely accepting malformed or invalid inputs on readers' part. Syntactic and semantic validity (incl. schema reading) or screen of death.

>otherwise people will make a to make up for the loss which is computationally more unefficient

This is because 98 % of people are retarded monkeys who cannot conceive of a document being used away from and abstracted from visual representation, in which contexts directions are meaningless: speech generation, as the other user said, but also general automated structural analysis by programs.

Also,

just say "less efficient." It's more efficient.

unefficient < less efficient < inefficient

/2¢

I agree, but 99% are conformists who accept whatever is there (lest they become "luddites" and outcasts etc.). How would you operate gopher/bbs services which are actually relevant these days and not just some nostalgic hobbyist thing?

I'm not trying to praise the web, but gopher sucks. Why not use something better?

I really love these well-argumented statements.

Attached: All_users_of_Windows_are_suckers.mp4 (1280x720, 598.46K)

Literally what?

Any hierarchical structure is basically web x.0.

Do you propose a different balance between linking and inclusion, different point of split between filesystem hierarchy and in-document hierarchy, intrinsic integration of conditional or loop structures into markup in terms of generating the final document tree, integration with networking resources, better integration with user interface, or what?

Any other changes are just redressing of same old HTML and thousand other formats.

As by a semantic describing user decision flow, think, off the top of my head,

>

You are what's wrong with web development.

Woah! You're quite the pushy cat, aren't you? Let's review what you said earlier:
And now:
Clearly you want a one-size fits all, universal perfect solution that's enforced everywhere. Well that's not what I'm looking for at all. No sir, and for starters I don't believe you'll ever get everyone on the same page as to what constitutes the ideal markup. Besides, it's just more complication than what I'm looking for. My kind of thing is the old school grab-bag of stuff like you can find in the "philes" section of textfiles.com, for example.
textfiles.com/directory.html
There's not even any attempt at any kind of uniform markup because those files all date from different years and were created on various software on a variety of competely different computers. And that's great. That's my fucking paradise, man! Halleluyah! No fucking bullshit shackles and all the crap you see in today's world with endless libraries and frameworks and all that shit, or the need for latest botnet hardware. Just plain text, simple, and perfect.

I don't need to be relevant tbh, and I try to avoid it as much as possible. But I guess you could always just start hosting important stuff on gopher instead of web (like move your open source projects and whatnot there, and hand out only gopher:// url to people).
I doubt BBS can be made relevant anymore, unless you somehow end up hosting a very popular door game or something. But perhaps if rampant censorship and witch hunts keep getting worse, then more people will start leaving the web. So it could just happen as a natural reaction of events already taking place. BBS are by nature more private than any web forum or chans. But probably something closer to Usenet would be more practical.

Attached: Osborne 1 with Gotek flop.jpg (641x474, 142.16K)

Hey, at least he had the imagination to propose his own elements. This is important; he's not beyond redemption.

no you
I would probably remove more unused or evil shit only evil people use than I would add.
-WebGL
-WebUSB
-ALL javascript referrer and popups
...
list is nearly endless

On wikipedia you wouldn't even need SSL when you're not writing/logging in/logged in since URLs aren't encrypted anyways.

He wants and tags, but he forgot and and about 3 dozen more visual semantics. We need to get this web development innovator to the WWW committees immediately so they can learn from his insights.

You don't understand. Semantic web would be the end of external CSS.

The modern web can be a nightmare if you're (color)blind.
Dumb fullstack numales make shit that runs on a bigass iMac in San Fransisco and nothing else.

>textfiles.com/directory.html
This is good website design.
Still looks perfect on my modern monitor.

No, it wouldn't and you can't tell anyone that is better than
External CSS sucks anyways and the server should put the site together.

I don't believe you'll ever get everyone on the same page as to what constitutes the ideal markup

No, I definitely don't think a single format (or rather, semantic) would serve everyone. I didn't point this out, but a namespacing solution of one kind or another would be necessary. This doesn't necessarily mean XML, of course; it would only need to provide a mechamism for expression of the general namespace-node-subnodes-parametres model.

Such a model could even allow for representation of binary formats, in fact, cf.

