Ho confrontato Google Chrome con la versione beta del nuovo Firefox 3.1, che include l'aggiornamento del motore di rendering (Gecko) e l'introduzione della tecnologia TraceMonkey.

SI tratta in sostanza un compilatore particolarmente ottimizzato per Javascript che velocizza l'esecuzione di tutte quelle applicazioni web che ne fanno uso: tutti i siti di social network (tipo facebook) e i motori CMS per i blog (come wordpress) e i siti (come Joomla!), ovvero tutto il cosiddetto web 2.0, nonché gmail, google docs e tutte le applicazioni che funzionano dentro un browser.

In altre parole, l'equivalente di V8 presente in Google Chrome, ma, allo stato, più veloce.

Il test su Javascript utilizzato è Sunspider, di Apple webkit (ricordo che webkit è il motore di rendering di Safari e di Google Chrome).

Inoltre ho sottoposto i due browser al test Acid3 per la verifica della rispondenza agli standard web. Maggiori informazioni su questo test sono reperibili su Wikipedia.

Le versioni confrontate sono:

Google Chrome Beta come distribuito da Google

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/525.13 (KHTML, like Gecko) Chrome/0.2.149.27 Safari/525.13

Firefox 3.1 beta1pre, nightly build del 5 settembre 2008 (con jit attivato)

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1b1pre) Gecko/20080905031348 Minefield/3.1b1pre

Entrambi, ovviamente, in versione Windows.

Ed ecco i risultati dei test:

ACID3:

Da notare anche che durante il test, Google Chrome ha notevoli problemi di performance.

Passiamo ora al test di velocità Javascript:

Firefox batte Chrome per la seconda volta.

E' necessario però abilitare Tracemonkey, che di default è disattivato.

E' semplicissimo.

1. scrivere about:config nella barra degli indirizzi di firefox

2. scrivere la parola chiave jit

3. cliccare sulle 2 opzioni che compaiono (se leggete chrome sappiate che non c'entra con Google Chrome :) )

Curiosamente, Google Chrome batte di molto Firefox 3.1 nel test dromaeo, di Mozilla, che però è  molto meno approfondito di Sunspider e difatti quasi mai preso in considerazione per i confronti, neppure da Mozilla stessa che, all'annuncio di Firefox 3, pubblicò i dati di Sunspider.

Da notare anche che, tra i browser in beta basati su webkit, Google Chrome è quello con la percentuale peggiore all'acid3 test (da marzo infatti, le versioni di sviluppo del motore webkit raggiungono 100/100).

Ho fatto anche un confronto con Opera.

Questi sono i risultati di dell'ultima versione di Opera (9.52.2):

Acid3: 83/100

Sunspider: 8816.4ms

L'ultima versione stabile di Firefox (quindi senza TraceMonkey), ovvero la 3.0.1, impiega la metà a superare il test Sunspider (precisamente 4440 ms) mentre per l'Acid3 si ferma a 75/100.

Installare Firefox 3.1b1pre (nome in codice: minefield)

Windows: scaricare l'installer qui ed eseguirlo

GNU/Linux: scaricare il file compresso, scompattarlo ed eseguire il file "firefox" in esso contenuto

Ubuntu Hardy Heron (8.04): Aggiungere i repository seguenti e successivamente installare da synaptic i pacchetti firefox-3.1 e firefox-3.1-gnome-support

deb http://ppa.launchpad.net/fta/ubuntu hardy main

Se non sapete come si aggiungono i repository guardate qui.

Ultima nota: le build di Firefox 3.1 per GNU/Linux sono un po' indietro rispetto a quelle per Windows, per cui le prestazioni sono minori per Javascript, ma comunque migliori di Google Chrome.

]]> http://joeb454.wordpress.com/?p=163 Thu, 04 Sep 2008 15:03:02 +0000 Joeb454 http://joeb454.wordpress.com/?p=163 There's been what I would describe as "mass press overload" on Google Chrome since it's release a day or 2 ago (depending on time zones). So I thought I'd join in ;) And yes - I'm currently posting this from Chrome, running on Windows Vista.

