Slashdot Mirror
Are 99.9% of Websites Obsolete?
citizenkeller writes "Zeldman is at it again: " Though their owners and managers may not know it yet, 99.9% of all websites are obsolete. These sites may look and work all right in mainstream, desktop browsers whose names end in the numbers 4 or 5. But outside these fault-tolerant environments, the symptoms of disease and decay have already started to appear.""
I think this percentage of the web sites that Iv'e developed over the years are obsolete. It's nothing to do with bad design - the owners of the site don't bother to use them effectively any longer and content becomes... obsolete.
It seems like someone has finally noticed that if you do not test your site using a wide range of browsers you do not know how your page is going to look... To most of us this problem is obvious.
I love stacking my barbecues in the shed at the end of summer - you can't beat a bit of grill on grill action.
I cant even keep OUR damn site up and compliant.
.. pages that work with older browsers - are choking up the newer ones.
It worked in all the current browsers a year ago.
but with IE 6 and the new netscape coming out - you would *THINK* there would be backwards compatability.
However, I get e-mails all the time from things that are now 'suddenly' broke.
And after verifying what browser/etc the user encountered this error with - amazingly enough
*go figure*
--Ne auderis delere orbem rigidum meum, non erravi pernicose!
Near as I can figure out, he's claiming "the web is broken, don't bother."
The book looks broken. Don't bother.
John
You can read the Webmaster World article, "XHTML -- is now the time?" if you want to read a debate among professionals. There are many pros, primarily developers of small sites, that are advocating dropping NN 4 for XHTML Strict and CSS, but most developers aren't going that route.
They are developing XHTML 1.0 trans or HTML 4.01, maybe adding CSS to go foward. NN4 will be around for a while, and few people are willing to write them off simply to appease the standards gods.
In the real world, we build sites for human composition. We separate content from display with our databases and content management. HTML may be an inefficient way to get the data to the browser (XML+XSLT would be ideal, XHTML+CSS would be easier on the browser), but it works. The browser parsers are done.
Sure XHTML+CSS is easier on the browser, and that may help rendering issues. However, the reality is that old browsers will be with us for a while. Maybe in 5 years this will matter, but not until then.
Alex
Now that the bubble has burst, fixing "obsolete" sites is not a priority. IT staffs have been cut, resources have been redirected into projects that actually turn a profit, or the "web guys" are gone all together. Nobody is around or has time to fiddle with the brochureware homepage.
And Jeffrey Zeldman will help us fix the errors or our ways! Anyone check Amazon for the price on this baby?
Who on earth is running a browser earlier than 4.x? Do you expect stuff to be rendered right if you use an older version of IE/Netscape/Opera? Do advertisers want to sell to people that refuse to use the latest and greatest thing? Don't you have to try real hard to even find an older version of any of these browsers?
Sounds like a cheap way to sell a book - and a little extra helping of FUD thrown in.
This space for rent.
Could this be because of the huge numbers of layoffs since the dot-bomb explosion? There are less people being paid to maintain and monitor the data, hence rendering it obsolete. Also, I am sure there are people who "maintain" to just keep the site alive and not actually doing anything as far a changing it since in most cases, it was not their site originally.
No!
(Hmm, I was tempted to leave that as is, but I think at least a little explanation is required. Zeldman disagrees with his own thesis in as much as he says that sites like Yahoo! are important because of what they offer not how they look. So QED a site that relies on it's content is not obsolete. Tadaaa!)
A little planning goes a long way...
Talk about sensationalism. The article just points out that many web sites have mark-up errors in them. Big deal. To go from that to saying that 99.9% of sites are obsolete is just dumb.
This is just a sensationist way to promote a book. Shame it got onto the front page of Slashdot. It will encourage more to do the same.
The web could do a lot worse than become a bit more strongly-typed, and a bit more like a programming language than a scripting language.
True, most folks don't need more than the basic mark-up for their websites, especially where personal websites are concerned. But commercial sites could stand for a much better design than they have. . . the author here makes a lot of good points when he calls out the faults of ZDNet and Yahoo for their HTML. The code is crap - thank God HTML doesn't have GOTO statements, or these sites would probably be chock full of those, too.
Let's do what we did with the blink tag. Don't just deprecate it--ignore it. Tell the browser, "Don't listen to the <font> tag, just skip over it."
Not too long ago, I re-wrote my own personal webpages using Cascading Style Sheets. It's tricky, since Netscape/Mozilla oftentimes has different ideas of how to interpret CSS than Internet Explorer. But it's easy enough to accommodate both, without too much effort. And I'm a lot happier now that my HTML code looks less like last night's dinner and more like something that someone else could read and understand.
!#@%*)anks for hanging up the phone, dear.
Backwards compatibility means it works in older browsers. As Zeldman mentions, it always has some cutoff point, such as Netscape 3 or IE 2.
Forwards compatibility means that it works in newer browsers. There is not necessarily any cutoff point, as long as you have constructed the website correctly. Structural problems and other typos in the HTML, proprietary and deprecated tags, and versioning can all limit the forward compatibility of the page.
Read the article and you'll see that Zeldman is arguing that web designers should be developing with forwards compatiblity in mind. Unsurprisingly, yours is one of the 99.9% of all sites that have not.
What a fool believes, he sees, no wise man has the power to reason away.
Correction .. I mean to say my employer's website, which uses asp/javascript/VB
.. my personal website uses PHP .which is just getting parsed into html for your browser}
.. if you would read the article .. even basic HTML can be corrupted ..
.. previous incarnations of browsers tolerated (and corrected) sloppy html.
.. got my CS degree in 1994 .. i never even learned visual basic in college] they are/were not always *aware* of things that html 'requires' but the browsers let them get away with.
{and technically
however
IE 5.5 will support nested tables up to 7 in depth. Netscape 6 will only support up to 4 in depth.
Netscape 4.7 does not require quotes around 'field' tags like width or height.
Netscape 6.0 can do unusual things if they are not there.
the problem (as stated in the article) is that becuase of the past 'browser wars' fighting for dominance
Now that everyone is trying (or at least saying they are) getting on the w3 bandwagon. These little 'faults' are starting to cause errors.
And since the vast majority of web pusblishers and early adopters out there have not received *formal* training in html [I for example
5 years of bad habits become 2nd nature.
sorry for the confusion.
--Ne auderis delere orbem rigidum meum, non erravi pernicose!
It seems so altough w3 offer a validator for free.
Maybe learning html in a weekend or in faster don't help keeping the quality of code at high level ; )
This is so stupid.
Do we start broadcasting TV signals in black and white again because a similar portion of viewers use b&w tv's?
Who ever uses an older browser ussually isn't a power user to start with and isn't looking for the latest fluff anyway.
You say things that offend me and I can deal with it. Can you?
Our website is not only obsolete (it was designed that way from the ground up), but it's ugly and almost entirely non-functional too! Mainly we use it to harbor and distribute viruses inside the company. It's been very effective.
Now that he's completely met his goal of total obsolescence, our webmaster spends every day looking for new ways to make our website even less useful, uglier, and more of a pain-in-the-ass to use. He's been very effective.
I can assure everybody that well over 95% of sites out there are in fact obsolite.
Lets take a closer look.
Overwhelming majority of websites out there are not HTML 4.0/XHTML 1.0 compliant. Even the sites that belong to members of w3c bend the rules which they help write. Sounds asinine? You bet.
Standards do not mean s**t anymore. Everybody is aiming for IE 5.x/6 compatibility nowdays. Cross platform understanding is dead, now that Netscape has lost the overall war. Vast majority of web designers do not even double check their sites in Opera/Mozilla nowdays, thinking they might have to do some extra compatibility coding/clean-up.
Most sites are NOT cross device/platform. You cannot view them on a PDAs of cellphones. Notice the word _MOST_
There are millions of other reasons, but I have to run to a meeting. I'll expand on this later today in more detail.
I've always considered Zeldman to be one of those self-proclaimed know-it-alls who has had little real industry experience in high volume, high technology web-sites. Most of his portfolio is brochure-ware that looks like it was done by a team of one. So I've always considered his belly-aching a little simplistic and, frankly, unrealistic in current web development scenarios.
It's easy to lament the fact that these sites aren't standard, but there are clearly reasons why most of these sites don't fit his vision of standards compliance.
For one, most sites don't have the budget to develop to standards. It's much easier to code to specifics and use non-standard work-arounds where possible then to boil everything down to the least common denominator (which standards are supported by whom). When I say easier, I mean that years of experience have instilled intimate knowledge in the seasoned web developer that almost comes as instinct now.
