Exiting new Firefox feature in the latest nightly build



  • They've been working on it for 8 years.  This should be awesome.



  •  I have no idea how that's supposed to work.



  • Where is it exiting to?



  • @El_Heffe said:

    Exiting
    Exiting? So it's about to land (after 8 years) and yet it's exiting at the same time? I suppose that's how it goes with software; one person wastes ages creating a useless feature, and in the very build that it is supposed to make its début, it gets pulled. C'est la vie.@toth said:
    Where is it exiting to?
    Damn ninjas!



  • I thought it was going to be the web-history-leaks-through-:visited bug.



    1. I think they're talking about cloning objects, or changing an object's parent, or something. I don't know for certain because I couldn't be arsed to read past the first 3 comments.
    2. It's "exciting" not "exiting", unless we're talking about what the half-decent software developers at Mozilla are doing.
    3. Further to the previous point, HOW FUCKING LONG DOES IT TAKE TO RELEASE FIREFOX 4 BETA 7? SSL via an NTLM proxy has been completely broken since beta 5 which came out like 2 months ago. (And no, I'm not going to use a nightly build because I barely trust Mozilla not to fuck up my bookmarks at the best of time.)


  • @moog said:

    people can read your memories
    Sounds good to me. Perhaps I can use that to read my own memory and work out all the things I forget. Like people's names. And faces. And that meeting I was supposed to be at. And where I left the keys. Seriously, if I don't see it written down, I won't remember, so this feature should become standard in all software.



  • @The_Assimilator said:

    SSL via an NTLM proxy has been completely broken since beta 5 which came out like 2 months ago. (And no, I'm not going to use a nightly build because I barely trust Mozilla not to fuck up my bookmarks at the best of time.)

    Well, why are you even using a beta then?



  • @El_Heffe said:

    They've been working on it for 8 years.

    8 years? They've implemented it in 3 months! Geniouses!



  • Actually, Comment #3 in the list makes the meaning of 'brain transplant' reasonably clear:

    "a very limited brain transplant (switch handlers) between the outer window (proxy) and the wrapper (proxy)."



  •  @Spectre said:

    @El_Heffe said:
    They've been working on it for 8 years.

    8 years? They've implemented it in 3 months! Geniouses!

    Yes.  Pure genious.  I am looking forward to a brain transplant.  Maybe then I'll remember how to spell "exciting".



  • @moog said:

    I thought it was going to be the web-history-leaks-through-:visited bug.

    That's not a bug. That's an oversight in the spec. Firefox could hack the functionality of :visited, but then they'd technically be the buggy one.

    Off-topic: I was reading a great article the other day, about how a lot of things that web devs perceive as bugs in IE8 are actually faithful to the spec, and Mozilla, WebKit, get the spec wrong. Simply because Microsoft isn't part of the browser "cabal." (WebKit: "How is this HTML4 feature supposed to work?" "Just do what Mozilla does." Mozilla: "How is that HTML5 feature supposed to work?" "Just do what WebKit does.") Microsoft's independent implementation of the W3C specs is actually a good thing, and the closest it gets to actual real-world testing, but they're going to get roasted over the flames for IE9 not following the HTML5 spec the same way the cabal does, even if they're equally accurate. There's no justice.

    Note that the HTML5 spec is supposed to be "final" once it gets two complete independent implementations. The problem is that if the two implementations are in the cabal, it doesn't really prove anything about the spec because the cabal browsers are all just borrowing from each other.



  • @blakeyrat said:

    I was reading a great article the other day, about how a lot of things that web devs perceive as bugs in IE8 are actually faithful to the spec,
     

    What's the article?

    I ask professionally. I expect to find a blow-by-blow, which will help me in my work.

    Currently, I find that there's basically nothing at all that I write that IE8 does differently from the cabal. I write shit for FFX (because firebug is best tool evar), then when I got to test the rest, I pre-emptively put IE in 7-mode because I've come to expect IE8 to work the same.

    There have been very, very, very rare edge cases in the past two years where my stuff had a bug in Chrome/Safari that looked fine in FFX and IE8.  Both of them were related to the wrapping behaviour of/around floating elements and easily fixed.

    ...

    Oh, I lie: the third one was related to Sarafi 3-'s nonstandard and Mac's smooth antialiasing, causing a 1-pixel difference in final glyph height for some text. This was when we still used a Mac Mini to test Safari.



  • @blakeyrat said:

    Microsoft's independent implementation of the W3C specs is actually a good thing, and the closest it gets to actual real-world testing, but they're going to get roasted over the flames for IE9 not following the HTML5 spec the same way the cabal does, even if they're equally accurate. There's no justice.
    Justice may be hard to come by, but reputation is not.  There's a reason IE is always assumed to be wrong when it disagrees with the vast browser-implementation conspiracy.



  • @dhromed said:

    Currently, I find that there's basically nothing at all that I write that IE8 does differently from the cabal. I write shit for FFX (because firebug is best tool evar), then when I got to test the rest, I pre-emptively put IE in 7-mode because I've come to expect IE8 to work the same.


    Well, I've found one "buglike standard-compliancy" in IE8 myself. To be honest, it's not HTML5, it's Unicode. The U+200B zero-width space belongs to some character category (I don't remember the details) that, in short, specifies the word break should NOT happen if the ZWSP is put next to a "punctuation character" (yeah, seriously... what have you been smoking, Unicode authors?). IE8 honours that, so long links that had the U+200B insterted after each dot in the displayed value for wrapping did not wrap in IE8.