Unfortunately, I can't try it out on Ubuntu just yet, as there's no Linux builds, and trying to build it from source won't work, many have tried. People were hopeful that it would run in wine...No dice. So we'll just have to wait. Until then, I can continue merging and moving threads about on the forums. Which have been literally inundated with threads regarding the browser, so there is now a "Google Chrome MegaThread" :p

I have to say, I actually quite like it...now that the EULA has been revised that is. And it renders very quickly (Yes - faster than Safari, which according to Apple is "the fastest, easiest-to-use web browser in the world" - perhaps they should rethink that statement?).

As this blog seems to be getting a lot of hits still concerning Acid3, I tried Google Chrome from the same acid test website. And it gets....78/100! I think that's pretty respectable, though it fails the LinkTest (screenshot supplied below).

Anyway - more on this later, possibly when there's major improvements or a Linux version :)

Update: I forgot to add the screenshot of the Acid test, here it is!

Here is my contribution to the buzz.  How does the new browser behave in terms of standards compatibility? Let's take the tests.

ACID2 gives

Which means 100% compatibility.

ACID3 gives :

which means 78% compliance. Were it 100% it should look  like this:

How do these results compare to the browsers on the market? I have run the same test on all the known browsers that pass the ACID2 test  and here are the results. All the test were run on a Windows XP with SP2.

So Chrome does a fairly good job in ACID3, but still trails Opera.

Update: I just spotted an article on CNET about Chrome's standards compliance.Instead of comparing IE8 it compares IE7 on the Acid3 test and gives a result of 14% which I could not confirm.  The acid servers were very slow to respond but when they did I just got a 12% as in IE8.

On the surface, Chrome doesn't feel much different from our beloved Firefox 3. Sure, it looks a little funny (and very blue), but it handles much the way any normal browser does. It has a phishing/malware blacklist, just like Firefox, a download manager, tabbed browsing, and a "smart" address bar that does everything under the sun.

Chrome does sport a couple features that set it ahead of Firefox: multiprocessing for tabs, a super-quick Javascript virtual machine, and a task manager. So are people going to be leaping out of their chairs to download Google Chrome? Or will it sputter and die because Firefox is "good enough" for most people? I managed to convert my mother from IE7 to Firefox 3 not by telling her how much better Firefox is, but by telling her that you don't have to put up with any crap in Firefox. That is, she was converted not because Firefox is so good, but because it doesn't suck.

So I feel like much of Google's potential market for Chrome is already taken and unwilling to make a change. Why should someone want to change from one perfectly good browser to another that they've never used before? Multiprocessing might seem neat to us techies, but to average Joe it's not that important.

In light of this, we have to wonder how much effort Google will spend advertising their browser. Harry McCracken asks a good question:

Just how hard will Google push Chrome on the Google homepage? Like no other company on earth, Google has an opportunity to get hundreds of millions of people using its browser in a relatively short amount of time. You gotta think that it’ll use the Google homepage to drum up interest. But will it check to see if you’re using IE, Firefox, or another browser and attempt to convince you to switch?

Currently, the Google homepage displays nothing at all related to Chrome (Edit: Nevermind, it's up there now - 9/03). Of course, hype has done a significant amount of advertising for them (sounds hypocritical, but I'm a Google fanboy, so I don't mind).

I'd love to report on Chrome's performance in the Acid2 and Acid3 tests, but they're either down or operating very slowly right now, perhaps because everyone thinks like I do. Edit: Chrome seems to pass Acid2 and scores a 79/100 on Acid3.

Source: Genbeta

Target: Mozilla Firefox y Microsoft Internet Explorer 8

La guerra de navegadores es algo tan antiguo como la propia Internet. Desde el primer navegador (ViolaWWW), el mítico Mosaic que el estudiante Marc Andressen realizó como proyecto fin de carrera, pasando por la guerra entre Netscape y Microsoft, hasta la actual guerra por los usuarios entre Microsoft y Mozilla. Hasta ahora Microsoft ha conseguido ganar claramente esta batalla desde que sacó del mercado a Mosaic, en gran medida por el dominio de su sistema operativo Windows.

