Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Opera Google

Opera's Haakon Wium Lie On CSS, Web Standards, and More 107

Posted by timothy from the have-you-read-our-libretto? dept.
mikemuch writes "The standard that eventually became CSS was originally submitted to Tim Berners-Lee et al by Haakon Wium Lie, who continues to have new ideas for the web formatting language. The latest proposal from the current CTO of Opera Software is the CSS Generated Content for Paged Media Module. Lie sat down with PCMag to discuss not only this scrollbar-free browsing initiative, but a wider range of Web topics, including thoughts on powers like Apple and Google. A teaser from the story: 'At Opera, we sometimes wake up in the morning and see a new Google service that could have been optimized if we could have worked with them in the development phase. It seems they're more eager to put out things and see what sticks.'"
This discussion has been archived. No new comments can be posted.

Opera's Haakon Wium Lie On CSS, Web Standards, and More More

Opera's Haakon Wium Lie On CSS, Web Standards, and More

Comments Filter:
  • But is he referring to optimized in general, or specifically for opera. Because honestly if he means just for opera why would google even bother? I anticipate that he meant in general however.
    • I was thinking the same thing; why should Google care if something is optimized for Opera? They have their own web browser to optimize for.

      • In the context, it sounds far more like optimized from a design point of view, ie instead of just slinging stuff out to see if it'll stick, they could focus on the design and create something that they know people will love. Apple really go the whole hog when they add even the most trivial features or concepts to their software, making everyone think that stuff like a lame version of multitasking is manna from heaven even when other platforms already have proper multitasking and so on.

        I'm having to learn th

        • Re: (Score:1)

          by Anonymous Coward

          Sounds like you're doing well for the most part, but coming up short in documentation. Pressing F1 for help should bring up an index listing such program features. (There are a lot of programs out there which are half decent, but many of their features come across to the user as "easter eggs" since their documentation sucks.) Some software also has a "tip of the day" screen on start. It can be annoying, but most (if designed well) can also be easily disabled. Tool-tips should also be a possibility enabled t

          • Well, it's mostly web apps that I do, so more tooltips would make more sense than F1 style loading of a help document.

            I've only recently started considering that I should write up some documentation myself, even before putting down any code. Documentation is obviously is beneficial to new users, but while reading other people's views on documentation, I read that writing up the documentation often helps you to think more clearly about the design and purpose of your software. I hadn't considered that before.

        • Re: (Score:1)

          by Anonymous Coward

          Optimized in two fashions. One, to make sure it's a *good* implementation of something, and two, to make sure it works across all browsers, not just Chrome. (So, Opera, Firefox, Safari, IE...)

          Also, if they worked with him - as he's also on the W3C - they could, possibly, get buy in from the people responsible from those other browsers.

          Paraphrasing: It's unfortunate Google is making nearly proprietary things for their browser when they wouldn't exist without the open web. If they release an "open" standar

    • But is he referring to optimized in general, or specifically for opera.

      I think he means "optimized in general"; Google has a preference for throwing something that works into the wild (usually, as a clearly designated beta, demonstration, etc.) and then getting outside opinions on how to make it work better rather than tossing out abstract proposals with no implementation.

      Lie seems to prefer people discussing the proposals, refining them, and then putting together an implementation.

      There are arguments for b

  • This is how rumors get started.

  • As you can see from the specification page, Bert Bos also worked on the CSS spec. Bert and Håkon also wrote a book together "CSS: Design for the Web" covering CSS. It's not as practical as some CSS books, but it certainly covers the spec and explains why things are the way they area. (especially the first edition of the book)

    http://www.w3.org/TR/2008/REC-CSS1-20080411/ [w3.org]

  • The whole idea of CSS was to separate content from presentation. But it never lived up to that promise for me. It would be more accurate to say that it separates content from *font* presentation. What would be REALLY useful to me is a way to separate out the actual layout of the page from the content. I can do this now with php (and I do it on most of my sites now), but it would be nice to have it native to html/css. The way I have it set up is that the header of the page (with all the header graphics, page

    • Re: (Score:1)

      by Anonymous Coward

      Good god man, are you living in the past?

      Are you a man from the past??

