FireFox non-standard event object property breaks KLM navigation in Opera

This makes me really annoyed: after fixing bugs and patching and in general trying hard to make things work, KLM's website is broken AGAIN. The navigation at the top does not unfold in Opera, making most of the site inaccessible.

The reason turns out to be an error happening here:

Navigation.prototype.open = function (e) {
if (e) if (e.originalTarget.innerHTML == "Home") return;
if (e) if (e.originalTarget.id == "navigation") return;

(from this JS file )

What, I ask myself, is this originalTarget property that we don't support? It's nowhere in the W3C DOM events specification. Even a Google search doesn't throw up much information. It's obviously a Mozilla-thing though.

So, we're again playing catch-up with a new, unknown and badly documented extension to the standard. We have to reverse-engineer it with test cases to understand what is going on. Meanwhile, important sites will be broken because they insist on relying on non-standard code whatever we say. Users might abandon Opera because of it, or not switch to us in the first place.

You know Mozilla, this sounds an awful lot like the game IE has been playing with us for all those years. If anything you're worse than Microsoft because MSDN usually has at least a minimum of official documentation. And you still claim to be "an advocate for standards on the Net"?? Yeah, right.

Advertisements

43 thoughts on “FireFox non-standard event object property breaks KLM navigation in Opera

  1. don't you guys know some people over at mozilla such that you could just make a phone call instead of doing the time consuming reverse-engineering?

  2. This is a time where a defineGetter or defineSetter would be nice :pOriginally posted by Kelson:

    Why someone might use it for a web page, I couldn't tell you.

    Well, Mozilla's core is all made of JS. The problem is that that entire core is exposed to the regular webpage environement. If you try Google Calendar and Mask as Mozilla it'll simply fail because Opera doesn't support the core method of the Document interface getBoxObjectFor(). Not quite standard uncompliance, but stupidity !

  3. Hey ! Look what I've found ?! 😀

    function addEventHandler(element, type, handler) {
        try {
            element.addEventListener(type, handler, false);
        } catch(inferiorBrowserException) {
            if(element.attachEvent) 
                element.attachEvent('on'+type, handler);
            else 
                element['on'+type] = handler;
        }
        return [element, type, handler];
    }

    They're Mozilla evangelists ! :DIt's all explained !

  4. I might be very wrong here (can't check this right now), but doesnt target always point to original element (target) that invoked the event? At last in standard compilant browsers?So maybe it would be enough, for problems like this, to implement this originalTarget property by just making a "symlink" to target property?Ignore me if I'm talking nonsense.

  5. It leave me wondering: is it enough to only support and encourage the use of web standards, or should the use of non-standard things be discouraged and/or blocked too (where practical)?I don't think there's any reason for Mozilla browsers to expose their inner workings to the public Internet – I'd almost be considering that a bug to be fixed.BTW, originalTarget appears in old W3C tech note of XBL (which looks to be a copy of the original Mozilla documentation), but appears to have been replaced with just target in the current draft.

  6. I'm wondering, does Opera community do anything beside of attacking Mozilla? I'm very sure that in last 2 months you attacked way more times Mozilla than Microsoft. Thanks, friends.Anyway. If the author would be enough smart, to ask the sources before attacking them, he'd probably know how stupid his post is. But it seems that you're not interested in any answers, you're just interested in attacking. So, sorry for interrupting, keep doing what makes you feel better.

  7. @zbraniecki The author searched already and didn't find anything, does that explain it to you?Something on a site that is hardly available through a google search of in the w3c can cause problems, at least all or most of microsoft standards actually get documentated and actually searchable. If alternate browsers should stay with standards, then its obligatory for mozilla and other browsers to restrict proprietary code outside of rendering pages.

  8. Originally posted by zbraniecki:

    I'm wondering, does Opera community do anything beside of attacking Mozilla?

    Sure. But when Opera makes mistakes, Mozilla trolls (including Mozilla employee Asa) will never let Opera and its users hear the end of it. So when Mozilla fouls up, such as trying to trademark someone else's feed icon or including proprietary and undocumented web featres, they must accept being criticized for it.

    I'm very sure that in last 2 months you attacked way more times Mozilla than Microsoft. Thanks, friends.

    So when Mozilla acts evil, everyone should cover their eyes and ears, and pretend nothing happened? Sorry, but not everyone is a Firefox fanboy who thinks Mozilla can do no wrong.

    Anyway. If the author would be enough smart, to ask the sources before attacking them, he'd probably know how stupid his post is. But it seems that you're not interested in any answers, you're just interested in attacking.

    The problem is, this would be expected from Mozilla. They pull this crap all the time. They even spread lies about Opera through Asa's blog and other more or less official propaganda outlets.If Mozilla exposes proprietary functionality to web pages, they messed up. Attacking Opera won't change Mozilla's actions.

  9. > So when Mozilla fouls up, such as trying to trademark someone else's feed iconIt would be really great if (why do I have to repeat myself?) you'd decide to get more knowledge about the issue, before making such stupid statements. Mozilla did not "fouled up", if you really have problem to understand what were the reasons of Mozilla move with icons, please, read all threads about it on Frank Hecker's blog and Marcia Knous. Then we can talk about it once more.> or including proprietary and undocumented web featresYou see. I really feel sad whenever I read such dumb statements. Tell me please, is XUL "undocumented web feature" too? Hey, you can use it in the website! So? Of course you know that Gecko beside of being a browser engine that supports web standards also is a full-featured programming platform with tons of programming API's, right? You obviously know that in Gecko people write VoiP apps, mail clients, web browsers, database management tools, music managers, IM clients, File managers, rich text editors, HTML editors etc.? You must probably already know also, that current web standards does not provide solutions for most of required technologies for such apps. Yes, of course Mozilla is trying to standarize what's not standard yet (like XBL, XUL) and I'm sure that it'll happen soon, but I think that it's obvious that beside of web standards that allows Gecko to properly display web pages, Gecko is much more. And if some dumb author is deciding to use platform feature instead of web standard, it's his mistake, not the authors of the platform. I feel disappointed by this post because the author of the post didn't try to think or ask Mozilla people. He automaticly decided that lack of his knowledge allows him to attack someone and blame him.If you'd take a look at developer.mozilla.org, you'll find that we strictly split what's the part of web standard world which we encourage people to write websites in, and the Gecko RAD platform. I'm still very disappointed to see how much you all focus on attacking Mozilla, as your new main opponent. You just listed two false accusions from last weeks, but reading Opera's planet in both EN and PL shows much more attacks, none of which has any logical reason. It's either false (like this one) or simply empty, like "let's compare Opera 9 and Firefox 1.5 to once more prove what I believe in, that Fx 1.5 is featureless comparing to my sweet, beloved Opera". I didn't see any comparsion with IE6, IE7, I didn't see any post about new holes in IE (including IE7.0), but I saw a lot about old holes in Mozilla that were either fixed or never exploited (there were never any evidence of successfull attack via Fx!). It makes me feel that you silently accepted that you have no way to reclaim more web from IE than Firefox did, so you decided that the only way to expand is through "easier" to convince Fx users. It's ok, I'm not afraid of Fx, and I'd prefer users to use the best browser they want, and if it'll be Opera, than it's almost OK (I'd prefer Opera to go open-source), but I'm simply sad that we're alone in the anti-IE crusade now (well, not THAT alone, there's Camino, Firefox, Seamonkey, Flock… ;)).If you'll have any problem like that in the future, and you'll suspect that Mozilla decided to use it's open source power to be evil, just mail one of the newsgroups or join IRC channel and ask. It'll be so much simpler.

  10. > is it enough to only support and encourage the use of web standards,> or should the use of non-standard things be discouraged and/or blocked> too (where practical)?Non-standard things that have standardised or cross-browser-compatible equivalents should certainly be discouraged.zbraniecki: I don't pay much attention to zealots on either side so I didn't expect you to see this as part of a pattern of anti-FireFox attacks. I've patched KLM.com in browser.js, it's one of my pet causes to have it working. Naturally I'm very annoyed by this problem and was posting in the heat of the moment when I finished debugging it and found the issue. However, I stand by everything I said in my rant and don't consider it an un-deserved attack on the Mozilla crew and software.* Implementing something in a browser is an invitation to authors to start using it* If it is a feature required for XUL, it should work only in XUL documents* If Mozilla wants to support it in HTML documents, it should be submitted to W3C or WHATWG for documentation and standardisation* In real life the effect of Mozilla's originalTarget feature is exactly same as Microsoft's "Embrace and extend": something we don't know we have to support until sites are broken. We both have exactly this problem versus IE. Hence, I'm very disappointed that the standardisation-friendly Mozilla-guys actually add to the incompatibilities.Yes, KLM's webmaster is doing a silly thing here and a tiny and sensible change would make it work. Unfortunately if they do not test their site in Opera they won't make that change. The only way Mozilla and Opera will be interoperable is to follow the same standard to the best of our abilities.

  11. Originally posted by zbraniecki:

    It would be really great if (why do I have to repeat myself?) you'd decide to get more knowledge about the issue, before making such stupid statements. Mozilla did not "fouled up"

    They certainly did. When only a few days were left until Opera 9 was to launch, Mozilla decided to send Opera a letter demanding that they sign a paper to be allowed to use the icon.