Pero Microsoft tiene un problema importante desde hace ya algún tiempo con Firefox. El porqué hay que buscarlo sobre todo en los complementos, esas pequeñas aplicaciones que le podemos añadir a nuestro navegador y nos permiten hacer cosas que el navegador no permite de serie. Firefox tiene muchos más complementos que Explorer, adicionalmente éstos son más fáciles de instalar y son siempre gratuitos. Todo esto a pesar de que los complementos han sido una de las prioridades en el desarrollo de Explorer 8. Un golpe bajo de la nueva versión de Microsoft es que cuando se instala pregunta si el usuario quiere que se busquen extensiones similares a las que se tienen instaladas para Firefox (el caso es que Firefox tiene muchas más extensiones que Explorer, así que el resultado será, en cualquier caso, cuestionable). Partiendo de la base que para el usuario experimentado los complementos resultan indispensables, y que para el usuario medio los complementos son cada vez más importantes, Firefox tiene cartas ganadoras en esta partida que dura ya años.

La gestión de las descargar y los marcadores es otro de los aspectos importantes de la comparativa. Explorer 8 prácticamente no se ha preocupado de este aspecto, por lo que Firefox sigue conservando una clara diferencia. Lo único que podemos destacar en este sentido es la incorporación de la tecnología de webslices, que permite descargar partes de una página web.

Con respecto a las pruebas para desarrolladores, Explorer 8 ha progresado al pasar el test Acid2, pero está muy lejos en Acid3 (18/100) de llegar al nivel de Firefox (71/100). El Acid3 es un test desarrollado para ver si un navegador respeta las especificaciones de los estándares web desarrollados por el W3C. Es decir, nos indica si el navegador cumple con los estándares generales de desarrollo de web. En función de la salida de nuevos estándares, hay nuevas versiones del test. La versión actual del test es la 3. Que un navegador cumple total o parcialmente el test quiere decir que mostrará con mayor o menor fiabilidad la página web tal como fueron programadas exactamente (respetando estilos, fuentes, etc).

En definitiva, estamos ante un intento que empieza a ser desesperado por parte de Microsoft de plantarle cara a una aplicación que le está superando claramente y que va inclinando poco a poco la balanza de su parte. Esta vez Microsoft lo tiene mucho más complicado que en batallas anteriores, y de hecho no tiene cartas ganadoras esta vez. Su último intento está muy lejos de llegar al nivel de su rival, por lo menos en esta primera Beta 1 que está disponible en el mercado.

• http://www.nextmag.pl/index.php/news/nid/1300
• http://www.nextmag.pl/index.php/news/nid/1335
• http://www.nextmag.pl/index.php/news/nid/1319

Na początku porównam przeglądarki Opera 9.27 i Opera 9.5, bazując na danych z wymienionych wyżej testów:

Test                            | Opera 2.7 | Opera 9.5
ACID3 [%]                       |    46     |    83
Dojo 1.0.2 [ms]                 |   116     |    94
JQuery 1.2.3 [ms]               |   295     |   180
MooTools 1.2beta2 [ms]          |   304     |   116
Prototype 1.6.0.2 [ms]          |   406     |   180
JavaScript Performance Test [ms]|  1072     |   482.5
SunSpider Test [ms]             | 28775.4   | 10674.8
CSS3 Info selectors-test [%]    |    60     |   100
Dromaeo [ms]                    | 13593.5   |  2946.8

Teraz Firefox 2.0.0.14 vs Firefox 3.0:

Test                            | Firefox 2.0.0.14 | Firefox 3.0
ACID3 [%]                       |     52.5         |   71
Dojo 1.0.2 [ms]                 |    305           |  179.5
JQuery 1.2.3 [ms]               |    588.5         |  341.5
MooTools 1.2beta2 [ms]          |    241.5         |  173
Prototype 1.6.0.2 [ms]          |    393           |  312.5
JavaScript Performance Test [ms]|    890.5         |  317.5
Celtic Kane Test [ms]           |   1891           |  532
RockStar Test [ms] (średnia)    |      0.921416316 |    0.321248114
SunSpider Test [ms]             |  27172.4         | 5364.8
CSS3 Info selectors-test [%]    |      60.46511628 |   83.72093023
Dromaeo [ms]                    |   12589          | 2606.8

