Slashdot Mirror
Future for Web Standards Pondered
An anonymous reader writes "With the next version of Internet Explorer tied to the release of longhorn, and still years off, what hope is there for innovation in CSS, SVG, XHTML and other web standards? Is the future of the web similarly tied to Internet Explorer and Longhorn? This article ponders this gloomy future, and sees a ray or two of hope."
i hope that windows just uses mozilla firebird. it would be best if everyone could just simply code for the superior web browsing application. Though adware would become a fast issue due to the open source....
With the next version of Internet Explorer tied to the release of longhorn, and still years off, what hope is there for innovation in CSS, SVG, XHTML and other web standards?
None. Zip. Zero. Diddly squat.
<insert witty linux comment here>
Site is already slowing...
Plus ca change
In a recent post I reminisced about the early days of CSS, and a few of the people I recall as influential and important in the development of a standards based web.
But usually I am the kind of person who looks to the future. In the last few months Microsoft made a couple of very significant announcements with possibly quite negative implications for the future of a standards based web. Which has me thinking about that future, and wondering whether there even is such future.
Since the release of Netscape and Internet Explorer 4, there has been a steady movement toward the idea of standards based web development. In some respects the innovation both in the underlying standards and their implementation has been quite extraordinary. But as the kids in the back seat are always asking "Are we there yet"?
In a sense, there is no "there". Perhaps plateaus or way stations along the way, but no final destination. Right now it may seem like we are at one of those way stations. A reasonably large subset of CSS2 (soon to become CSS2.1) is quite well supported by most browsers.
CSS and xhtml support are markedly improved since the early parts of this decade.
But is it a way station, or are we just stalled?
Microsoft has in the last few months both discontinued IE for the Macintosh altogether, and let it be known there will be no new IE for today's generation of Windows based computers. The next iteration of IE will be solely for "longhorn" based systems (longhorn being the code name for the successor to Windows XP). Any such systems are unlikely before 2006, leaving a several year hiatus between major upgrades for IE, the single most pervasive web platform by a long way. And at present the platform with the most web standards "issues".
Which makes wonder - will we see standards based innovation in future?
Who cares about standards?
When it comes to commercial competition, standards are the friend of those without market dominance. The dominant player sets the "industry standard", as companies who dominate their niche tend to describe their software.
I believe that during the second half of the 1990s, during the most innovative time of the development CSS, commercial considerations did not play a significant part either in the development of CSS or in its implementation in browsers. CSS flew below the radar at Microsoft and Netscape/AOL/Time Warner. That won't happen again.
So what might the future hold? Let's turn the browsers for a moment. What happens here will determine what happens with CSS and standards more generally.
Where are we now?
Internet Explorer 6
When Microsoft did not dominate the browser market, open standards leveled the paying field for them. But now with IE dominant, will Microsoft be so supportive of standards?
Internet Explorer 6 is for Windows only. It supports much of CSS 2.1 though support for attribute based selectors, and more sophisticated selectors in general, such as the child selector is limited. It has some serious issues with the box model and positioning which cause many developers considerable frustration.
As noted before, IE 6 is the last version of IE which will be available until probably mid 2006, perhaps later, and the next version will never work on today's computers, not even on XP.
It's the end of the road for IE as we know it.
So, if things stay as they are, with Internet Explorer the benchmark, then say goodbye to CSS innovation for a long long time.
There are number of things which may affect this. First, CSS's design to allow forward compatibility means the user experience for more advanced browsers can be enhanced without compromising the experience of IE users. And there is even a simple way of hiding things from IE, using the child selector, which no version of IE on windows supports.
If not IE, who will innovate?
Opera? Mozilla? Anyone?
The more important question is who will innovate on the
1) I believe Konqueror is the best browser currently out there. Some will complain that it is not available for Windows. But then, why should, or since based on Qt, why shouldn't it be possible
2) The most important thing for standards is that not patented technology will be allowed to sneak into the standards.
Many of us have been conditioned to think that both standards and innovation are good things. And the latter is an overused word that Microsoft marketing has forced into the memestream. But really, standards tend to stifle innovation.
Standards will be partially incorporated, but slightly fucked up. Dreamweaver 2k7 and Frontpage Longhorn will output garbled XHTML with a raped form of CSS that fails to display/work properly on any non-IE browser. SVG will turn out to be a disaster in IE, making sure everyone in 2007 is still stuck using JPGs and GIFs. By then IE will have integrated .NET ( Or some other half-assed scripting language. ) scripting abilities tied into the browser to replace the now obsolete potential ActiveX vulenrabilities. People will cry, bitch, moan, whine and Linux is set to take over the desktop market in 2007 again. Blah.
Hate me!
Actually, there might be hope if Mozilla was available as an Internet Explorer plugin.. similar to adobe pdf and macromedia flash plugins. When a page wanted to use the mozilla renderer for advanced features, it would simply tell them to install a plugin, which most IE users don't think twice about. Eventually, these users may get tired of seeing most everything in a plugin browser, and may want to try using mozilla standalone.
It's time to tell anybody who asks you anything about their computer that they should download Mozilla or Firefox. I do, and most people who've done it have thanked me afterwards.
Need a Python, C++, Unix, Linux develop
There are plenty of alternatives available. In the early days of the web nobody cared about primitive browsers. Let's do the same now.
I don't think there is a need to get XHTML and CSS all gooped full of new features, so I hope it doesn't go in that direction. I know Microsoft will try and take it in that direction to compliment their overcomplicated Long Horn. In my opinion as a user of XHTML and CSS with PHP, I believe that what is required is simlification so that everyday users will want to use XHTML with CSS. Products could provide this but I still think the best way to code websites is by hand. XHTML and CSS are quite satisfactory at this point, but perhaps they may require some refinement. Please no more crazy features, because you can save that for DHTML and Flash (yuck, but good for some). Take a look at CSSzengarden.com if you are not yet convinced in XHTML with CSS is artistically pleasing enough for you. It's a better standard than many websites around.
The dangers of knowledge trigger emotional distress in human beings.
where there is no one dominating operating system?
XHTML-MP+SVGT11 Recommendations
Microsoft's whole goal in the IE/Netscape war was to make its webpages incompatible with Netscape. We still see crap like that today.
I think the only hope for actually implementing web standards lies in demonstrating the superiority of products like Mozilla Firefox. Don't expect any development from Microsoft on this front; the more exclusive they can make their browser, the better (in their eyes).
I don't expect to Longhorn/the new IE giving anything helpful to web standards.
Until people stop browsing with Netscape Navigator 4.07, standards will be impossible to enforce. The new IE won't change anything. As any designer knows, CSS-based designs are awesome and advantageous in so many ways compared to traditional table presentation. However, while enforcing compatability you sacrifice the visual quality of the site (for old browsers), and most businesses would rather single out handicapped people than certain browsers (makes sense % wise). The only thing everyone agrees on is that the migration AWAY from Internet Explorer would be the best for web standards.
Unfortunately, standards have come to mean very little in the browser world. Everyone touts XHTML to be awesome. But have you tried designing a site that uses an XHTML strict schema w/CSS for all your formatting? 3 different browsers can give 3 totally different results - to hell with that "standard." Right now, it's useless. Now, take JavaScript on IE and Mozilla. IE supports the "document.all" collection, while Mozilla relies on "document.getElementById." No problem there, and I know the "all" collection is not part of a "standard." But there are certain times when having the "all" collection can be beneficial. If people can make Mozilla support ActiveX, why can't they support the "all" collection? Clearly it's in the best interest of Mozilla to be compatible with the browser that "defines" unofficial standardization. And I can bitch about IE 6, too - who the hell came up with its selective and strange CSS support? And why did MS really stop developing IE for the Mac?
If we Ignore the attempts of Microsoft and others to make the web dependent on proprietary formats, etc. for a moment, the future of the web is quite exiting.
I'm talking about the Semantic Web, which is an attempt to deal with the IMO biggest problem with the web, and especially searching the web for information: you can only search according to syntax. Words, regexes, etc. is really the best you can do right now.
Searching would be so much better if we had semantics. Semantics would make searches and web pages in general much easier for computers to index and relate to what is actually being searched for.