    Tell me please, is XUL "undocumented web feature" too? Hey, you can use it in the website! So?

    So someone fouled up there too, as XUL doesn't belong on web pages. (Not to mention the fact that XUL is terrible.)

    You must probably already know also, that current web standards does not provide solutions for most of required technologies for such apps.

    That is obviously nonsense.

    Yes, of course Mozilla is trying to standarize what's not standard yet (like XBL, XUL) and I'm sure that it'll happen soon, but I think that it's obvious that beside of web standards that allows Gecko to properly display web pages, Gecko is much more.

    That doesn't make it right to tie people to proprietary Mozilla technologies.

    I feel disappointed by this post because the author of the post didn't try to think or ask Mozilla people. He automaticly decided that lack of his knowledge allows him to attack someone and blame him.

    Mozilla exposing their proprietary technologies to web sites certainly warrants criticism. If you can't handle criticism, you shouldn't push your proprietary technologies on the world.

    I'm still very disappointed to see how much you all focus on attacking Mozilla, as your new main opponent.

    Who are "you all"?

    You just listed two false accusions from last weeks, but reading Opera's planet in both EN and PL shows much more attacks, none of which has any logical reason.

    Really? What are those attacks all about?

    It's either false (like this one) or simply empty, like "let's compare Opera 9 and Firefox 1.5 to once more prove what I believe in, that Fx 1.5 is featureless comparing to my sweet, beloved Opera".

    First of all, it's true. Second of all, independent people comparing two competing programs aren't "attacks".

    I didn't see any comparsion with IE6, IE7, I didn't see any post about new holes in IE (including IE7.0), but I saw a lot about old holes in Mozilla that were either fixed or never exploited (there were never any evidence of successfull attack via Fx!).

    I fail to see how people talking about Firefox instead of IE is relevant.

    It makes me feel that you silently accepted that you have no way to reclaim more web from IE than Firefox did, so you decided that the only way to expand is through "easier" to convince Fx users.

    "You" being? This is not my blog, in case you didn't notice.

    I'm simply sad that we're alone in the anti-IE crusade now

    How? Lots of Opera users are still anti-IE for some reason. Why, I can't imagine. If people want to use IE, let them use IE.But this is a red herring, since we're talking about Mozilla's immoral acts lately (and through the years for that matter), and how you cannot explain them away, and especially not by attacking Opera users.

  12. > * Implementing something in a browser is an invitation to authors to start using itI strongly disagree, and I'm sure that in short time you'll forget your own words once Opera will start serving more API and extending their platform API.Implementing something in the platform engine (not – a browser), is just a method to provide required feature to the author of the application.And there are much more ways to use HTTP and Gecko than HTML page, really. You can create complex API for an application. And your statement means that it should be forbidden "If you wish to be the web browser, do not even try to do anything more" – that's what you wanted to say? If so, wait few months for next Opera's features and you'll be in a problematic position ;)> * If it is a feature required for XUL, it should work only in XUL documentsOnce again. Wrong. You still claim things but you still don't understand what you're talking about. Please, stop.> * If Mozilla wants to support it in HTML documents, it should be submitted to W3C or WHATWG for documentation and standardisationWhy? HTML is a standard. We can use this standard to provide any feature we need/want. It's like you'd ask Adobe that if they want to base their ActionScript on ECMA, any new feature they add for their own users, should be sent to W3C and standarized before using.And it's not the solution in many cases because Mozilla's goal is not to add new feature to web pages, but to add a feature to Mozilla platform, which (once more) is ALSO (but not ONLY) able to present web pages. Web pages is a SUBSET of what Gecko can do. Web author should write for web standards, and we'll work well with it. That's the structure.> In real life the effect of Mozilla's originalTarget feature is exactly same as Microsoft's "Embrace and extend": something we don't know we have to support until sites are broken.No, it's a bit different. Microsoft PROMOTED to use it in the web pages, because for them, webpages should be made of what they provide. In Mozilla's point of view, web pages are a SUBSET of what we provide, but we'd never suggest, want or ask anyone to use anything beside of web standards.Well, overall, our vision is to get web developers do not think about browsers when they create the web pages, but about standards. To make them ensure that the page works according to standards and then it'll work with all browsers that supports standards.The only problem here is that we need to ensure that we support standards with the browsers, because original problem with why web authors are doing pages FOR browsers is that they can write something according to standard, and because of the browser bug it wont work. And if the browser has high market share the web author needs to choose either to sacrifice standard or browser.> Yes, KLM's webmaster is doing a silly thing here and a tiny and sensible change would make it work. Unfortunately if they do not test their site in Opera they won't make that change. The only way Mozilla and Opera will be interoperable is to follow the same standard to the best of our abilities.Yea! Right! But stop saying "Opera" and stop saying "Mozilla". You still seems to live in the "for-browsers" webpage world. All browsers should follow the same standards everywhere. It would be best for web pages and web authors, so let's work on it!But in this case, Mozilla IS following the standard. You can write the page according to standards and, IIRC, it'll work properly in Mozilla. If not, then it's a bug and you should report it. And internal API has nothing to do with it. Web author shouldn't even know about it until he's interested in building stuff for Gecko. He should read W3C standard and use it. That's all…So yes, I'd say that your post is wrong, because you attacked Mozilla without a reason. More! You accused Mozilla of working AGAINST standards which is not truth and I'd be surprised if you really think that Mozilla is working against them (imagine how the web pages would look today if there would be no Firefox and IE6 would still have 97%…)Nilsen:> When only a few days were left until Opera 9 was to launch, Mozilla decided to send Opera a letter demanding that they sign a paper to be allowed to use the icon.Oh, another Urban Legend. Can you tell me more about it? I heard it a few times, but never see it. And what I saw in Opera's Changelog was that Opera decided to stop using the icon because Mozilla started working on what to do with this icon. No word about any "letter" "demanding" anything. So, can you show me some evidences, please?Also, ignoring the letter issue, did you read the whole story, do you understand what was the idea behind Mozilla concerns? What they wanted to achieve and what they worry about? Do you think that in such situation stating that "Mozilla wants to trademark logo" is a right statement to use? Because for me it sounds like a bullshit. We all know what Mozilla goal is and we all agree that the goal is perfectly clear. Their intentions about this icon are good, right? Or maybe you disagree? So, where's the "foul"?> So someone fouled up there too, as XUL doesn't belong on web pages.Oh, it's so wrong. How did you invented this sentence. Yes, XUL is not a part of web pages world. So what? Only web pages can use HTTP protocol? RDF XML is a standard but it's not a part of web page, really.You're so selfish. "Opera doesn't provide something so it shouldn't exist, and once we will start support it, then it should be"…> Not to mention the fact that XUL is terrible.I hope that you can prove it. Please, do so, because such statement sounds funny to me.> That is obviously nonsense.Really? Nonsense? Tell me more. Cause I claim it's true, and I heard so many times the word "obviously" used not to have to prove the statement that the author could not prove…> That doesn't make it right to tie people to proprietary Mozilla technologies.So, gotcha. You misunderstood the term "proprietary". It's really hard to use it for Open API, and Mozilla is open source. Anyone, anywhere in the world can implement XUL. Please, read the definition of "properietary" once more.http://en.wikipedia.org/wiki/Proprietary_software> Mozilla exposing their proprietary technologies to web sites certainly warrants criticism. Once more.1) As above, read about terms before using them, especially when attacking.2) Opera exposes "window.opera" DOM property to web sites, does it make Opera a bad boy? Should I list more examples or you can think of them on your own and ask yourself a question "what can be exposed and what not?"Also, once more. From our point of view there's no "Web world" and "other world". It's the world of web applications. We allow you to write them in the way you want. If you want to write a web page, you have to use W3C standards, and we promote it all the time. If you want to mix XHTML, XUL, XBL, RDF, E4X and Canvas, that's ok, but you need to know that it's not a web page. If you misunderstand it, it's your fault, not ours. > Who are "you all"?people from Opera community that I can read via Opera Planet, and those trolling all articles and news about any browser with "Opera rulez" satement.> First of all, it's true.And you're willing to prove it and discuss it, right? Beside, you must be a very frustrated person than, beliveing that your browser is such perfect, and seeing that so many people disagree with you. And such statements makes me think of you as one of the blind zealots that can spend any amount of time to walk around and scream that your browser is perfect while others are piece of shit. Congratulations. Black/White world is for sure easier to live in.> Second of all, independent people comparing two competing programs aren't "attacks".Never said of intependent people comparing two competing programs, that's one. Second, Opera is not competing with Firefox. Firefox is a browser, Opera is a Suite. Third, Firefox has one competitior. The browser that makes the web world a bad place. And we're focused on fixing this. Opera is a very good browser and we see no reason to compete with Opera. So it's a one way war (compare planet.mozilla.org amount of posts about Opera and Opera planet amount of posts about Firefox…)> Why, I can't imagine. If people want to use IE, let them use IE.Oh, that's simple. Because in the current model, Microsoft has no interest and is doing nothing to make the web a better place and make it innovate and progress. In fact they block the web progress by releasing and promoting an app that is not progressing at all.> we're talking about Mozilla's immoral acts latelyTell me more. You're making statements so fast. I'm jelaous, your world is so simple. You don't need to check or prove anything. It's enough that you think that it's like this and that's enough to claim it. Beauty.So, can you ask (yes, ask) about things that YOU think are immoral (in the meaning – that Mozilla is doing them to hurt someone/something) and I'll do my best to explain you the reasons and prove that you're wrong.But to do this, you'd have to assume at the beginning that it's possible that your accusions are wrong.> (and through the years for that matter)Same as above. I'm working in Mozilla project as a volunteer for 5 years now. I was coding, managing, promoting, never get a cent from them. I think that I can say that I'm clean, and I'm willing to help you understand how wrong you are.> and how you cannot explain them awayGive me a chance 🙂 Ask :)> especially not by attacking Opera users.I'm not attacking Opera users. I'm stating that I don't understand why they're so crazy about attacking Firefox, and my best two ideas are that either you're very frustrated that you simply don't know how to make a product that majority will want to use (while you create a great browser, on a great engine! It's just that I think that Opera doesn't understand what users want… not the Power Users, the ordinary majority…), or because you think that the best way to promote Opera is via black PR for Firefox.