I na koniec Opera 9.5 vs Firefox 3.0:

Test                            |    Opera 9.5    | Firefox 3.0
ACID3 [%]                       |    83           |   71
Dojo 1.0.2 [ms]                 |    94           |  179.5
JQuery 1.2.3 [ms]               |   180           |  341.5
MooTools 1.2beta2 [ms]          |   116           |  173
Prototype 1.6.0.2 [ms]          |   180           |  312.5
JavaScript Performance Test [ms]|   482.5         |  317.5
Celtic Kane Test [ms]           |   375           |  532
RockStar Test [ms] (średnia)    |     0.516834861 |    0.321248114
SunSpider Test [ms]             | 10674.8         | 5364.8
CSS3 Info selectors-test [%]    |   100           |   83.72093023
Dromaeo [ms]                    |  2946.8         | 2606.8

Opera 9.5 vs Firefox 3.0 - 7:4 .

Opera 9.5 wygrała z Firefox 3.0 w siedmiu testach, a przegrała w 4. Widać, że Opera lepiej przeszła m.in. test ACID3 i CSS3. Świadczy to o tym, że przeglądarka ta lepiej spełnia standardy. Mozilla zresztą nigdy nie przejmowała się standardami. Wystarczy choćby wspomnieć, że Opera wydała przeglądarkę przechodzącą test ACID2 20 czerwca 2006 roku (Opera 9.0), a Mozilla 17 czerwca 2008 roku (Firefox 3.0) (niecałe dwa lata później!). Firefox za to szybciej interpretuje skrypty JavaScript, lecz o poprawności ich interpretacji można dyskutować. W końcu gorzej radzi sobie z testem ACID3, który jest "wypchany po brzegi" skryptami JavaScript. Lisek szybciej także interpretuje, ogólnie, kody stron internetowych. Szybciej jednak nie znaczy lepiej. To również widać w testach.

Przedstawiłem tu wydajność, zgodność ze standardami i szybkość przeglądarek. A co z bezpieczeństwem?
O bezpieczeństwie nie można jeszcze zbyt wiele powiedzieć, bo minęło zbyt mało czasu od wydania obu przeglądarek. Można jednak powiedzieć już coś o Firefox'ie: http://www.theregister.co.uk/2008/06/19/firefox3_bugs/. Luka ta została w nim znaleziona zaledwie godzinę po wydaniu aplikacji.
W Opera Browser 9.5 nie odnaleziono żadnego buga.

Czy Firefox 3.0 jest tak dobrą przeglądarką, jak twierdzi Mozilla? Jeżeli porównać Firefox 3.0 do Firefox 2.0.0.14, to rzeczywiście widać duże postępy. Jeśli jednak porównamy go do przeglądarki Opera 9.5, to nic specjalnego w nim nie ma. Ja jednak zostanę przy Opera 9.5.

A co można powiedzieć o przyszłości? Niewiele. Nie będę pisał o zamierzeniach producentów przeglądarek, bo jest to bez sensu. Ja wolę fakty.
Weźmy ACID3. Firefox 3.1 preAlpha w teście ACID3 zdobywa 80 punktów. Jest to o 9 punktów więcej niż Firefox 3.0, ale mniej niż zdobyła najnowsza, stabilna wersja przeglądarki Opera 9.5 - 83 pkt.
Za to testowa wersja WinGogi zdobywa w tym teście 100 pkt.

Mozilla Firefox 3.0 is pretty fantastic.

I have to be honest, I was a little hesitant about upgrading. It's not that I wasn't looking forward to checking it out -- I'd tried out v.3.0 beta 5, and v.3.0 RC1 after all.  It's just that I really do rely on a number of key add-ons for my everyday browsing and Web development.

I needn't have worried. At work 1/20 didn't work with Firefox 3, and at home it was 2/17.  The offending add-ons were "Tails" (a microformats add-on) and "X-Ray" (which shows HTML tags in the page).  Nothing show-stopping, nothing that couldn't be replaced with something else or dropped.

Firefox 3 is so much faster, both loading and rendering pages. The new features on the address bar are fantastic. My only niggle is that I really don't like the default theme.  I loved the chunky look of versions 1.0.x, 1.5.x and 2.x.  So I'm using CrystalFox Modern until someone ports the old theme to version 3!