An example: searching for "a yellow car for sale in $CITY, with a cost between $VAL1 and $VAL2." would not give a lot of unusable results today, but the semantic web would return what is actually asked for.
Of course, all this is just theory, and a best-case scenario example. And there are lots of obstacles for the semantic web; many people are happy with the web as it is, and it will take a long time to implement it.
Probably, some ideas would be incorporated slowly into the web as we know it now.
Konqueror has a lot of CSS rendering problems.
At what point did the internet become reliant upon Microsoft for standards? I would go back to sgml and gopher before I'd allow MS to dictate standards.
I suppose there's nothing like a flamefest to get the circulation going.
"As a brief aside, if I were Google ..., who are rumored to be working on client side technologies for managing information, I'd put a lot of energy into Mozilla, and release a Google branded browser..."
Mozilla, and Open Source in general has an amazing window of opportunity right now. A product tie in like the one described in the article is exactly what is needed.
IE looks as if it will remain stagnant for at least another couple of years. If there is a Mozilla marketing arm, they should be jumping in with both feet.
Similarly, now is the time for Open Office to get the MS Word compatibility bugs sorted out and to mount a big attack on the corporate sector.
If the Open Source community waits another year or two MS will steamroller them with the latest and greatest MS OS and Office packages. If they jump now and can find backers to finance PR and advertising, groups like Mozilla could make major gains.
Three Squirrels
> But really, standards tend to stifle innovation.
Perhaps I could reword your statement to:
But really, standards have stifled innovation, and they don't have to.
The dangers of knowledge trigger emotional distress in human beings.
It wasn't that long ago that people made an active choice to download a browser. It's not an uncommon choice.
This article paints a gloom picture, but no one seems to see the light.
If Microsoft wants to wait to release a new browser then this merely opens a nice hole for increased market penetration.
The gap will fill, but not if people complain Microsoft is not innovating.
"You should always go to other people's funerals; otherwise, they won't come to yours." -- Yogi Berra
You can see it on the streets. "Damn I hate these popups." "Use Mozilla."
As long as we keep telling everyone that there is an alternative superior to IE, they will begin using it. Eventually, people will have to build websites for Mozilla, and then we will be back to the IE/Netscape wars. Except this time, nothing new will be coming from Microsoft for several years.
I strongly suggest we build our websites with XHTML and CSS and ignore IE. We can put a message on our sites "We have detected that you are using IE. We require a standards based browser. Please download Mozilla, Firebird, or Opera."
The radical sect of Islam would either see you dead or "reverted" to Islam.
That said, XHTML is fucking stupid.
And that statement is based on what?
As a developer, I find XHTML to be a huge improvement on HTML - it just makes more sense. No more half-assed guesses as to whether or not a tag needs to be closed or VARIATIONS in tag name CASES that SEEM to BE randomly switched BETWEEN by CERTAIN web designers.
Tables are discouraged which means XHTML code written by a competent developer is much simpler, presentation and content are easier (but IMHO not yet easy enough) to separate so designers have an easier time of things, the structure of XHTML is consistent, unambiguous and - assuming you avoid going crazy with namespaces - easier on the eyes of a developer, and much more easily parsed.
So what exactly was your gripe with XHTML?
You can bet the second a competitor gains a single percent of desktop penetration with some new browser MS will suddenly release a ream of IE updates despite what they've said. So I say bring on the innovation, either way IE users win.
This article brings up many good points about IE's potential of totalitarian rule over the internet in the future, but I feel that it lacks insight on certain predictions, especially those regarding longhorn.
For one, the time it will take for longhorn to be widely adopted isn't factored into this hypothesis at all. It's 2004, that means its something around 4 years since the release of Windows XP. But is it as ubiquitous as this author claims it is? Absolutely not. It costs a lot of money to upgrade a whole mess of computers to a new MS operating system, and many people just don't need to for whatever reason, so in many fronts, it hasn't been done. My high school has some 100-200 computers: some are brand spankin' new dells with XP, others are Windows 2000, and there are more than just a few OS 9 macs floating around there as well. M$ can't assume that longhorn's release - and subsequently the release of XAML, etc - will take web dominance even within four years. It will take much, much longer.
So do the math. We've had a year or so heads up on the threats that longhorn posits to the Interweb, we have 2 1/2 more at least until the sucker actually comes out, and then over 4 years for reasonable ubiquity of the OS to make developing all future websites in technologies like XAML, etc worthwhile. That's nearly a total of eight years for standards to be utilized and improved upon. There is no reason why technologies like XUL, CSS2.1 (or even 3), and SVG can't be the accepted norm before then. The word just needs to get out somehow, but that's another post altogether...
On another note, regarding his mentioning of a Google-branded mozilla or something thrown into the forray, that's just overkill. Just imagine if, instead, Google merely placed these words on the bottom list of links on its homepage:
Really, they'd only need to have it up there for what... a month? two weeks? for it to make a HUGE impact in IE's dominance. Imagine......
eric http://www.ericdfields.com/
"With the next version of Internet Explorer tied to the release of longhorn, and still years off
Heavens thats actually a good thing. It means that the other popular browsers, Mozilla/Firefox, Opera and others. Can continue to gain ground setting the standards that Longhorn+IE will have adhere to.
, what hope is there for innovation in CSS, SVG, XHTML and other web standards?
I'd say there is a good hope, Longhorn/IE will undoubtably break / embrace and extend web standards, probably offering some "revolutionary technology" which is infact a rehash of an existing standard butchered and twisted to work only in IE.
Is the future of the web similarly tied to Internet Explorer and Longhorn?
I sincerely hope not. Now is the time for web-developers to start building with upcoming standards and tools. Id like to see all browsers fully supporting SVG for a start. In this interim period of no new IE versions we have the ability to build and popularise the technologies that are available to us before they get the IE poisoning. It is, after all the tools developers decide to use that drives the future, and by pushing boundaries innovations can be realised.
Electronic Music Made Using Linux http://soundcloud.com/polyp
I'd rather like to see browsers that can handle SVG natively first. Plugins don't count because of their operational problems. (Automatically deploy a security update for a plugin from Adobe? Good luck!)
- Do you have iTunes on their Windows machine
Literally millions of people use a big chunk of Safari on Windows. It's the browser built into iTunes. It works today.
So arguably the quickest, most standards compliant browser around, which by the way is based on the open source
Blah. I was following this guy's argument until I came to this part. Are there seriously "millions" of windows users really using the iTunes browser ? That number seems a bit high given how many songs Apple has sold. Also, i'd challenge that Safari is even close to being the "most standards compliant browser" around. If you're working off a W3C checklist, I'd say Mozilla has it beat by a longshot, and makes a much more meaningful dent on the web applications side of things than Safari does, which is another big battle against IE altogether. I just can't believe that anybody really thinks for a minute that the whole future of the web and the battle of winning the hearts and minds of "millions" teeters on whether or not a browser supports CSS text shadows..-- Maciek
Quick, bring on more stuff to make things go even slower!
I'm happy with web pages with pictures on them. In fact, uninvent Flash and I'll be even happier!
I think improvement can still happen as far as internet standards go with Longhorn. Mozilla is making far too much head way in the browser market for Microsoft to ignore. I think they will have to "develop" new tech. before Longhorn to keep anyone buying it.
Why? The only chance Opera has to compete with Internet Explorer and Mozilla on the desktop is by inventing and adding new features to their browser.
Opera has many great coders and inventors working for them, such as Håkon Wium Lie the creator of CSS. How can mozilla compete with such a experienced and professional team?
Opera has included many features in their browser, like email, newsreader, rss feed and irc, and still their installer is only 3.5 Mb, and the fastest browser out there.
Wasnt it opera who invented mouse-gestures?
In my eyes the main challenger for Opera would be Google if they decided to make a browser, but their first rule in their philosophy say:
"It's best to do one thing really, really well. - Google does search"
So i guess they will focus on search.
Simpsons 11:11: Lisa: They must have programmed it to eliminate the competition - Bart: You mean like microsoft?
This article's viewpoint is pretty puzzling to me. It says in essence: "How can there be change and progress in Web standards during a long gap between versions of IE? Oh no!"