Secondly, all of these "standards" are interpreted differently by the different browsers, so you can't insure consistent look and feel without kludges.
Third, most of the foundations for these sites were layed out before coding to a standard was even possible, and when the mindset was not focused on any sort of standards compliance.
Finally, I've always thought that they made writing to standards compliance sound easier then it actually is, because even though it's called a standard, it rarely exhibits standard and consistent behavior across the various platforms. Most art directors and graphic designers - specifically those that migrated from print or traditional design - tend to be exteremly unyielding in the way their designs are interpreted on the web, leaving developers with few options that are fully supported by these so-called standards.
Personally, I think Zeldman needs to spend some time in the trenches working on a large site with a large development team under real deadlines for real clients. Things are rarely ideal in these circumstances.
What is it they say about armchair coaches?
How many variations of 'standards' should one have to comply with to make a usable, functional, Web-based information node? That I have to test against huge numbers of browser/platform/OS variations is a massive waste of time and energy, when I should instead be able to focus on making the information clear and the functionality flawless.
I'm not saying that we as a collective need to move back to HTML 1.0, but there has got to be a solution to increasing complexity in Web information spaces. Companies that intentionally cripple some browser/OS combinations are doing the greater community a vast disservice.
The majority of Web pages are not necessarily broken, but reflect limits on the time and energy of those who create them to keep up with 'standards' that seem to shift every other week.
It's harder to play one note and have it be perfect than it is to play a thousand and have them be close. Most people choose the latter, and hope that one note hits home.
Here's a condensed version of the article for those who don't have time to slog through it:
1. Standards are good.
2. Bad code that happens to work in current browsers is bad.
3. Buy my book.
It's Slashdot's evil twin... SlashNOT
Yeah, whatever. 83.7% of all roads are in need of repair. 99.9% of all sewers contain rats and cockroaches. Things in society are messy and are nearly always far from perfect. Someone trying to make a buck doesn't make it anymore interesting or news.
-Sean
The main problems that I see are that
1. Web standards bodies move slow and specifications are obsolete before they are approved. Take SVG. (please) Flash is a superior format with a large installed base, quality authoring tools, platform scalability, and open but expensive architecture. SVG took five years to become a reality, and is still VERY immature.
2. It's about the user stupid! For the most part, users sit at a computer desktop, with a commercial browser (IE), and use the internet. It needs to look right for THEM. The .001% of users on cell phones are doing specific activities with mostly packaged content. These users are novelty users. Portable devices have no standards as to how they display, and without this, nobody can expect a useful cross platorm "standard" that works everywhere. It's a microsoft world whiner. There is no doubt that IE is the only browser that matters. If someone else wants to make a competitive browser, it needs to be IE compliant, not W3C compliant. Microsoft took it upon themselves to create a language that works, no matter how it's written. Who cares about sloppy coding? Bandwidth is hardly an issue, and if a browser renders correctly, it should LOOK right.
in conclusion, the web standards project and w3c have failed due to their manegerial impotance, and can be safely ignored.
People who think they know everything really piss off those of us that actually do.
I agree with you, and hope someone mod's this post up.
.. where people are HORRIBLY disillusioned with the internet. I work for a fortune 500 which produces power tools - and it has been kicked around previously the idea of actually SCRAPPING our web-based projects. Hows that for a scarey morning meeting to walk in on :(]
.. my real point is .. standards change, and 'mega-powers' in the browser world ignore them anyways.
.. is maybe 80% now. [good and bad .. means that html is more versatile .. but means that you have to recode that stuff.]
.. geez .. I have been using it at work for about 3 years now .. and for a 'universaly standard' language .. its sure been through the damn wringer.
.. and put it on a unix box and watch it puke. [and vica versa]. The standards on this 'universally adaptable' langage have changed so many times in the past few years my head is still spinning. .. i dont mean the 'Official Top Shelf writtin in stone' standards .. I mean the ones that are in the real world .. MS for example. Its not a surprise they tweek things .. but when a major player in the software dept {yeah yeah} produces something sub-standard .. how long before it BECOMES part of the standards? even if its unwritten?]
.. I think your insights are dead on here.
All the folks out there who are slamming web developers/authors really need to step back a second. [I'm amazed that my first post in this topic already has 3 "You should code better" responses.]
I have been working with 'web' pages professionally since late 97.
And man has stuff changed.
Anyone who works in the real world (not academia) understands that not only is there the pressure of a 'real world' environment - but the need to show value for a company.
Understaffed departments, unreasonable demands, HUGE goals. Those are the factors that REALLY limit the 'good code' out there. Its very hard to make sure your 100% compliant [no matter how hard you tell the board/your boss/your dept/the finance people that you SHOULD be] when at the end of the day - you have more 'new' projects in your inbox than ones you have finished.
[and before folks cry - TELL THEM ! TELL THEM ! We are in an economy now
but I digress
HTML that was 100% w3c 4 years ago
XML
I can write some xml/xsl for IIS
[clarification
So yeah
--Ne auderis delere orbem rigidum meum, non erravi pernicose!
Zeldman asserts that the problem plaguing web developers is a desire for backward compatibility. In fact, that desire seems unfortunately missing in most websites. The real problem making websites suck is the desire to view the web as a graphic design medium.
Designers want to control every pixel of a page's layout, completely ignoring what the web was designed for. If everyone used logical markup to describe their data, later adding CSS to attempt to influence the layout, the web would be a much friendlier place. It may not look exactly the same on every browser (which, come to think of it, may be Zeldman's point), but with proper testing, it should look similar on popular browsers, and at least be LEGIBLE on others.
People need to be convinced that the web is not a graphic design medium. That's what PDF files are for. People don't try to build their sites solely from PDF files, because that just wouldn't fly. Instead they try to use the web to achieve the same goal, completely oblivious to the fact that it's a really poor tool for that purpose. Rather than embracing a new paradigm, they try to contort it to look like what they already know. To me, that's just incompetence.
He made no such mistake. He places the burden of interoperability on the producers of the software, not the designers of the sites. You place the burden on the designers, not the producers. From his perspective, the software companies should make sure that their software does not make unnecessary deviations from standard, thus breaking older sites. You think that the designers should predict change and design their sites to take this into account.
I don't know which philosophy is more unreasonable.
Hello,
the world wide web is about what ever you make it. I could make my own meta language that the uses http servers. coming soon- rEml - randomErr markup language. it won't meet your standards, but it meets mine.
forcing everyone to do things your way is so... microsoft.
You say things that offend me and I can deal with it. Can you?
I have to laugh at the assertion "For a beginner, XHTML is easier to learn than HTML precisely because its rules are consistent"--what wishful thinking! XHTML is harder to learn because there are so many more rules. Newbies, even ones who manage to make some interesting content think HTML already has too many rules...
Can someone tell me, is
<b> go and <a href="somelink">click me</a> now</b>
illegal in XHTML? Does it need to be
<b> go and </b><a href="somelink"><b>click me</b></a><b>now</b>
because A HREF tags aren't part of the valid contents of the bold tags?
SO YOU'RE GOING TO DIE: The Comic for Dealing with Death
I work for a mid sized company but I know the web site is very out of date and has incredibly poor content. In my mind I can pinpoint this to one thing. The inability for the people who write content to get it to the site.
I know for fact there is more than enough good stories and photographs in the organization that can be published but most of the technicians who would write it (or at least the first draft) don't have the time to learn a web design program. The solution I believe is a good content management system. I've been looking into Typo3 and a couple of other content management systems. I believe once we make it easy to update then content will be less likely to be obselete.
Content Management Systems are right now the best place I can start introducing open source software at my work. We've looked at Microsoft's Content Management Server which is highly over priced for our needs and its hard to argue with the documentation and self-help community that open source software provides. I know there are other content management systems out there but the point is that for content to stay current publishing capabilities must be pushed to the people who will author it.
My servers' web stats show 96.4% of all browsers visiting the servers are Internet Explorer and/or Netscape. The only thing surprising in this article--other than the clearly fudged percentage sited--is that the author advocates, with a straight-face, that because 3-4% of a site's visitors use incompatible browsers this translates into a 99.9% obsolence rate.
Still, it's always amusing to see someone suit up, gird their horse, and charge at the windmills while proclaiming the revolution.
Rants against Netscape 4 tread well beyond the scope of CSS, but it's commonly known that any webpage that implements a fair amount of CSS1 will not be supported correctly on NN4. Better yet, if the webpage implements ANYTHING from CSS2, it's very likely that Netscape 4 won't support it. And there's much, MUCH more:
NN4 doesn't support <DIV>. It supports <LAYER> instead.
