Opera 10.50 Coming, Carakan, Presto 2.4 Rumored

By | December 17, 2009 | 37 Comments

Opera 10.50 Coming, Carakan, Presto 2.4 RumoredToday, Opera Software has announced, that on December 22nd, they will release a pre-alpha version of Opera 10.50.

Although no details were revealed, expect speed, that’s what was said: “We don’t want to say too much and ruin the gift, but expect speed.”

What is Carakan?
Carakan is the upcoming Opera’s JavaScript engine, which was already more than 2.5x faster during early development stages, when compared to Opera 10.

What is Presto 2.4?
Presto is Opera’s new layout engine, which is still under development but already used by Opera Mobile 10.

[digg-reddit-me]


About (Author Profile)


Vygantas is a former web designer whose projects are used by companies such as AMD, NVIDIA and departed Westood Studios. Being passionate about software, Vygantas began his journalism carrier back in 2007 when he founded FavBrowser.com. Having said that, he is also an adrenaline junkie who enjoys good books, fitness activities and Forex trading.

  • nvm

    Wouldn’t it be embarrassing if Carakan didn’t beat V8 after all this time! :D

  • Pingback: [Fav] Opera 10.50 Coming, Carakan, Presto 2.4 Rumored - Overclock.net - Overclocking.net

  • nobody

    i hope that by the speed they mean speed, not speedy launching widgets that noone cares about.

    opera lacks compatibility, compatibility and compatibility.. and some dev tools (that are dead now, dragonfly is dead..).

    and i would like to hear their excuses if that new js engine fails to beat chrome/safari results in popular speed tests. will they devise their own one to prove a point?

    unfortunatelly, most probably, given their history of (not)suiting people needs, instead of something from long ‘should do to catch up’ list, opera will deliver something incredibly un-usefull and buggy, develop it further for two years, never actualy closing the gap between it, and rapidly running away competition.

    to anyone talking about ‘big picture’.. big picture is that opera has minimal stagnant desktop share and had recently got seriously threaten by various webkit browsers on mobiles.

    • nvm

      Even Google supports Opera’s “all apps will be web apps”, so that’s what’s gonna happen! No wonder Opera is betting on widgets.

      Also, check out Opera’s mobile market share. Whoa!

      http://gs.statcounter.com/#mobile_browser-ww-quarterly-200803-200904

      • nobody

        i do not believe in that chrome os paradigm, that everything you do is on the internet. maybe if you are unemployed and have social issues, then you spend lots of time facebooktweeteringflickring etc. if you have real live friends, families, and good and satisfing job, you dont need social substitues for more than ociasional linkedin account for headhunters or contacting potential business partners.

        and if you take out that social-stuff sites, chrome os/opera widgets have almost nothing to offer. browser, as an another layer will never be able to outperform native OS applications, just because it is another layer. with all penalties that it produces. compare java and C++ performance.

        another problem is that, market for IDEs for developers is saturated with high quality tools for native applications. but there are not so many GOOD tools for browser, and none for opera.

        you HAVE to use such tools to create applications more complex than weather widget. how? opera does not deliver such tools. noone will write them, because why bother. this is egg and chicken syndrome. opera hasnt got users, because it hasnt got users.

        i simply cannot stop laughing when someone is reinventing the wheel creating html/javasctipt photo editors.. they will never be better than native apps, and in fact, most probably will never be even enough. or, you create that with flash/silverlight, and then it will work wonderfully.

        btw, how is opera going to do what chromeOs wants, if almost all popular social sites do not work properly in opera?..

        • nvm

          Widgets are local apps, not everything on the internet.

          What makes you think the only thing you can do on the web is social networking? That’s what it sounds like. Chrome OS gives you a browser. You can do anything with a browser.

          The need for IDE tools for web apps isn’t that pressing because it’s much easier, faster and cheaper to do web apps than native apps.

          Also, hasn’t history told you that you don’t need to be better (as in advanced) to win people over? All you need is to be more convenient and accessible. Notice how Chrome will be kicking Firefox in the butt because it’s such as simple and smooth experience in comparison.

    • http://www.trygve-lie.com/ Trygve Lie

      nobody; why don’t you present your self with your real name?

      A widget is a soon to be W3C recommendation (http://www.w3.org/TR/widgets/) so please stop referring to them as Opera Widgets. Yes, I know, Opera use that therm them self at the moment. I think they should stop doing that.