I don't get the "Oh no!" part. I see this as good news. If the article's author sees IE's "monopoly" on Joe Q. Normal's computer as a problem, he should be jumping for joy that Microsoft has left the arena open for competition. Quoting the article's comments on Mozilla:
But will it, even with continual improvement, on top of what is already a fine platform, be sufficiently compelling to have Windows users replace IE 6 as their browser of choice? Some certainly, but enough to worry Microsoft?
Why do we care if MS is worried? The question is only whether we can get enough people to adopt an alternate browser, and MS is clearly not worried enough to compete anymore. Perhaps they're HOPING someone will take the browser market away from them. Perhaps they want to train computer users onto a less frequent release schedule so they don't have to release code that's as buggy. I don't care.
We should see this as an opportunity, as should the developers of Mozilla, Opera, and the rest, to remind Joe Q. Normal that he doesn't have to accept a crappy, stagnant, buggy browser as his one and only window to the Internet.
A hen is only an egg's way of making another egg. -- Samuel Butler
I'm guessing you're not actually a developer? Most developers I know don't have to make "half-assed guesses" about the language they work with, they know it. Also, what does case have to do with anything? Your main arguments against HTML don't really make much sense.
I'm guessing you're not actually a programmer? If you've ever written a program which parses HTML or decorates it in some way, you'd appreciate what a god-send it is to have a simple, consistent syntax for operating on the document tree. XHTML is the deal.
The author of the article thinks that Safari is the best browser. Safari is a derivative of the open source browser Konqueror. Therefore the parent poster is definately ON topic...
So briefly stated, this is likely to be tagged as troll or flamebait, but there's a lot of truth behind this.
It is inarguable that a lot of the best innovation in the history of any industry has been made by people who go outside current standards ("Here's to the crazy ones...") and build something that is the best that they can make it first, and worry about the other considerations later.
[Note that "best" can have many contradictory meanings: best in some narrowly defined performance criteria (fastest, highest, biggest, smallest, etc.), or broad appeal (most general utility, most sell-through), or most efficient, least polluting, cheapest/easiest to manufacture, etc.]
Sometime these evolve into "de facto" standards, and it can be difficult to turn those into "open standards" where there's a level playing field for others beside the first-to-market to gain traction.
As a response, there have been many efforts to develop standards in advance of actual product. In my experience (CAD interchange languages in the 70's and 80's, XForms today), progress on these standards is relatively glacial, and they are often passed over by the industry at large.
I submit that both approaches are good, and that we ought to strive for a healthy tension between them. This argues for moderation by those who cling to the "purity" of their ideals as circumstances change out from under them, and for a willingness to exercise enlightened self-interest and surrender proprietary advantage, vs. rapacious exploitation of current dominance. (We know who we're talking about here...)
To that end, I'd rather see some of the browsers take some risks in advance of accepted standards, at the risk (and expense) of requiring a few willing innovators to perform some extra work ("click here for a non-fizbin version of this site").
Just for a couple of examples, why not re-think where some of later innovations are supported? Can the concept of tabbed browsing by pushed up to the server, so a web designer can deliver a set of related tabs to the client? Could support of the portal/portlet structure be pushed into the client, so that the work of rendering and compositing a page full of portlets can be offloaded from the server, and servlets can execute more autonomously when appropriate?
Of course there is the occasionally shooting yourself in the foot when inexperienced with a language or a particular implementation of a language. But that's why QA cycles are integral to solid engineering practice. Even "experts" have brain malfunctions from time to time.
LRC, the best-read libertarian site on the web
This would be an amazing idea, and could probably work if it ever managed to pick up momentum.. After all, it seems unlikely that flash would have become so ubiquitous if developers hadn't seen how much power they had using it, and encouraged people to install it all over the place.
It seems to me that if developers had the option of using things like proper PNG alpha support, and embedded SVG in HTML by just specifying that pages should render in a full-page Gecko ActiveX control, a lot of them probably would.
Many businesses want a good HTTP-friendly GUI standard. The current crop of standards is designed more for fancy online brochures than for serious data entry and data browsing. When doing business web applications, it is clear that the "customer" really wants applications that are as easy to develop and as flexible as Visual-Basic, Delphi, and Power-Builder-like arrangements. I don't think the slight time delay and bandwidth limits between client and server are a big problem if the protocol is designed well.
Table-ized A.I.
I'd like to know a little bit about the guy who wrote this article. He links to some of the usual standards gurus in his sidebar (Eric Meyer, Jeffrey Zeldman, CSS Zen Garden) but I can't find any background information on him.
I'm not saying that his musings aren't valid, but I'd like to know where he's coming from and what sort of relevant work he does that involves web standards. This would give the article more context and help me to understand better why he says what he does.
Your fantasies contain the seeds of important concepts.
Before y'all get up in arms, I'm not disputing that there's uglyness to HTML, and CSS is a huge step forward. However, CSS is a huge, bloated beast, and I can't really see how SVG advances the web. IMNSHO, the web should be:
- An easy way to access information
- Simple, adhering to the lowest reasonable common denominator that works across all common browsers (HTML 4, limited CSS, etc.)
- Not filled with bloat and fluff that doesn't help me access information (such as flash intros, flash menus, Java menu crap, etc.)
Many of the webmonkeys I've known in my company that complain about such things not working are the same people who couldn't do HTML by hand if they wanted to, insist that beauty should take priority over functionality, and develop IE-only pages because they never thought to test any other browser and then blame those browsers for not supporting the latest, greatest standard. Here's a tip: if you want people to use your stuff, you have to provide it in a format their tools can understand. You can't expect everyone to upgrade, so you have to work to your audience.
Granted, I, too, would like to shoot everyone using NS 4.x, but there are still people out there running it and viewing my site at 640x480. I don't know how they can stand it, but it's their choice. My choice is to continue to support them as well as possible, for the moment. So I don't really concern myself with the new standards. Besides, for me, I have little to no use for them at the moment anyway.
IMHO, mis-applied Java and Flash are the worst two things that ever happened on the web. And those were both "innovations", especially the Java bit. So understand if I'm wary about any so called "improvements" to what already works pretty darn well and is just now starting to truly work the same (mostly) in most mainstream browsers.
That said, I run Fire(name this week) and, yes, I don't have the Flash plugin installed. F@#$ing hate flash. Bane of my existance.
Pinky: Wuh, I think so, Brain, but if we didn't have ears, we'd look like weasels.
I have to disagree about Konquerer being the best browser because I've seen a ton of rendering problems. In fact, I would almost say it feels more like netscape 4 to me when I use it.
Oddly enough the Mac version (safari) doesn't seem to have as many rendering problems.
The man who trades freedom for security does not deserve nor will he ever receive either. - Benjamin Franklin
The article is wrong about the iTunes browser. It has no relation to Safari whatsoever other than being made by Apple. iTunes is not based on WebCore (nor does it use HTML in fact). SafariTunes seems to be a really common misconception among a lot of people.
The article suggests that Safari is the quickest, most standards compliant browser around. Quickest, it may be, but most standards compliant is Gecko, because that's Gecko's raison d'etre and they do it better than anyone else.
Read all the notes on w3schools.com, and use google religiously when you have questions. Also, be sure to look at the CSS source code on csszengarden.com, because it can save you a lot of time to learn through example. Keep it simple, too.
> I've managed to get my head around XHTML, but when I try to use CSS, I have trouble doing even the most basic layouts that could easily be achieved with s.
I had the same problem, until I ditched tables for div tags and css classes. Using the id tag is the key to getting layout right, and nesting your divs correctly will help too.
Start with one container div that holds everything, and that's your page. Give it an id class like: id="container", and in CSS, use the # symbol to identify it.
for example (in the CSS file or style tag):That would be for an id tag in your div: > I can understand why Slashdot still uses them.
They kinda have to at this point. The Slash system is too entrenched in HTML to change direction. Why? Because many comments would break XHTML, and there is no point using CSS without using XHTML, IMHO.
> With CSS, nothing seems to 'just work' on every browser. The W3C specs are confusing. And there's no decent HTML/CSS editor (as in the Dreamweaver kind, not the Vim kind) that I know of for Linux, so it has to be done by hand or elsewhere (Wine/Windows, et cetera).