  13. So yes, I'd say that your post is wrong, because you attacked Mozilla without a reason.

    No, he criticized them for a good reason: For exposing their XUL nonsense to normal web pages.

    Oh, another Urban Legend. Can you tell me more about it? I heard it a few times, but never see it. And what I saw in Opera's Changelog was that Opera decided to stop using the icon because Mozilla started working on what to do with this icon.

    Rewriting history, are we? 🙂Desktop blog:"Mozilla would like us (and other users of it) to sign an agreement on the use of the feeds icon."OperaWatch

    Their intentions about this icon are good, right? Or maybe you disagree? So, where's the "foul"?

    They stepped across the line in trying to force competitors to sign an agreement to be able to use the icon (which wasn't even created by Mozilla apparently).

    You're so selfish. "Opera doesn't provide something so it shouldn't exist, and once we will start support it, then it should be"…

    Um, no. The point is that XUL is not a web standard, and XUL should not be exposed to normal web pages.

    I hope that you can prove it. Please, do so, because such statement sounds funny to me.

    XUL reinvents the wheel. Badly.

    beliveing that your browser is such perfect

    What do you mean by "my" browser? I don't own Opera.

    Never said of intependent people comparing two competing programs, that's one.

    So who were the people comparing them, then?

    compare planet.mozilla.org amount of posts about Opera and Opera planet amount of posts about Firefox

    What's your point? The Opera Planet thing includes blogs from regular Opera users. When Mozilla fouls up and exposes its XUL nonsense to normal web pages, causing problems for standards compliant browsers, why shouldn't Opera users raise the issue? When Mozilla sends a nasty letter to Opera shortly before release and wants them to sign an agreement, why shouldn't Opera users raise the issue?

    I'm not attacking Opera users. I'm stating that I don't understand why they're so crazy about attacking Firefox

    You interpret valid criticisms as attacks, which is really not everyone else's problem.

    either you're very frustrated that you simply don't know how to make a product that majority will want to use (while you create a great browser, on a great engine!

    Why do you keep talking to me as if I represent Opera or as if I created Opera? You are aware that my.opera.com is a site mainly for normal Opera users, right?

    It's just that I think that Opera doesn't understand what users want…

    Oh, so now you resort to insulting Opera Software… ;)Unfortunately, this is not the first time Mozilla officials have made derogatory remarks about Opera. I won't get into Asa Dotzler and his attacks against Opera, but there are other examples too, such as Doug Turner's claim that:"We can be ported to many platforms that Opera can't"Funny, then, that Opera has been ported to a lot more platforms than Minimo, and that there's really no way Turner can make such a claim without access to the source code. So either he is lying, or he is breaching an NDA. I don't know which is worse…"We can be ported to many platforms that Opera can't"

  14. > * Implementing something in a browser is an invitation to authors to start using itI strongly disagree

    So Mozilla implements things just for fun, and hope that nobody uses it? Doesn't make sense to me :pI'm splitting hairs – I know what you meant to say – this:

    Implementing something in the platform engine (not – a browser), is just a method to provide required feature to the author of the application.And there are much more ways to use HTTP and Gecko than HTML page, really.

    Naturally. But the context here is klm.com which is a HTML site, throwing in some AJAX and DHTML effects but not what I would call an application. There is no XUL in sight..

    You can create complex API for an application. And your statement means that it should be forbidden "If you wish to be the web browser, do not even try to do anything more" – that's what you wanted to say?

    No. I wanted to say "a feature required for XUL should work only in XUL documents". I'm not saying "don't invent XUL" and I don't understand why you seem to say these statements say the same thing.As far as I can tell, event.originalTarget is in HTML pages merely an alias for event.target (probably makes sense in XUL apps, which I know nothing about). In the context we're discussing it's not doing "more", it is being a proprietary alias to a standard feature – which means "embrace and extend, damn the others".

    > * If it is a feature required for XUL, it should work only in XUL documentsOnce again. Wrong. You still claim things but you still don't understand what you're talking about. Please, stop.

    Would you mind explaining to me what is wrong? I'm the first to admit that I know very little about XUL.

    > * If Mozilla wants to support it in HTML documents, it should be submitted to W3C or WHATWG for documentation and standardisationWhy?

    So that other user agents can support the same feature in an interoperable way. Anything else means fragmentation of the Web and sites that are inaccessible if you happen to use the wrong operating system or user agent.Fortunately many people at Mozilla do understand this (no wonder with all the problems IE's implementations have caused everyone else over the years) and say plain and simple "this is bad" when it turns out that websites have started using something meant to be internal:https://bugzilla.mozilla.org/show_bug.cgi?id=340571That bug shows they are aware of the problem and I hope they will do a much better job at keeping internal APIs off-limit to web pages in the future.

  15. > For exposing their XUL nonsense to normal web pages.His criticizm is not tied to anything related to XUL, once more you make statements that shows your lack of knowledge of Mozilla platform.> "Mozilla would like us (and other users of it) to sign an agreement on the use of the feeds icon."So, there's no word about a letter.> They stepped across the line in trying to force competitors to sign an agreement to be able to use the iconIf that's all you know about that case you should learn more before saying anything. Mozilla started discussion about how to keep an icon (not only feed icon) still be meaningfull in open source world. The threat was that anyone can use this icon for anything (like, will use it to represent Reload button in his product) and make the icon lack of the meaning for users. You really, really, really should know this before judging the case. You should know much more then what I just wrote, and it's all in blogs, you probably have no will to read.> XUL reinvents the wheel. Badly.Which wheel? XML based UI language? Which should be used instead in your opinion? And why badly? :)> When Mozilla fouls up and exposes its XUL nonsense to normal web pagesThe discussion culture of human being says that once you start discussing issue with someone else, until you both agree, you stop using the statement that someone claim is false. Especially if this is meaningless, highly emotional statement like "nonsense". Beside, your words still means that you don't understand what I just wrote and you have a huge problem to open your mind for other forms of technologies used via HTTP protocol. Please, try.> sends a nasty letter to OperaI questioned that fact, and you didn't prove your statement. You're repeating unproved statements that lacks of meaning value and are full of emontional abuse. It's called FUD. Would be nice if you'd stop doing this.> You interpret valid criticisms as attacksOnce more, I just questioned if the criticizm is valid. It would be nice if you could stop pretending that wrong statements until you prove them. FUD is pointless.> Why do you keep talking to me as if I represent Opera or as if I created Opera?When I say "you" I mean "Opera community".> Oh, so now you resort to insulting Opera Software Insult? Can you tell me where I insulted Opera?> Unfortunately, this is not the first time Mozilla officialsI'm not a Mozilla official, in the same way you're not an Opera official.hallvors:> So Mozilla implements things just for fun, and hope that nobody uses it? Doesn't make sense to meMozilla is implementing things to allow people to create rich applications for Gecko platform. Gecko platform is designed for applications that work via web. It's not intend to replace web pages, it has different goal and just reuses standards so in result people like you have problems to see the split line.> But the context here is klm.com which is a HTML site, throwing in some AJAX and DHTML effects but not what I would call an application. There is no XUL in sight..So KLM just made a mistake. The problem is that you decided to share the guilt and accused that Mozilla is "equally" guilty, has "bad intentions" and is playing "against the web standards", while what we agreed up to this point is that Mozilla is just allowing people to create rich apps and that KLM made a mistake. See the huge difference?> No. I wanted to say "a feature required for XUL should work only in XUL documents".But Gecko apps can be created without XUL. In XML, or XHTML, or HTML, or SVG… That's what I'm trying to explain to you both, that it's not possible to split things to "HTML" world and "XUL" world and expose different API's to both.> it is being a proprietary alias to a standard feature – which means "embrace and extend, damn the others".Point of view is related to the point of sitting. Mozilla uses it's own API for it's own features, there's nothing properietary in it (you can reuse it, reuse the sources etc. – hint: look at the definition of the word proprietary).We support standards. You can use standard. You should use standard and Gecko will work well. You and any web author should never interes in internal Gevko API's as long as he's not going to work on Mozilla app.> Would you mind explaining to me what is wrong?As I explained above, Mozilla platforms allow you to create apps using various technologies. Most of them are standard ones (Firefox application UI is styled in CSS and every button you click fired JS function for example), some of them are extendend due to lack of standard (XBL, XUL, E4X, etc.). We want to standardize them. For example the author of JavaScript – Brendan Eich is working in Mozilla on JavaScript 2.0, that will base on ECMA 4 standard, we also try to standardize XBL 2.0 and XUL 2.0 (will be ready for Mozilla 2.0 platform).We offer tons of methods to the users for using our platform. Those methods are not for web page authors, but we can do nothing in my opinion to guess that this is a web page and not expose any API beside of standards :(Please, open http://developer.mozilla.org. Do you see the split between what's for web authors and what's the internals?Now, look at http://developer.mozilla.org/en/docs/Special:Search?search=originalTarget&go=GoReading this will give you a wider knowledge. It's a part of XBL API. XBL is often used with both HTML and XUL to prepare applications, so it needs to be available from HTML context. That's why it's exposed.> Anything else means fragmentation of the Web and sites that are inaccessible if you happen to use the wrong operating system or user agent.No, you mixes two different things. One of them is a web sites world. The world created by Tim Berners-Lee. Mozilla supports this world but does not only limit to this world. Another world is a world of Gecko based applications. Those two worlds might be tied together at some moment when standards will catch up the needs. It would mean that we could write Firefox as web app and I'd love it 🙂 And I feel that Google is trying to force this way, cause they want to be able to create rich UI's. But for the moment those two worlds are not the same. The first part is also (beside of creating the Web Pages world) a subset of Gecko app world.We promote the first one to go standards, and we offer the second one for NON web pages world so that for example you can write a rich text editor and put it in your intranet in your company and journalists will just open an intranet URL to work on their texts 🙂