      A widget consists of HTML, CSS, SVG and JavaScript. The widget standard in it self provides packaging and distribution and the posibillity to set the browser in a application modus. It does also remove a couple of things like a browsers back, forward, home button etc and it does also bring a couple of things to the table like cross site AJAX calls, persistent storage client side and file i/o etc..

      This means that all web sites which today are build upon AJAX, Comet and soon to come WebSocets as transport carrier between client and server can be packaged into a program which can run as a application instead of in browser mode. Google Docs, Google Wave, Meboo, Gmail, Hotmail etc.. etc.. and even MS’s Office Web Apps are examples of such applications and can be turned into widgets and with the offline capabilities widgets introduce they will be able to operate as standalone applications even when you are offline. I believe many of them will and I do also think many of these applications will use the W3C widget standard.

      As an example (I have many more); Google have today proclaimed they will focus more and more on developing the mobile Gmail client as a web application instead of developing different native OS applications: http://googlemobile.blogspot.com/2009/12/iterative-web-app-feature-rich-and-fast.html

      I do also recommend taking a look at this video regarding the choice between building a web application vs a native OS application: http://scobleizer.com/2009/12/16/iphone-developers-abandoning-app-model-for-html5/

      Regarding tools for developing web applications it looks like you have not done very complex web application development if you think the tools is not there.

      First of all; Dragonfly is actually a very neat tool. It is the only debugger where you can debug on a remote client (its very nice to hook up the debugger to a mobile phone). You can not do that with FireBug or any other tool. Dragonfly does not lack much behind FireBug on other functions these days. If you think so, please specify what the differences is and not just proclaim it as a fact.

      You should also look into IntelliJ (http://www.jetbrains.com/idea/) as IDE for web application development. It has become a very good tool for JavaScript programming. Its also very good for HTML, CSS and SVG too.

      You do also have tools such as JSDoc (http://code.google.com/p/jsdoc-toolkit/) which will provide documentation for your code, you have JSLint (http://www.jslint.com/lint.html) which will guide you into good practice JavaScript coding and you have tools such as JS Test Driver (http://code.google.com/p/js-test-driver/) which give you unit testing. All these tools can be run standalone, in a IDE such as IntelliJ and Eclipse, as an Ant task or as plugins to Maven. You do also have Watir (http://watir.com/) which is a very good web application testing tool. There is a flora with many other tools too.

      Yes; there is still difference between browsers. Many of them frustrating but this is getting better and better. To deal with these differences you should use a well known framework such as, as an example, Dojo Kit or jQuery and you will be much more effective and have less pain. Choosing such a framework is the same thing as you select frameworks in other languages such as Java, .Net, PHP, Python to deal with the problems in these languages.

      You say Opera lacks compatibility; can you please specify?

      Opera is actually the browser which is most true to the W3C standard. That is a good thing for us developers. Opera is also the browser which support most W3C standards (a bit outdated, but still valid since Opera has implemented more than any other browser since the last update of this table: http://en.wikipedia.org/wiki/Comparison_of_web_browsers#Web_technology_support).

      • tst

        I’ll respond, even if it is not me you are asking questions.

        Yes, there are lots of webdeveloper’ tools/developer environments. Eclipse or Aptana as a base and lots of plugins. But problems start when you look into the details. Some of these plugins rely on 3rd party debugger. Mainly IE (visual studio) or Gecko. That means, that they are worthless when debugging other browser’ issues. Syntax highlighting isn’t exactly something that defines IDE, it is the ‘INTEGRATED’ part that is most important. It is hard/impossible to cover all browsers with one seamless environment. And it is very impossible to have tools of the same quality for all browsers, mainly because Opera choose to lock-in and does not offer API and does not offer tools or in fact even up-to-date detailed documentation of its product.

        WATIR AFAIK doest not work with Opera, Opera being too strict and paranoid with its Netscape Plug-in Architecture implementation.

        Dragonfly _IS_ dead now. It will be dead until it will remove _ONE_ fatal flaw it has: need to reload a page (completely destroying bug you were to inspect) to start using it. This is a complete no-go behavior, and until it is there, noone will try to use it. It also lacks almost everything Firebug has, starting from _EASY_ in-place CSS editing, ending with tracing, resource preview and raw http responce inspection.

        As for stuff that Dragonfly doesnt have, but others have?
        http://googlewebtoolkit.blogspot.com/2009/11/new-insights-into-web-application.html
        http://ejohn.org/blog/deep-tracing-of-internet-explorer/

        This second link is _AMAZING_. Just try it, and think how many years untill Dragonfly start to offer just 5% of this.