I recommend doing everything by hand. You'll learn more and your code won't break as much, and you can quickly repair it if you know your system well. Or you could just download a package that lets you quickly post news to your site without having to change your templates every page. I've created one at sourceforge called Gemsites that will be releasing a 2.0 version soon, and while Gemsites used to be a Slash clone, it's now a standards compliant blog/photoblog package.
> What's the best way for a n00b like myself to learn and use CSS in the real world, where some people use Mozilla, some use Opera and Konqueror, and a lot of people use Internet Explorer?
Talk to people like me over email and I'll help you.
The dangers of knowledge trigger emotional distress in human beings.
The sad thing is Java is one of the better-supported lagnuages for Mozilla. XUL would ideally be a great platform to hack on, but it's very badly documented and there isn't much in the way of language bindings or mindshare out there.
Ditto for OpenOffice. UNO is great in the abstract but in practice it's extremely forbidding. It would be nice if they could work together on a single, XML-based cross-platform app delivery framework - they even have similar licenses - but...hell, you know the story.
Google confirms: Ruby is the world's most beloved programm
Would it be impossible to make web standards a browser plug-in? Something like an XML DTD that would be automatically downloaded every month and contained the lastest standards rules, so that all browsers would support the latest features as soon as they are published? Or maybe it's not as simple as that. Maybe there are rendering engine issues, etc. Still, it would be nice.
Where he errs, IMHO, is in the assumption that innovation will be incremental. He seems to be implying that the most we can expect from the future of the web are some (gasp!) cool new CSS features.
I beg to differ. The future of the web will ride on the wave of two related trends, both of which have revolutionary rather than evolutionary implications:
- Increased client computing capacity - back in the mid-90s it was all the average PC could handle to render a complex HTML page. Nowadays PCs are at least one order of magnitude faster, and a lot of the processing currently relegated to the server could be offloaded to the client. The reason that this hasn't yet occurred is that no browser has the appropriate plugin architecture. It is possible to develop plugins for major browsers, but there is no proper framework to integrate these plugins into a cohesive whole. Instead, an increasing number of networked apps are eschewing the web browser altogether in order to provide a better user experience (e.g. IM, P2P file sharing, online gaming, VoIP, etc.).
- XML - for all the hue and cry, the only significant impact of XML on the web since its inception 6 years ago is RSS. RSS is certainly cool, but it's just one XML-based language, and the whole premise of XML is that it enables the creation of multiple vocabularies. So there's a huge opening for someone to create a browser that intelligently processes XML vocabularies. This would include managing the relevant XML schemas (perhaps using a centralized repository), rendering the XML in various ways (perhaps including HTML templates and autogenerated forms) and persistent storage/retrieval. This is basically the goal of RDF, but besides taking what I consider to be a number of unfortunate design decisions, the RDF designers have essentially ignored the need for a new browser architecture to make XML use on the web an attractive alternative to HTML.
None of this is easy, of course. But considering the potential rewards of owning the new new browser architecture, I have no fear whatsoever that innovation will stagnate just because Microsoft decides to take itself out of the game for a while.Nonetheless, most of these applications would be that much more valuable if they were integrated together. To achieve this, a platform is needed that permits inter-plugin communication: a shared data model, a high-level framework for UI development and way for plugins to exchange messages. Think Eclipse for networked apps instead of development tools and you'll be on the right track.
Peer Pressure
MSFT won't do anything until they feel some pressure from the market.
The idea of a Google branded browser based on Gecko would work. Especially if the Google desktop tools work best with this browser.
Getting Google to rank pages based on standards compliance would work (XHTML/CSS2+ design = higher page rank = more sites wanting compliance = less sites holding onto IE6 only designs.)
A Windows version of Safari might work. If an iTunes install put it on the system (like it does now with QuickTime) then people might use it -- hard to say if that would provide any market pressure though.
If something doesn't come along to shake up Microsoft (and it's got to be big, like the Internet in 1995) then things will not change in Redmond. At this point in time, Google is the only thing big and successful enough to rattle their cage.
-ch
If the web designer does it right, you should only have to download the CSS once (until your browser cache expires, at least)
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
> The CSS standard is crap.
Statements like this illuminate a kind of ineptitude that is too revealing for a place like Slashdot. If you dislike CSS or you have had a hard time using it... if you are frustrated with it: ask for help, or just simply state that you are frustrated. Don't bash the standard because you have had a hard time with it.
The templates on csszengarden.com are all created by graphic artists who believe in CSS and what it can do. They don't spend months on each template. In fact, I find it easier to create fast, graphically appealing websites with XHTML/CSS than I have ever created with HTML and Microsoft-friendly tag attributes. It all comes down to compliance and follwing the rules. Maybe CSS needs some refinement, and that I won't debate, but to bash the whole standards seems rather uninformed.
> A good example of the futility of working with the CSS standard is Jeffrey Zeldman's site www.zeldman.com. This site has been through so many redesigns yet inevitably each new redesign breaks in some major browser or other.
Maybe he's redesigned it so many times because it's fast and easy to do so? Part of the problem with many standards is when designers try to take it too far. They should all just keep it simple and the results will be better; there will be less trouble. The web is for information distribution, and therefore it's quite possible to create an appealing website that doesn't break browsers.
The trouble with standards, starts and stops with the browsers that try to change the standards to support some kind of corporate domination theory. When browsers support standards, the way they were meant to be supported, browsers wouldn't break when reading sites designed with standards.
The dangers of knowledge trigger emotional distress in human beings.
> I've been making web pages for 7 years, but it's only in the 18 months or so that I've really become so convinced and enamored of the wisdom of css for real separation of content from presentation.
I have to really agree with you on this. The wisdom of separating content from format is critical to any single application, and it's a standard that should spread through the rest of the programming world, and not be limited to web design.
Games, for example, run faster when they have been coded with proper standards similar to CSS and XHTML. Keep the logic and the rendering separate, keep the audio separate, and the resources separate. Join them at runtime with hooks into each from the main engine. That's why games that are designed with a good set of standards all play faster, with better framerates, than those that mix things up too much, or are coded sloppily.
I think that the most important development on the web was separating logic from data, and data from format. PHP, XHTML and CSS make a killer combo!
The dangers of knowledge trigger emotional distress in human beings.
The Apache and Mozilla projects should work together to create useful extensions to the browser. For example: persistent network connections so the browser can behave more like a terminal when writing applications (see: Flash).
Two things:
First of all, compact discs, like all the formats before them, were developed for one major purpose: to make you rebuy the music you already own. The record industry does this. Every few decades they switch to an entirely new format and make people buy their music all over again. We're due for a new format shortly, expect them to start pushing music dvd's like crazy (we're already seeing this, but expect it to get bigger and for dvd players to become essential components in a home stereo system).
Secondly, the W3C does not ignore common sense. Yes, there are things which should have been easier to do with CSS-based layout (mostly stuff that emulates other media, like footers below a multicolumn layout). However, you can not escape the notion that CSS-based design is vastly more powerful and time-saving than old-style design. Now you can finally separate content from presentation, and redesign your site without having to rewrite content. That's a major win for any large site. Another benefit is that CSS-based design saves bandwidth by producing smaller pages and allowing the presentation to be cached between browser sessions (if you link your stylesheet externally).
You can go look at the discussion about CSS in the W3C mailing list archives and see the reason behind every single feature and quirk of the language.
The main problem with CSS-based design right now is that the browser with the largest marketshare has really poor support for the coolest stuff in CSS, ensuring that what the standard says should be the right way to do stuff often doesn't pan out when you try it in real life. Moving the web forward from IE6 is desperately needed, but I'm not going to hold my breath until it happens. Maybe if the alternative browsers manage to get enough marketshare to make cross-browser design a must market pressures will cause microsoft to respond with IE7, despite earlier claims they're not going to do that. But it's doubtful microsoft will respond to market pressures anytime soon. It's just not their shtick.
Sure, IE's marketshare may be shrinking - however, will the drop in marketshare really be enough to give an incentive for a mass exodus from IE?