      If you think all CSS can do for you is style your text... yikes

    • Comment removed (Score:4, Informative)

      by account_deleted ( 4530225 ) on Thursday October 27, 2011 @09:22AM (#37855246)
      Comment removed based on user account deletion

      • Re:CSS and why I never bought into it (Score:5, Informative)

        by Jim Hall ( 2985 ) on Thursday October 27, 2011 @09:29AM (#37855338) Homepage

        I also recommend the poster look at an example of how just changing the CSS can dramatically change the appearance of the size: CSS Zen Garden. [csszengarden.com] Their HTML has maybe a few too many divs and spans (they did this to make it easier for designers to apply new styles) but it's a great demonstration of what CSS can do for you.

      • Re: (Score:2)

        by dabadab ( 126782 )

        I think this separation of content and presentation fails in many cases.

        One is inhertiance: you can't just say ".style2 extends .style2" because there's no such thing, you have to either hack with copy'n'paste or adding some more divs to your HTML.

        The same goes for absolutely ordinary things like adding a border around a region: the CSS border property is absolutely underpowered so you have to add multiple divs to your HTML.

        And it's not like that these issues are unknown or something, these are widely discu

        • Re: (Score:1)

          by searlea ( 95882 )
          With respect to ".style2 extends .style1", that's pretty basic:
          .style1, .style2 { ... common properties ... }
          .style2 { ... override properties ... }

    • I'm not sure if you're being facetious or not, but CSS is far, far more powerful than you seem to think. Check out this site to see what can be done without changing a single line of HTML:

      http://www.csszengarden.com/ [csszengarden.com]

      Also, if you're still writing php spaghetti code, I would highly suggest you check out some modern templating engines. They will make your life so much easier. I don't use PHP that often anymore, but I understand Smarty [smarty.net] is still popular.

      • Re:CSS and why I never bought into it (Score:4, Insightful)

        by QuasiSteve ( 2042606 ) on Thursday October 27, 2011 @09:51AM (#37855658)

        I'm not sure if you're being facetious or not, but CSS is far, far more powerful than you seem to think. Check out [CSS Zen Garden] to see what can be done without changing a single line of HTML.

        That's all good and well, but I think the point GP was making - in jest or not - is that CSS is still at the whim of the structure presented by the HTML.

        For example, if your header has a single div, you can use CSS all you want but you're not going to get 4 separate texts in each of the corners of that div (actually might be possible, but no cheating by using js)

        Thus the reason that the CSS Zen Garden website works as well as it does is not just because of CSS, but because the structure is well thought out as well.

        So what GP is saying is very much true. You'd still want to keep your headers and footers and many other things in separate files that can easily be included by a great many other pages, so that you need only update that included file to adjust all those other pages' results.

        And, in a way, that might make some think "well then I don't need CSS if I only need to update that single file.. why specify styling attributes for the header when I can adjust them in the header include file itself?".

        But they then miss that headers often share visual aspects with the rest of the page. A background color change, for example, might be simple enough by editing your header, sidebar, content and footer pages.. but even simpler still is editing the CSS that governs all four.

        CSS wasn't meant to make structure organized, just style (which does include positioning, but I'm wishfully thinking GP wasn't referring to that) - in effect, exactly what he's saying.. separating content from presentation - but that presentation still relies on the structure that may or may not be considered part of the content.

        • I appreciate your point of view, but I think the OP is conflating the two issues. I completely agree with you that CSS is at the mercy of good or bad HTML. CSS Zen Garden shows what kind of presentation magic you can do with well structured HTML.

          Remember back when people talked about "the semantic web"? A lot of it was dreamer bunk and stuff that is not likely to be implemented ever, but I think at some degree, it transfers to just simple HTML. If you have a list of something, put it in an ol or ul. Even if

          • Separation of content and presentation can never be fully achieved so long as the structure of the content is bound to certain constraints. For example, at least in western regions where top-to-bottom representation of lingual content is the norm, artificial importance is placed upon the top-to-bottom parsing of HTML documents. There are just certain things that you can't do in CSS if one element precedes another; the exact ordering of the elements is important for the rules to be interpreted properly by

            • For example, at least in western regions where top-to-bottom representation of lingual content is the norm

              Is there anywhere else in the world where top-to-bottom representation of content ISN'T the norm? Obviously right-to-left is fairly common (Arabic and derived scripts), Chinese can occasionally have a kind of modern boustrophedon with right-to-left and left-to-right, a few (traditional Mongolian) is vertical columns of top-to-bottom in left-to-right order, etc.