NN4 doesn't like inline styles.
NN4 doesn't fully support the height attribute (e.g., table cells).
NN4 doesn't allow onclick events on every object, such as <img> and <div> (or, layer, if we want to be technically correct).
NN4 uses its own Document Object Model, which results in very poor DOM Level 1 support, and virtually no support for Level 2.
NN4 supports the onunload event, but it does so quite unconventionally. This results in strange behavior when resizing a window: content unloads and refreshes, which is very undesirable for persistent objects, such as applets.
I guess that's a good stopping place. The list goes on, but I hope you see my point. In fact, the word "unconventional" suits NN4 quite well.
Web developers who are serious about dynamic or heavily stylized content will quickly realize that full NN4 support requires either an insane dedication to little hacks and gimmicks or a text-only version of their website. The way to present cross-platform, stylized content today is to use Shockwa^H^H^H^H^H^H^H a plugin.
The fact that 5th and 6th (and now 7th) generation browsers are 95-99% standards compliant means that bleeding-edge content will target newer browsers, and Netscape 4 will be left to rot. Five years is an insane lifespan for a browser, and if you remember correctly, Netscape 4 was just getting off the ground five years ago. Internet life moves at the speed of normal time ^2, so your five years is really like 25.
Maybe I live in a parallel universe, but in my reality, NN4 is already dead. Or, at least it has a really bad case of leprosy.
I want to do a nice little page, and do it in XHTML because it's The Way Of The Future (or I want to display a little math, which only XHTML+MathML allows without resorting to ugly inline images). The tag soup itself isn't a problem, I just close all my tags and make sure the doctype declaration says XHTML instead of HTML, as prescribed by the standard.
However, is this enough? The document is now XML, and therefore should have a <?xml declaration, if only to specify its encoding. Except that said XHTML standard says it is optional if the encoding is UTF-8 or UTF-16, or has been otherwise determined (think HTTP headers), which contradicts the XML standard, sec. 4.3.3, the last two paragraphs, one which says that no declaration and no other information means mandatory UTF-8, and the next one "It is also a fatal error if an XML entity contains no encoding declaration and its content is not legal UTF-8 or UTF-16."
So I need a declaration no matter what. But according to this page about the different layout modes in current browsers, MSIE will react to an XML declaration by switching to "quirks" mode, which is precisely what I wants to avoid by sticking to the standards... And I wouldn't want to lock out 85% of WWW users, wouldn't I?
But wait, this is only if the page was served with a text/html content-type. The right answer would then be to use the standard content-type for XML/XHTML... which should be application/xhtml+xml! Yes, "application"! Now if I use that content-type, all browsers I have at my disposal except Mozilla (MSIE5, Konqueror, Links, Lynx...) either consider the page an application and offer to save it to disk, or display it as-is! Same with the second-best, text/xml.
Okay, am I the only one experiencing this? Any point in not using good-ol' HTML4 and avoid doing (yet another kind of) horrible bugware?
Everyone who I have ever worked with has generated invalid HTML that has made even current browsers crash or behave erratically in different browsers. When I realized that I was also making these mistakes, I finally learned my lesson and started using the W3C validator to make sure my web pages are valid HTML. Since then, I have not had any problem with my pages not working in any browser. This is exactly what Zeldman is asking web developers to do.
What a fool believes, he sees, no wise man has the power to reason away.
it's lost its meaning. It's been degraded by marketing drones and morons to mean 'anything thats not the cutting edge'.
:P
Here's what it means: http://www.dictionary.com/search?q=obsolete
Hell, I still use lynx when all I want to do is snag a tarball. My linux boxes dont even have a GUI. If the content there has meaning, who cares if the web page uses the latest 'nifty tricks'. Is an ASCII text file obsolete? No, not if the information it contains is valid. Is EBSDIC (sic) obsolete? Probably. I cant even remember the acronymn
I'm constantly hearing how my P3 600 is obsolete. There's nothing that doesn't run on it. Hell, I have a router box running a P90.
Is my original NES obsolete? Or my Atari 2600, for that matter? Not as long as I enjoy playing them.
Is a 2001 model vehicle obsolete because the 2002 line is introduced? It does have a bigger cupholder, after all.
If people want to push their agendas, sell whatever they're selling, go for it. Just quit trying to redefine perfectly cromulent words in the english language to do so. Make up new ones, like cromulent. I propose 'obsolastweek' to mean everything that wasn't shrinkwrapped within the last 24 hours.
This article should read "99.9% of websites are obsolastweek because they haven't been redesigned because some propellerhead made a new widget"
Propellerheads (I can use that word because I am one), dont realise the cost of doing business. The world doesn't start over at 0 just because they invented something 'slightly better'.
I don't need no instructions to know how to rock!!!!
Why not VMware? A web developer where my dad works uses this with some copies of Netscape 3-6, and IE 3-6 on his box so he can test and see what the pages look like all on one box.
Do you have any idea how old Netscape 6.0 is?
For goodness's sake, upgrade to Netscape 6.2, 7.0 or Mozilla 1.1! 6.0 is so old and has so many bugs, while 6.2 is almost infinitely more stable/faster/better in rendering.
.... But outside these fault-tolerant environments, the symptoms of disease and decay have already started to appear.
Tell me about it. I just checked my webpage, and all my <br> tags had decayed into <blink> tags....
...who don't understand what HTML is.
You're not supposed to be able to. That's not what HTML does.
HTML is a content language. The whole beauty of it is that the final presentation is NOT THE DESIGNERS RESPONSIBILITY. No web site will look the same on all platforms - that's the point.
The people you are talking about are not 'web designers' - cannot be, because they don't have a clue what the web is. If you cannot accept the fact that your content can be presented different ways (including to blind people) as appropriate to each individual client, you have no business on the web. Make .pdf files or something.
I know someone will interpret this as flamebait, and someone else will probably tell me to 'get with the real world' or the like, but in fact I am just telling you the truth, and I'm quite grounded in the real world. There has been no shortage of people explaining these simple facts about what HTML and the Web are, in simple terms and moderate tones, from the very beginning - and sadly there has been an overabundance of self-styled 'designers' that refuse to understand the medium and insist on trying to make it what they want it to be, instead of what it is. REAL designers work with their medium, they take the time to learn how it works and why, and they produce designs that are appropriate to it, rather than insisting that every media work the way their favourite one does and breaking it every time they touch it. And that is something that every decent art teacher in the world tries to teach his students. Sadly, the students, particularly the ones that go into web design, don't often listen. I'm not trying to pick on you personally, but your clueless post makes an excellent example I must admit.
'Designers' that couldn't be bothered to understand the medium of the web before proceeding to dump their work on it have done great damage to the web, and that's something I happen to care about quite deeply. Your ad-hominen attacks and dismissals of Zeldman aside, he makes a point that is absolutely true, and will have real economic consequences. All that patched up proprietary spaghetti code of mal-formed HTML-abuse IS coming down. While standards compliant pages from the very earliest days of the web still display perfectly in the latest nightly builds of Mozilla, the pages written by people with the philosophy your post shows ARE becoming obsolete, very quickly. In a way, the 'designers' that can't be bothered to learn their medium have won - the new standards will allow them to do what they always wanted to do, and what HTML was never designed to do - to specify layout and 'look and feel' issues. But it will require them to do it in ways that consistent with the underlying philosophy of HTML and the web - something they've never shown any interest in doing before. I expect to hear a lot of whining from that corner in the coming years, but don't look to me for sympathy.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
People tend to knock down geeks who have become popular or well-respected. As for sample chapters, I think they are great! Not only New Riders, but Oreilly does a great job in letting readers sample chapters. What a wonderful thing that anyone can download chapters before a book actually comes out. In book publishing, there is an enormous lag time between assignment of the book and publication date (just look at the review of the blogging book from yesterday). By the time a book comes out, the examples are irrelevant and the standards have changed or improved.
The essay gave a good analysis of tradeoffs that web programmers have to make when planning websites. Some of the code examples here were particularly hilarious (if only because I know my websites have code that is equally ugly). This chapter, as I see it, is not advocating anything radical or controversial; it is merely restating the problem in as dramatic way as possible.
Book Previews reduce the "obsolescence" of technical books. I say, let's have more of them!
rj
Robert Nagle, Idiotprogrammer, Houston
You're talking about forwards compatibility of the HTML code (being able to render properly on future browsers, where the onus of compatibility is on the HTML author).