Average Joe will probably be using Internet Explorer for Windows, and he definitely doesn't care about web standards (nor does he probably know what they are, anyways). If you tell him to download a different browser, he'll simply shrug off your suggestion, since he believes that he'll be inconvenienced by the fact that he has to download...a different browser.
In order to change the future of web standards to a much more optimistic one, we, the coders, designers, developers, etc. must perform some kind of action. We must evangelize web standards, and educate other coders, and even users, about them. Eventually, users will have an incentive to change their browsers, simply on the basis that their favourite pages won't work in IE anymore.
It won't be the users who will directly cause the change. It will be us, the ones who actually use these standards. We have a 2-3 year timeframe before Longhorn comes out; the opportunity to increase the efforts to spread the word is now.
Is far from true.
.Net thrives on it.
.Net framework. It can do everything you have expressed.
XML is integrated so deep into almost every technology available for internet development it is considered a ubiquitous skill for any level of developer.
WebServices are run by it.
Databases talk in it.
Office applications communicate with it.
Many large websites use it to render their entire sites.
And BTW the lofty platform/framework of which you speak is completed and needs only widespread adoption. It is the
blah
...if you are not yet convinced in XHTML with CSS is artistically pleasing enough for you. It's a better standard than many websites around.
Artistically is one thing, but how about simplifying the development side? The ability to insert separate blocks of HTML seamlessly into a single document without relying upon a perl script or some unlikely-to-work DHTML hack would save people a lot of time, and would let them be a lot more creative with their designs. CSS touches upon this separation of content and layout, but doesn't quite get the concept that layout consists of content.
There are still good reasons to improve the standard.
The ______ Agenda
It is not in microsoft's interest to do so. The vast majority of web users are complete morons. They think IE *is* the internet.
Here are some recent browser statistics over a couple day period:
Internet Explorer - 93.65%, 391,965 hitsMozilla - 11.16%, 54,654 hits
Netscape - 1.64%, 8,022 hits
Safari - 1.06%, 5,193 hits
Everything else - less than 1% each
Now how many of those are really going to see the light and realize that IE sucks? Not many.
i second that, great idea.. of course they would be (at least for a certain time) be providing poorer results to their customers, which would require that the weight be relatively quite small. Nonetheless, good thought.
-ashot
I have "converted" several of my non-techie friends and a few of my co-workers to Mozilla. I installed the plug-ins for them so their experience was seamless. My friends now swear by Firefox as their browser and refuse to use Internet Explorer. Word of mouth is Mozilla's best friend. Once people start to use the browser and realize how much better it is (no $41Tware via ActiveX, no pop-ups, and tabbed browsing to name a few) then this may be yet another door to open on the path to Enlightenment.
has anyone else had similar results w/ their immediate and not-so immediate humanoid contacts?
There is still work going on to further web standards. SVG 1.2 is coming along and, according to Dean Jackson here at WWW2004, a working draft for XBL should be forthcoming (after being separated out from SVG 1.2).
Another interesting thing is the upcoming workshop on web applications and compound document which will be addressing the issues of mixed namespace documents and also the things needed for the development of sophisticated web applications using SVG, XHTML, etc.
I'm guessing you're not actually a developer?
Uh, yes I am - this is about my third professional year now.
Most developers I know don't have to make "half-assed guesses" about the language they work with, they know it.
Then I envy you. Perhaps you can get one of them to recite the HTML 4.01 spec to me then? Or hey, get them to write me a standards compliant HTML parser. I mean, what IS "knowing"? What's good enough?
I'm not the best software developer you'll come across, but fact is that most developers (indeed, quite a few of the ones I've come across) are relatively clueless.
This makes sense really, since they're dragged along in one direction by whatever technology $EMPLOYER chooses with no real opportunity for expanding their horizons.
And when I say "clueless" I don't mean "stupid bastards" - they're intelligent people, they've just been left out in the technological cold.
Also, what does case have to do with anything?
Exactly what I was talking about: Conformity. Readability.
Your main arguments against HTML don't really make much sense.
Considering I didn't mention font tags, didn't really go much into the evils of table based layout and the wonderful (if sometimes frustrating) benefits of CSS.
Don't get me wrong, CSS is as horrid as it is great due to the whole cross platform deal. But it's leap years ahead of The Old Way.
So let go. Learn something new. Appreciate it, despite its flaws.
The solution is simple; switch end users away from IE, to *any* other browser (just as long as said browser does not use IE as the rendering engine). It doesnt neccessarily have to be Mozilla/Firefox, it could be Opera. Any browser that has all the basic features; web standards complicance, popup blocking, extensibility, etc. Theming isnt neccessary, but its always nice.
I work as field networking tech for a local SORC (Small Office and Residential Computing) company. Every chance I get, I urge users to ditch IE and move to something better. Usually, they ask me to install it for them, and I gladly do so.
Most end users dont understand spyware, dont know or care what web standards are, but they *DO* know about popups... That in itself is one of the biggest selling points for alternative browsers. If it stops popups dead, the end users will like it.
It is unfortunate, though, that most end users are unaware of browser hijackers. They just assume that random porn popups and huge annoying toolbars are "part of the web experience". Most users dont realize that the only reason thier web experience sucks is because of the browser they are using. Heck, for most end users, the web *IS* the browser.
For any tech who has contact with end users; I urge you to reccommend them to switch away from IE. Granted, doing so will effectively remove them from your list of regular customers for browser spyware removal, but it will greatly improve thier web experience, and, eventually, the web itself.
Please post your "fix" for html. I'm sure we'll all be interested to see what you come up with.
"Our interests are to see if we can't scale it up to something more exciting," he said.
The browser wars are over. MS won, they achieved an absurd marketshare. A new war began while the smoke was still rising from the browser battlefield: the standards war.
I've noticed that all the ire, hated and derision that web developers held for Netscape 4 has in the last 18 months shifted to IE. Developers finally realize standards not only allow for cool things to be done, but also that those things only have to be done once. Chances are it won't work in IE. Avalon (the IE rendering engine) has barely changed since IE 5.0. Mozilla, Opera, and KHTML continue to implement standards released as far back as 1999 while IE arrogantly takes a nap within sight of the finish line. All of us need to stand along the race course with gatorade for the tortoise.
How to do that? Joe Public needs a reason to download a modern browser (which IE certainly isn't). When I tell people I haven't seen a popup in almost 3 years, the invariable gape is followed by some question akin to "How is that possible?" I've been using Mozilla as my regular browser since .8 was released. I point the soon to be former web victim toward Mozilla (not Firefox, because the next step is telling them how to avoid mail virii by not using outlook), and not once has anyone ever looked back. Evangelism is how web standards will be able to sneak past the sleeping hare to win the race. Or war, however one wished to view it.
In my opinion the role of a standards body is to codify existing practice, not to create new ideas.
The fact that the W3C tries to innovate is exactly why it is becoming less and less relevent in the real world.
XHTML is not fucking stupid, but it can be quite annoying, when you have to deal with
On the other hand, it's neat to edit XHTML in nxml-mode in emacs.
Are technologies such as CSS or XHTML really standards at all? The W3C doesn't even refer to them as standards, they're called 'recommendations.' As far as I'm concerned a technology can be considered a standard by one of two ways. Either its endorsed by a standards organization such as ISO or it becomes used and accepted by the greater majority. New recommendations such as CSS2 or XHTML meet neither of these requirements.
I've logged onto a Windows XP machine a couple of times, and what I have found striking is that they try hard to make the whole user interaction thing look like a web page. Looks weird, but that seems to be their thinking.
Except that you can't really have any advanced interaction with a user with just HTML forms, so the next step is clearly to get the rest of the GUI elements of Windows onto the web.
People who are just talking about "the browser" is therefore missing the point: It is not a focus of MS at all: It's applications running over the web, and the browser is not very relevant in this.
The funny thing is that I think it looks like they are allready behind the Free Software and web standards worlds: We've got Mozilla, XUL, we've got Apache and the best XML Application servers, we've got very clean separation of logic from markup (and layout), and we've got XForms, SVG etc. to make it widely standardized.
