Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

Does ANYbody follow HTML 4?

By Miniluv in Internet
Tue Oct 03, 2000 at 07:53:50 AM EST
Tags: Internet (all tags)

After reading another discussion here on K5 I got to playing with the Validator that the W3C consortium has made available. I'm noticing that very few "popular" sites really follow the transitional spec. The other site doesn't, K5 as already mentioned and commented on doesn't. Not even the standards whiners Mozilla do.

If this is true, why does everybody raise such a stink about it? I know I'd like to take the time to make my site compliant...but I suck at HTML..which is why Dreamweaver writes my code.

Is anybody in the world of web design particularly concerned about the fact that no one follows the specs? Should W3C even bother with specifications anymore? It seems like in this new Internet age we just don't give a damn about following the rules.


Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure


Related Links
o Validator
o The other site
o K5
o Mozilla
o Dreamweave r
o Also by Miniluv

Display: Sort:
Does ANYbody follow HTML 4? | 56 comments (55 topical, 1 editorial, 0 hidden)
pragmatism (3.25 / 8) (#2)
by bobsquatch on Tue Oct 03, 2000 at 12:58:28 AM EST

Almost all of the web designers I've worked with have been pragmatists, not idealists. Their only concerns were:

  • Does it look good on $MOST_POPULAR_BROWSER?
  • Does it look good in my portfolio?
In short, nobody gives a damn about "standards compliance." They only give a damn about the de-facto standards set by NS and IE. That's because their job description is not "Make the anal-retentive geek boys happy with 100% W3C compliance," but rather to "Make a website that looks good on the browsers that 90% of our target audience uses."

Re: pragmatism (3.66 / 3) (#26)
by fb on Tue Oct 03, 2000 at 09:30:25 AM EST

This does not explain why - according to my not so limited experience - most professionally designed web pages tend to be completely unreadable and to look like shit in all browser/OS combinations but - at best - one.

Standard compliant pages are at the very least readable in all browsers and they can look good with a little effort. Even considering the blatant bugs in browsers - and there are quite a few - and the limited support for CSS and the advanced features of HTML 4, I reckon that following standards never hurts and it almost always helps to improve the overall look of a site. This is my personal opinion, of course, but I got it after several years of experience.

Anyway, many web designers are grossly overestimating the influence of the MS-Windows audience: in the end the IT professionals are often the one doing the reviews and addressing people to sites.

For example, if I cannot read some page under Solaris because I do not have the latest "coolest" Microsoft supplied fonts, I am then not bloody likely to give that page a positive review.

On the other hand I often find out, when I have the occasion, that web pages that are unredable on anything else than Windows would have been best left unread.

On yet another hand not every page in the above-linked web site currently validates, although discrepancies are fairly minor and I am slowly getting there. Just so I do not look like a complete hypocrite.

P.S. No particular Microsoft bashing: I was taking Windows as an example because it's actually an OS I seldom use (given that I work in a field where it's next to useless).
[ Parent ]
Re: pragmatism (5.00 / 2) (#31)
by WWWWolf on Tue Oct 03, 2000 at 11:43:06 AM EST

Here's the problem with this "pragmatism" scenario: It isn't pragmatism at its best.

Because, one thing is missing from your list:

  • Works with the next browser versions or not, who cares, I'll be working elsewhere and as sure as hell I'll be better paid.

Because being standards-compliant, you can make sure the pages work in future so you don't need to take the trouble to redesign it for the new browsers.


-- Weyfour WWWWolf, a lupine technomancer from the cold north...

[ Parent ]
I tried ... (2.77 / 9) (#3)
by strepsil on Tue Oct 03, 2000 at 01:24:48 AM EST

I really had a shot at doing this a while back. I set out to redesign my site with the goal firmly in mind that it would look OK, and be 100% HTML4 according to spec.

It was a nightmare. In the end I decided I cared more about 'best viewed with any browser' than I did about the specs. You can't keep to the rules and still be cross-platform (unless you're happy with the most basic of sites).

It sucks, but what can you do? Just try to produce neatish code that works and is readable in most browsers. Test in Opera, NS3 and 4, IE 4+ and Mozilla. Point Lynx at it. If in all cases you can read what your site was intended to say, you've done damned well and deserve a pat on the back.

fsck the standards. The poor slob who who just wants to read your site will be happy if it works, not if you tell him his browser sucks because it doesn't follow some standard he'll never read or understand in his life.

Using standard HTML (2.83 / 6) (#4)
by Suanrw on Tue Oct 03, 2000 at 01:30:07 AM EST

The problem, as I see it, stems from the wishy-washy specification that if the browser doesn't understand a tag or element, it ignores it. The intent here, I think, was to let less fully featured browsers work with complex input, not to allow mal-formed HTML

But, this opened the flood gates to sloppy HTML from day one. Before validators became well known and readily available, the best check author had available was to display the page in their target browsers. If it looked OK, it was accepted as good HTML. By the time validators became readily available, it was too late.