" rel="nofollow" target="_blank">saveformat.of.some.obscure.80s.game'>
2
126
...

Yes, less complication on your egoistic local part, you disingenuous ass. Your local simplification of your document server software would be easy on you, but will necessitate global redundancy in terms of people having to write their own parsers for every single hand-written document that you host which otherwise could have been formatted consistently and parsed easily. Ethical software does not refer only to its liberty.

This thread is by very virtue of its community about communal, consistent, socially viable solutions to data sharing. If you want to host completely inconsistent data, feel free, but you contribute nothing in the popular sense.

>you can't tell anyone that is better than
My dude, you don't need to put a fucking class attribute on every element you want styled. The location of the span tag is enough in a well-structured document.

To be fair, it would be trivial to make a browser add-on that would understand style wrapper elements.

Hey!
being equivalent to
Hey!

Hello world!
being equivalent to
Hello world!

etc.

Ah, so you think I'm egotistical for not wanting to play by your made-up rules, huh? Everyone must do as you say, because you're the king, all-knowning and wisest of them all.
Protip: you parse text philes with your eyes. Why would we make it easy for the likes of google and other cianiggers to crawl our stuff? Fuck them.

Worse. tags are literally never necessary. They have no semantic by definition, and even in space-saving sense, they are longer than an XML namespace plus one-letter tag name, eg.

foo

Fuck off to /b/ or whatever the containment board is on this website with this noise.

found the retard.

Look at this website. It often occurs mid sentence

These are for text. It doesn't make sense to use them for layout. THERE IS NO text-align:float-left

more unnecessary shit. Do you know how over complicated it already is? Don't make it more shit.

Right, good day to you too. May you bow forever to the WEB AUTHORITIES.

Absolutely superfluous for documents that aren't embedding behavior.

So? That doesn't change anything I said. You can still use the structure of the document to select it, especially with proper use of the "id" attribute. This is still not a reason to abuse classes.

Okay, you retard. Just don't let the realization hit you as you one day type, off the top of my head, "RMS Titanic" or "New York" and remember that as soon as you type "RMS" or "New" respectively, you are literally using a namespace.

I literally just gave a case in which they prevent superfluity in the most countable of senses.

That's even more stupid. It would mean the server would have to generate custom CSS for text parts like this: TEXT

feels good to not be american

Found the over-engineering autist.

I said "proper" use of. Nice scarecrow. The fact is, span tags in the middle of text can still be structurally selected in a well formed document.

You never defined what "proper use" is and it does not work with user generated content.

I'm not saying that. I'm trying to talk sense into a guy who considers namespaces unnecessary without qualification.

It is good to remember about :first-letter and :first-word, too, which also take away certain uses of .

(Quoting .)

Unlike namespaces I think those are pretty useful and can be used for those traditional newspaper capitals and stuff like that. Without them the page stays perfectly viewable.

Are you asking me for a HTML tutorial?

No, HTML must be purely semantic, putting presentational tags into HTML is killing usability for people who rely on assistive technology. Tags like or should never be used. What does it mean for text to be italic? Is it emphasized, is it a Book Title, is it a quotation? Italic means nothing. What does it mean for a paragraph to be centered? Is it a quote or do you simply like the way it looks? Using semantic tags allows the interpreter to know what the meaning of the element is, so it can for example be pronounced accordingly by a screen reader for blind people.

While we're at it, all the , and so on tags should be deprecated as well. Instead there should be one generic tag and the nesting of the containing sectioning tag should determine the level of the heading.

Visual styling of the document can be done with CSS. You should not use classes, instead let the CSS style elements based on their parents or siblings. A well structured document can go by without any presentational hacks in the HTML.

Hell no.

Italic means emphasized in a quote sense
Bold means emphasized as in emphasized/important
This is cultural knowledge you retard and can be represented perfectly.
Go back to >>>/reddit/

The web is a shitty place for people with no eyesight to begin with.
Pretending it's not and the website authors suck won't help them.

I just noticed that a blind person doesn't even need to know what's bold or italic since it rarely offers extra information.