The only thing we seem to miss is a lot of hackers thinking in terms of using it. Ballmer will be chanting this to his droids when longhorn comes out, so they will use it uncritically. We can start doing it long before MS where it makes sense now.
Employee of Inrupt, Project Release Manager and Community Manager for Solid
I wouldnt mind so much if IE was consistently crap accross versions and systems, but after tweeking a css based site to work on IE for mac i found out the hard way. There is absolutely no excuse for the type of rendering screw-ups that microsoft (and afew other browsers) make, the css box model is clearly defined and very simple and if you cant implement it you certainly shouldnt be making mission critical applications (well since microsoft cant even implement paintbrush without a buffer overflow exploit thats pretty much a given). The big problem comes because a browser fucks up so the web-developer specifically designs the site for that bug (im guilty of this) which means now the browser cant be fixed without breaking the site and that builds up until you get total chaos.
I dont want to sound like a facist dictator but you must follow W3C to the exact letter! If they tell you to jump out the window naked then you bloody well do it! that goes for making a browser or a website, if people had made the effort right from the start then we wouldnt be in this mess, websites would all work on all browseres, blind people would be able to use them better, pop-ups would be non-existent (using the idea that the browser is in control of the site) and i would be much much happier. Personally i think microsoft is 80% at fault here, it just seems logical that the same people who neglected to add afew lines of code to stop pop-ups back in the 90's and who stupidly allowed vb-script to have free run of your email, they are most likely the suspects who screwed this up.
This comment does not represent the views or opinions of the user.
I am starting down this road myself with several sites. The problem I have a as a page designer (which I am hoping you can comment on) is that all of the examples I have seen do not accomplish the separation of content and layout which CSS seems to promise, or rather, that the separation is very much one way.
From my point of view, one of the biggest promises of CSS is that I can cleanly separate my page content from my page layout. I can write content and then decide how to display and decorate it via the CSS at a different point, possibly changing my mind at different times or even with different audiences. CSS Zen Garden demonstrates this very well. I can view the exact same content through multiple stylesheets and get a very different experience. I can switch on the fly, and I can even view the raw (X)HTML on a low capability device like my phone.
This is a big improvement over tables-for-page-layout, which royally screw up low capability devices. With tables, I forced into client-side include hacks which allow me to send a low-capability device to the individual chunks.
There is more to separation than just this however. In particular, I have to wonder what will happen to CSS Zen Garden if the content of the page is changed, say a paragraph is suddenly twice as long or goes away or an extra paragraph appears. From my read, all the style sheets break. There is even a warning to that effect on the site. The style sheets include knowledge of the content and depend on it to fit inside certain boundaries. The prevalant use of ID tags (rather than descriptive style tags) really seems to underscore this.
The problem here is that I want to be able to update my content with minor or even major edits and not have to redo all of my stylesheets. If I add a new paragraph or reorder some text, this should not throw everything off. I build up my content through multiple includes to centralize boilerplate text. The purpose is defeated if my stylesheets depend on it for positioning.
It seems like there should almost be three files: (at least) one for the content, one for general style rules, and one for local tweaks for the current page (laying out specific graphics). Can this be done well in CSS? Are there good examples which do this kind of thing right? Or am I back to manual positioning and font dependence?
The "Semantic Web" is mostly an academic pipe dream, and here's a great explanation why:
While I think that Shirky is taking a bit of an extremist stance, I do believe that the web as we know has been successful for the very same reasons that the standards brigade finds it frustrating. A profoundly easy learning curve, a well-defined though flexible standard, and few requirements for semantic correctness all go a long way. The success of google is proof that useful meaning can be strained from a sufficiently large corpus of disorganized data, and I think it's generally the right approach: provide a few obvious places for semantic meta-information to go (title tags, hyperlinks, headers), and let the masses decide how far to take it. The entire SEO field is based on providing economic rationales for good meta-data, but it's not a pre-requisite.
Think we can start a trend?
Mail? Put "slashdot" in the subject to pass the spam filters.
> all of the examples I have seen do not accomplish the separation of content and layout which CSS seems to promise
The very best use of XHTML and CSS is by adding PHP into the mix. With PHP you can quickly generate your content using a template, without having to grow your XHTML out of control.
I'm going to be releasing the source code for my blog/photo-blog site over at zenbuzz.org very soon, at sourceforge.net/projects/gemsites.
This project lets you quickly design an XHTML template that goes with a CSS stylesheet, and then just leave it alone, running your site much more efficiently.
Features include, but are not limited to: photo/image uploading, safe image authorization by an admin, link submission, story submission, comments, user registration and validation, password changing if you forget your password, XHTML 1.0 Strict, PHP generated CSS (so you can write CSS using PHP), templates, admin contact form, FAQ generation, autologon, and a fairly good tracking system to know where links are coming from. That's just a start, really.
Images will display in articles if they are authorized images. The way the image code works is that images are uploaded and stuff into a database for approval. When they are approved they are written to the FTP and snuffed out of the database. When an admin is looking at the images that are not approved, the image validates the admin's session and only writes the image if the admin is valid. That keeps people from spamming your site with crappy images, and it prevents ftp uploading that could result in legal action.
> There is more to separation than just this however. In particular, I have to wonder what will happen to CSS Zen Garden if the content of the page is changed, say a paragraph is suddenly twice as long or goes away or an extra paragraph appears.
The trick is to create a template that supports dynamic text, using either classes or id tags based on the type of data. To do so is not easy; it's hard, but when you get it right, it's good to go. It's fairly simple to expand div tags so they can support dynamic content; the hard part is making a really nice looking site that conforms.
I did a page for a company before that resulted in some problems with the background image they wanted to use, as some pages dropped past the threshold of the image, and using another scroll-bar is never the answer (ie: don't use the CSS overflow attribute or use it very sparingly).
The solution? I posted the image in the main area, and simply designed the website around it. The background image became a seamless concrete texture.
So you have to work around problems with how we understand the web, and how the web works best. It's our problem, not the standards, IMHO.
> The style sheets include knowledge of the content and depend on it to fit inside certain boundaries. The prevalant use of ID tags (rather than descriptive style tags) really seems to underscore this.
Well we do have boundaries to work within; the browser window, but apart from that, we could really learn how to use the class attribute better so that the boundaries are more dynamic. Using the percent % width value, works wonders, but it's hard to sometimes get it right. I define my sheets as static size and let the end user worry about scrolling over if they have a small resolution; I just design it to work nicely with 800x600 and go from there. I too had many problems with CSS in the beginning but I'm far better at it than I once was. PHP created CSS works wonders.
> It seems like there should almost be three files: (at least) one for the content, one for general style rules, and one for local tweaks for the current page (laying out specific graphics).
PHP generated CSS will do this nicely. I can help you do some if you want to contact me, just click my site link. =)
The dangers of knowledge trigger emotional distress in human beings.
I think that google could make a easy move against windows. They only need to put a link to firebird. It is the hard way but harder and much nastier will play microsoft.
google needs a fair internet market so that longhorn does not finish them. Just a link in google can change a lot of things.
Any google-worker there?
P.D: If anybody thincks that it is not fair, think about how fair is to include Explorer and Windows Media Player in windows.
Many of the stylesheets on the Zen Garden's site break when using a non-standard font size, yet alone if you change the document.
I am sure that this kind of things can save us from microsot control.
Yeesh. If Microsoft can come up with a better system, then great! I'll gladly design pages with it. Thanks to PHP and javascript, it's possible to alter tags and code according to which browser the user is on, and unusable tags are ignored anyway, so a skilled developer can take advantage of the qualities of all browsers. If they can come up with something better than CSS then great! Other browsers would likely catch on. Nobody's losing here.
g eLoader() ), but if they don't include CSS, SVG and so forth, web pages simply won't display properly, everyone's going to realize 'this sucks', and they'll all use a more competent browser.
But of course, if they don't support the standards and shape up a little, Internet Explorer will simply become less popular and other browsers will look more appealing. They've already lost some following with the lack of javascript speed, position:absolute and PNG alpha (which by the way can be solved using a DIV and filter:progid:DXImageTransform.Microsoft.AlphaIma
Users and developers can make their own decisions, they don't have to use something they don't like, and Microsoft isn't going to change that, so we'll win either way. If Microsoft can build a better browser, let them! If it's not up to par, then too bad. We've been playing the browser game long enough to know natural selection plays the biggest hand in it.