The Mozilla page checks very well, contrary to what was suggested in the orginal post. The /. page was a surprise.

In defence of soft-validating user agents (4.00 / 4) (#5)
by Dacta on Tue Oct 03, 2000 at 02:14:51 AM EST

One of the main reasons the web took off the way it did was that it was/is so easy to write HTML. Even if you had no idea what you were (are) doing, you can write plain text, and maybe put a H1 tag here and there, and you get something that will display and be readable in most browsers.

That is the way it has always been, and that is good. It is useful to have robust user-agents that correct authors mistakes, because it make the user experience more satisfing, and make barriers to publishing less.

There is a place for hard-validation of web content. I think XML browsers should validate properly (as IE4+ does), and I would listen to an arguement that if the document defines a DOC-TYPE it should validate. Apart from that: Error correction rules!

[ Parent ]
W3C standards (2.90 / 10) (#6)
by charter on Tue Oct 03, 2000 at 02:15:46 AM EST

No one follows the W3C standards because they are, in a word, insane. They're probably good rules to teach newbies, but in the real world we have to be a little more pragmatic about our HTML.

On a slightly different tangent, I ran my index page (which is quite well-coded IMHO, thankyouverymuch) through their validator and received several ludicrous error messages, including this one, which I think is absolutely priceless:

<p>Ah, the documentary. The flip side of movies, documentaries can ...

Error: document type does not allow element "P" here; missing one of "APPLET", "OBJECT", "MAP", "IFRAME" start-tag

So they're saying that the tag that kicks off my paragraph is incorrect, because it needs to start with either "APPLET", "OBJECT", "MAP", or "IFRAME"? Since when?

I would now like to make a public declaration of shenanigans against the W3C.

-- Charter

Re: W3C standards (2.66 / 3) (#14)
by Simon Kinahan on Tue Oct 03, 2000 at 06:40:14 AM EST

I'm not sure about HTML, but the new generation XML specs are definitively crazy. For instance, I have the schema spec printed out in front of me. Its 500 pages long. It has an 83 page primer. I keep it in a lever arch binder.

Now, length is not necessarily a problem, but on reading the thing, I realise not only is it long, its vague, and silly. There are a hundred way to do everything, none of which works properly. There's a ****ing mechanism for specifying primary keys, for ***'s sake.


If you disagree, post, don't moderate
[ Parent ]
Re: W3C standards (4.42 / 7) (#20)
by fredrik on Tue Oct 03, 2000 at 09:04:56 AM EST

Well, you'll have to check the specs to figure out what the validator means:
The validator output tells me that you have declared a block element (the P tag on row 26) inside a inline element (the FONT tag on row 25).

So if you move all FONT declarations inside the P declarations you won't se the messages again. And btw, don't forget to close the FONT declarations.

Have fun editing a structured language :)

[ Parent ]
More block elements inside inline elements (2.66 / 3) (#21)
by AndrewH on Tue Oct 03, 2000 at 09:13:39 AM EST

Interestingly, another page has the same problem.

John Wilkes Booth, Lee Harvey Oswald, John Hinckley Jr — where are you now that we need you?
[ Parent ]
Re: W3C standards - you did the basic error (4.50 / 4) (#25)
by NKJensen on Tue Oct 03, 2000 at 09:28:54 AM EST

You did not include a doctype. This is the actual output:

Below are the results of attempting to parse this document with an SGML parser.

Line 1, column 1:


Error: Missing DOCTYPE declaration at start of document (explanation...)

Have a look at http://www.htmlhelp.org/tools/validator/doctype.html.

Best regards, NKJensen
From Denmark. I like it, I live there. France is another great place.
[ Parent ]

Netscape Must Die (3.81 / 11) (#7)
by Sunir on Tue Oct 03, 2000 at 02:22:11 AM EST

You couldn't really adhere to standards even if you wanted to. It's quite simple, Netscape must die. It is a horribly non-compliant browser. A lot of people give Internet Explorer flak for not being compliant enough, but it is way more compliant than Netscape. Unfortunately, since it is popular, people have to break their pages for it.

Anyway, I care about making my pages as standards compliant as possible for several reasons:

  • I'm a perfectionist. ;)
  • Accessibility (it's hard to read degenerate HTML).
  • Lynx friendliness.
  • The dilution of the browser market is just around the corner. Unbelievable numbers of devices will (or do) have an embedded browser in them and even desktop applications are becoming "web-enabled". Sticking to standards is your only hope.

So, in other words, I aim to be inclusive instead of stylistic. This, on the other hand, excludes flashy technology like, well, Flash which I abhor. I suppose website designers who focus on style instead of content also break the rules. They don't really care, do they?

You know, in the old days, the simple plaintext gopher protocol pushed people to worry more about content. Not much else to worry about, really. But as HTML "advanced", it added a lot more style tags. Funny that, the style tags are now being replaced by style sheets.