Oh, and it passes the Acid2 test and scores 71/100 on Acid3.

But hey! if that's my only criticism of Fx3 then they must be doing something pretty darned well!

Right, now I'm off to bed!

Firefox 3 was just officially released, Opera 9.5 was released about a week ago, Safari 3.1 has been available for a while, and IE8 BETA and Safari 4 BETA are available.  It also just so happens K-Meleon 1.5 BETA is available as well.  I will give a run through Firefox 3, Opera 9.5, Safari 4, Internet Explorer 8 and K-Meleon 1.5.  These tests will be for how well a browser is fit to what I personally want, though, so mileage may vary on agreement in features.

First thing to note are my complaints with K-Meleon 1.15:
-Pop-up blocking is inadequate and circumventable.
-JavaScript speed is a bit slow
-It closes itself to reopen a tab (jumps around on the taskbar)
-Rendering engine is dated (early Gecko 1.8 - 52/100 on Acid3)
-View source sometimes reloads page without checking cookies (so you can't see the source if you need to be logged in)

I've been giving K-Meleon 1.5 a go, and there are improvements:
-Pop-up blocking is fixed
-It no longer jumps around on the taskbar

However, all rendering is the same.

Here's my hypothetical perfect browser, in order of priority:
-Tabbed
-Option to have URLs from address bar open in new tabs
-Fast and low memory
-Adblock and Flashblock
-Custom tab-size/reasonable tab size
-Option for tabs ONLY (no extra windows)
-Compactable (Forward/Back and address bar all on the File/Edit/etc. bar)
-Clear JavaScript error reporting
-Double-click to close tabs
-Firebug or similar tool
-Built-in XHTML validation or quick access thereof
-Good view source highlighting
-Can send any page to any other browser through menu / right-click

All browsers are tabbed, so that's covered immediately.

For my testing, it's going to be quite informal.  I'm not running benchmarks, just my personal subjective feelings on how things are going.

I'll start through the technical tests first:

First test is the basic speed test.  Here's the results in order:
1. Safari 4
2. Opera 9.5
3. Firefox 3
4. K-Meleon 1.5
5. Internet Explorer 8

The difference between Safari and Opera are relatively negligable.  I find Firefox seems a little slower rendering, though not bad by any stretch.  K-Meleon 1.5 actually comes across as a little slower at times than 1.1.5.  I'm not sure why.  It's slow enough that there's a desireable and noticeable difference between it and Safari.  Internet Explorer 8 is pathetic.  It is so incredibly slow, it's hard to believe.  I'm really giving it a chance here, but it's terrible.  There are constantly lag points where the browser doesn't respond, the JavaScript performance is awful.  It's just not good.

Evaluating the speed, I'd say Safari and Opera are big enough win to warrant an advantage in this area, while Internet Explorer is so brutally behind that it deserves severe penalty in this area.

Second is standards compliancy.  This is how accurately a browser renders a page with regard to standards set forth by the W3C.  Each will take the Acid3 test

1. Safari 4 - 100/100
2. Opera 9.5 - 83/100
3. Firefox 3 - 71/100
4. K-Meleon 1.5 - 52/100
5. Internet Explorer 8 - 18/100

The pattern continues in terms of technical prowess with Safari 4 leading the pack, followed by Opera, and IE8 giving a new name to failure.  In practice, though, Acid3 isn't a perfect test as it doesn't account for human direction.  Safari is so accurate in standards that I have found it is compatible with sites,  though Opera 9 (though I haven't tested enough with 9.5, tends to have some problems on occasion).  This is due to the market mainly being a IE/Firefox market.  However, IE8 can't even excel here because its standards introduce so many new problems that pages that worked in IE7 are broken in IE8.  At least there's an emulate IE7 button.