I am NOT a number! I am a - oh wait, I'm number 761710. Look! 761710!
Isn't one supposed to research demands first before
introducing new technologies?
To the article poster: I'm sorry, did you just mention how IE is tied with the release of Longhorn, and therefore there won't be any innovation in CSS, XHTML, or any other web standard?
May I remind you that IE does not adhere to CSS or XHTML specifications. You also make it sound as if Internet Explorer defines web standards. Wrong again. The World Wide Web Consortium defines XHTML, CSS, SVG, and many more. W3C is the leader of web innovation and standards production (yes I know that makes me sound like a W3C fanatic).
In order for standards to work, they must implement all the necessary features and allow users to follow the standards easily.
XHTML, CSS, SVG are OK for document presentation, but most webpages are not static documents, they incorporate interactivity. The aforementioned standards do little for interactivity, so authors turn towards Flash, ActiveX. And no, Javascript doesn't help, Javascript is a pile of shit that's too simplistic for programmers and too complicated for non-programmers.
A successful web standard must incorporate one well-defined and easy-to-use language to implement basic GUI elements and operations. It must also implement one other well-defined and complete language to implement more complicated programming tasks that may be needed for complex web pages. These two standard languages are missing.
A successful web standard must also be easy to support by IDEs geared towards non-programming authors. XHTML and CSS are great, but few web authoring tools generate code in these languages in a manner consistent with their original intent. Same goes with Javascripting (or what should be replacing it). Standards must be usable by IDEs.
If you want to avoid a single-company web, you must have everything availale: Standards that are complete and usable (i.e. easy to understand by non-programmers), Several web browsers that implement complete standards, Several IDEs that allow authoring by non-programmers.
And if you think web-authoring should be the sole domain of highly-skilled programmers, think again. It's that attitude that lets MS take everything over. MS will create products that are usable by casual authors and these authors will use MS products only and flood the web with MS-only documents.
... being the tech-avantgarde community it says it is, doesn't upgrade to modern standards, I don't know how can I expect other people to do it.
Check Slashdot's HTML code. It's been said before, and I think it's important to underline it again. Why is Slashdot's HTML code not modernized?
I may be wrong, but I think it shouldn't be so difficult to CSS-ize the code. I'm sure it would be leaner, more elegant and it would save a lot of bandwith to the ./'s servers too. It would also
transform gracefully.
The ability to add a src to any tag would be welcome, this would let you set a division with a remote source and of course use scripting to change the src the same way you can dynamically change images right now. Sort of like frames but without any of the limitations.
Of course border: rounded would be a nice addition for rounded corners on a bordered element but you can hack your way around that right now if you know the tools of the trade.
Aside from a few bugs in IE (:hover not working for non-link CSS elements, etc) there really isn't all that much left to do. We're pretty much to the point where if you can imagine it, you can implement it.
Since VRML failed we're pretty much done with HTML/CSS until we get a full motion video internet and need a better way to manage all those phat video files.
Im going to get flamed by all the MS hating bearded linux evangelists who still live in their parents bedroom but...
I think IE6 is a better browser..sure it has some bugs but have any of you actually bothered trying to lay out complicated pages using CSS in mozilla?? Its a freaking nightmare, it implements the css standars yes but if you do something a little weird rather than degrade gracefully and do what you intended all along it will generally just refuse to render the element at all.
At the moment IE is the ONLY web browser which is remotely usable in terms of performance, ease of use and quantity of web pages which are actually rendered correctly. Mozilla for me doesnt even render slashdot correctly about 1 time in 5, quite why it is rendered non-deterministically is also slightly beyond me.
Instead what i would rather see happen is IE's standards opened up so that the other browsers can implement IE's formally proprietary DOM. This way most of the battle is allready done since IE is compliant 99% of peeps are viewing through a compliant browser.
Basically what needs to be done is stop thinking of the self appointed W3C as the standards body but rather the company which actually implements the most succesful browser as having allready set the standard.
And no it doesnt mean we need to return to the bad old days of IE/Netscape 3. Theses days browsers are written by very professional teams who are good enough to ensure that that kind of divergent craziness doesnt happen any more.
In conclusion you all suck
I have discovered a truly remarkable sig which this post is too small to contain.
> Actually, tech like PHP/ASP/JSP/CF/Perl/.NET/etc has killed a lot of the *need* for seperating content and presentaiton on the client-level.
I strongly disagree. You're starting to discuss programming standards and it's paramount to modularize code from data and data from formatting; it's just so much faster to make system-wide changes if you do!
> It's already being done on the server-side, where it matters. Content is in the database, presentation (html soup) is generated by the templating engine. It's ugly but it works.
Sure, but without conforming to standards, you're going to lose out on the cross-browser ease you'll find with XHTML/CSS. Just keep it simple and you'll be fine.
> For example, if CmdrTaco wants to search for a particular comment, he'll find it much easier to run a SQL statement than to do an XPath query on generated XHTML.
I think it's good to use SQL for all your data needs. But that's not really what we're talking about. I think we're talking about standards compliance and any system with strict standards will trump a loose system any day for ease of use, ease of design and ease of alteration.
> If slashdot had clean, structured HTML, it wouldn't make the comments any better. (although the page might load faster)
And that's the point; the point is to lighten the load on all your pages, keeping them clean and smart. Doing so will save you money, and impress your customers.
The dangers of knowledge trigger emotional distress in human beings.
Umm... IF someone made a nice clean standards adhereing tabbed browser for MacOS 9.1 *I* would love you forever.
:(. Anyone wanna send me a G4 addon card so I can bring my PPC 8500 up to spec?
I kinda noticed that after OSX came out everyone abandoned poor people like me
Deltron 3030 - Virus (music video)
It seems to me that this move by Microsoft to withhold future versions of IE from their current operating systems might backfire. I've never been a fan of having to change the OS completely every couple years, and the hardware specs and integrated "security" (that is, RIAA controls) make me skeptical of Longhorn. For myself, this will likely be enough of a push to get me to say "forget it" to the next iteration of Windows, and spend the time to properly learn to use Linux.
It occurs to me, as I read all of the posts here about Microsoft failing to release an update to Internet Explorer until Longhorn is released in 2006, that many people are missing a few key points.
One is that Windows XP Service Pack 2 is slated to add a few features to Internet Explorer (a pop-up blocker, for example).
Another is that 2006 is not quite "several" years away, anymore. In fact, it's only a year and seven months (and one week and two days) away.
I'm surprised that nobody (at least who's been modded up enough) seems to have said anything about two particular projects that attempt to deal with a couple of the main problems that developers tend to have with MSIE.
"IE7" is an Internet Explorer hack that parses standards-based CSS that you provide in a page, and mangles it so that earlier versions of IE display it how it's supposed to be displayed.
"PNG in Windows IE" is a hack that tells IE to use a separate ActiveX control to load any PNG's in the page, instead of the internal image display code. This causes it to get alpha blending right. (I think there are a few variations of this hack around the web besides the one I've linked to.)
Both are javascript hacks that you can include at the top of a page and add the appropriate construct around them so that only IE will see them. Clearly they're not perfect, and I'd be edgey about using them in important websites without a lot more testing.
But has anyone actually used them effectively? How useful are they?
I've managed to get the PNG hack working, but I still haven't been able to get IE7 going. (Possibly something to do with the server sending the wrong MIME type.)
The future of the World Wide Web? It has no future. The question is meaningless. It's a bit like saying "What's the future of email?". Email is and always will be simple text with the useful extension of being able to attach files. Anything else is just tinkering (think about the true utility of HTML email and auto-executing scripts). That's it. There is no more to see here. Likewise the WWW is simply hypertext documents with the useful addition of images and tables. In the same way that email has been/is being extended unnecessarily, so too the web: tedious flash; javascript pricking around and now the joy that is CSS.