I guess I just wish people would focus more on content over form.

"Look! You're free! Go, and be free!" and everyone hated it for that. --r

Re: Netscape Must Die (3.00 / 2) (#13)
by AndrewH on Tue Oct 03, 2000 at 05:28:09 AM EST

The imminent production release of Konqueror means that there will be far less buggy and more robust downloadable browsers on all platforms, finally including Linux/X11. This is a key step to change Netscape 4.X from a necessary evil to an unnecessary one, as it no longer matters that AOL/Netscape have made such heavy weather of replacing it.

John Wilkes Booth, Lee Harvey Oswald, John Hinckley Jr — where are you now that we need you?
[ Parent ]
It's a pity (3.25 / 8) (#8)
by Dacta on Tue Oct 03, 2000 at 02:27:30 AM EST

I wish more sites were written in XHTML. That would make intergrating applications into them very easy with conventional XML tools.

For instance, I wanted to produce a RSS file of Google search results. Unfortunaltly, Google doesn't produce XHTML - although its HTML isn't too bad. In order to make it work, I had to run the seardch results page though the W3C's "Tidy" program to turn it into XHTML, then use a XSLT stlyesheet to convert it to RSS.

It works, but it's not exactly neat or simple. I'd love to be able to write more applications like this, but it is just so messy!

Oh well... at least I close my paragraph tags on K5!

the only site that actually follows html4... (3.77 / 9) (#9)
by abe1x on Tue Oct 03, 2000 at 03:31:53 AM EST

The only site I've ever found that follows it is icab.de the home page for the worlds best browser. It mac only and built entirely by one person! Kinda make you wonder what the fuck is up with Mozilla when they can't do shit with hundreds of contributers and one German dude can build an amazing browser by himself. how good is it, well its stable, fast, has full plug-in and java support, decent javascript support, plus amazing image and cookie filtering built right in. Great well thought out interface too, lots of little stuff that makes you realize how much care and thought went into it.

In any case what I really wanted to mention was the fact that iCab has a builtin feature that checks the html on the page you are on. If its clean you get a little smiling face icon in the corner, non standard code a frown. Other then iCabs own page and I few pages that are linked to from iCabs page specifically cause they make it smile, I have never seen the smile, never. Of the tons of pages I view everyday, every single one of them uses non standard HTML. So much for that standard.

iCab would be the greatest browser... (4.40 / 5) (#17)
by madams on Tue Oct 03, 2000 at 08:24:43 AM EST

...if it had CSS support. I see this as the crucial missing ingredient. Once this piece is in place, I may switch to using it all the time.

But, honestly, the best browser in the world right now is Internet Explorer 5 for Macintosh (the Win version is not as good). It's fully CSS1 compliant, has a clean interface, manages bookmarks well, and overall does a bang up job.

IE5 for Windows sucks because it's trying to break the Internet for everyone else. Mozilla is total bloatware. Galeon looks promising, but it's missing a lot of things (no right-click on links, no copy-and-paste, wtf?). Opera is good but I find the MDI a little hard to work with.

And I love to make iCab smile. :-)

Mark Adams
"But pay no attention to anonymous charges, for they are a bad precedent and are not worthy of our age." - Trajan's reply to Pliny the Younger, 112 A.D.
[ Parent ]

Re: iCab would be the greatest browser... (3.00 / 2) (#34)
by FFFish on Tue Oct 03, 2000 at 12:27:02 PM EST

No, the best browser is Opera v4.x.

It's *more* HTML 3.2 and 4.0 compliant than MSIE.

It's *more* CSS1 and CSS2 compliant than MSIE.

It has a *more* flexible user interface.

It has *more* cookie-handling/privacy features.

It has *more* tweakable settings.

And, finally, the company is *more* responsive to customer feedback.

Can't be argued: Opera is a mo' better browser.

[ Parent ]
Re: Opera the best browser (none / 0) (#56)
by Moghedien on Sat Oct 07, 2000 at 07:16:48 AM EST

at first when reading your post, I was going to start complaining that Opera didn't support the DOM very thoroughly, but then I decided look up the technical spesifications, and indeed, all it takes to make my current IE5-specific website both Opera- and NS6-compatible is to start using the getElementById()-method. simple as that. so thank you, mr FFFish, for making me start thinking about Opera.

[57 68 6F 20 63 61 72 65 73 2E]

[ Parent ]
Re: iCab would be the greatest browser... (3.00 / 1) (#37)
by abe1x on Tue Oct 03, 2000 at 01:07:19 PM EST

I use iCab 95% of the time. Lack of CSS isn't much of a problem, except for ironically on Jacob Neilson and Bruce Tognazzini's sites. Lack of full javascript support is what usually forces me into IE5. IE5 is a pretty good browser but the lack of cookie and ad filtering makes it nearly impossible for me to surf with it for long periods of time. What really puts iCab over the top though is the click and hold contextual menus. Can't live without it, beats right clicking any day. Couldn't surf on a wacom at all without this feature, so iCab's saving me from tendonitis on top of everything

[ Parent ]
HTML 4 is stupid, that's why! (2.45 / 11) (#10)
by RadiantMatrix on Tue Oct 03, 2000 at 03:38:11 AM EST

In my experience, HTML 4 doesn't get followed because it is a foolish standard.