  •  8 years?

    2010-07-19 till 2010-10-20 is 8 years fot You ?



  •  @spamcourt said:

     8 years?

    2010-07-19 till 2010-10-20 is 8 years fot You ?

    Yeah, I had another tab open with a bug that  started in 2002.  I really need that brain transplant.

     

     



  • @blakeyrat said:

    Off-topic: I was reading a great article the other day, about how a lot of things that web devs perceive as bugs in IE8 are actually faithful to the spec, and Mozilla, WebKit, get the spec wrong. Simply because Microsoft isn't part of the browser "cabal." (WebKit: "How is this HTML4 feature supposed to work?" "Just do what Mozilla does." Mozilla: "How is that HTML5 feature supposed to work?" "Just do what WebKit does.") Microsoft's independent implementation of the W3C specs is actually a good thing, and the closest it gets to actual real-world testing, but they're going to get roasted over the flames for IE9 not following the HTML5 spec the same way the cabal does, even if they're equally accurate. There's no justice.

    Note that the HTML5 spec is supposed to be "final" once it gets two complete independent implementations. The problem is that if the two implementations are in the cabal, it doesn't really prove anything about the spec because the cabal browsers are all just borrowing from each other.

    And yet, somehow...

    @[url]http://www.theregister.co.uk/2010/11/01/w3c_html5_conformance_tests/[/url] said:
    The Worldwide Web Consortium has released the results of its first HTML5 conformance tests, and according to this initial rundown, the browser that most closely adheres to the latest set of web standards is...Microsoft Internet Explorer 9.

    The W3C tests — available here — put IE9 beta release 6 at the top of the HTML5 conformance table, followed by the Firefox 4 beta 6, Google Chrome 7, Opera 10.6, and Safari 5.0. The tests cover seven aspects of the spec: "attributes", "audio", "video", "canvas", "getElementsByClassName", "foreigncontent," and "xhtml5"


  • @Zecc said:

    @blakeyrat said:

    Off-topic: I was reading a great article the other day, about how a lot of things that web devs perceive as bugs in IE8 are actually faithful to the spec, and Mozilla, WebKit, get the spec wrong. Simply because Microsoft isn't part of the browser "cabal." (WebKit: "How is this HTML4 feature supposed to work?" "Just do what Mozilla does." Mozilla: "How is that HTML5 feature supposed to work?" "Just do what WebKit does.") Microsoft's independent implementation of the W3C specs is actually a good thing, and the closest it gets to actual real-world testing, but they're going to get roasted over the flames for IE9 not following the HTML5 spec the same way the cabal does, even if they're equally accurate. There's no justice.

    Note that the HTML5 spec is supposed to be "final" once it gets two complete independent implementations. The problem is that if the two implementations are in the cabal, it doesn't really prove anything about the spec because the cabal browsers are all just borrowing from each other.

    And yet, somehow...

    [quote user="http://www.theregister.co.uk/2010/11/01/w3c_html5_conformance_tests/"] The Worldwide Web Consortium has released the results of its first HTML5 conformance tests, and according to this initial rundown, the browser that most closely adheres to the latest set of web standards is...Microsoft Internet Explorer 9.

    The W3C tests — available here — put IE9 beta release 6 at the top of the HTML5 conformance table, followed by the Firefox 4 beta 6, Google Chrome 7, Opera 10.6, and Safari 5.0. The tests cover seven aspects of the spec: "attributes", "audio", "video", "canvas", "getElementsByClassName", "foreigncontent," and "xhtml5"
    [/quote]

    Yeah, isn't it hilarious? See the teeth-gnashing on Slashdot: http://tech.slashdot.org/article.pl?sid=10/11/02/1851255 The amount of contorting and twisting required to get that community to somehow correlate "standards are good" with "Microsoft has best standard support" with "oh God we hate Microsoft and particularly IE", well, it's just hilarious to read. (I particularly liked this one: http://tech.slashdot.org/comments.pl?sid=1851100&cid=34104622 paraphrased: "if IE9 has good HTML5 support, it must have shitty HTML4 support!" and this one: http://tech.slashdot.org/comments.pl?sid=1851100&cid=34106074 "IE9 only passes the test because MS wrote a lot of the test-cases.")

    *ahem*, anyway. I wasn't trying to give the impression that the W3C was biased against IE (although I firmly believe they used to be, I don't think they are anymore), or that the other browsers banding together in a code-sharing cabal meant that IE was always going to be behind standards-wise.

    But you're right, this kind of invalidates my point... hrm.



  • @Spectre said:

    @The_Assimilator said:
    SSL via an NTLM proxy has been completely broken since beta 5 which came out like 2 months ago. (And no, I'm not going to use a nightly build because I barely trust Mozilla not to fuck up my bookmarks at the best of time.)

    Well, why are you even using a beta then?

    Until Firefox 4, I'd never encountered beta software that had some of its fundamental functionality be completely broken. It's a fucking web browser; if it can't browse the web, it's useless. Beta software is supposed to be feature-complete with potentially some small, non-show-stopping bugs... Mozilla's "beta" is more like an alpha because every new build adds new features and breaks at least one existing one. Why they didn't just add Aero support to a 3.x release I don't know.

    TL;DR Mozilla are doing an excellent job of flushing the Firefox brand down the toilet by forgoing QA and testing in favour of MOAR FEATURES.


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.