Taking CSS as an example (to counter the pro CSS bias in the article): if I want to separate formatting information from content then I'll make that choice and write in that manner locally. Just because you can't think of websites where the style and content are unified doesn't mean that I can't. Further, does CSS solve the most important Web related issue? Does it enable your non-techno-obsessed friends/relatives onto the web? No. It simply makes it more confusing and makes them further reliant on software that they simply have no interest in buying. I am bored sick of reading web pages by the kind of people who would read the CSS specs. *I've* read the specs and I already know all about people like me. I want the web to be filled with cool stuff written by non-tech literate people. Further more, if that means that I have to wade through sites written in #ff0000 on #ffff00, then fine, bring it on.
The deeper point here is that the whole idea of creating setting web-standards from on-high is a bit silly really. You cannot and should not be able to control what happens out there. You certainly can't do it in an autocratic (top-down) way, especially when you used bottom up as your initial model. The harsh truth is that things just 'happen' to be great/useful whatever. It of course helps if the originator's been clear/accurate in his description, but it isn't strictly necessary. You most certainly *do not* just don't get this kind of stuff as a product of 5 years of committees.
I have to say the original article strikes me as whining from someone who's just realised that no-one can be arsed to fully implement all the CSS specs (and other W3C dross) he wants. My reaction: "Great!". I for one am well pleased that MS is leaving the Web alone for a while. If we have a new version of IE every year, we wouldn't be able to go to any website without getting some dumbass virus. Also, I am *really* pleased that the guys who have to actually fork out for the coding time are realising that there simply isn't any return on their investment anymore. The Web *is already* what it was designed to be: hypertext pages (and URL). It's had some nice additions - a *long time ago*- but stop already!
The next big thing? The thing to replace the WWW? Here's a clue: it ain't the Semantic Web.
Sorry about any attitude, I've got a mother of a cold today.
The link you gave is bad. There is a DNS lookup failure. Maybe you meant to link here instead.
now we need to go OSS in diesel cars
Your explanation looks just came out of a colledge class, so I'll add a bit more to what exactly semantic web is like.
/>Cheapest : $5</p>' with bunch of other text with meaning less tags.
Right now HTML means quite almost nothing to computers. Yes 'title' and 'h1' may mean something important about the site, but who uses 'h1' properly? I guess there are 'p' and 'div' all over the place with some CSS put on it or the obsolete 'font' attributes...
Now, human looks at HTML not by code, but by visual browsers and being the human on readers' side as well, we can guess what the page is trying to say, even it looks dumb ugly or all the content is tagged with only 'p'.
If a machine tries to read, such as through php or whatever crawling web pages, how are they supposed to know which word is important in that page? or what it wants to tell? The only thing right now that is viable probably is to get how many times a word had come out, or something like that, all guess work. In semantic web, we tag stuff in a more understandable way. From above example, we tag price in a probably '<price></price>' tag and the name of the stuff in '<item></item>' tag. That's a custom XML but, far better than '<p>We have jeans<br
If all web sites become like that(won't happen for many years), according to the rules of those web pages, you can gather the actual data stored on web pages. Right now, even google can only get text information, not like price etc the real data.
Making it easier to read by machines, will save search engines and whatever wants to go through your web information.
Currently HTML really is a bunch of tags that can sort of do anything in terms of text handling, but NOTHING professionally. This has to change soon.
The blurb mentioned CSS, SVG, and XHTML as example web-standards. Truly, those standards are in a waiting-phase. SVG less so than CSS & XHTML. I think the genuinely interesting protocols & standards are the ones less tied to the browser and Internet Explorer. For instance, Semantic Web standards like the recently-approved RDF and FOAF will be interesting to watch. Also interesting is how the instant-messaging protocols will shake out. Personally I want to see the XMPP standard used by Jabber win, but I honestly don't know where it will end up. The point is, there are many many web protocols and standards. The innovation will continue for some time longer.
Sure there will be those who abuse SVG to do cheap Flash imitations, but you can kill somebody with a hammer too.
What you need to think about is areas where many people still find a pencil and paper or a whiteboard easier to use than a computer, and that starts with diagrams. Words alone often have a lot of trouble telling you what a simple diagram can make perfectly clear.
It is an unfortunate accident of history that when the original Macintosh came out, Bill Atkinson's neat trick (MacPaint) beat the promised object oriented MacDraw to market by many months and the Mac offered independent Font, Style and Size menus unmediated by styles, thus setting expectations about graphical computer interfaces back for 20 years and counting. Then we got the Web with two raster formats and no object graphics and Web representation of useful graphical information was left stranded in the too hard basket.
I recently bit the bullet on an esoteric website and put as SVG diagram on the front page, along with other things like Unicode characters that seriously test the limits of browser and system support, but that SVG for now must rely on a plug in for viewing and is only one tenth of one percent of what I'd be trying to do if the tools were in place.
-- Our systemic servants do not good masters make.
Plain old HTML is OK. XHTML + CSS is better. But the real future of the web is XML.
.Net relies heavily on XML and is strongly oriented towards web services. EI was very early in supporting XML + XSLT (but, of course, not-quite-standard, the pricks!). Microsoft, through VB, has historically been successful in selling tools for client-server style development, and that model is strongly intrenched in the community of corporate developers on the MS platform, (and older platforms like CICS).
Unlike HTML and XHTML, which are essentially document presentation languages, XML (used semantically) gives you complete separation of content from layout, style, and formatting. This gives the browser more freedom to render a given chunk of content in different ways -- radically different visual layout, braille, speech synthesis, etc. This also gives you the ability to write client-server applications using XML over HTTP as the communications protocol.
This is, almost certainly, where Microsoft is heading.
This kind of web app has real technical advantages over an html based web app. More work is done on the client. A richer GUI can be used. Smaller downloads per page-hit. There's greater decoupling between the server and the client platform. An XML based web service could support browser based and non-browser based clients. Easier to automate (screen scraping made easy!).
As to rendering XML content on the browser, my feeling so far is that neither CSS (in its current form) or XSLT is an optimal solution. CSS is limiting, and tricky to get basic things to work. (vertical centering, anyone?) More importantly, CSS is tied to the assumption that the thing you're formatting is a document. What if it's not? Think arbitrary XML here -- database records, spreadsheet cells, a stock ticker, a graph of a mathematical formula, whatever. XSLT is more general, but just plain quirky and weird. Functional programming is foreign to most of us who cut our teeth on curly braces.
It's strange to me that no-one on slashdot seems to recognize this. This isn't that far outside the box folks! It's not all div tags vs. table tags. Zoom out to the big picture. Think a little.
-cbare
> Most of the people who jump on the "XHTML is the future" bandwagon somehow get mixed up into thinking XHTML doesn't contain all the presentational crap that HTML does. That is the difference between Transitional and Strict, not the difference between HTML and XHTML.
You have to start somewhere. I think you have a good point here, but for me, the use of XHTML 1.0 Strict and CSS has been nothing but a benefit to my websites, and I happen to also think that it's perfectly feasible to create well-formed XHTML, and beneficial.
Many of the html tags you're pointing at are depricated with CSS. Why use them? Do you know how much processing power and bandwidth is wasted by websites pushing font tags instead of properly using CSS? It's wasteful.
The dangers of knowledge trigger emotional distress in human beings.
webstandards (CSS2, DOM2, XHTML1) barely work amongst all browsers (Konquerer/Safari being the buggiest), and without intuitive tools that support the standards, no one really cares. GoLive and Dreamweaver are started to get with the program, and tools like freeway by softpress are nice. But for other stuff like SVG, there's nothing comparable for Flash. So SVG is mainly in the realm of the geeks.
.... *silence* ...
Now, as for newer standards like XHTML2, W3C is in outer space. There's no browser going to touch it with a stick. Got to be a rocket scientist to use it, and where are the tools for non-geeky designers? Where's the browsers to do it? *pauses*
Yep. That's what I thought.
I hope designers get off the table design with spacer.gifs littered everywhere and just pretend Netscape 4 and IE 4 don't exist. In fact, Netscape as a browser doesn't exist, as after AOL got bling bling from MS, the whole Netscape team (sans the portal people) got pink slips. C'est la vie Netscape.
So where does this stand with Microsoft. Well, now they are starting to get it. But for Microsoft customers to get it, they need to dish out the dinero.