Those of us who are used to using <center>, <b>, and the rest are now supposed to change to the new <p align=center> (despite that it skips a line) and <strong> tags. And that's just two. HTML 4 is inefficient, overly strict, and massively deprecates HTML 3. Thank god there's a "transitional" version, or the intranet sites I develop (which MUST be HTML-4 compliant) would be a pain in the ass.
I'm not going out with a "meh". I plan to live, dammit. [ZorbaTHut]

Re: HTML 4 is stupid, that's why! (3.40 / 5) (#11)
by vsync on Tue Oct 03, 2000 at 04:35:47 AM EST

Um, you were always supposed to use <strong> and <em>, the point being that browsers could use other attributes to display them. Lynx does that, for example. Emphasized (or italic) text is usually underlined, and strong text can be bold or bright or whatever.

I don't see why they included <b> at all...

"The problem I had with the story, before I even finished reading, was the copious attribution of thoughts and ideas to vsync. What made it worse was the ones attributed to him were the only ones that made any sense whatsoever."
[ Parent ]

Re: HTML 4 is stupid, that's why! (5.00 / 2) (#46)
by factotum on Tue Oct 03, 2000 at 03:41:39 PM EST

<blockquote type="cite">Those of us who are used to using <center>, <b>, and the rest are now supposed to change to the new <p align=center> (despite that it skips a line) and <strong> tags. And that's just two. HTML 4 is inefficient, overly strict, and massively deprecates HTML 3. Thank god there's a "transitional" version, or the intranet sites I develop (which MUST be HTML-4 compliant) would be a pain in the ass.

You haven't understood what it's about. HTML isn't a page description language. It isn't mean to tell the browser what the page looks like, but what its content is. Earlier versions of HTML have contained unfortunate elements such as those you name, and they say nothing of relevance about what the document is as opposed to how it's supposed to look. Imagine a speech engine has to render <i>... how does it do that? <em>, on the other hand, is obvious.

[ Parent ]
Re: HTML 4 is stupid, that's why! (1.00 / 1) (#47)
by RadiantMatrix on Tue Oct 03, 2000 at 04:47:39 PM EST

HTML isn't a page description language. It isn't mean to tell the browser what the page looks like, but what its content is.
My understanding is that HTML was designed to be a Text Markup Language that was both human- and machine-readable. There are cases, yes, where %lt;strong%gt; and <em> should be used, but there are also times when I want assurance that text will be italicized. The fact that I can't do that and conform to HTML4/strict is annoying.

Besides, HTML 4 is growing toward obsolete anyway.
I'm not going out with a "meh". I plan to live, dammit. [ZorbaTHut]

[ Parent ]

I would hope EVERYbody would use XHTML (2.50 / 6) (#12)
by chaotic42 on Tue Oct 03, 2000 at 04:47:09 AM EST

Why not use it? It's not a terribly large amout of work to translate your valid HTML 4 documents to it. Unless they're Geocities quality.

Of course they do. (4.36 / 11) (#15)
by Inoshiro on Tue Oct 03, 2000 at 07:49:10 AM EST

I certainly follow HTML 4 transitional. HTML 4 strict requires more things be CSS than I am willing to use at the moment, because of the lack of compliance in browsers. However, I expect that once Mozilla releases a stable version (and Galeon/KMelon become more popular), I'll go to HTML 4 strict with heavier CSS. Beyond that, XHTML!

[ イノシロ ]
butchering the standards (4.25 / 8) (#16)
by madams on Tue Oct 03, 2000 at 08:18:29 AM EST

I try my best to make the web ages I publish HTML 4.0 Strict. However, a couple of things stand in my way.

I've made perfectly compliant webpages using HTML 4.0 Strict and CSS that look great in IE, Opera, Lynx, etc but are totally butchered in Netscape 4.x. (because of it's half-assed CSS implementation). Netscape can make even the simplist CSS pages totally illegible.

The solution? Make a seperate CSS file for each browser. The HTML is exactly the same, but a cgi script on the other end of the CSS link detects the browser and sends the appropriate CSS file. This lets you leave out CSS that makes Netscape choke (like background color anywhere except the body) and go on with your life. Sure, the page doesn't look as pretty as it does in IE or Opera, but at least it's usable.

Mark Adams
"But pay no attention to anonymous charges, for they are a bad precedent and are not worthy of our age." - Trajan's reply to Pliny the Younger, 112 A.D.