        “Google Docs, Google Wave, Meboo, Gmail, Hotmail etc.. etc.. and even MS’s Office Web Apps are examples of such applications ”

        Well, most of these applications do not work with Opera anyway. And it is not because they are not standards complaint. It is because Opera that isn’t ‘real world compliant’. W3C specs are so full of holes, gray areas and simple examples of reckless stupidity, that both browser and web page can be standards compliant and both could fail. This is the case of Opera, that decided to swim upstream and promote _OPERA’_ vision of how to interpret gray areas, instead of joining the crowd. Until that changes (or Opera gains 20% marketshare) common scenario will not change: the more advanced the page (standards compliant ofc) the most likely it will not work in Opera. This is the case now.

        As for the widgets if I might add, I haven’t seen one useful, same with Vista Gadgets or Win7 Gadgets, so I couldn’t care less.

        • nvm

          How would the Netscape plugin API help WATIR in any way? Does it specify browser UI integration? If so, why isn’t Firefox using NPAPI for extensions?

          Dragonfly isn’t dead, you just need a new Presto version or something.

          Most of those do work in Opera as far as I can tell.

          What gave you the idea that Opera isn’t “real world compliant”? If you read their “compatibility” blog they are detailing how they are being compatible with various browsers and stuff like that. Opera isn’t swimming upstream, it’s trying to be as compatible as possible, but it isn’t easy due to all the browser sniffing and stuff like that.

          • nobody

            almosta all i wanted to tell has been said above, but:

            dragonfly cant do sh.. now. i dont care if it is perpetum mobile in their internal builds, but what i and others can see is a development dead cold and a product, that shouldnt be called ‘workable’. releasing it to developers is f.. offence.

            yay, it can debug on mobiles! woow. so it is equaly worthles on mobiles as it is on the desktop.

            if you say, that dragonfly can do stuff like this:
            http://ejohn.org/blog/deep-tracing-of-internet-explorer/

            then id like you to check your sight sir, as it is clear that you cant see properly. dragonfly cant do anything similar.

            it cant do also much simplier tasks, like live css editing (like firebug on html tab), it does not support console object (opera swam upstream generating opera.postError – knowing only too well, that firebugg used console, and webkit uses console.. this is swimming upstream in action), its debugger more often than not simply doesnt work generating empty call stack and one line worthelss error message.

            and yes, its main fault is that it is written in JS, seemingly ignoring deep-hooking possibilities that all other dev tools utilize – firebug, visual studio, chrome/safari dev tools etc.

            reloading page to start using dragonfly is just a sign of pure retardness..

            btw, watir doesnt work in opera, why do you ask about firefox? answer yourself why it doesnt with opera. maybe it is the number of automation api’s released by opera (0) so far.

          • nvm

            I didn’t think they claimed that Dragonfly is a final product? It’s an alpha isn’t it? If you can’t handle alpha software, stay away from it, LOL.

            I asked about Firefox because someone said that Watir wasn’t working because Opera’s NPAPI implementation was too strict. If that’s the case and NPAPI allows for UI access and all sorts of stuff, why isn’t Firefox using it for extensions?

            BTW: http://my.opera.com/core/blog/2009/03/06/test-automation-with-operawatir

          • http://www.favbrowser.com Vygantas Lipskas

            It’s been an alpha for ~1.5 years, that’s quite a long time.

          • nobody

            http://my.opera.com/core/blog/2009/03/06/test-automation-with-operawatir

            operawtir isnt there yet. it is internal tool, that isnt helping webdevelopers even a tiny bit.

            this news is from march 2009. we are now entering 2010 and there is still no release.

            it is great that they have this inside, but until it is out, it doesnt exist – it doesnt help people code for opera. thus all new big and complex webapplications arent tested in opera (how?) and found bugs arent fixed (how?). so here you are – standards compliant pages not working in opera. because why bother, if opera seems to not bother at all.

            if half of effort put into widgets was put into dev tools, we would have firebug killer now. (reminder – firebug was written by ONE man in time opera managed to release this crap).

          • http://www.trygve-lie.com/ Trygve Lie

            > if you say, that dragonfly can do stuff like this:
            > http://ejohn.org/blog/deep-tracing-of-internet-explorer/

            No Dragonfly can’t. So can’t FireBug either. So can’t the dev tools in IE or in Safari. Does the IE devtools, Safari devtools and FireBug suck then also?