Third and last in technical is memory management.  I opened 5 tabs of GG Weekly, waited for them to load then checked the RAM usage, then closed them all and opened 1 and checked.  Here was how that worked:
1. K-Meleon 1.5 (54MB/42MB)
2. Opera 9.5 (83MB/53MB)
3. Internet Explorer 8 (92MB/68MB)
4. Firefox 3 (100MB / 92MB)
5. Safari 4 (102MB / 94MB)

This has always been a big advantage for K-Meleon, which it still maintains.  Where Firefox 2 had significant memory issues, K-Meleon offered the same rendering at a lower bloat.  If you look at tech site tests, they declare Firefox 3 the winner in memory manage between it, IE, Opera and Safari, but why?  It's due to that under heavy load, Firefox does an incredible job at maintaining the same memory useage.  So at lower use, it's not as efficient, but high use, it is.  K-Meleon is the big winner here.

On to features.  URLs in the address bar to open in a new tab:
Success:
K-Meleon 1.5
Firefox 3 (with an extension)

Partial Success:
Opera 9.5 (functional work around)

Failure:
Safari 4
Internet Explorer 8

K-Meleon has this built-in, while Firefox 3 requires the Tab Mix Plus extension.  Opera 9.5 I was able to edit the keyboard shortcuts so that when I press CTRL+L to select the address bar, it also opens a new tab.  Not perfect, but very similar in function.  Safari and IE I could not find any way to have addresses open in a new tab.

Tab-size formatting:
Success:
K-Meleon 1.5
Firefox 3

Failure:
Opera 9.5 (Opera 9 skin.ini hack doesn't work)
Safari 4
Internet Explorer 8

Another built-in function of K-Meleon.  Firefox 3 has an available adjustment in about:config, but can also be adjusted using Tab Mix Plus.  Opera 9.5 may be able to get smaller tabs via an alternate skin, but the adjustment that worked in Opoera 9.2 doesn't appear to work here.  Safari 4 and Internet Explorer 8 don't seem to have anything of the sort.

Tabs only:
Success:
K-Meleon 1.5
Firefox 3

Partial Success:
Opera 9.5

Failure:
Safari 4
Internet Explorer 8

For all intents and purposes, Opera is tabs only since links that are intended to open in new windows open in tabs, but technically you can always still open more than one window.  Safari and IE8 painfull always allow a new window including any target="_blank" or "window.open()".

Adblock and Flashblock:
Success:
K-Meleon 1.5
Firefox 3
Internet Explorer 8 (ok, I didn't download the add-ons, but they're out there)

Partial Success:
Opera 9.5

Failure:
Safari 4

K-Meleon has them both built-in, Firefox has the ever-popular add-ons, and IE8 has its own imitators.  Opera 9.5 has the content filter, and user-made JavaScript file tricks to get adblock and flashblock, so it works well-enough, if not bothersome to set-up.  If the extensions exist for Safari on Windows, I can't find them while filtering through the Mac offerings.

Capactable:
Success:
K-Meleon 1.5
Firefox 3

Failure:
Opera 9.5
Safari 4
Internet Explorer 8

The usual suspects.

Double-click to close tabs:
Success:
K-Meleon 1.5
Firefox 3 (with an extension)

Failure:
Opera 9.5
Safari 4
Internet Explorer 8

Here, I actually suspect that there's some way to do this in Opera, but I cannot find it.  And don't think I haven't tried.

Onto the debugging and development section.
JavaScript error reporting, HTML display:
1. Safari 4
2. Firefox 3
3. K-Meleon 1.5
4. Opera 9.5
5. Internet Explorer 8

Once I found it, Safari's error reporting and debugging tools were very nice and clear.  Firefox 3 is also quite nice, while K-Meleon can be occasionally troublesome, and Opera's I just find hard to read (it is rather thorough, though).  IE8 makes an error, but still gives cryptic error messages and the debugging tool just isn't very user-friendly.

Firebug or similar tool for DOM inspection
Success:
Firefox 3
Opera 9.5
Internet Explorer 8
Safari 4

Failure:
K-Meleon 1.5

This is an area in which K-Meleon is obviously lagging behind, including the 1.5 release.

Summation:
K-Meleon in terms of options in the browser has everything built-in, which means when you install, you know you can get the features.  Improved speed and standards compliance is what it mainly needs, but certain development features are lacking while competitors are improving quickly.