Yes. (4.12 / 8) (#18)
by eann on Tue Oct 03, 2000 at 08:33:42 AM EST

HTML 4.01 Strict is impractical because neither Netscape 4.bloat nor IE 5.x for Windows supports enough CSS to use (IE for Mac does a pretty good job). However, it's not particularly difficult to make a document that conforms to 4.01 Transitional (with CSS) and still "degrades gracefully" in the common browsers. It feels like a kludge to me every time I do it.

Web professionals don't generally complain that no one sticks to the standards. We complain that the browsers don't let us. We complain that we have to spend hours and hours to get interesting design that displays even somewhat close to how we intend it because Naggravator and Exploiter can't even agree on how they're going to suck.

What to do? Stick with the standards, and put pressure on software makers to do the same. Join the Web Standards Project while you're at it.

Wow. I just noticed this: the two front-runner browsers suck, and the two front-runner U.S. presidential candidates suck, but in both cases, there are many third-party options that don't. I wonder if this is a trend or something.

Our scientific power has outrun our spiritual power. We have guided missiles and misguided men. —MLK

$email =~ s/0/o/; # The K5 cabal is out to get you.

Re: Yes. (4.00 / 1) (#38)
by keltor on Tue Oct 03, 2000 at 01:35:28 PM EST


i get stuck writing lots of pages. i refuse to write to exploiter or naggivagator, but i do make sure that with what i do write, the content is visibile in either. i use Opera as an initial viewer as it gets closest to the spec (no mac here with exploiter), make sure it actually views in both ie and netscape and lynx, make sure I hit the major encoding for the blind points (alt tags), compress images, set height and width on images, dont use tables unless necessary, dont use frames unless it is strucurally much more sound (intranet) (frames are much more modular, and modular designs are much more flexible-- ditto for usinc CSS), i don't use fonts (unless i have a damn good reason and the font is downloadable)...

i seldom run each daily through a validator due to simple time constraints

and always always KISS (Keep It Simple, Stupid)

A picture had better be worth a 1000 words-- it takes longer to download (this comment posted from a debian X-Box)
[ Parent ]

(3.75 / 8) (#19)
by n8f8 on Tue Oct 03, 2000 at 08:39:58 AM EST

One of my favorite HTML editors is HTML-KIT. It includes HTML Tidy, a code formatter and HTML validator. One of the validations is checking for the use of "summary". Having never used this table tag attribute I was quite supprised to learn that it is intended to be used by blind websurfers to examing the contents of a table before "opening" the table to read it. Neat. The only catch is that I tend to use nested tables a lot and this makes it tough for blind browsers.

Aside from that I find the W3C spec all but useless as a reference document.

Sig: (This will get posted after your comments)

I was unimpressed (2.00 / 1) (#28)
by AndrewH on Tue Oct 03, 2000 at 11:07:09 AM EST

HTML-KIT crashed on me once too often when I tried it last year. Is it any better now?

John Wilkes Booth, Lee Harvey Oswald, John Hinckley Jr — where are you now that we need you?
[ Parent ]
Re: W3C Useless? (none / 0) (#52)
by wendall911 on Wed Oct 04, 2000 at 02:35:07 AM EST

>Aside from that I find the W3C spec all but useless as a >reference document.

I find this quite the norm unfortuately. 99.9% of the time in my experience, it has been because people don't know or take the time to learn how to read the documentation. I'm sure most people think a dictionary or thesaurus are useless because they already know how to talk and have the 3R's down. Let's get real though. Just like a good writer, would I really want to leave my dictionary on the shelf collecting dust? After all my spill chucker does just fine right?

Just a thought.

remove my account
[ Parent ]
Tidy your html-pages! (3.10 / 10) (#22)
by hph on Tue Oct 03, 2000 at 09:19:50 AM EST

Tidy will check your pages and print out valid html. It also translates html into xhtml.

You use Dreamweaver. What do you expect? (2.60 / 5) (#23)
by paranoidfish on Tue Oct 03, 2000 at 09:25:06 AM EST

So how does your dreamweaver site compare with the code at /. or k5?

I mean, if you are going to complain about a subject, at least make sure you are informed first. You don't get mathematicians saying "This proof of Pythagoras is Crap! I mean, not that I can add two numbers without a calculator but...".

Webpeople do want to follow the specs. Just look at the Web standards Project to see that. The problem is that the specs sit in an alterate reality. Just because our sites throw up errors, doesn't mean we don't know about them, and now exactly why those errors were introduced to the code; and what effects they have on browser x. For example, I remember ages back on /. when cmdrtaco had to choose between supporting Opera2 or Netscape

Now, if you were to bring up the problem with most sites (including/especially dreamweaver sites) not working in browser y or if you were complaining about accessablilty, usability or compatability then I'd agree with you. But you're not. You are bing pedantic about a standard that cannot work on todays web. HTML 4 is a step in the right direction, but you can still code crap sites that are standards compliant and don't work in the real world.