              Does anything go, eg, bottom to top?

              • Japanese (and I'm assuming Chinese as well) is traditionally written in vertical, top-to-bottom columns that are arranged right-to-left. I think it's almost unheard-of online, probably for the reasons stated above, but common in print, for example newspapers [geotypografika.com].

                Take a page of English text and rotate it 90 degrees clockwise, and you'll get the idea.

                Ooh, I wonder if you could write a tablet program which, when you rotate the device, only rotates the characters instead of the entire page.

      • I don't use PHP that often anymore, but I understand Smarty is still popular.

        Ugh, Smarty is a bloated POS. It certainly is popular, but you're better off using PHP to template itself. Less is more!

    • XSLT.

      • Agreed. My reading is that the GP is narrowly correct: CSS can't "completely" reskin a site, because it depends on the underlying html which does have some structural semantics to it. XML data + XSL templates (possibly several, for different end-devices and/or user agents) + CSS would more closely fit his goals, but the XSLT approach wasn't nearly as widely publicized as CSS, so I don't completely blame him for missing that boat. Even today's popular toolkits encourage "backing pages" for data, but the laye

        • Re: (Score:2)

          by putaro ( 235078 )

          A large percentage of my company website is XML that gets translated into HTML via XSLT. On the plus side it works fairly well and getting our non-designer employees to update information in the XML files is straightforward.

          On the minus side, we've found it nearly impossible to find web designers who can wrap their head around the idea of making a template page or to write an XSL stylesheet. Typically I wind up with a monstrosity in HTML that then has to have the template ideas teased out of it. It might

    • Re: (Score:2)

      by REggert ( 823158 )

      document structure != layout

      The header and footer are part of the structure of the document.

      The fact that they appear at the top and bottom of the screen is part of the layout.

    • The whole idea of CSS was to separate content from presentation. But it never lived up to that promise for me. It would be more accurate to say that it separates content from *font* presentation.

      If by "more" you mean "less", then this is true.

      What would be REALLY useful to me is a way to separate out the actual layout of the page from the content.

      Content: use semantic HTML (in HTML5, that's pretty easy, since even the old physical tags that have been retained are now semantic tags, but you may need to revie

  • Whats a Wium ? Some new Nintendo console

  • Opera's Haakon Wium Lie On CSS, Web Standards, and More
  • You lie!

  • I'm not saying it's impossible, but I haven't seen multi-column layout with images spanning columns done in JavaScript. You reach some walls in JavaScript.

    Can anybody figure out what he's trying to say there? You wouldn't even need Javascript - you'd do that with some very basic CSS. I don't see the problem he's trying to point out.

    • I'm not saying it's impossible, but I haven't seen multi-column layout with images spanning columns done in JavaScript. You reach some walls in JavaScript.

      Can anybody figure out what he's trying to say there? You wouldn't even need Javascript - you'd do that with some very basic CSS. I don't see the problem he's trying to point out.

      *And* you can use JavaScript to alter CSS on-the-fly. So it's a pretty bizarre answer.

      • Yes you can do it in Javascript, but the code would be convoluted. Someone may want to create a library for that. It's like writing macros for LaTeX.

        Maybe the walls he was talking about was the walls he was talking about refers to the limit of how complex web authoring are allowed to be. At least that's how I see it, but it's easily mitigated by introducing a JS library for it.

        • Yes you can do it in Javascript, but the code would be convoluted. Someone may want to create a library for that. It's like writing macros for LaTeX.

          Maybe the walls he was talking about was the walls he was talking about refers to the limit of how complex web authoring are allowed to be. At least that's how I see it, but it's easily mitigated by introducing a JS library for it.

          http://en.wikipedia.org/wiki/Inner-platform_effect [wikipedia.org]

        • Yes you can do it in Javascript, but the code would be convoluted.