            DynaTrace AJAX is a good tool and I’ve been using it since almost the day they did release it. But it’s a pain to use in multiple IE versions since it’s not possible to run several IE’s on the same Windows installation. That’s not DynaTrace’s fault; it’s IE’s. And; DynaTrace AJAX does not integrate with any other browsers than IE. That does also suck.

            Another tool which is pretty similar is Speed Tracer (http://code.google.com/webtoolkit/speedtracer/) for Chrome. The problem; it only works against Chrome because all browsers has different API’s for integrating such tools against them.

            As I’ve already sad; the real problem is that all browsers have different API’s. They should agree one common API for this purpose and I’m pretty sure Opera would implement such an common API.

            > and yes, its main fault is that it is written in JS, seemingly ignoring
            > deep-hooking possibilities that all other dev tools utilize – firebug,

            FireBug is also written in JavaScript: http://code.google.com/p/fbug/source/browse/#svn/branches/firebug1.6

        • http://www.trygve-lie.com/ Trygve Lie

          > It is hard/impossible to cover all browsers with one seamless environment.

          That’s because browsers have not agreed on one common API for this purpose. The world would be better if they did. That does not make Opera suck more than any other browser. It does make all browsers suck.

          Btw; Opera has a debugger API: http://dragonfly.opera.com/app/scope-interface/index.html

          > And it is very impossible to have tools of the same quality for all
          > browsers, mainly because Opera choose to lock-in and does not offer API

          So; from what you are saying; since Opera does not offer a API; IE, Safari, FireFox and Chrome has chosen to not make it so we can have one seamless environment to build debugger tools on?

          You realy do not see the problem here. The problem is that all browsers has different API’s for this. If there where agreement on one API for this purpose I do think Opera would implement it.

          > WATIR AFAIK doest not work with Opera, Opera being too strict and
          > paranoid with its Netscape Plug-in Architecture implementation.

          If that’s the problem. Then I’m wondering why it’s so that WATIR 2.0 will get support for Opera – http://rubyforge.org/pipermail/wtr-development/2009-October/001313.html

          > Dragonfly _IS_ dead now. It will be dead until it will remove _ONE_
          > fatal flaw it has: need to reload a page (completely destroying bug
          > you were to inspect) to start using it.

          I do not have this problem. Try to switch off the reload function under settings and it will not reload.

          > W3C specs are so full of holes, gray areas and simple examples of
          > reckless stupidity

          Nothing is perfect. Your are free to help W3C out with the specs. That’s the beauty of W3C and also why it does take so long to write the W3C specs. The public can contribute to the W3C specs. The public means you as a private person can contribute free of charge. W3C does listen to all suggestions made to the specs and all comments are taken under consideration before a spec is made final. I suggest you use your energy on contributing on making better specs than complaining about how poor they are.
          http://www.w3.org/2005/10/Process-20051014/intro

          • nobody

            “Btw; Opera has a debugger API: http://dragonfly.opera.com/app/scope-interface/index.html”

            you try to make me look like fool, while anyone serious knows how dragonfly sucks. yes, it is called debugger, yes, it SHOULD do some stuff, but no, it CANT do them properly now. it like naming old rawing boat ‘yaht’ and waiting for magic transformation. it wont happen.

            “So; from what you are saying; since Opera does not offer a API; IE, Safari, FireFox and Chrome has chosen to not make it so we can have one seamless environment to build debugger tools on?”

            difference is, that all other browsers have some kind of automation API, yes, different, but it is there – thats why waitr can be used with all other browsers, while it cant with opera, as it hasnt got any.

            “> Dragonfly _IS_ dead now. It will be dead until it will remove _ONE_
            > fatal flaw it has: need to reload a page (completely destroying bug
            > you were to inspect) to start using it.

            I do not have this problem. Try to switch off the reload function under settings and it will not reload.”

            you pretend to not understand. dragonfly on first start HAS to reload debuging scope, it works that way. opera event comented on why firebug doesnt have this problem, and they as for now decided to not use firebug’ approach (being always on) due to performance issues that dragonfly still has. just try to understand harder. start opera with no DF running, start DF, DF WILL reload a page for you, or youll have to do it yourself before debugging javascript. DF works this way now. and it sucks badly.