  16. @zbraniecki"Gecko based applications" — You just said it, that proprietary to mozilla's rendering engine in itself(being open source doesn't mean every browser can just agree to implement it, its propreitary because its not standardised anywhere else), you think other rendering engines(Webkit(Safari, konqueror),Presto(Opera),Trident(IE & its shells) ) are going to try and implement XUL and other mozilla based languages? XBL & XUL etc needs to be closed from normal web pages unless the so called "standard" is actually recommended. The way I see it, is that Mozilla are being arrogant and forcing non-cross browser technologies upon web developers, which causes them more stress to users when it doesn't work in any other browser, it just simply breaks the web even more.

  17. Now, look at http://developer.mozilla.org/en/docs/Special:Search?search=originalTarget&go=GoReading this will give you a wider knowledge. It's a part of XBL API. XBL is often used with both HTML and XUL to prepare applications, so it needs to be available from HTML context. That's why it's exposed.

    Thank you for a good explanation. Can I then argue that XBL should have been a XUL-only feature? I want to hear Mozilla tell authors: "so, you want to write a web application in HTML? Fine. Use the existing standards. You've hit the limitations and need more? Fine, learn XUL and get the power of XUL and XBL." See – no problems! No danger of damaging interoperability in web pages and scope for innovating in web applications.To further understand my point, please take part in a thought experiment: Imagine that we see two-three years into the future. Microsoft has released Windows Vista and promote applications built with XAML. It just so happens that IE 9 internally doesn't separate the XAMLElement interfaces and the HTMLElement interfaces cleanly. It just so happens that some web page authors and JS libraries, eager to play with any new toy, start using XAML features in HTML pages. All of a sudden 10 000 cutting-edge cool sites do not work in Gecko-based browsers (nor in Opera/Safari/Konqueror et al) because they use XAML-based interfaces. Would you blame Microsoft or the authors?

  18. > being open source doesn't mean every browser can just agree to implement it, its propreitary because its not standardised anywhere elseNo, you still misunderstand the meaning of the word. Read: http://en.wikipedia.org/wiki/Proprietary#Proprietary_softwareProprietary software is not free software or open source software as end-users generally do not have the ability to: * Run the software for any purpose – You can do this * Study and modify the software – You can do this * Copy the software and provide it to third parties – You can do this * Make and release improvements to the software – You can do thisYou misuse the word. You mean "it's not an open, independent standard", and yes, we all agree here. But it's not proprietary. The fact that nobody else decided to use it doesn't make it proprietary since we never blocked anyone from doing the above things.> you think other rendering engines(Webkit(Safari, konqueror),Presto(Opera),Trident(IE & its shells) ) are going to try and implement XUL and other mozilla based languages?If it will become a standard of W3C, than yes, until that happens, no.> forcing non-cross browser technologies upon web developersCan you show me one example of this "forcing"? I proved above that we're "encouraging" users to use standards when they write for web. Show me the prove for your statement or stop using it please.> it just simply breaks the web even moreyou still seems not to understand what I'm writing about. It's not for the thing that you call "web". It's the same kind of technology as Java, .NET etc., but the difference is that we use it VIA the web, so you misunderstand it, assuming that it must be "part of the web" since it uses HTTP protocol. It's not.> Can I then argue that XBL should have been a XUL-only feature?No. Cause HTML is sometimes a better background for an app than XUL.> I want to hear Mozilla tell authors: "so, you want to write a web application in HTML? Fine. Use the existing standards. You've hit the limitations and need more? Fine, learn XUL and get the power of XUL and XBL." See – no problems!There are problems. Maybe we'll manage to fix them with XUL 2.0, but for now, there are cases when it's better to use HTML (for example, Credits window in Firefox is made in XHTML+XBL, some parts of preference windows too).> No danger of damaging interoperability in web pages and scope for innovating in web applications.I agree and I see this danger too. But I find different reasons. The reasons are that people use the wrong technology only because it's possible to use it. And your solution is that we should forbidden to use this technology only to avoid people making stupid things. It's like avoiding window.open only because some stupid people may use it badly.> All of a sudden 10 000 cutting-edge cool sites do not work in Gecko-based browsers (nor in Opera/Safari/Konqueror et al) because they use XAML-based interfaces. Would you blame Microsoft or the authors?Great question. To answer I need to ask others:1) Did Microsoft promoted the usage of XAML-based interfaces for web?2) Did Microsoft opened the standard of XAML interfaces?3) What superior over HTMLDocument XAMLDocument has?4) Why HTML does not progress to support what XAML has if it's so needed by the web that it uses it so widely?I'm happy that you seems to understand my points, and we simply do not agree, it makes it easier to discuss 🙂

  19. Cause HTML is sometimes a better background for an app than XUL.

    That's probably true (particularly given the number of HTML apps that are popping up..) but then the counter-question is why these apps can't be written with the actually standardised parts of HTML/CSS/JS?

    I agree and I see this danger too.

    Thank you 🙂

    But I find different reasons. The reasons are that people use the wrong technology only because it's possible to use it.

    History (and particularly the history of Internet Explorer) teaches me that web authors use anything just because they can. A very pragmatic look at history shows "implementation causes usage, which causes fragmentation and "works-in-one-UA-on-one-OS" syndromes, which limits competition between UAs, creates natural monopolies and makes it insanely hard to catch up with the market leader". A more fragmented, less inter-operable web..

    And your solution is that we should forbidden to use this technology only to avoid people making stupid things.

    Yes, my solution is that UA vendors should show self-restraint and be concerned about interoperability and the state of the web as a whole. As implementors the responsibility rests with us even more than with authors. Thus we should support the standards, develop the standards, and avoid things that may cause incompatibility – or find ways to implement and standardise things in ways that cause minimal breakage in other browsers. (For example the way Ian Hickson's WHAT specifies INPUT type=date so that pages remain entirely usable to UAs without specific support for WF2).

    It's like avoiding window.open only because some stupid people may use it badly.

    Personally I'd prefer if window.open had NOT been invented, but I must admit popup blocking gives us a competitive edge over IE, so.. 😉 Well, that's a digression. It's more like avoiding adding another three commands named window.showModalDialog, window.openChromeless and window.openOperaBrandedPopup just because we need them internally 😉

    > All of a sudden 10 000 cutting-edge cool sites do not work in Gecko-based browsers (nor in Opera/Safari/Konqueror et al) because they use XAML-based interfaces. Would you blame Microsoft or the authors?Great question. To answer I need to ask others:

    Damn, where did I leave my crystal ball last night? OK, can't find it now so if you make MSFT investment decisions based on my response I accept no responsibility for the outcome 😉

    1) Did Microsoft promoted the usage of XAML-based interfaces for web?

    This is the core of our disagreement. I say that simply making something available invites usage, you say that it's OK to make it available if it is also made clear how it is intended used, even if part of the potential use cases may be considered harmful. Makes me want to refer to bugzilla's 340571 again :)I guess the answer two years from now will be "no" and the answer 10 years from now will be "yes". If Microsoft is concerned about political correctness versus the web they would not promote it, though not publicly discourage it. Once sites have started using it the demand for documentation on that would grow and Microsoft would "give in" and document it.

    2) Did Microsoft opened the standard of XAML interfaces?

    It will necessarily be specified in detail since they want people to develop with it. The spec will be twice as large and twice as complicated as SVG's. I don't know enough to even guess at how they will licence it.