(Oops, that was a bit of a rant. Sorry)

Re: You use Dreamweaver. What do you expect? (3.00 / 2) (#24)
by paranoidfish on Tue Oct 03, 2000 at 09:28:42 AM EST

sorry, bad editing. This was in there somewhere:

[...]when cmdrtaco had to choose between opera 2 and netscape 2. He could have had one, or the other, or standard compliant code. The opera users appreciated him not following the standards route.

[ Parent ]
Beating a Dead Horse... (4.00 / 7) (#27)
by ScottBrady on Tue Oct 03, 2000 at 10:35:14 AM EST

This argument has been spewing threads on every damn HTML message board and newsgroup since 1.37 seconds after TBL invented the Web.

Standards are the basis for communication. Without them we'd still be running around in animal skins hitting each other over the head with rocks. HTML is simply another means of communication (albeit, VERY popular). Once we allow browser makers (Microsoft, Netscape, Opera, etc.) to set the standards, the lines of communication start to break down because it's not in their best interest to play fair. Meaning, fully documenting their spec and maintaining backwards compatibility.

You have to remember that browsers are inherently forgiving creatures. If you misplace a semicolon in C++ your compiler will croak. Misplace a dump truck load of tags and attributes and your page will probably still render (although not as you intended). This is a Good Thing. It is meant to be. This is a communication system that we want it to be forgiving of errors.

Unfortunately, this is a double edge sword and the problem lies with the developers. Far too many have a "it rendered in browser X it must be ok!" attitude.

Oh, and FWIW, I make a living designing W3C compliant websites. I only know of one other professional web dev who designs to W3C standards like I; Mr. Art Sackett of Co, USA.

I find that sad.

Scott Brady
"We didn't lie to you... the truth just changed."

Re: Beating a Dead Horse... (4.33 / 3) (#33)
by analog on Tue Oct 03, 2000 at 12:18:34 PM EST

Oh, and FWIW, I make a living designing W3C compliant websites. I only know of one other professional web dev who designs to W3C standards like I; Mr. Art Sackett of Co, USA.

Well, you can add another to the list, and I suspect there are quite a few more out there. I also suspect that, like me, they tend to keep quiet about it because of the amount of invective hurled their way when it gets brought up.

That said, the ones doing the shouting tend to have a lot of good points. However, I tend to blame the browsers rather than the standards. I find it quite easy and painless to write compliant html; making it work in more than one browser at a time, OTOH, can be a serious PITA. I can't find it in me to blame the W3 for that.

Right now, I have two sites that don't validate; both of them could be made compliant with style sheets, but since you can kill style sheets in NS4.x by turning off Javascript and they look really broken then, and 20% - 50% of the people coming to those sites use Netscape (depending on the site), I don't feel safe doing it. Again, I won't blame W3 for that.

So, I guess what I'm saying is, it's all Netscape's fault!!! ;)

[ Parent ]

Re: Beating a Dead Horse... (1.25 / 4) (#36)
by Tin-Man on Tue Oct 03, 2000 at 12:54:34 PM EST

This argument has been spewing threads on every damn HTML message board and newsgroup since 1.37 seconds after TBL invented the Web.

Please check your sources again. Everybody knows that Al Gore invented the internet.
The future sure isn't what it used to be!
[ Parent ]

Re: Beating a Dead Horse... (1.00 / 1) (#53)
by AndrewH on Wed Oct 04, 2000 at 04:36:25 AM EST

Everybody knows that Al Gore invented the internet.

I think you meant to say Bill Gates :-)

John Wilkes Booth, Lee Harvey Oswald, John Hinckley Jr — where are you now that we need you?
[ Parent ]
Re: Beating a Dead Horse... (3.25 / 4) (#39)
by four12 on Tue Oct 03, 2000 at 01:54:17 PM EST

Without them we'd still be running around in animal skins hitting each other over the head with rocks.

'Cept for the non-compliant types wearing rocks and hitting people with animal skins.

[ Parent ]

I do. (4.50 / 8) (#29)
by WWWWolf on Tue Oct 03, 2000 at 11:33:12 AM EST

Well, I follow the standards. And I definitely think everyone should.

I haven't done many "big webdesigns" myself, but my <a href="http://www.iki.fi/wwwwolf/">home page is as standards-compliant as possible. I've made one bigger page for pay (A small Finnish <a href="http://www.kuhmo.fi/talvisota/index.html">war museum page), but I wiped it from my CV as soon as some idiot touched it with FrontPage. (Yes, I'm serious. Actually, I mention in the CV that the brokenness of the front page and technical stuff that still plagues the page are not my fault. =)


Because writing standard-compliant code is easy. Everyone with half the brain can do it. If someone is unable to write standards-compliant HTML, I call them lazy for a reason.

A very quick guide to standards compliance: Make sure you use all required attributes, keep end tags properly matched, add the DTD declaration, and validate early and often. Most often, the validators seem to complain about missing ALT= attributes for images, and such. Laziness, as I said.