            “Nothing is perfect. Your are free to help W3C out with the specs. That’s the beauty of W3C and also why it does take so long to write the W3C specs. The public can contribute to the W3C specs. The public means you as a private person can contribute free of charge. W3C does listen to all suggestions made to the specs and all comments are taken under consideration before a spec is made final. I suggest you use your energy on contributing on making better specs than complaining about how poor they are.
            http://www.w3.org/2005/10/Process-20051014/intro”

            eeee what? you try to dissolve the problem: w3c specs are bad NOW, and all talk about ‘standards’ NOW doesnt have much sense. defending opera because it is standards compliant is quite senseless, because standards compliant isnt everything browser needs to do. it also has to be compliant will leading implementation – this means firefox’ now. opera isnt, and seemingly doesnt want to be.

            “No Dragonfly can’t. So can’t FireBug either. So can’t the dev tools in IE or in Safari. Does the IE devtools, Safari devtools and FireBug suck then also?”

            but i CAN do this stuff with other browsers (at least in IE with dynatrace, and if you can find bottlenecks in IE, you moset probably found 95% of perfofmance problems in all browsers), because other browsers are open to 3rd party modifications. opera locked its product out, so it is OPERA that has to provide tools. other browsers can use 3rd party tools. opera doesnt do much to ship tools that are up to date or any of any value whatsoever.

            “As I’ve already sad; the real problem is that all browsers have different API’s. They should agree one common API for this purpose and I’m pretty sure Opera would implement such an common API.”

            this problem can be bypassed by hiding tools behind IDE and IDE taking care of ‘integrating’ it out of users view. you cant do anything to opera not providing tools nor API. waiting for browser vendors to organize and decide upon something is laughable excuse.

            “FireBug is also written in JavaScript: http://code.google.com/p/fbug/source/browse/#svn/branches/firebug1.6”

            yes, but it uses deep hooks into rendering engine that gecko exposes. IE does the same (to the extreme) and webkit does good job exposing new features. opera? locked in even for their developers. debugging and tracing javascript with javascript is rather stupid, when rendering engine knows very very well (and fast) how long certain action took, why to measure it again (with performance penalty?).

        • http://www.trygve-lie.com/ Trygve Lie

          > As for the widgets if I might add, I haven’t seen one useful

          There are several useful widgets out there. The therm useful is very verbose but you can for instance look at the Vodaphone 360: http://www.vodafone360.com/
          All 360 clients running on the phones are widgets.

    • OperaGuy

      I think you are just spreading fud. I stopped reading at “DragonFly is dead”. You obviously have no idea what you are talking about.

      There are current builds for even the “yet to be released” Presto 2.5

      https://dragonfly.opera.com/app/core-2-5/cutting-edge/zips/

    • http://my.opera.com/IceArdor IceArdor

      Widgets are the way to go. If you’ve spent a little bit of time figuring out how to make an App for the iPhone, you’d realize how much better the widget platform is. As described by Popular Science, designing an App for the iPhone is a financial and time investment. With widgets based on HTML and Javascript, you won’t have to worry about downloading some iPhone App SDK and a Cocoa environment that lets you code in Objective-C. And then you have to recompile and load it onto your iPhone to test it, with no guarantee that future firmware versions will like your App. When you stick to web standards, you know your widget will last as long as the web standard is supported, which is pretty much forever.

  • nobody

    “Widgets are local apps, not everything on the internet.”

    so what can you do with them? browsing trough opera’widget catalog i couldnt find one single item that has any sense at all. how many clock/calendar/b-day widgets can there be?

    “What makes you think the only thing you can do on the web is social networking? That’s what it sounds like. Chrome OS gives you a browser. You can do anything with a browser.”

    watch chromeOS advert once again, they imply, that almost all you do is social-zombying and some online (why?!?!) document editing.

    “The need for IDE tools for web apps isn’t that pressing because it’s much easier, faster and cheaper to do web apps than native apps.”

    you’ve NEVER designed anything more complex than widget then.

    it isnt cheaper, it isnt easier (supporting even 4 browser (IEs, ff and webkit) is a horendal task, you need for each browser entire testing and development environment – because browser vendors implement so much stuff differently not documenting it. you need completely different set of tools for each browser (or lack of them for these 3 people that still code for opera), you need to know MORE about entire process as a developer than native app coder, that codes for one platform, with one IDE and has various wonderfull tools like unit tests or refactoring helpers.