    3) What superior over HTMLDocument XAMLDocument has?

    Probably fancy 3D and animation stuff given what I've heard about Vista demos :). Perhaps also interesting approaches to managing security of markup-based apps from less trusted sources. Definitely stuff web application authors will be interested in playing with, and stuff they might then carry over to their next web site job.

    4) Why HTML does not progress to support what XAML has if it's so needed by the web that it uses it so widely?

    UA vendors make 3 un-coordinated attempts that make authors tear their hairs out before finally attempting to get the W3C to standardise alternative interfaces. Authors hate them when the new interfaces are different from even the 3 preceding UA-specific attempts and have very long names with awkward capitalisation ;)Anyway, I think this is becoming so out-on-a-branch speculative it's hardly worth following up the thought experiment. I think you evaded the question with your counter-questions, but we've learnt more about our disagreement so all in all it was a useful exercise.

    I'm happy that you seems to understand my points, and we simply do not agree, it makes it easier to discuss

    Indeed, easier and way more interesting than an average flamewar.

  20. His criticizm is not tied to anything related to XUL

    The point is that it's a Mozilla specific technology.

    Mozilla started discussion about how to keep an icon

    No. Mozilla contacted Opera and wanted them to sign an agreement if they were going to use the icon. Whether it was done by letter or phone or by nailing a Post-It to an Opera skull is besides the point. The point is that Mozilla fouled up.

    Once more, I just questioned if the criticizm is valid.

    No, you started throwing around accusations because someone pointed out that Mozilla fouled up. Criticizing Mozilla for exposing their internal implementations to normal web pages is not an attack. Criticizing Mozilla for not allowing Opera to use the feed icon without an agreement just before release is not an attack (that Mozilla later turned after a storm of criticism all over the web is besides the point).

    When I say "you" I mean "Opera community".

    The Opera community does not create Opera.

    assuming that it must be "part of the web" since it uses HTTP protocol. It's not.

    Mozilla certainly tried to make it part of the web by allowing normal web pages to use it.

    And your solution is that we should forbidden to use this technology only to avoid people making stupid things.

    That's not what people are saying, is it? No, people are saying that Mozilla's internal languages should not be available for normal web pages.

    Great question. To answer I need to ask others:

    No, you do not need those questions answered to come up with the answer that non-web stuff shouldn't be available to web sites.Instead of answering the question, you conveniently dodged it because you know that it pokes a huge hole in your arguments.

  21. > the counter-question is why these apps can't be written with the actually standardised parts of HTML/CSS/JS?No, please, read Wikipedia entry about, for example, XBL to know why it cannot be substituted with HTML/CSS/JS. I feel that wherever we can use standards, we use them and we're doing a lot to keep it and fix all remaining issues where we use own implementation in places where we can use standard.> History (and particularly the history of Internet Explorer) teaches me that web authors use anything just because they can.No, web authors used IE proprietary solutions because Microsoft proposed them to use those, and even enforced them to use those by not providing standard solution support. It's completely different.> A more fragmented, less inter-operable web..I'm wondering how far you live from what Mozilla does and how it progress to make statements like this about them. We spent years on promoting standard support, we were the first that made a public campaign of sending letters to authors of non-standard-support based sites (6 years ago!). The Mozilla Evangelism project was a community effort organized by Mozilla. We are activly developing most of the current standards, we employ the author of JavaScript, we are open source offering our alorithms to anyone in the world, we help creating many different browsers, we rise the support of standard in each release in the same way as Opera does, and you accuse us (as well as Nilsen) of "foul games" and "trying to destroy standards" because you can't understand what are the reasons for this case. :(> Yes, my solution is that UA vendors should show self-restraint and be concerned about interoperability and the state of the web as a whole.Self-resistance to what? Once more. Gecko is, as primary, NOT a "web browser engine". It's a platform. Like Java. Like .NET. You didn't see it when you spend time comparing size of the Firefox vs. Opera bundle, and you don't see it now. What Opera is just starting doing – creating a platform – we're doing for 8 years.Please, read it slowly:We are a platform. This platform allows user to write applications. This platform also gives an easy way to render open standard compliant web sites. In result this platform can present websites. Thank you for listening to this message. :)> For example the way Ian Hickson's WHAT specifies INPUT type=date so that pages remain entirely usable to UAs without specific support for WF2I hope that I don't have to explain it also to you (I had to explain it to Quiris) that Ian Hickson is working on Mozilla and is helping us to make Mozilla a better platform. Is your trust to this person high enough to assume that he'd not work on the project with such a bad will that you (and Nilsen) accuse them of?> just because we need them internallyYou wanted to propose it to the users who create Opera platform based applications. I see nothing wrong in it.> Makes me want to refer to bugzilla's 340571 againPerfect example. So we have something that web sites started using. We don't want them to use it AND we're not sure if it should overall exist for our platform.We'll not remove it only because the websites started using it, but if we can get two birds with one shot, we'll switch it to any standard we'll agree on (even quasi-standard that is shared between WebKit and Gecko first).Seeing how we're discussing bugs like this, how can you still keep your sentence with accussions of Mozilla playing against the world and having "bad intentions"?In this case, we also can try to limit it only to XUL, in others we can't so it makes the game harder, but we try to keep it clean and out of web world.Nilsen:> The point is that it's a Mozilla specific technology.You can find PHP-XUL bindings, Java-XUL ones, XUL is used in some other places, sometimes according to the specs, sometimes not.> No. Mozilla contacted Opera and wanted them to sign an agreement if they were going to use the icon. Whether it was done by letter or phone or by nailing a Post-It to an Opera skull is besides the point. The point is that Mozilla fouled up.Once more your assuming something you don't have knowledge about and derriving a statements that are wrong. And they're accussions. Please, stop.Mozilla published a series of blog posts about the icons including the post about feed icon. THEN, Opera decided to remove it. It's quite possible for me that Opera simply read the thread and reacted. Also, Opera wrote "we're removing the icon until we'll clear the situation with Mozilla". It doesn't mean "The nasty Mozilla forced us to remove it", but "we're not sure what's the status (possibly – after reading the blog posts) so we'll contact Mozilla and ask".So, once more, Mozilla did not fouled up. (until you'll be so megaloman to try to prove that Mozilla did it specially against Opera…)> Criticizing Mozilla for exposing their internal implementations to normal web pages is not an attack.Well, it is. The question is if it's a valid attack. And the much nicer way would be to first discuss it with Mozilla, and then attack. Not before.> Criticizing Mozilla for not allowing Opera to use the feed icon without an agreement just before release is not an attackWell, it is. The question is if it's a valid attack, and your knowledge seems to be pretty low if you was unable to recreate the timetable of the issue.> The Opera community does not create Opera.I never referenced "you" to creating Opera, rather to focusing on attacking Firefox with FUD.> Mozilla certainly tried to make it part of the web by allowing normal web pages to use it.Did Opera try to make window.opera a part of the web by allowing normal web pages to use it? Hey, normal web pages are allowed to use -o-link-source and -o-link in Opera. Do you think that Opera's goal is to make it part of the web?> you conveniently dodged it because you know that it pokes a huge hole in your arguments.I really appreciate your self-convinience about how good human mind reader you are :)> non-web stuff shouldn't be available to web sites.1) It's not possible if you want to be a platform2) It will touch other platforms as well as they'll extend their platform feature set (including Opera and WebKit).3) I'd agree that we could try to standarize the per-platform-prefix for JS methods and objects like we did with CSS (see, every platform has it's own CSS settings EXPOSED TO THE WEB. Oh, how malicious Opera is, right? ;)). For XML-based languages we can use DOCTYPE (and Mozilla does it).4) Your statement bases on the false assume that the problem is (like hallvors described it) in that people WILL use things just because they are there. For me it's like removing the DIV element because author can misuse it and create a huge red DIV in the middle of the page making it unreadable. It's HIS fault, not browser authors fault.

  22. @zbranieckiIf Mozilla exposes some of its internal methods, fine.If people want to use those exposed internal methods, for a gecko-specific web page/app, fine.The question is, how do we prevent (or greatly reduce) people using those exposed internal methods with the assumption that they'll work across platforms/browers?It would be a good gesture if Mozilla put more effort into stressing that the exposed methods are gecko-specific or specific to only certain platforms.For KLM specifically, it doesn't look like anyone has had any luck contacting them. zbraniecki, have you tried? Know someone that might have better luck?Does originalTarget have any more use than target in the security scope of a web page, or is originalTarget's useful methods only available to extensions etc.?It seems that if orginalTarget does not provide anything extra than target for a regular web page scope, then it doesn't *need* to be exposed. If so, is the Mozilla security model not advanced enough yet to keep originalTarget out of a regular web page scope?Thanks

  23. web authors used IE proprietary solutions because Microsoft proposed them to use those

    Oh please. KLM uses a Mozilla thing in spite of you not promoting it. Their site is less interoperable as a result. The only question is how much of the blame is KLM's and how much is Mozilla's for putting it in in the first place. I do respect Mozilla, standards development and advocacy and all – but originalTarget is proven to be harmful on the web. KLM menus being broken proves it.Opera's platform work is (so far!) different – what functionality is exposed to Opera widgets through JavaScript APIs is NOT exposed to normal web pages. If I have any influence on that we will keep it that way.