:) Either term works for this application as long as you are looking from the correct side of the issue.
... that works, but only if you write your code compliant 100% to standards. That means leaving out all the proprietary cruft (which became especially prevalent in the "4.0s" of Netscape and IE) -as well as- all of the stuff that doesn't work in a cross-browser environment.
:).
The parent was talking about backwards compatibility of the browsers (being able to properly render old HTML code in a new browser, where the onus of compatibility is on the browser author).
It's semantics, but I didn't start the nitpick
As for the parent that wanted browsers to be backwards compliant
This is very hard to do if you want interactive sites, or at least was until recently when most browsers began to pay more attention to standards such as the DOM (document object model).
Again, we're back to a very basic problem. Do you write your page to work in old browsers or do you use the latest standards? I'm less concerned with this (as the author of the book seems to be) than I am with the idea of writing code to today's standards and having it work in future browsers.
I as a user understand that I'm taking my experience in to my own hands if I try to load a modern page into Netscape 1.0 (but it is fun some times
However, words can't express my frustration when I have the most modern browsers available and I can't load a page because it was written for an older browser. This happened to me yesterday when trying to sign up for a service from my phone company. The reps kept saying "I see that option, you should have it to". 30 minutes later I decided to load the same page into a 2 year old browser and it worked fine. It had used some tags that were horribly broken, not in any standard, and later abandoned by all involved.
If the modern browsers had had to be compatible with everything since the dawn of the web, they would be twice as large and 4 times as buggy. I would much rather that web authors stick to published standards and not rely on proprietary tags for public pages.
From what I see, this is what the book's author meant by "obsolete" and I agree. Most websites, if locked down and not changed for 3 years, would no longer render in the browsers that are new in 3 years.
While they will naturally work to fix these issues as the new browsers are released, they would not have to if they wrote to the basics. And the problem with fixing things as they evolve is that some pages (like that damned phone company page) get ignored and by the time they're found no one knows how to fix them.
It is more productive to voice thoughtful opinions (reply) than to judge (moderate) others.
More like ...
<Personal Shill mode>So now you all go out and buy my book and your HTML will be cleaner, 20% whiter, your breath will be fresher, and you'll get this lovely set of steak knives</Personal Shill mode>
But the problem is that most designers are NOT following these standards ,they keep using non standard features of the older browsers, thus the software writers now have a dilemma of the own making i grant.
They have two choices, Only render the pages that follow the standards and have 99% of sites non functional in there browser or allow it to work so there browser can be used today.
The only company that could currently force the updating of many sites is our favorite company Microsoft and even then I'm sure there would be resistance to a browser that only followed the standard.
So the burden had to be on the designers of the site to pull them into line with the standard, once the browsers can render strictly to the standard such as mozilla and opra etc.
Last time I checked lynx wasn't going to show images anytime soon
Lynx handles images by using an external program - essentially like a plugin. Plus, for maximum accessibility you should be providing textual alternatives to rich media types anyway - thats a priority 1 checkpoint of WCAG.
So lets all just use HTML 0.1 with only <br> tags and <a> tags. Whine whine whine...!
No. Well structured HTML (as in _this_ is a heading, _this_ is a paragraph, _this_ is a quote), and using CSS to style the presentation (whatever the output destination: screen, printer, aural devices, holograms).
At one point durring the heyday of the .com gold rush, people threw money at companies which claimed the ability to draw increadible proffits at some undetermined point in the future. Some onsider this long term thinking, while others consider it foolishness.
Website designers have learned this lesson well. They strive to serve their business clients by allowing them interact with the largest customer base possible by using clunky non-standard, bandwidth-consuming techniques to get outdated browsers to render their stores in the desied fashion.
You really can't blame website designers for this, nor can you blame site owners. The designers are working to meet their client's requirements, which is to make money, by being accessible to the largest percentage of the available customer base.
The fault, dear brutus, is in ourselves. Website visitors are at fault, for using browsers which promote this non-standard architecture. Certainly no one will use a browser which is strictly standards complient such that any non-standard website would not be visible, because that would diminish the user's internet experience; but this is what's required. We need to force site owners to become standards compliant, which will in turn improve efficiency throughout the net.
If only, bandwidth were more expensive, this problem would already have been fixed, as the bandwidth costs of ineficient non-standard site design would be far mor visible.
It really is a foustian bargain. Reduce revenue by modernizing your website thereby making it inaccessible to older browsers and thus reducing your potential customer base and save money on bandwidth usage, then wait for web users to upgrade their browsers so as to be able to view your site, and build up your custoemr base once again; or, cater to every antiquated browser in existance, so as to maximize your potential customer base, and accept the increased bandwidth costs.
In the long term, with a little short term pain, this problem will be resolved, but in the short term, there really is no good answer.
--CTH
--Got Lists? | Top 95 Star Wars Line
"99.9% of all websites are obsolete."
Ergo,
"0.01% of all websites render nicely in Lynx."
Seems to me there's some confusion between "obsolete" and "usable." Those websites that will be obsolete with fubar 6.x are the same ones that cram a lot of visual shit down your throat, making you work very hard to extract the useful information out of the noise.
Fight designed obsolescence, and write text-based web content with a minimum of static content. Otherwise, don't bitch when fubar 6.x fubars your site.
I use XML output from PHP ran through my XSL stylesheets to produce the final output. The stylesheets get fed the users language and user-agent along with the output and easily produce custom output for all devices without any significant coding. I will be glad when (if?) HTML is finally replaced by good XHTML support but overall keeping up with these things is not difficult if you design your site well. Also since XSL checks the HTML output it produces it eliminates many of the problems commonly found in output code. The biggest problem is trying to deal with user inputted data and that is more of a language problem than a formatting problem.
At what price learning? At what cost wisdom? The price is a man's peace of mind, and the cost is his life.
So? At least if you specify a DTD you are showing that you understand there are issues and have made a rational choice in selecting one.
If you bother to adhere to a particular DTD, the odds on your site being viewable accross a range of browsers increases dramatically.
Yes, you will never get 100% compatibility, but you will get damn close.
If you insist on features outside current DTDs, then use server side browser detection to serve either the site as you intended, or a heavily stripped down totally cross platform version.
.02
cLive ;-)
cLive ;-)
-- Trinity in high heels carrying a whip: The donimatrix - there is no spoonerism
Let me get this striaght... he sez:
- people who adopt IE only standards are stupid because the piss away 25% of potential users.
- people should abandon older standards for W3C
What is logically inconsistant about those two statements?
Authors want and write backwards compatibility in order not to piss off the friggin users who use older browsers! Get a clue pill dude.
IE only extensions force people on other platforms to change platforms. Standards compliant HTML forces people to upgrade their browsers. Which would you rather do??
Fascism should more properly be called corporatism, since it is the merger of state and corporate power.
Yeah, that's right. It was the fault of all those developers who didn't have the forsight to see the standards that would eventually be approved years later. What were they thinking?
It didn't have anything to do with the standards process being slow, or diverging from the needs/demands of the market (HTML 3.0). And even after the standards were finally approved with buy-in from the browser makers, no blame rests with both Microsoft and Netscape for serious bugs in their 4.x browsers, often causing their browsers to crash on many CSS features.
Yep, those developers were at fault. They learned bad techniques, when those techniques were the only way to accomplish what their customers wanted. They continued to use them when the 4.x browsers would crash on standard-based markup. Even after the really serious problems were cleared up in IE5.x, they still used their old tricks. And now, damn them, that 6.x browsers have been available for only a year or so, they haven't redesigned all the world's websites to be fully standards compliant (and broken on 4.x and some 5.x browsers which are still in heavy use).
Yep, if anyone's to blame, it's those developers.
PJRC: Electronic Projects, 8051 Microcontroller Tools
I do not think designers should predict change. I think designers should simply use recent standards and ensure that they adhere to these standards by using validators such as the W3C HTML validator. Absolutely no predicition is necessary!
Please re-read the article, as it is very clear on these points.
What a fool believes, he sees, no wise man has the power to reason away.
I run a fairly sucessful website.
:
Like many businesses money is tight so guess where I'm goign to spend it when it comes to testing, certainly not on Sparc/Solaris9 combination.