    only thing that is easier is deploymnent and ‘always-current-version’ effect. all other is very comparable. unfriendly development environment being clearly biggest problem. you know what ms did great? documentation for .net, win etc platforms. there is NOTHING comparable if it comes to web applications. even yahoo’ YUI documentation (the best of the best now) falls short compared to MS effort and quality.

    writing native apps is kid-easy task nowadays, try .net for starters, or even ruby/python. it takes less than few hours to start working (given that you know something about general software engineering). it takes years of painfull experience to swim accross buggy sea of browsr incompatibilities.

    yes, better not always wins. but usefull? always. opera widgets arent usefull now. and given that noone will write them, if there is comparable chromeOS around, they’ll never be. opera tries to win users, by appealing to carriers and phone manufacturers. it doesnt work that way, not with strong competition.

    • nvm

      You can do with widgets whatever you do with web sites presumably! And with hardware acceleration, crazy fast JS and all that it’s going to be a nice ride for both online and offline web apps.

      I don’t have the same impression of Chrome OS that you do, but if they do focus on those things, so what? It’s not like Google is holding a gun to your head to keep you from doing work on the web, LOL.

      Opera widgets seem to be useful to Opera, who are scoring lots of nice deals involving them. Also, there are lots of widgets at widgets.opera.com.

    • TTT

      “The big picture” is: Opera invented most of the features that now most browser have copied. From tabs on! This means that the things you think nobody cares about could become pretty popoular..
      All that Opera has always lacked is good marketing, that’s why it has a small user base…

      Python and Ruby apps are not native apps: they are interpreted just like js so the problems are the same.. Real native apps are clearly more difficult and time-consuming to develop than html and js because native languages are NOT cross-platform and so you need to develop them for all the OS.. and each OS has its nice package of incompatibilities that make browser incompatibilities look like kids-play..

      Widgets made with html,js&co with super-fast standard-compliant browsers could really be the next best thing for lots of small apps.. you would really understand if you wanted to work in the same way on multiple machines and multiple OS: internet and web-related technologies are the main common factor and that’s not just Opera or me that think this..

  • bingo007

    i dont care about speed….even with windows 7 ultimate i am not noticing significant speed differences….the fastest is opera turbo mode which beats any other browser in speed by atleast 4 times but isnt loading most pages in it format and most of the time various sections are misplaced….things i want changed in opera 10.5 is the browser shouldnt go on loading the final element i.e. 245/246th element loading…..if they can fix it it ll be awesome…

  • daniel h

    This will have Vega I assume because of a tweet from Mr. Hicks talking about rounded corners.

  • somebody

    “All this work going into the next Opera, and I get the impression that some folks will only care about border-radius #bloodyroundedcorners”

    Seems like they have lot of stuff coming with 10.5.
    But maybe there will only be a few things in this release. After all, this is just a pre-alpha build.

  • http://my.opera.com/IceArdor IceArdor

    If Carakan has a new backend and a new version of Presto, wouldn’t we expect it to be v11, not v10.5?

    • somebody

      I think Desktop version numbers are not necessarily reflective of the core version. Normally, they increase a point or so in the version number if there is no other change except the core. And it has been confirmed by one of the developers that Opera 10.5 will have Presto 2.5, not even 2.4.
      I’m very excited to see what goodies they have in store. Vega and Carakan are pretty much confirmed i guess.

  • somebody

    I think the title needs to be updated. It is most probably Presto 2.5, not 2.4

    http://twitter.com/pepelsbey/status/6773799764

    http://twitter.com/dstorey/status/6779895865

  • lucideer

    “I think the title needs to be updated. It is most probably Presto 2.5, not 2.4″

    Yup. Some have even mentioned it could be 2.6 or 2.7 (though they were probably joking).

    Another possible inaccuracy in the article – Carakan 2.5 times faster? Where was this number taken from – their initial article said 5-50 times faster in early dev… which sounds a bit more than a mere factor of 2.5

    As for the comments above mine:
    @nobody
    “Dragonfly is dead” – why? Substantiation?

    @Trygve Lie
    You make an excellent point that widgets are not just “Opera widgets”, but a W3C standard. To bolster this, someone has actually already implemented an XUL based widget engine that is completely compatible with Opera widgets – so we should hopefully have cross-browser widget development on the way! Which I’m really looking forward to. I don’t know if it will be possible with webkit though…

    @tst
    “Dragonfly _IS_ dead now. It will be dead until it will remove _ONE_ fatal flaw it has: need to reload a page”
    This FEATURE is toggleable in settings and has been for a VERY LONG TIME. You should try actually verifying claims before commenting ignorantly.