Personally, I was out of validation services for a long time. During that time, I updated my page kind of often. What was the result when I got the validator again? 5 HTML errors, I guess. And I had many pages there that I had changed several times.

And remember: Validity often helps a lot when you make the pages to be as accessible as possible.

Standards compliance it's not fascism, it's common sense that everyone should listen to.

-- Weyfour WWWWolf, a lupine technomancer from the cold north...

Ack. (3.00 / 3) (#30)
by WWWWolf on Tue Oct 03, 2000 at 11:36:05 AM EST

Sorry for the kind of irony that the above comment seems to have. Blame Mozilla for the lack of A end tags. Or K5. Whatever. Those were in the original comment allright, they disappeared between Preview and Post for some reason.

-- Weyfour WWWWolf, a lupine technomancer from the cold north...

[ Parent ]
Re: I do. (1.00 / 1) (#40)
by blameless on Tue Oct 03, 2000 at 02:52:40 PM EST

as standards-compliant as possible

What exactly does that mean?

Is it impossible for you to close your &lt;LI&gt; tags?

[ Parent ]
Re: I do. (none / 0) (#41)
by blameless on Tue Oct 03, 2000 at 02:55:46 PM EST


It previewed fine.

[ Parent ]
Re: I do. (4.00 / 2) (#48)
by WWWWolf on Tue Oct 03, 2000 at 06:51:32 PM EST

HTML is an SGML application, which means that each HTML file must conform to the HTML Document Type Definition (DTD), as defined by W3 Consortium.

As for <LI> tags, the DTD says that the end tag can be omitted, so it's up to the user to decide whether or not to leave them out. (Of course, if you're already worried about moving to XHTML, you probably want to put the end tag there anyway.)

Sorry, can't make any more throughout answer, it's been a long day. Consult the documentation if this doesn't satisfy your endless curiosity.

-- Weyfour WWWWolf, a lupine technomancer from the cold north...

[ Parent ]
A convert to compliant HTML (4.62 / 8) (#32)
by Eliezer on Tue Oct 03, 2000 at 11:46:46 AM EST

I always used to laugh at sites that closed all their <p>'s and had DTD declarations, etc. It didn't seem terribly important to me.

In the course of mucking around with Emacs (and not liking the HTML modes I found for it), I came across PSGML, which parses the DTD to determine what you are and aren't capable of in any particular place. Suddenly I found that I _had_ to write compliant HTML to be able to work at all. This was annoying at first.

As time has gone on, however, I have found that my HTML (especially if I stick to 3.2) is extremely readable and surprisingly portable now. I am a convert; if you write it right the first time, you can always point to the standard when someone doesn't see what they're supposed to see. If browsers were _wildly_ noncompliant, things might be different, but for the basic stuff they work. After a certain point, one wonders if the medium is not in fact the right one (HTML and SGML were meant to convey meaning, which is translated differently by different viewers; most web designers now use it like it was PostScript).

Re: A convert to compliant HTML (none / 0) (#54)
by wee on Thu Oct 05, 2000 at 03:49:32 AM EST

The closing <P> tag is optional (although not a bad idea). Take a look at http://www.w3.org/TR/REC-html40/struct/text.html#edef-P for more info.

Just fyi. (And my first K5 reply... :-)

[ Parent ]

Re: A convert to compliant HTML (none / 0) (#55)
by Eliezer on Thu Oct 05, 2000 at 01:42:39 PM EST

Ha! You're absolutely right. It's been a while since I looked at those pages (I've been relying on PSGML to handle this stuff for me, lately, and it automatically adds the end-tags). Oh, well, bad example, but my point's still valid.

[ Parent ]
Leave Everything Alone! (1.66 / 6) (#35)
by OKolzig37 on Tue Oct 03, 2000 at 12:48:42 PM EST

Actually, I wish the browsers would stick with where they are now. I spent hours rewriting my DHTML JavaScript after looking at my sites in Netscape 6 Preview. They're switching horses in midstream, and it's screwing up our sites!

And the worst part is that the 4.0 specs are incomplete anyways! They don't even support margins: something which almost all of my sites have now! So the W3C Validator screams at me for using marginheight="0" marginwidth="0" topmargin="0" leftmargin="0", but it looks fine (except for NN3 and Opera, less than 1%).

Forget standards. Just stop changing stuff.
Oldy moldy, history mystery!
Maybe you should be using CSS. (4.00 / 1) (#43)
by madams on Tue Oct 03, 2000 at 03:21:28 PM EST

margin-top, margin-bottom, margin-left, margin-right. It's all there.

Mark Adams
"But pay no attention to anonymous charges, for they are a bad precedent and are not worthy of our age." - Trajan's reply to Pliny the Younger, 112 A.D.
[ Parent ]

UK Government follows XHTML ! (2.50 / 2) (#42)
by new500 on Tue Oct 03, 2000 at 02:57:44 PM EST

Open Gov The UK overview government site complies with XHTML spec

So at least the UK Government seems to take standards compliance seriously - or else the contractors for the site do or won a beauty contest pitch to get the business.