So far from 350,000 hits this month I've had the following Browsers
MS Internet Explorer (Versions) 94.9 %
(MSIE/3.xx 0 % MSIE/4.xx 1.9 % MSIE/5.xx 56.6 % MSIE/6.xx 41.3%)
Netscape (Versions)No 2.7 %
(Mozilla/3.xx 1.1 % Mozilla/4.xx 55 % Mozilla/5.xx 43.7 %)
Unknown 1.9 %
Opera 0.3 %
Konqueror 0 %
ANT Fresco 0 %
iCab 0 %
WebCollage (PDA/Phone browser) 0 %
LibWWW 0 %
Microsoft Mobile Explorer (PDA/Phone browser) 0 %
Lynx 0%
Using the following OSs
Windows 37.4 %
Windows 2000 17.3 %
Windows XP 17.1 %
Windows Me 10.9 %
Windows 9.4 %
Windows 4.9 %
Mac OS 1.2 %
Unknown 1.1 %
Linux 0.2 %
Sun Solaris 0.1 %
HP Unix 0 %
Warp OS/2 0 %
Windows 3.xx 0 %
OSF Unix 0 %
Irix 0%
RISC_OS_4.03 0%
Thats at least 15 browers on 17 OSs.
How am I supposed to test my pages for all those expectations?
My HTML passes 4.01 Validation but I can't be sure it displays on those browsers.
I know it displays in Lynx okay so that's about the best I can offer.
I've had one email in the past year saying 'your site doesn't display properly' and that was IE on NT4. A product I can't buy and test with even if I wanted to. (except through warez of course). HP Unix presents a better challenge.
What would you suggest is my *obvious* solution?
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
"if yahoo was just a single column of words and link, it would have the same use of it currently does. No way,"
Have a look at http://www.paulgraham.com/ for "layout" considerations. Certainly struck me as different, and very usable.
"yahoo still is tring to support non-CSS enabled browsers, why should they write two versions"
Which bit of what-CSS-is did you not read? Nobody *has* to support it - I regularly browse in links-2.x, myself, which doesn't, and standards-compliant sites look just as good without. CSS is *exactly* what they want.
~Tim
--
Rushing on down to the circle of the turn
But actually Mozilla is still quite forgiving as long as you don't specify a doctype you haven't actually written to. IE isn't really so bad, either.
HTML should return to its original strength of simplicity. HTML code should have a minimum of noise, and maximize content. Good design is not the same thing as gimmicky fluff. Plan words with a few pictures can tell almost any story worth telling among human beings. Good design gets out of the way and lets the words and pictures speak. It's only the gimmicks which go obsolete, never good design.
"with their freedom lost all virtue lose" - Milton
Well, if you coded to standards the first place (with CSS, try relative positioning, and define the "left" attribute by how many pixels to the left you want it to be) it would look great with Mozilla, IE4-IE6 and Opera 5-6. ..
You miss my point. Yes, I know there are other solutions. It doesn't matter. Twice recently I've gone to pitch to new clients and naturally they've wanted to look at websites we've designed using their own computers, and those computers have had Netscape 4.7 on them. What do I do? Do I say, "Hey, can you wait ten minutes whilst we install a new browser on your machine? Our sites will look shit in Netscape 4.7 but it's your fault because nobody else uses that these days. It's really easy to install another"??? No, personally I prefer to breathe a sigh of relief that I've designed the site in a way that works on all browsers, and win the job.
Guess which one Slashdot chose.
:
Well let's see
Attractive - nop, butt ugly
Compatible - well I can't test it on multiple platofrms but according to http://validator.w3.org the (logged in) homepage has over 100 HTML errors.
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
The font tag, used properly, CAN BE a great tool. Many HTML-generating programs don't use it properly. First, they don't specify just the elements they want in the font, they specify all the optional values, they specify absolute font sizes, and so on. Some wrap every line of text in their own font tag. Some programs wrap every line in EXACTLY THE SAME font tag. That's where it is wasteful.
The gripe with Yahoo! and others is (if you look at their code) every line of text has a font tag around it. Most of them are size=+1 and size=-1, and several lines that say: <small> <small> <br> </small> </small> . in fact, if you look at the yahoo! main page today, there are over 130 pairs of font tags, and about 200 words. That would generally be considered bad code-to-content ratio for formatting text.
Those are what is fairly wasteful.
As for the comment that they are standard, they are not. Tags like <big> and <small> are standard, but <font> was removed from the standard when CSS was introduced several years ago. The way you are supposed to do it is: <P style="font-size: 95%; color: blue" > or use whatever style tags you want. As part of the style, you could specify any of a number of font faces or styles. In this case, the paragraph should have a slightly smaller than normal font, in blue.
Frob.
//TODO: Think of witty sig statement
Because neither Netscape or IE meet the stringent requirements of standards compliance needed to force site owners to bring their sites into compliance.
The key to such standards compliance is NOT permitting display on non-compliant sites, but you're right that it would be tough to get users to adopt a browser which intentionally prevented access to non-standards compliant sites.
--CTH
--Got Lists? | Top 95 Star Wars Line
99.9% of new web browsers are obese. What this guy essentially wants is for me to upgrade my browser so he can use his fancy tricks. But he doesn't want to help put pressure on browser makers to get their browser under 8MB.
now we need to go OSS in diesel cars
The problems this article discusses were created by browser implementation problems and limits of the earlier HTML versions. Netscape in particlar, was terrible to write HTML for. While Microsoft actively rev'ed IE, Netscape did little, and the problems of building Web sites to support the current users increased. When NS6 arrived, it was actually worse than NS4!
There is much redundant code because NS and to a lesser degree, IE, didn't do things like inheritance of formats correctly. Developers were forced to try various hacks until they found something that worked. Having gone through the pain, and with new stuff to do, the developers were not willing to remove what worked. Browser developers made certain that the old pages worked, even if they were incorrect, because to fail to do so was to lose users and gain a terrific amount of ridicule in various publications and online sites (including Slashdot).
The issue is if you run a public Web site, you have to support what the public has, not what is convenient for the developer. And the public takes time to update their browsers. The pace of update has quickened over the last 12 months, but before that you had to code for NS4.0x or some real per centage of users couldn't visit your site. IN particular, the South American and other foreign markets were very slow to upgrade their browsers. Sites like Yahoo, who are truly global, must support just about all of the terrible, broken browsers that exist.
With the cutbacks in IT spending, little money exists to make changes to Web sites that are not absolutely required. Changes are made to fix terrible problems and do things to bring in new revenue. That is it. I also think this author really underestimates the effort to build a great site that supports all the required browsers and is cmpleeing to users. Anyone can make a home page, making a great site is hard and expensive. Look how few great sites there are.
I noticed the same thing in Netscape 4.78. I'm not sure if he is trying to make a point (though he doesn't say anything) or if his own website is "obsolete" also.
It is by the juice of the coffee bean that thoughts acquire speed, the teeth acquire stains. The stains become a warning
Now we get to the web where in many cases the presentation is creative and graphical. Zeldman gives an example a site's "Join Now" text being inside font tags which are inside a table. He suggests replacing this with <h2>Join Now</h2> and add CSS stuff for h2. This is wrong! The text is not a heading, and <h2> should not be used. It is simply some text. The correct markup way to do this is to define a class p.joinnow in the CSS, and in the body use <P CLASS="joinnow">Join Now</P>.
This is utter insanity. You could end up with a separate class for almost every element. A markup language just doesn't make sense for complex graphical presentation where most elements are one-of-a-kind. It is just wrong. Most of the web doesn't fall into this category, and html-css is the right way to describe the content, but some of it does, and trying to squish it into a markup language causes many of the problems we see now.
And most of the time, it's not doing anything useful. Who needs that "abstraction"? Programs that do something with web pages other than render them (and I've written a few) don't find this stuff helpful, because you can't rely on it.
What we needed was a standardized way to download fonts, not all this CSS crap.
That's what HTML is supposed to be, not what it is. I agree that it would be sort of nice if everyone could use the standards (not so easy, since browser support is poor) and use it as a content-markup langauge, but really we want to just make pages that look nice and display correctly. Standards don't help with that until they're implemented and working. HTML right now is a loosely collected set of folklore about what you can do to get a consistent look across browsers.
The last time I checked, HTML is still an SGML application.
HTML 4.01 is an SGML application. However, the newer versions of HTML are XML applications. They go by the name XHTML.
Will I retire or break 10K?
Can you say "Old Machines" and "Default Install"?
.... I finally got semantic layout/CSS when I saw it, and the upside was that it degrades perfectly.
On some computers, you'd be insane to even try running a 5.x browser. Browsers are some of the most memory and processor hungry applications your average user can come across. On one of my machines (and old Performa 6116), even running a 4.x browser is foolish. I'm using Netscape 3.x when I run anything on that thing. Or Lynx. Same with a Win 95 PC.