    I feel that wherever we can use standards, we use them and we're doing a lot to keep it and fix all remaining issues where we use own implementation in places where we can use standard

    Great. Perhaps a result of that work will be to find a way to remove Event.originalTarget from the JavaScript environment available to normal web pages. I'll patch KLM again and hope you will achieve this before I have to patch originalTarget problems elsewhere..

  24. Aside.. KLM has made one fix. It now says:

    if (!isSafari()) {
        if (e) if (e.originalTarget.innerHTML == "Home") return;

    Aaaargh. Heard of object detection, guys? It's SO simple to do this correctly. 😦 Well, I hope our Open the Web guys get a nice contact over there soon.

  25. zbraniecki,First, supporting an element that shouldn't be supported in certain context (normal web-page) is supporting a non-standard behaviour, so working against standards ment for open web. Period. Supporting "standards and more" means supporting incompatibilities and problems raising from it like ShowModalDialog() makes some pages work only in IE. So Firefox/Gecko engine should recognize when it works with application and when with a normal webpage to prevent danger of improper use of functions ment only for web applications.Second, as hallvors noticed, KLM site is a proof by itself, that people do not need to be encouraged to use XUL elements on normal web pages – so leaking this XUL function is already proven to be harmful. Wheather Mozilla promotes usage of it or not, will only have impact on the scale – number of such pages on the web – will there be hundreds, thousands or millions of them, sooner or later.Third, agreed that this problem is harmful, we should consider how to solve it.Your claim that it can't be solved – 'cause it is not possible to serve different behaviour to normal HTML pages and web applications – is not entirely true, though it indicates a fundamental misdesign of XUL/Gecko-web-applications-support-concept. Opera widgets are also web applications but designed in a way to not interfere with normal web pages. Fine, they do not support much more languages (like XUL) than the Opera browser itself but support for them can be added without risk of breaking normal sites.It certainly is possible to solve the problem, though some solutions may be painful, as the fundamentals of the misdesigned concept must be fixed. First few solutions that came to my mind are:- activating WebApplication-mode in Gecko only if in comment of HTML page is some tag like – split Firefox into two kinds: Firefox browser and Firefox application runtime or at least make the applications mode off by default and available to turn on when someone wants to (minority of users need it so it's along with Firefox philosophy)- at least for a start, when Firefox enters a page using XUL etc. display a clear informative-warning message on the title-bar or anywhere "This page is a Gecko invented hyper advanced web application", mainly for purpose of webmaster. Of course the message can be much more informative and explaining than what I suggested.If Mozilla doesn't try to prevent using XUL on normal web pages in ways mentioned above or any others which are effective, they will be supporting violating standards and (with silent satisfaction?) gaining profits from fact that more and more sites are using XUL/Gecko-only extensions, closing the web again, this time not to IE-only but XUL-supporting-browsers-only.Thank you for your attention.

  26. This reminds me of companies that wrote extensions for Mac OS X using undocumented kernel APIs and APIs that were specifically marked as "this will change," and then claimed to be totally surprised and outraged that their programs stopped working in the next version of the OS when the API changed.Given the lack of documentation on this "feature," I can't figure out where KLM's developers even got the idea to use it. I certainly wouldn't shed a tear for them if, say, Firefox 2 dropped support for an undocumented, already-deprecated function.

  27. burnout426:> It would be a good gesture if Mozilla put more effort into stressing that the exposed methods are gecko-specific or specific to only certain platforms.Yes, I feel that we do a lot to explain this difference and make people not try to use them in the web. Our Developer knowledgbase strictly signs which parts are Web Standard world and which are Mozilla only-for-internal-use ones. We're also participating in every possible project focused on promoting standards, including polish Osiolki.net. If you have any suggestion or you found a mistake in our policy (like, in some document not stating clearly if it's web standard part or not) please, report it as a bug, or simply fix if it's on a wiki.> For KLM specifically, it doesn't look like anyone has had any luck contacting them. zbraniecki, have you tried? Know someone that might have better luck?No, I didn't. I was sure that hallvors did before writing his post 🙂 If not, we should. I'm currently in contact with Yahoo and Wall Street Journal on fixing their sites using bad JS.By the way. All you who discuss with me here. You should know that I'm a Flock employee, and I'm working on different to Firefox browser, so the example of Yahoo and WSJ is about those sites using if() statement to find if the browser has "Firefox" in UA name and present standard JS to "Firefox" and "IE-only" to all other browsers. It hits all browsers, not only Opera and Safari but also Camino, Seamonkey and Flock. We're on the same side, believe me :)> Does originalTarget have any more use than target in the security scope of a web page, or is originalTarget's useful methods only available to extensions etc.?My knowledge is too low in this case 😦 I pointed you the Devmo articles about it. I see no reason against fileing a bug in bugzilla.mozilla.org asking to deprecate originalTarget and not exposing it to web content because websites may use it (example – KLM). CC me to such bug please (gandalf at aviary dot pl), and I'll try to figure out more.> but originalTarget is proven to be harmful on the web. KLM menus being broken proves it.Same here. Please, file a bug, ask to remove it and give your reasons. We should try to evaluate if we can remove this from the web context and if we want. If not, we should keep poking KLM to stop using it.> Perhaps a result of that work will be to find a way to remove Event.originalTarget from the JavaScript environment available to normal web pagesIt's possible.Overall. Please, take a note that during this whole discussion I didn't try to say "We'll do this, KLM's fault, we don't care what's going on". We do care, we do want KLM not to use it. I just wanted to explain WHY we do this, and explain that the reasons you think (Mozilla being malicious, vendor-lock-in, properietary, evil company trying to takeover the world for years, and this is another evidence) is very unfair.> Opera's platform work is (so far!) different Opera is at very, very early stages of moving from browser engine to platform.Nasty:I'm happy that you decided to step in and take part in this discussion, but I don't see why you decided to repeat once again the already proven false statements.> First, supporting an element that shouldn't be supported in certain context (normal web-page) is supporting a non-standard behaviour, so working against standards ment for open web.You misunderstand what the web standard is. Web standard is not enclosed. There's no word in JavaScript 1.5 documentation that parser cannot support ANYTHING beside of given list of objects, methods and properties.It means that it's not the "Standard->Engine support" it's "Some platform that may support standard".It's like with rails. To be able to run on the rails, my vehicle needs to follow some specification. But it doesn't mean that this vehicle CANNOT support other methods of movement. I can create a vehicle that will be able to fly, swim, run on mud, roads AND rails. And than I can say that my vahicle supports rail-transport. Some "standard".You look from "rail-centric" point of view and you see anything beside (for eg. flying) as "malicious usage of rails".> So Firefox/Gecko engine should recognize when it works with application and when with a normal webpage to prevent danger of improper use of functions ment only for web applications.Yes. I'm not speaking for myself, not Mozilla, but yes. I'd agree with you, but, as we discussed it previously (why you're making such statements if you read the whole thread and saw me saying that it's impossible?) it's impossible.I cannot know if you're writing a web app, or web site or mozilla app. It's a bit like if you'd try to guess if the website is a clear HTML website of Web App. You can see the difference between Wikipedia article, my home page, or stock exchange data site and Gmail, Google Maps, Netvibes, but it'd be impossible to describe it, write algorithm to guess which one is which and present different set of API to each of them (through, I'd love to see such thing).> will there be hundreds, thousands or millions of them, sooner or later.I understand that you have bad experience with what Microsoft did, but I'd like to remind you that:1) We promote standards. Once user uses standards, there's no problem with it.2) Firefox will never get the amount of market IE has. We don't want to.3) Microsoft gave you no choice but to use non-standard solution. When you were writing your website for IE4, there was NO support for getElementById.It made possible the IE-lock-in that happened a few years ago. I don't think that there's such threat now. Yes, we should try to prevent users from doing this, but we shouldn't get crazy about it. There are tons of other problems with web standard support and browser-racism (like the WSJ and Yahoo examples I gave above), so please, don't pretend to make this issue the "World Biggest Problem Right Now". :)> Opera widgets are also web applications but designed in a way to not interfere with normal web pages. Fine, they do not support much more languages (like XUL) than the Opera browser itself but support for them can be added without risk of breaking normal sites.You don't understand what Mozilla platform is. Opera platform is something totally different, completly different and has no intentions to serve in the way Mozilla works. Mozilla allows you to create an application and deploy it accross whole company (even if your employees are located around the globe) and use it through the web. You can use this app, it's very easy to maintain it, etc. Giving a widgets as an example of "almost like extensions" is naive. But trying to say "See? We have widgets and they work, so why your platform can't?" is a fundamental mistake.> – activating WebApplication-mode in Gecko only if in comment of HTML page is some tag like It would work pretty well for JS and CSS files.> – split Firefox into two kinds: Firefox browser and Firefox application runtime or at least make the applications mode off by default and available to turn on when someone wants to (minority of users need it so it's along with Firefox philosophy)You prove to know nothing about future of Mozilla. Please, read more before making statements like this.> – at least for a start, when Firefox enters a page using XUL etc. display a clear informative-warning message on the title-bar or anywhere "This page is a Gecko invented hyper advanced web application", mainly for purpose of webmaster. Of course the message can be much more informative and explaining than what I suggested.This statement proves lack of knowledge of Mozilla platform and it's goals.> If Mozilla doesn't try to prevent using XUL on normal web pagesThis statement proves lack of knowledge about XUL, "normal web pages" and Mozilla.> they will be supporting violating standardsThis is simply false> with silent satisfaction? gaining profits from fact that more and more sites are using XUL/Gecko-only extensions, closing the web again, this time not to IE-only but XUL-supporting-browsers-only.This is impertinent. I'm shocked how you (after Nielsen) make it easy to accuse Mozilla of such things. How dare you? What did you make for web world to say things like this? What do you know about what Mozilla did to allow yourself to make such stupid sentences. You must be really crazy about how much you hate Mozilla if you simply use such discussion to say such bastard things like this. It makes me really feel that I don't want to talk to such rabble person. Think twice before accusing people of something, because otherwise I can start for example saying that maybe you're satisified with people dieing in Iraq, and maybe you're a plain bastard?… In Poland there was a guy who used questions instead of statements to be able to make accusions without having to feel the responsibility for his words. It was Andrew Lepper. Congratulations for being same smart…We do not use, we never used, and we will never use vendor-lock-in to promote our browser. You have absolutely no prove of your accusions, and once more. Saying such things without a prove, just before you CAN say them, and you have some problem with what Mozilla is and how it works, is very pathetic.It was my last discussion with you Nasty. You just crossed the thin line between discussion and being rude. You just accused the tens of houndreds of volunteers around the world who created open source browser, and changed the web world more than anyone in last 10 years and in the exact direction you can benefit of, making it open, standard compilant, full of new technologies and alive. So what you just did is a simple FUD, you behaved like a zealot, fanboy who needs to share his frustrations and is not responsible for anything so can say everything.You're rustic and until you stop using things like this, I'm not going to talk to you anymore. It's so pathetic, that you're entering such a long discussion after so many words said, and you make a comment where you first prove not to read what was said before or not to understand it at least, then you prove not to understand what Mozilla is (while you allow yourself to speak about it), and finally you accuse it of things that we're fighting against for many years. I have no words to describe it.