There is an issue of maintainability here. For a particular document, relational hooks may suffice to style it* fully, but as further documents are added to the site, a certain relational selector may suddenly begin to catch too little, or too much. Then you also need to rewrite the document all of a sudden to add the necessary class attribute after all. Sometimes, proper -- semantic, non-redundant with respect to HTML -- classing is called for for sake of being future-proof.

*Or operate on it in any other automated manner.

tl;dr do not intrinsically avoid classes. Use them when they're, well, justified. Anticipate your users' needs; think if a particular element is likely to need a simple hook that would save users from tedious traversal.

It means it came from Italy you moron.

(I think I understressed the fact that styling is only one and rather particular application of element addressing. While CSS is certainly ubiquitous, and can be used outside of the web and outsite of XML family in fact, there are countless parsers that need to traverse the tree in more direct ways. Use classes for their sake as well; in their case, the alternative would be really painful.)

ROFL, you could say the exact same about HTML and javascript but it makes zero sense.

No some thing DO NOT CHANGE like this: BOLD because I defined it to be bold. If I definied it to be a quote then a class would be justified.

Not even XHTML is XML complient

What makes zero sense about it? JS, or whatever its standard is, is a general-purpose language. Likewise, it is quite hard to think of syntactical elements of CSS that are tied to the XML family beyond the model of named elements plus named attributes. The only case I can think of is dedicated CSS namespace selector.

Please explain.

>Hell no.

(By which I mean, CSS can be used for addressing in any hierarchical named structure.)

Using any of them outside the web is stupid even on the server side. Windows help documentation was in HTML but old HTML without CSS or JS
The people which use something like NodeJS are pajeets with low pay producing slow shit websites.

means emphasized too does not make it better. IT ONLY MEANS MORE TEXT
no it doesn't what those people would need is the correct use of s and not endless garbage tags.

The proper way is to use emphasis or hard emphasis. How you render these is up to the style sheet. Usually it's displayed as italic and bold respectively, but it could also be red text and underline.

Blind people do know what emphasis is, you pronounce emphasized words differently from regular words. When your markup says "this word is in italics" it doesn't mean shit, but when you say "this word is emphasized" the interpreter can style it accordingly, either by typesetting it in italic or pronouncing it differently.

While you are correct generally, you ironically chose the shittiest possible examples for your point. Signifying by color is disapproved of for sake of color-blind people, and underlining is easily confused with links. There is a reason that emphases are signified by slant and boldness; those are two most visually distinct and failproof methods.

>emphasis or hard emphasis
This makes zero difference and if you need hard emphasis you could make a class.
In fact it's better to have a low common denominator! Such as
It makes it easy to recognize and interpret with special software
But it's normally unimportant and doesn't give the text a new meaning.
no it means: "this word is emphasized"

Unironic suggestion for you.

Attached: 51LsOroONoL._SX430_BO1,204,203,200_.jpg (432x500, 52.72K)

If you want to have special red text that would be the best method
It's also used right on this page for red text.

I agree. This is why large websites don't write raw CSS. They use an abstraction layer on top of it which allows for mixins, such as LESS.

en.wikipedia.org/wiki/Less_(stylesheet_language)

How can we turn bullshit into more bullshit?
The answer is Less.

So what if I want to use italic text to style the title of a book like Structure an Interpretation of Computer Programs, is that that supposed to be emphasized as well? You can use italics for different purpose. This is why the distinction of semantics (what it is) and presentation (what it looks like) is important.

Then replace red by ALL CAPS. That's a reasonable thing to do if the font lacks a bold typeface.

The difference between the use cases isn't as important as you think. If the device says its italic it's italic. Then everyone lives in the SAME reality and there are no misunderstandings.
Almost all humans have eyesight. There is no purpose in this. The ones which don't are better off adapting to our reality than the other way around.

t. self-taught PHP programmer

LESS can be precompiled. You don't have to use less.js, and if you had actually ever done web development, you'd know how handy dynamic compilation is during debugging.

I humbly suggest that you overlooked a funnier example. Namely, emphases may be nested.

>I really really really really really really fucking hate you.

This makes sense. Nesting or (nearly) doesn't.

Also, stop replying to bad bait.