The real point shouldn't be to abandon older browsers: it should be to fight layout complexity. XHTML + CSS is a wonderful tool in this game. I love Blue Robot
Wisely designing to standards doesn't abandon the older browser... rather, it means you can use Netscape 1.0 or Mosaic or Lynx to use a page, and when you use Mozilla 1.0, you get all the purty layout and bells and whistles.
Libertarianism is rich wolves and poor sheep playing gambler's ruin for dinner.
In different words, browsers will just have to deal with it. New information becomes old information, and new media become old media. Unless there is a really good reason, tools for accessing the old information better be able to cope with it. Sorry, guys, but "bad HTML" is here to stay. Maybe the badness can be isolated by making it a separate program that gets invoked by browsers when needed and translates bad, old HTML to shiny, new HTML.
One thing web designers need to do is get off this "consistent look and feel" drug. It doesn't work that way because it is not supposed to work that way. For example in HTML, the <h1> tag starts a header at level 1. That doesn't mean the enclosed text has to be rendered in some particular size or some particular font or some particular color. It just means it is a header at level 1. The look and feel is up to the browser or the windowing system of the user. Even applications that get ported between different systems (for example on both Microsoft Windows and on Macintosh) are supposed to have the look and feel of the respective system so the user can access it in their own familiar way. When I go from one web site to another and find the buttons are always different, that is where consistent look and feel has failed.
now we need to go OSS in diesel cars
(* [so you can't insure consistent look and feel without kludges.] You're not supposed to be able to. That's not what HTML does. *)
I have told the boss(es) multiple times what the tradeoffs were. They didn't care. They were focused on the here and now and target what the CEO will see, and NOT what the consumer will see because the CEO has more pull than comsumers WRT career advancement.
What PHB's *really* want is a coordinate-based brochure-building language/protocol. The problem is that it is really really tough to design a brochure that will look okay on multiple monitor sizes unless you make a seperate design for each size, which is expensive. Plus, most browsers don't tell their screen/window size because it would be a privacy violation (small == poor in some e-store's minds).
If you do it right, it would be very expensive, and they don't want to fork over the bucks. The choices are:
1. Do it right.
2. Hack it up so it looks pretty in majority of browsers, screwing the rest.
3. Keep it simple (lowest common denominator)
To PHB's, #1 is too expensive. #3 does not please the CEO (who only checks it in his/her own browser). So, they usually go with #2.
I don't know if *any* technology can be sufficient. It is a people problem, as the parent attests.
Whenever the tradeoffs are sufficiently complex, PHB's screw it up badly. That is a fundimental rule of business that we must learn to live with.
Table-ized A.I.
Who on earth is running a browser earlier than 4.x?
Me. Lynx anyone? Not anyone around here who uses a shell is there? Also, old Macs - SE, SE30, etc - can dialup, and there are ethernet adapters for them. They make good, cheap, space-saving machines for simple access. Use Nifty Telnet for shell access, older versions of Fetch and Netscape 2.0.
But the important messge here is that:
The web is about content, not format.
Remember this. The whole point to html is that it's a *markup* language, not a *forced formatting* language. The browser takes the content and displays it in the manner of the user's choosing.
This seems to have been lost in the corporatization and control of the 'net.
Remember the good old days? When the web was about content and not about spam and marketing? That's where I live. I don't want to see blinking and flashing and animated ads and popups. If I can't see your content on lynx or with a 4.x or pre 4.x browser, you have lost my eyeballs and any potential to recieve my money. No popups on lynx.
The same goes for html formatted mail (there is a special place in hell reserved for people who send html formatted mail.) If I can't read it in pine, I don't even care what it says. Send me text if you want me to read it. (No web bugs and stuff that way too.)
In short, the goal is to get your content to other people, stop being such control freaks about how it is displayed. Write to the lowest common denominator, be creative with what is available there and you save much time, aggravation and money. -- And I'll be able to see your content.
NEVER FORGET --
The web is about content, not format.
Join the Any Browser Campaign and make your pages 'content enhanced'.
War is Peace. Freedom is Slavery. Ignorance is Strength. - George Orwell or George Bush?
The clients of my HTML application(s) are primarily school systems. Big rich ones with the latest greatest mostest wonderfullest hardware that money can buy (or that corporations can donate) down to dirt-poor schools with 3 Mac 030's in the back running the Oldest Browsers Known To Man. My job is to insure they they ALL can properly access the system. It is not my job to tell my clients "you have to upgrade or you can't play". I'm not being paid for that. I'm being paid to develop a system they can all use as-is.
I guess my argument with Zeldman's "conform to the standards or die" approach boils down to the fact that the browsers used by my clients often do not conform to the standards. Hey, it would be nice to be able to use CSS or XHTML. I'd love to. Make my life a WHOLE lot easier. But then I'm not meeting the requirements of my clients, which is the whole reason I'm doing this in the first place.
--matt
The websites I design contain links to the W3C HTML and CSS validators. The links might look something like this
XHTML 1.0 CSS
and I put them in the site template, so they appear on every page. These are referer links, which mean that they check the page you are linking from. When I finish making changes to a page, I click those links in sequence, and if my page doesn't pass, I fix the XHTML or CSS that's causing the problem.
Depending on the type of page, I might make them bold and obvious, with the checkmark graphics that W3C offers, or I might hook them to a bullet or a period so they're obscure and don't become a design element.
I use absolute positioning to do layout that people often do with tables, and my sites look fine in anything from IE to lynx to Mozilla.
Ellen
mods metamodded as "Unfair"
My web site is designed for reading
NN4, IE5, IE4, WebTV, etc. my customers use them. I hate jumping through hoops, but at this point we can write an HTML 4.01 transition + CSS 1 page that renders perfectly on IE and Mozilla browsers, really well on Netscape 4, and should degrade nicely on WebTV.
The more people using a modern browser, the better an experience the users will have. My site isn't big enough to warrant separate NN4 pages, just separate stylesheets.
However, people like you help me out. I won't pass on their business, it isn't my place to tell them what to use. Webmasters like yourself make the web more painful for NN4 users. HOPEFULLY that will cause them to upgrade (although it is more likely that they'll stop webbrowsing, which would suck), but who knows.
I'll leave the upgrade war to others.
Alex
Hello, world!
I heard Zeldman speak at a conference, and he was quite aware that in many situations it wasn't possible to code to standards (if you had to worry about layout in older browsers, etc.) He knows that the job ultimately dictates what choices you have to make.
I think he feels his role is to try to push folks towards something better than what we're used to working with. When you're leading the charge like that, you don't emphasize everything you *can't* do yet.
Personally, while I'm not going to ditch stuff like table-based layout for quite a while, smaller changes like moving to CSS for fonts can offer a substantial benefit in the short term.
That said, I can tell you from several years experience in high end web development that it just isn't that difficult to write a web server piece in any number of back end languages (CGI, Perl, PHP, Python, JSP, Server Side Java Script, ASP, to name the main ones...) that first checks the incoming http request and then responds with content tuned to match the rendering abilities of 90-95% of the user agent variations out there. It's even easier and bandwidth saving than using "graceful degradation techniques", because it relies on template programming to "pick" the right page to be returned. For example, if the "user agent sniffer" detects a Browser supports CSS? the back end program should use page X built from the CSS template... Alternatively, if the user agent only supports , use the font only template... Etc. Etc. Etc. and this can be taken to ridiculous extremes.
However, most variations of this technique which I have seen essentially require the page author to make numerous versions of the same page, which is also a waste of time and data space. Fortunately, there is a middle road, which is to design a site so that the content is author driven, but the formatting is programmatically driven. [If this sounds like I am describing something alot like the /. code then you're catching on...]
So I have no mercy in my analysis for large companies who don't do this on their content driven sites, nor patience for authors who seem to think that the latest greatest US-centric browsers are what we should all be writing to.
...Open Source isn't the only answer -- but it's almost always a better value than the alternatives...
Unfortunately, the industry really doesn't have time to wait for standards bodies. This isn't just in browsers - TCP/IP itself was a hacked together protocol that became a standard while ISO spent years devising and getting people to implement OSI and by the time they had it working, they were 5 years too late and everyone was using TCP/IP. I won't even bother mentioning DECnet - OK, I did - it was closed source and short lived.
One of the reasons many game developers only support M$'s proprietary Direct3D technology is because the standards body for OpenGL takes too long to get new features in (although GL extensions and the new pluggable APIs from nVidia and ATI may solve this problem).
A lot can be said about being first to market. If you have enough of a head start, you really can leverage the industry (although occasionally Microsoft comes around and uses their other overpriced products to destroy your market by offering theirs for free).