  28. zbraniecki: filed https://bugzilla.mozilla.org/show_bug.cgi?id=343032 and CCed you. I hope it will be considered carefully.Please check that I've done things correctly, I'm not *that* used to Moz's bug tracker.

    We do not use, we never used, and we will never use vendor-lock-in to promote our browser.

    Thanks. I don't think I've convinced you that being careful with what you implement/expose besides the standard is an integral part of that approach but we can agree to disagree here.Of course Opera's "platform"-ness has much shorter history than Mozilla's, but the design goals also seem very different even if I don't know much about yours.The new browser.js patch went live today, adding the object detection KLM failed to add themselves so the menu will work again. browser.js makes such a difference from how we in the past were just knocking at doors and shouting when important sites made stupid errors.. Perhaps Flock will copy that feature eventually ;)I'll end the discussion here, any further debate probably belongs in bugzilla 🙂

  29. Sigh.. just found this:

      Line 136 of inline#76 script in http://mail.google.com/mail/?view=page&name=js&ver=2e81b753cff5147
        if (Bx || c.explicitOriginalTarget.nodeName != "scrollbar")
    

    Event.explicitOriginalTarget ? One of the reasons GMail chat isn't enabled in Opera, whatever else it is..

  30. Flock was considering reuse of browser.js thing. We'll get back to it in Danphe cycle. :)About Google. No, Google and Yahoo have strange policy. They simply think the old way. They white list browsers they're sure about. Thay check their work against browsers, not standards. Yahoo blocks Flock even if it's 100% same engine to Firefox 😦

  31. By the way:"However, it must be noted that Mozilla implements a variant of XBL 1.0 which does not quite match the specification."

  32. Mozilla published a series of blog posts about the icons including the post about feed icon. THEN, Opera decided to remove it.

    No, it was because Mozilla specifically contacted Opera. This has been explained plenty of times. It was even mentioned by Mozilla representatives.Opera didn't react because of blog posts, but because of nasty letters from Mozilla lawyers.

    So, once more, Mozilla did not fouled up.

    Yes they did. They sent their lawyer to bug Opera just before the release of a major new version!

  33. Google and Yahoo have strange policy. They simply think the old way. They white list browsers they're sure about. Thay check their work against browsers, not standards.

    Sad state of affairs. Well, you can see where they are coming from.. In a nutshell: when you're a web developer shipping production websites, implementations matter, standards don't.Ironically the opposite policy ("code to standards, blame the implementations for any problems") would force UAs to focus even harder on spec compliance and thus make it progressively easier for authors to write web applications and sites. But then again, I've been told that enforcing spec compliance on the client is user-unfriendly (said often enough in the debate about draconian XML error handling), not to mention that even us spec evangelists must admit that specs do contain errors, bugs, omissions, incomplete instructions..Yahoo's "graded browser support" is a step in the right direction, but I think an "allow-by-default, enhance-where-possible" philosophy is best.

  34. It just so happens that IE 9 internally doesn't separate the XAMLElement interfaces and the HTMLElement interfaces cleanly.

    Ie 9 ?? :p probably by year 2020 …BTW, klm no longer blocks Opera accesshttp://www.klm.com/travel/klm_splash/static/js/global.jsOpera 8's browser.js file doesn't have the originalTarget fix… only the file for Opera 9.Also the padding+wdith css override isn't needed for Opera 9.Last but not least, why not using DOMContentLoaded for Opera 9 ? :p for sure it would improve things a bit !

  35. @zbraniecki:

    > First, supporting an element that shouldn't be supported in certain context (normal web-page) is supporting a non-standard behaviour, so working against standards ment for open web.You misunderstand what the web standard is. Web standard is not enclosed. There's no word in JavaScript 1.5 documentation that parser cannot support ANYTHING beside of given list of objects, methods and properties.

    Oh really? You just precisely defined "Embrace and extend" attitude. I can add any feature I like to my browser in addition to what is defined in original specification, "because it doesn't say it's enclosed". And when someone makes use of this "enhancement of standard" in my browser, others browsers will fail to understand the "enhanced" code. Simple as that. Didn't I yet put it clear enough?

    It's like with rails. To be able to run on the rails, my vehicle needs to follow some specification. But it doesn't mean that this vehicle CANNOT support other methods of movement. I can create a vehicle that will be able to fly, swim, run on mud, roads AND rails. And than I can say that my vahicle supports rail-transport. Some "standard".You look from "rail-centric" point of view and you see anything beside (for eg. flying) as "malicious usage of rails".

    This comparison is precise enough to have the same problem as described above. So, if your vehicle supports not only rails but also say swimming, the creator of railroad can in some places leave water, assuming your vehicle will swim through it anyway. The problem is that only your vehicle will deal with such rail-water-road, while other vehicles will not be able to swim.Exactly the same as when other browsers are not able to render XUL-enhanced website, which works well in Gecko though, because of its "enhancing" of standards.

    > will there be hundreds, thousands or millions of them, sooner or later.I understand that you have bad experience with what Microsoft did, but I'd like to remind you that:1) We promote standards. Once user uses standards, there's no problem with it.2) Firefox will never get the amount of market IE has. We don't want to.3) Microsoft gave you no choice but to use non-standard solution. When you were writing your website for IE4, there was NO support for getElementById.

    None of these arguments make by a bit untrue what I pointed out – that sooner or later there will be hundreds, thousands or millions of sites using browser-specific, non-standard functions if browser vendors will expose them for use. There will always be plenty of people eager to use them, from various reasons. And browser vendors should behave responsible and prevent this where possible, by not exposing these functions where they should not be available.

    It made possible the IE-lock-in that happened a few years ago. I don't think that there's such threat now. Yes, we should try to prevent users from doing this, but we shouldn't get crazy about it. There are tons of other problems with web standard support and browser-racism (like the WSJ and Yahoo examples I gave above), so please, don't pretend to make this issue the "World Biggest Problem Right Now".

    The worst biggest problem is to make browser vendors conscious, that they should be very careful with adding new, non-standard functions to their browsers for use of webmasters.I am primarily replying to you for this purpose, to react before things get worse and you (and your colleagues, Mozilla etc.) add and expose a bunch of other mozilla-specific functions for webmasters, enabling further degrading standarisation of www.

    > So Firefox/Gecko engine should recognize when it works with application and when with a normal webpage to prevent danger of improper use of functions ment only for web applications.Yes. I'm not speaking for myself, not Mozilla, but yes. I'd agree with you, but, as we discussed it previously (why you're making such statements if you read the whole thread and saw me saying that it's impossible?) it's impossible.(…)> – activating WebApplication-mode in Gecko only if in comment of HTML page is some tag like It would work pretty well for JS and CSS files.

    So… after all, it turns out to be possible, despite previous claims that there is no way. 😀

    > – split Firefox into two kinds: Firefox browser and Firefox application runtime or at least make the applications mode off by default and available to turn on when someone wants to (minority of users need it so it's along with Firefox philosophy)You prove to know nothing about future of Mozilla. Please, read more before making statements like this.