That's a great example! I'm sure the 90's web dev shitting this thread up with his "sage is downboat" 'sperg out will ignore it though.

This is some retardism it's like the precompiled version of JQuery of CSS.

That's even more horrible and I'm pretty sure there is NO interpreter that can make use of that in existence because tags get turned into generic objects and their kind into "default CSS" and then interpreted from there.

No it delivers static CSS to the browser. It's no different than server-side scripting. Damn, you're not even a PHP programmer, you're like... some sort of retarded techno-LARPer. Just sad.

Sad.

It's almost like updates to the HTML specs, such as the one that introduced "em" and "strong," move on from limited presentational semantics to expanded functional semantics for specific purpose of encouraging interpreter makers to come up with treatments of all possible configurations of those tags, that are as useful to end user as possible.

tl;dr the fact that interpreters (incl. majority of stylesheets) fail to represent a certain configuration of tags (such ad nested emphases) in an appreciable manner only speaks of the interpreters, not of the specs.

(Another example would be, for instance, a hypothetical scenario if nested tables had been forbidden for some reason. Then, allowing them in a subsequent version of the specs would not be useless, but rather encourage markers-up to remember that there are valid reasons for such nesting.)

The white man uses plain HTML. CSS is nigger technology.
I mean if you want to increase the size of the trusted code base, yeah it makes perfect sense to require CSS in order to do anything useful beyond what can be done with a plain ASCII or UTF-8 file.

Because it's easier and more practical to type
kill yourself
than
.le_emphasized { font-weight: normal; .moz_pos: lol;}kill yourself
And a white man's browser like Links will just ignore CSS anyway (which means every document looks the same, instead of pink and green text with 300mb of fonts and pointless cringly looking laggy animations with screen tearing, input lag, and motion blur), while supporting simple tags like

There are console browsers that support CSS.

I wrote: precompiled
I clearly understood but I still think it's overcomplicating the problem
fuck off
It's good you autist. Otherwise people would start designing there website that way because it's "the latest hype"

now that's sad

Couldn't have said it better.

Also: as soon as you say "but has acquired the semantics of emphasis anyway," you literally agree that creation of " was a good idea.

No, you didn't and don't understand. You compared precompiled LESS (server-side) with jQuery (client-side).

I don't really follow that subdiscussion, but now I'm looking it up, I see that that Less thing can be interpreted client-side too, and likewise I think there is nothing inherently impossible about a script that would translate jQuery-written scripts to vanilla JavaScript for reuse elsewhere.

You wouldn't deploy to production with the interpreter, but it's definitely useful while debugging. Propping up a poor use case as reason for denigrating the technology itself is a strawman fallacy.

Why on Earth would you do this? It doesn't eliminate client-side JavaScripting like precompiled LESS does?

>literally agree that creation of " was a good idea
No I don't it's redundant and reduces compatibility.

Precompiled always means server side and I just wanted to say it's horrible by comparing it to JQuery.
Abstracting the abstracts things imaginable is fucking autism. I'm not sure you'll actually save computation on user created content when you have to recompile the css after each post.

Seems like a backpedal, unless you literally just use jQuery as an arbitrary point of reference for anything that's shitty. As far as LESS use cases, you wouldn't want to use it on every project. Here's a website that uses LESS:

www.uverse.com (I was on the dev team there)

The little tiles representing video content have to appear consistent on many different pages, in many different containers, all of which have their own styling rules. Using LESS let's you do this pretty easily. With pure CSS, it would be a nightmare, with a shitload of duplicate code (and weird rules/workarounds which would stress the browser and cause performance issues).

I don't dislike Less. Calling it "thing" didn't mean disrespect, but my unfamiliarity with it. I was just pointing out it's similar to jQuery.

Yes, but it could still remove framework overhead. Don't read too much into that post; I was just stating those tools' likeness.

According to your "reasoning," literally any innovation, for instance, introduced to increase usability and reduce ambiguity as in case of b -> em, "reduces compatibility" because of lack of adoption of the updated version at first. Better start campaigning for termination of all software development on that basis.

Fair enough, sorry if things are a little heated lol