Just as Ford defines what a Taurus or a Focus is and does, so does MS as to what IE is and does. The internet is a celebration of individuality. Everyone has their own 'way' of doing things.
Particularly Microsoft. I applaud their attempts to encourage individuality by setting their own standards. This proves that Bill Gates loves us all (I think).
The Internet didn't become what it was today through standardization- thank God that pesky TCP/IP plan never took off.
"Slashdot - News and Chat Sites Deviant". (Click "homepage" link above for details).
The consistency should be between different web sites when viewing content on the same browser with the same OS under control of the same theme. Don't assume that <h1> gets you 18 point times new roman, even on Windows in Explorer, because as you mention it being the default, it is certainly changable, and may well not be the default when you change browser or even OS (though such changes should only happen when you change whatever is appropriate for a look and feel change).
Many web designers come from the graphical arts field, which is full of brochure-on-paper production. They get to pick (or are told specifically) the paper size, font theme, color theme, etc. That's a fixed environment and quite easy to work in. Now along comes the web, and it isn't that way anymore. Some people have smaller browser windows, or are limited to 640x480, or to 256 colors. Others have 1600x1200 and 16 million or even 1 to 4 billion colors. Some have 28.8k or worse download speeds and others have T1 and better. Some have plenty of CPU power and RAM to run the latest browser, and some are stuck with an older browser on a hand-me-down computer. The world is full of diversity and these brochureware designers still can't cope with it. They could build the whole site in a big image file and piss off the 28.8k dialup users and be scrunched over on the left side of the 1600x1200 users. Even with advances in new protocols and formats, so many problems remain that their assumptions still don't work. They might be better off with PDF, which was designed more for the kind of thing they are doing.
now we need to go OSS in diesel cars
However, when the HTML is standard, it's a bug in the browser, which needs to be addressed.
Your logic is flawless, but notice where you're left now.
The browser is branded buggy and non-compliant.
Say the browser is IE 4 or Netscape 4.
Great - the browser creators come out with a new version of the browser that fixes those bugs.
IE 6 and Netscape 6 are in greater compliance with standardized HTML 4.01, CSS, DOM, etc.
Now you come to the end of the road:
"Provided by the management for your protection."
Hum? What do you mean? How is XSLT in any way a replacement for CSS? You might use XSLT to get from XML to XHTML, but it's not for adding style.
If you meant XML:FO, well, that's horrible for the web. Most XML documents have no real meaning to a UA; they're just an arbitary collection of nodes, so unless your FO's cover every possible media type, the web becomes conciderably less accessable.
Given the author's recommendation of FrontPage (ugh) as well as designing for a particular display resolution (resolution should be irrelevant), I'd think that throws the authoritativeness of the entire page into question.
20 January 2017: the End of an Error.
Zeldman, why don't you drive on down to friendly South Carolina, and help me upgrade all the teachers, principals and parents from Netscape 4.02 or somewhere around there to a browser that doesn't rely on the font and table tag to get decent layout.
Give me a break. Netscape 4.x supports CSS. You can easily write a HTML 4.01 strict doctype page that works in Netscape 4 WITHOUT A SINGLE FONT TAG. The variations from HTML 4.01 that are needed to support browsers back to Netscape 4 are TRIVIAL.
Zeldman is RIGHT.
You mean, we request that the majority of weirdos who choose to ever upgrade their web-browser or OS stop doing that at once. Maybe we should forbid any further development of browsers too at the same time?
Badly written websites dont merely fail on non-standard browsers, they also often fail on every new release of IE or Netscape.
But, *if* you code to standard, then it should work with every browser, and it should work with every new release, and if it doesnt, it's the browsers fault, not yours, and you can tell whatever minority of upgraders or weird-browsers to go whine at their vendor instead.
This works in most cases, but not all. Netscape 4.* treatement of CSS is so awful, and degrades so poorly, that I simply had to make a choice knowing that my stuff would be ugly for NS4 users. But for the most part, Conservative and Standard works.
I guess that this makes my site (shameless plug) one of the 0.1% of sites with old content that isn't "obsolete" (well, the content may be).
Prime numbers are exactly what Alan Greenspan says they are -S. Minsky
Look at the code Frontpage spews. Many people have snickered at me for using text editors to edit HTML. Then they fire up Frontpage. Later, I check out a file they have editted and find all sorts of goofy tags in it. So, it ain't just the browsers. Some of the authoring tools intended to dumb down the process of designing and maintaining web pages share the blame for the garbage code. Yessir.
Wansu, th' chinese sailor
They need to use the right tools for the job. If the job says make it look exactly like this paper brochure, then they have to choose something like a giant image or PDF. Sure, it can be approximated in HTML, but they need to not whine when it fails because a different environment with a different look and feel ends up showing it differently. Do you know how many pixels wide an 18pt font is on my display?
As for XHTML and CSS, they might not be there. Sure, feel free to use them where they are, but don't expect me to have to deal with the problems of some new browser just for your benefit. What if the CEO's cousin's computer ... the one that is so old that it only has 256 colors ... can't even run the new browser at all, and therefore has no XHTML or CSS?
If the designer loses his job working for a PHB, at least now he has a chance to get a decent one. Sure, I can understand he would rather keep the job and make a real honest attempt to accomplish the task in a way that keeps everyone in the company and West BFE happy. But who is to blame when someone reports to the company that the site shows up lousy on some other computer?
You can design for the middle or you can design for the CEO's cousin. If management makes the decision, then the web designer is just doing what he can with what he's allowed to do. The real culprits are the ones that don't allow the web designer to design it as content.
Do you see any XHTML or CSS on http://linuxhomepage.com/? Does it look on your browser today like it looked in Netscape 3 back when I originally designed it?
now we need to go OSS in diesel cars
I know of at least one place that uses Lynx on terminals for multiple clerks to enter orders over the net! Last I heard, they provided about eighteen percent of the supplier's revenue. The supplier has tried many times to get the client to move away from Lynx, even to the point of giving them a computer and program to try out. No dice. Now go tell the CIO of the supplier about Zeldman and watch him sputter and fume about having to include 1.0 standards for one client!
Yes, times do change, and standards do change. But everything is according to the standard. The site is PHP driven, and PHP doesn't seem to handle DTDs. But it works now because the default HTML still works.
When will features in HTML be obsoleted from the standard and removed from browsers? I don't know yet. That time has not come. Hopefully there will be an adequate transition period to a decent browser that can do the new replacement features before expecting everyone to get off the old browsers. Unfortunately, the browser situation is getting worse, not better. What browser no longer accepts HTML 4.0 or even HTML 3.2? If you find one, then that browser has jumped the gun because HTML is not yet depricated. But those sites you mention using XHTML and CSS come up a total mess in NS 4 because unlike the goals for XHTML and CSS, they don't work in NS 4 (CSS is there, but it's horribly broken). When a new browser can do some of the things (including X standards) that NS 4 can do, then let's talk about browser upgrades. But you have to help try to get decent browsers implemented and deployed if you don't want to have to face people who are having to make hard choices to find the least-broken browser. I don't deny NS 4 is broken. It just happens to be the least broken right now. Konquerer appears to be poised to edge it out soon. Maybe the next version will. One can only hope. Still, Konquerer is more obese than I would like.
If you at least admit that browser programmers are doing a bad job of getting standards out to users, then we'll at least have something to agree on.
now we need to go OSS in diesel cars
Yeah, if the W3C standard ment a damn. Microsoft co-writes the standard, and then writes a non-compliant browser. Unforunately, since they have 99.9% of the market, then 99.9% of the websites will be "obsolete". I render my web site with the W3C standard, yet I have to deal with IE's EXTREMELY annoying lack of support for CSS and proper JavaScript. (I've had to deal with plenty of "Works in Netscape, but not IE" bugs.) Nevermind that this property isn't supported in CSS or that command in JavaScript reports something much different in IE than the web standard.
Give Netscape all the grief you want, but at least they stuck to standards. Propeirtary tags be damned because all of the real tags were there and did when they needed to do.
(Not to mention that the W3C validator is extremely anal about "obsolete" tags. Maybe I want a FONT tag in one piece of text that I'm never going to change. Why do I need a CSS name for EVERYTHING?)
Zodiac Survey
(Not to mention that the W3C validator is extremely anal about "obsolete" tags. Maybe I want a FONT tag in one piece of text that I'm never going to change. Why do I need a CSS name for EVERYTHING?)