          Proof-of-concept demo [tinyurl.com]. Click and drag the empty image frame to dynamically relocate it.

          (Only works in browsers which recognize the data: URI scheme. Tested in FF and Opera.)

          • Re: (Score:2)

            by Ster ( 556540 )

            ... (Only works in browsers which recognize the data: URI scheme. Tested in FF and Opera.)

            Works in Safari 5.1.1 (on OS X Lion) as well.

            -Ster

            • The only browser that I know probably won't support it is IE (even the newer versions, I think - support for data URIs is limited to certain parameters, e.g. backgrounds), but I was also unsure of the mouse events since clientX, clientY are not computed the same over all browsers. Although since it doesn't care about the actual x,y of the click, just the relative distance between two clicks, that shouldn't cause a problem.

              It is handy, though, to have a free host for a few k worth of HTML to do a demo like t

    • Say for example, I want an image at a between certain paragraphs, and it spans two columns, and any text on the second column should wrap around the image instead of written over it. How do I go about doing that? In fact, someone else was also asking the same thing here: http://stackoverflow.com/questions/4577380/css3-columns-and-images [stackoverflow.com]

  • The standard that eventually became CSS was originally submitted to Tim Berners-Lee et al by Haakon Wium Lie

    which was in October 1994 [google.com] BTW.

  • CSS is an awful standard with nebulous overrides and unclear behavior. The fact of the matter is you basically have to use a visual editor just to get the CSS correct and even then it looks different or has some tweaks on every browser. Mystery white space and unclear inheritance, the fact that "height: 100%" basically never works for anything, and to add to that the fact it's not really a format like anything else used in web development (maybe a little JSON esque?) just emphasizes how bizarre it is. Oh, a

    • Re: (Score:2)

      by kmike ( 31752 )

      Hear, hear!

      I absolutely agree. It feels like an elaborate thought experiment carried out for the sake of experiment, and artificially handicapped by many limitations to the boot. I'm surprised a more sane thing wasn't proposed at the time.

    • Fully agree with this. In many ways, CSS is the hardest computer language I have to work with regularly, and creates the most headaches.

      Yes, it got annoying, random limits, too much of what you want to do is non-obvious, some of the changes are more or less unpredictable, it's too dependent on the HTML structure, and then you got varying browser support on top of that.

      SASS does a good job of bringing some sanity to it, but it only adds some basic stuff that should have been standard.

      But, on the other hand,

  • "It seems they're more eager to put out things and see what sticks" I guess he's not aware that this is Google's standard mode of operation as a business. Does this man live in a box, developing a browser nobody needs?

    • Re: (Score:2)

      by jgrahn ( 181062 )

      "It seems they're more eager to put out things and see what sticks" I guess he's not aware that this is Google's standard mode of operation as a business. Does this man live in a box, developing a browser nobody needs?

      Perhaps he's questioning Google's standard mode of operation as a business? Did that occur to you?

      • No. He's whining about how they develop software faster his team can adapt. The small players adapt to the bigger ones. This is regardless (usually) of who invented what.

    • Re: (Score:3)

      by UpnAtom ( 551727 )

      The browser that all other browsers are based on you mean?

      The browser that was out 10 years before Firefox?

      He's being polite about Google. He's basically accusing them of behaving like Microsoft.

      • "He's basically accusing them of behaving like Microsoft." You mean like a corporation? I think he's whining. He wants people to come to him for approval. "...if we could have worked with them in the development phase" Do companies have teams of product specialists who go around asking "What outside parties can we involve in the development phase to make this project take longer?" I don't think so.

        • Re: (Score:2)

          by UpnAtom ( 551727 )
          Not all companies out there exist to screw us out of as much money as possible.

          Google is supposed to be a company that Does No Evil. I think it's kinda obvious that's no longer true.
          • The philosophy of a corporation doesn't always align with its strategy, as in this case. To increase market share the way Microsoft does is what I would consider 'directly evil'. Google's strategy of "throw it at the wall, see what sticks" is maybe evil as a byproduct of its being somewhat irresponsible.

Slashdot Top Deals

Variables don't; constants aren't.

Close