    “It also lacks almost everything Firebug has, starting from _EASY_ in-place CSS editing, ending with tracing, resource preview and raw http responce inspection.”
    It’s tracing is far superior to start, Opera’s native CSS editing (something it’s had since before Firebug was invented) is and always was far superior to Firebugs or any alternatives. Resource preview is a nice frill, but hardly a game changer. And AGAIN Dragonfly has had raw http response inspection for a VERY LONG TIME. It really sounds like you’ve never actually used Dragonfly.

    @IceArdor
    I think Opera’s major release cycle is every 0.5 – otherwise they’d be nearing version 20 at this stage!

    • lucideer

      Correction: I accidentally confused “tracing” and “trace” in the above post. Ignore that paragraph.

    • nobody

      “This FEATURE is toggleable in settings and has been for a VERY LONG TIME. You should try actually verifying claims before commenting ignorantly.”

      wrong, it HAS to reload a page. start opera, load a page, start dragonfly – it WILL reload the page. it simply works that way. what you are refering is in fact here for a long time (reload documents auto on selecting a window – it has nothing to do with behavior when DF is closed and you start it), but it has nothing to do with that MAJOR flaw.

      “Opera’s native CSS editing (something it’s had since before Firebug was invented)”

      tell me where is the option – available in all other tools – to disable a line in css by clicking it (it firebug it sets a ‘no go’ sign and makes it gray. i can only delete a line or make it nonsensial to make it go away. but where is that basic option?

      where is the right-click menu of many possibilities, like add attribute to node?

      “Resource preview is a nice frill, but hardly a game changer.”
      well, it is there just because it is needed in real world applications. theoretics like you might not see the point.

      “And AGAIN Dragonfly has had raw http response inspection”

      oh really? oh lol.

      Raw response
      HTTP/1.1 200 OK
      P3P: policyref=”http://googleads.g.doubleclick.net/pagead/gcn_p3p_.xml”, CP=”CURa ADMa DEVa TAIo PSAo PSDo OUR IND UNI PUR INT DEM STA PRE COM NAV OTC NOI DSP COR”
      Content-Type: text/html; charset=UTF-8
      Content-Encoding: gzip
      Date: Fri, 18 Dec 2009 21:02:15 GMT
      Server: cafe
      Cache-Control: private, x-gzip-ok=”"
      Content-Length: 1959
      X-XSS-Protection: 0

      this is what dragonfly returns. where is the BODY of this response? the f.. xml file?

      ““Dragonfly is dead” – why? Substantiation?”

      what i see is what i can use. i cannot use a tool that is aavilable only inhouse for opera employees. for me dragonfly is in version 0.7 alpha-4 that hasnt changed for a year now. it is unusable for anything bigger than mere fooling around. if it is better inhouse – lets see it, until then, DF is dead.

      it is you who knows nothing what he is talking about

      “I don’t know if it will be possible with webkit though…”

      yes, webkit can handle opera widgets with no problem. in fact it is a bad thing for opera, as webkit can root opera out really fast of mobile market.

      “widgets are not just “Opera widgets”, but a W3C standard.”

      there are not standard yet. and it is a good thing, because opera’ specs that they’ve submited has few pages and is so far from being strict and precise as a blob of mud.

      • nvm

        LOL, that logic dictates that Windows 7 was dead because it was in internal development for a long time before anyone outside MS got to have a look.

        • http://www.favbrowser.com Vygantas Lipskas

          So you assume that Dragonfly and Windows are on the same difficulty level?

          • nvm

            Not really relevant for my point.

        • nobody

          and that is your entire answer?

      • nvm

        “webkit can root opera out really fast of mobile market”

        Making a browser based on webkit is really, really hard. RIM bought an entire company to get an existing webkit browser, but they had to keep hiring because it wasn’t enough.

        Webkit is more of a threat to Mozilla than to Opera. Chrome will own Firefox.

    • http://www.favbrowser.com Vygantas Lipskas

      lucideer,

      http://my.opera.com/core/blog/2009/02/04/carakan

      So how fast is Carakan? Using a regular cross-platform switch dispatch mechanism (without any generated native code) Carakan is currently about two and a half times faster at the SunSpider benchmark than the ECMAScript engine in Presto 2.2 (Opera 10 Alpha).

«
»