Use a SPAN tag with a STYLE attribute, e.g. <SPAN STYLE="font-size: 125%">
I pine for the days when we just had bullet lists of black text against gray backgrounds.
No images, no tables (which are just abused as layout devices), no background colors, no fonts, no frames, no Javascript, no popups.
Just bare-bones INFORMATION, boring to look at but useful and efficient. Hardly a byte that wasn't relevant, and easily rendered on any browser you can imagine.
Always keep a sapphire in your mind
Lite browsers do make up less then %10 of your visitors. However if you owned a small bussiness, would you tell 1 out of every 10 or 1 out of every 15 customers to leave or go to hell?
Sure bandwith and development costs money but turning away potiental customers also costs money and probably alot more of it. I heard the same argument for years on why major software companies should support the mac platform. Microsoft in return would make their own tools more proprietary to make software more expensive to port. However most of the big major players in software have mac versions of their products for this reason. I do not mean games but MS-Office, quicken, turbo tax, photoshop, corel photopaint, word perfect, etc. Why turn away potential customers?
Also if yahoo wastes gigs of data a day but pays for itself in just minutes then its worth it.
I do agree on convulted html. Yes html can be ugly if not coded properly. If I were the owner of yahoo, I would plan to redesign the whole website with easy to read css after the majority browsers that support it are %97 or %98. I do not know if the palm's browser supports css(i assume not) but that may be the future of internet browsing as consumers buy more appliances with computers inside them.
http://saveie6.com/
So I should limit myself to design in the areas that are supported by all of those browsers only?
If you look at the charts with an open mind, you can see quite clearly that there are large areas of compatability with Netscape 4.x and more modern browsers. The bad actor is IE 3, not Netscape 4 as you stated in your previous message.
If you really do have a client base that really requires you maintain support for IE 3, you have my sympathies. But the reality is that IE 3 accounts for less than 0.1% of web traffic today. Most stats sites aren't even tracking this browser any more. It simply does not make sense to be writing IE 3 code when the same code causes problems with the other 99.9% of your users.
Myself, I work as a web programmer at an multimedia agency, and we have not had a single client ask for IE 3 compatability for 3 years now. For two years we have been building professional quality sites using CSS that are compatable with all post IE 3 browsers using HTML 4.01 strict with a very small number of tweeks for Netscape 4.x. Eventually I imagine we will drop Netscape 4.x compatability, but the fact is that right now it only causes us minor problems.
At this point if we had a client that mandated IE 3 compatability, we would build a separate set of pages and charge the client accordingly.
The remaining .1% of websites which are not "obsolete" and "convoluted code" includes the following sites:
;-)
www.zeldman.com
That is it. All other sites are bad. End of story. Well, probably the author believes this at least
There is an interesting website called css/edge which attempts to explore all the possibilities of doing "neat stuff" using only standards-compliant HTML and CSS. There are some really stunning demonstrations, such as the complexspiral demo. This demo shows a page that has a two column menu/content layout, complete with alpha-blended translucent background that seamlessly glides over a fixed background image as you scroll, with the translucency changing for mouse-over events on the buttons. The text size can be gracefully sized, and the layout works for any window size. This is done only with pure HTML, a stylesheet, and four JPGs -- no javascript, alpha-channel PNGs, half-screen GIFs, etc.
Thing is, if you visit this site with Internet Explorer on Windows, the above demo and most of the other demos look like crap. This really opened my eyes to IE's lack of CSS conformance. But visit the page in Mozilla, Konquerer 3, or IE5/mac, and it's beautiful.
Very well put. The popularity of Google's interface further backs your claim. However, I pay the bills and keep a roof over my head with my design and development skills. Most of the people who are signing my checks live in the corporate "ooooh... look... shiny things" world. They like cute, million dollar Superbowl commercials and they want to hang their name on a website that dazzles them. You have to remember, marketing is about packaging, not content. Call me (and the legions like me) a sellout, but man... Ramen noodles every night for dinner ain't the life I want.
Entrepreneur : (noun), French for "unemployed"
A couple of points.
i =h ttp://www.thebigchoice.com/
Although http://www.thebigchoice.com/ validates as good html 4.01, the CSS file it uses doesn't validate as correct CSS:
http://jigsaw.w3.org/css-validator/validator?ur
You might want to make sure both your HTML and CSS are correct. cyan and darkblue are not valid standard colors. Try aqua and navy respectively or use full rgb values.
Also, the site seems to be hardcoded to use X pixels horizontally. This could annoy users of devices that can only display way less than X pixels horizontally.
Also, it could actually annoy future users with very high resolution displays because the whole website will take up a tiny fraction of the screen. I see there are already 3840x2200 22" monitors. This would make your website appear about 5 inches wide on that screen at full resolution.
We disagree on which browser is the current best. Mozilla is not working out for me. It is noticeably slower than NS 4. And NS 4 was too slow for me compared to NS 3 until I upgraded CPU and RAM. Mozilla will have to wait until my next hardware upgrade. Mozilla also has some bugs whereby it won't even function correctly in my desktop system at all. NS 4 had problems but I found ways to work around it. Those ways don't work with Mozilla, and until I have the time to deal with it, it won't be my default browser.
now we need to go OSS in diesel cars
we instead request that the minority of weirdos who choose to use non-standard browsers simply start using either IE or Netscape?
What makes you think that only browsers are allowed to use the World Wide Web? Surely Googlebot and another scripts have as much right to use the information on the World Wide Web? Why not approach Google and tell them only to use IE or Netscape when indexing websites and see how long it takes them to stop laughing.
I ignore the CSS, FONT and other stuff and stick to H, H1, H2 and P tags for most everything.
Sounds like you know what structured markup is. Why not use CSS to suggest the presentation for browsers that can tolerate it, and keep the same HTML for the rest of the browsers.
Its the old Web authoring rule: Start with valid HTML, then enhance.
Did it ever occur to this guy that perhaps it's 5 percent of browsers that are obsolete, rather than 99.9 percent of web sites?
Probably not, since that 5 percent includes Googlebot which brings in traffic to websites; Yahoo editors who add your site to those directories which again brings traffic to your website; Pocket Internet Explorer that runs on the Pocket PC (so you can read your mail while waiting for the Tube), cell phone to check directions to tonights game.
Maybe you should block Google and Yahoo from traversing the net -- all in the interests of the other 95% of your audience?
I have P3 800, so shouldn't it be even faster for me? You'd think. Maybe it's just the rapidity of my vision. Even on this P3 800 I can see a difference in the speed between NS 3 doing rendering and NS 4. It's just that on my previous Celeron 333 the NS 4 speed was not tolerable, so I used NS 3. I remember doing some timing tests way back then between NS 3 and NS 4 on some complex HTML tables, and NS 4 was 22.7 times slower than NS 3 at rendering them (the page was made sufficiently complex that it took NS 3 a total of 3 seconds to render, while NS 4 took 68 seconds. That was just quantifying what I was actually seeing visually on much less complex pages.
NS 4 is still slow at rendering multi-block GIFs. NS 3 did them nearly instantly. Maybe this is because NS 4 has too much per-block startup overhead. Or maybe it is because NS 4 mistakenly thinks a multi-block GIF is an animated GIF and artificially inserts a delay. Download this image [184,565 bytes] to disk first, so you aren't seeing any network delay, then test some browsers on it referencing the local file and see how quickly it renders (BTW, certain older builds of Moz and Konq won't even render it at all). You'll need to do this on X running in 24 bit or more per pixel, with a video card that can do that (what can't these days).
now we need to go OSS in diesel cars
CSS didn't even exist way back then (it was being talked about as a great solution). Even today, CSS is broken in Netscape 4, and not everyone can, or is willing, to upgrade beyond that, yet. As soon as some leaner browser shows up, I might. As soon as 99% of people have upgraded, then I will switch to using CSS.
now we need to go OSS in diesel cars
Even today, CSS is broken in Netscape 4,
Netscape 4 doesn't support the @import directive, so why not hide the stylesheet with
<style type="text/css">
@import url(/path/style.css);
</style>
Then Netscape 4 doesn't see the stylesheet, but other CSS supporting pages can use the stylesheet.
Tell it to Zeldman. I'm not the one who is doing CSS. I've put that off until at least 98% of users are using a browser that does CSS correctly so I don't have to do hacks.
now we need to go OSS in diesel cars
The problem is that it's not backwards-compatible. Sure, everybody should be using HTML/4.0, but it was only recently (correctly) implemented with today's browsers, so a FONT tag (and other "obsolete" tags) is still useful for the 4.0 browsers and such.
Zodiac Survey