    Thank you a lot for reprimand, I admit I do not follow closely to what Mozilla plans, but whatever they plan, I cannot imagine reason why people who want to use "just the Web", could not get "just the Web browser" (with all XUL extensions on opened web sites disabled), while only those needing XUL applications have their Firefox Enhanced.It's not easier for Mozilla, but it's better to enforce on webmasters being closer to web standards.

    > – at least for a start, when Firefox enters a page using XUL etc. display a clear informative-warning message on the title-bar or anywhere "This page is a Gecko invented hyper advanced web application", mainly for purpose of webmaster. Of course the message can be much more informative and explaining than what I suggested.This statement proves lack of knowledge of Mozilla platform and it's goals.

    Another harsh reprimend. 🙄 First, I'm just dropping ideas to help. Second, goals of Mozilla platform have little to do here. I realize such move might not be the perfect solution from perspective of "Mozilla goals", but it would help support web standards.

    > If Mozilla doesn't try to prevent using XUL on normal web pagesThis statement proves lack of knowledge about XUL, "normal web pages" and Mozilla.

    Not at all. I realize Mozilla renders XUL implemented on "normal web pages" but this does not justify mixing of support of web standards with Mozilla enhancements.

    > with silent satisfaction? gaining profits from fact that more and more sites are using XUL/Gecko-only extensions, closing the web again, this time not to IE-only but XUL-supporting-browsers-only.This is impertinent. I'm shocked how you (after Nielsen) make it easy to accuse Mozilla of such things. How dare you? (…) You must be really crazy about how much you hate Mozilla if you simply use such discussion to say such bastard things like this. It makes me really feel that I don't want to talk to such rabble person. Think twice before accusing people of something, because otherwise I can start for example saying that maybe you're satisified with people dieing in Iraq, and maybe you're a plain bastard?… In Poland there was a guy who used questions instead of statements to be able to make accusions without having to feel the responsibility for his words. It was Andrew Lepper. Congratulations for being same smart…

    zbraniecki, next time, before you start calling people names, take a deep breath and think again if you at least have any reasons.I only said facts – that if Mozilla doesn't fix it, they will benefit from it. No doubt about it. Only by the way (hence in brackets) I was wondering if it would be intentional – and hence the question mark not a statement! Do I really have to explain you usage of basic written language elements? For Pete's sake, relax a little and don't over-interpret or put in others mouth what they didnt't say.

    We do not use, we never used, and we will never use vendor-lock-in to promote our browser. You have absolutely no prove of your accusions, and once more. Saying such things without a prove, just before you CAN say them, and you have some problem with what Mozilla is and how it works, is very pathetic.

    Pathetic is not noticing it wasn't an accusation or a statement, but just a conditional sidenote wondering whether it is possible.And I am happy that condition seems to evaluate to false 😉 for the time being (but nothing sure yet – until this bug get's fixed and implemented, I will keep an eye on the progress).

    It was my last discussion with you Nasty. You just crossed the thin line between discussion and being rude. You just accused the tens of houndreds of volunteers around the world (…). So what you just did is a simple FUD, you behaved like a zealot, fanboy who needs to share his frustrations and is not responsible for anything so can say everything.

    This is just so pure nonsense. :rolleyes:I think you should really restrain yourself from writing on the web while being (from a trivial reason) furious – a state where people are half out-of-mind, not controlling themselves.

    You're rustic and until you stop using things like this, I'm not going to talk to you anymore. It's so pathetic, that you're entering such a long discussion after so many words said, and you make a comment where you first prove not to read what was said before or not to understand it at least, then you prove not to understand what Mozilla is (while you allow yourself to speak about it), and finally you accuse it of things that we're fighting against for many years. I have no words to describe it.

    Seems like you have way too many words to describe it. :right::left: Much more than it deserved, as I did nothing to deserve slinging mud at me.

  36. > And when someone makes use of this "enhancement of standard" in my browser, others browsers will fail to understand the "enhanced" code. Simple as that. Didn't I yet put it clear enough?Wrong. If somebody misuse my feature, he makes a mistake. I assume that you're also against fork and knife cause somebody can use it to kill other people and it'll be fork&knife author fault.> the creator of railroad can in some places leave water, assuming your vehicle will swim through it anyway. The problem is that only your vehicle will deal with such rail-water-road, while other vehicles will not be able to swim.No, you misplace the error cause. If the autor of the rails does it, he makes a mistake. He can also enforce the user to install a widget to use the website, and it's not Opera's fault.> The worst biggest problem is to make browser vendors conscious, that they should be very careful with adding new, non-standard functions to their browsers for use of webmasters.I am primarily replying to you for this purpose, to react before things get worse and you (and your colleagues, Mozilla etc.) add and expose a bunch of other mozilla-specific functions for webmasters, enabling further degrading standarisation of http://www.You know what? That's really funny that you decided to "warn us". If I'm right it's Opera that holds the crown of most IE-compatible of modern browsers, and it's Opera who made more than anyone ever effort to expose non-standard features directly to the web authors to make sure it raises the user experience.And now you warn "Mozilla" who was the last browser to accept marquee, and still holds an ongoing debate to remove it.Mozilla did not expose any mozilla-specific function in recent history, this example is rather very old thing. No needs to worry.> So… after all, it turns out to be possible, despite previous claims that there is no way. :DNo, I didn't say so. Read once more. I said that it'd work with JS and CSS. That's all.> I cannot imagine reason why people who want to use "just the Web", could not get "just the Web browser" (with all XUL extensions on opened web sites disabled), while only those needing XUL applications have their Firefox Enhanced.Oh, you understand that pretty well, but prefer to pretend you don't. It wouldn't make sense to get people install a web browser and switch if they want to use a web app. Please, remember that Mozilla is aiming to compete with IE and their IE7( hint: XAML). I see that Opera is more interested in competeing with Fx, but please, try at least not to block Mozilla in their goal.> but it would help support web standards.Not much. People who want to violate web standards will do this. The goal is to educate them so they want to follow web standards. You're very exclusive now, when you want to focus on proving that Mozilla is sooo wrong (and forget the Opera's past) about this thing, and claim that web authors are not important, it's all about browser vendors.> but this does not justify mixing of support of web standards with Mozilla enhancements.Yes it does.> Pathetic is not noticing it wasn't an accusation or a statement, but just a conditional sidenote wondering whether it is possible.You and I pretty well know a great example of question form used to accuse. (Lepper about PO politics) Conditional form can be used in exactly same way and you know that.> I think you should really restrain yourself from writing on the web while being (from a trivial reason) furious – a state where people are half out-of-mind, not controlling themselves.Thanks for telling your story. I can ensure you that even when I'm very pissed off by someone writing lies, I stay calmer than most others because of my nature. Ask any of my friends about how it looks when I'm arguing – I'm always calm. So sorry, missed shot. You said did accused people, and I find it totally bastardish. Are you going to say that it's because I'm furious? Can you accept that it's just my judgement of what you did date to say?> Much more than it deserved, as I did nothing to deserve slinging mud at me.You did. I keep my words.

  37. If I'm right it's Opera that holds the crown of most IE-compatible of modern browsers, and it's Opera who made more than anyone ever effort to expose non-standard features directly to the web authors to make sure it raises the user experience.

    The comparison is invalid, since Opera didn't make up those non-standard features. It just had to implement them or lots of sites would break because they were already in use.On the other hand, Mozilla is now carelessly promoting its own proprietary features to sites.

    Please, remember that Mozilla is aiming to compete with IE and their IE7( hint: XAML). I see that Opera is more interested in competeing with Fx

    This comment is obviously just meant as trolling. All browsers on the market today are competing. You are either trolling or being completely paranoid by thinking that Opera is somehow aiming to compete with Firefox more than IE.

    You're very exclusive now, when you want to focus on proving that Mozilla is sooo wrong (and forget the Opera's past)

    I don't know what you are referring to in Opera's past, but it does not justify Mozilla's careless and dangerous attitude.

  38. I just read throught this thread…. my two favorite quotes from the whole thing:Originally posted by hallvors:

    History (and particularly the history of Internet Explorer) teaches me that web authors use anything just because they can.

    …andOriginally posted by zbraniecki:

    [the page author] can also enforce the user to install a widget to use the website, and it's not Opera's fault.

    Great read.

  39. Originally posted by author:

    [the page author] can also enforce the user to install a widget to use the website, and it's not Opera's fault.

    Not exactly.Scenario #1: A widget must be used to access the service. Answer: Well, then it's just a widget, not a website so they're not hindered.Scenario #2: A widget must be active to access the website.Answer: I'm pretty sure this one can't be done, or at least the author must jump through some pretty narrow hoops to pull it off. One way to do it could be to offer the widget ONLY as a compressed archive (lest the user could simply open it as a webpage and the scripts would be run) in which the widget set a cookie on the parent page that the user was now allowed to enter.It's pretty clear that scenario 2 will not happen accidentally because of lack of knowledge on the author's part, and even then there are SOOOOO many easier ways to do it. Like "if(!window.opera)document.location='foobar-off.html';" for instance.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s