More seriously AAA W3C org acessibility is there too. Though I've not taken advantage of anything but the keyboard shortcut aspects of this.

Question is, strict quotes and tags aside, would XHTML be used or usable for "brochureware"?

With respect to the increasing dependance I find personally on information, public and commercial via the internet, maybe it would be a good idea to *require* accessibility and coding standards.

A capricious part of me wonders if representations via a website could become part of a contract and thereby be abrogated or diluted in effectiveness or enforcability as a result of inadequate rendering.

I can see it now, the "Nutscrape" defence . . . "Sorry Your Honour, this part of the click through contract is invalid in performance as a results of

versus Layers".

On the flip side of this could we actually see online subscriptions or whatever actually requiring users to read with a particluar browser arising from legal concerns.

That's my Lot for the day :-)

== Idle Random Thoughts. Usual disclaimers apply. ==
Netscape 4.x (2.50 / 4) (#44)
by CompUComp on Tue Oct 03, 2000 at 03:35:28 PM EST

Everyone is complaining about Netscape 4.x but remember it is a legacy browser the only things being changed on it are security issues, for all intents and purposes it is dead. NS six is in devel now, as i recall it just forked from the Moz tree. and Moz has been stablizing over the past few weeks. The focus there from now till release is blokers, size, stabilty, and foot print.

Howard Dean 2004

Browsers.... (2.00 / 2) (#45)
by Miniluv on Tue Oct 03, 2000 at 03:40:44 PM EST

Part of the point of this post is that browsers and designers are in a never ending spiral to write the least compliant implementations of HTML, etc.

In my experience Netscape and Mozilla both are horrible about supporting 4 either transitional or strict...IE is better, but then has a bunch of custom tags. Neither of these is a Good Thing.

The reason I mentioned Mozilla specifically in the story is that their webpage contains a link to said validator, and a really pompous declaration that before you submit a bug to them, verify the code follows the W3C 4.0 spec cuz their browser renders it perfectly

Need I really say more...?

"Its like someone opened my mouth and stuck a fistful of herbs in it." - Tamio Kageyama, Iron Chef 'Battle Eggplant'
Re: Browsers.... (1.00 / 1) (#49)
by khallow on Tue Oct 03, 2000 at 07:31:54 PM EST

You forgot the standards committees which are composed of members trying to backstab each other. It's part of the reason virtually everything is noncompliant with W3C standards. And look at the hype for XML (thankfully it's dying now). I somehow got the impression that with the advent of XML, we'd no longer have to program, merely run that validation checker. And in the process we get world peace, immortality, and a decent American beer.

Don't forget they have XML schemas for everything. Pretty soon all personal interactions will be expressed in terms of validated XML documents. But gentle readers, the most annoying thing of all was that it took me several weeks (given all the "helpful" documentation) to figure out what XML was and did (or more accurately didn't do). *:o)

Hmmm, feels like a rant is about to come. Better take some more of those blue pills.

Stating the obvious since 1969.
[ Parent ]

People are missing the point (5.00 / 2) (#50)
by driptray on Tue Oct 03, 2000 at 09:28:47 PM EST

Even those who have said that they make (or try to make) standards compliant sites seem to be missing a fundamental truth about HTML - that the rendering of the web page is controlled by the reader, not the author.

This misunderstanding is evident when people refer to pages "not rendering as they were designed", or "looking different in different browsers". Well thats the point - when you mark up a document you are not giving instructions about "layout" or "rendering" (leaving aside CSS and physical markup such as <b>), you are attempting to accurately describe the structure of the document.

Here's my web authoring tip. Conform to XHTML Strict. Do it honestly, for example, a <BLOCKQUOTE> is to denote a quote, not to indent text. Forget about how the page renders - you can't control it anyway, and your efforts will fail in a significant number of cases.

Extreme advice, and I don't always follow it myself :) You can see my compromises at http://www.doco.net.
We brought the disasters. The alcohol. We committed the murders. - Paul Keating

W3C is a good thing (5.00 / 3) (#51)
by wendall911 on Tue Oct 03, 2000 at 11:26:24 PM EST

Being "old school", I have always followed W3C standards. There are several reasons for this. The standards W3C create are a great resource and guideline that is practically unheard of in any technical field. I can always fix my code if I follow the standard, because mistakes are more glaring. (Try fixing something with a WYSIWYG editor when your site won't load in NS.) With languages moving more toward SGML, syntax becomes more important every day. By sticking with the W3C standard, I have a huge leap on the WSIWYG dependent public. I like my users to be able to view my site (even on a 56k modem). The standards set by W3C help me achieve this.
remove my account
Does ANYbody follow HTML 4? | 56 comments (55 topical, 1 editorial, 0 hidden)
Display: Sort:


All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!