Firefox's both strength and weakness is in its extensions, as while they are good to have, if Firefox gets an upgrade, you don't know that the extensions will keep up.  The advantage, though, of Firefox's extensions is that there are so well organized, while for other browsers it's all over the place.  Firefox was the only browser I could get built-in XHTML validation.

Opera's built-in functions, particularly shortcuts, are incredibly in-depth.  Additionally, it is the only browser for which I could get an Open With option to open a page with another browser by right-clicking (K-Meleon only has the option for Internet Explorer, and I can't get any others on there).  Opera continues to feel "almost there".  Feature-rich, quick and good with resources, but just minor annoyances spread around.

Safari gets its foot in the door with top-notch standards compliancy and lightning fast JavaScript execution and speedy HTML parsing.  Its development tools, and generally just the touches it adds to page, visually look great, and the tools very functional.  After that, though, things quickly go downhill.  The features just aren't there, and neither is the customizability.

Internet Explorer 8 finally gives some kind of an effort in terms of JavaScript error reporting, and the Firebug clone is nice.  The best thing to developers in IE8 is IE7 emulation, and in Developer Tools, Quirks Mode (I believe IE5.5 emulation, either that or IE6).  The world still uses Internet Explorer heavily, so there are add-ons also, but far more dubious in terms of malware than anywhere else.  In terms of browsing speed and features, IE continues to fall flat on its face.

Verdict: Inconclusive.
Opera's improved compatibility and Firefox's improved performance combined with Tab Mix Plus both give a run at K-Meleon.  More time is needed to settle on what I'll use most, but it seems hard to go wrong between them.  Safari in performance delivers, but little elsewhere.  Internet Explorer 8 does nothing to improve its standing, and by standing still, loses ground to the ever-improving competition.

판올림 한 두번 정도 더 한 다음에 정식으로 3.0으로 런칭되지 않을까 싶네..

ps. 아직까지 acid3 test 점수는 71점 :)

I thought I give Firefox 3 Beta 5 a try. Keep in mind, Firefox 3 is still in beta.

[UPDATE][4/29/2008 at 21:26]

So, I was a bit disappointed after Mozilla Firefox 3 Beta 5 received a 71 on the Acid 3 test. I thought I go and download the latest Opera (9.50 Beta 2 Build 1933) for Linux.

Speed, Security, & Performance is being advertised on the Home page of Opera. I clicked download & few seconds later, I installed it onto my IBM T43 running Ubuntu 8.04. First thing I did was to fire up the Acid 3 test, thinking Opera should receive a perfect score 100/100. but to my surprise it only received 78/100. Performance wise, I do see a faster response when I try visiting a few websites that I frequently visit.

Congratulazioni agli sviluppatori del port GTK+ di WebKit: il tanto chiacchierato motore di rendering è infatti il primo engine libero a superare completamente il test Acid3 su piattaforma Linux. Questo risultato non porta vantaggi immediati all’utente finale ( visto che i CSS3 non sono ancora ufficiali ) ma testimonia l’impegno degli sviluppatori nella creazione di un motore di rendering rispettoso degli standard.

Al momento solo Epiphany ha annunciato la sostituzione di Gecko ( il motore nativo di Firefox ) con WebKit.

Il test Acid3 analizza la compatibilità dei motori di rendering con un’ampia gamma di tecnologie web come SVG, CCS, DOM Level 2, XML ed alcune funzioni JavaScript. Giusto per fare un po’ di confronti, una recente beta di Internet Explorer 8 ha superato solo 18 dei 100 test, le nightly build di Firefox 3 71 test ed una versione di sviluppo di KHTML 73 test. Opera ha superato i 100 test su tutte le piattaforme supportate.

[via: arstechnica.com || ossblog.it]

Hace mucho que no me pasaba por aquí, debido a falta de tiempo y otras cosas que verán dentro de poco, pero como ya hable sobre la versión anterior no podía dejarlo pasar. Hay varios aspectos que quiero recalcar sobre esta nueva versión, y que tampoco he podido de la anterior, por lo cual lo haré todo en este. Aquí va mi análisis y mi opinión sobre Firefox 3 BETA 5.

En primer lugar, veamos las nuevas mejoras que trae esta versión:

Mejor integración con sistemas operativos: ahora Firefox utiliza los temas, iconos y estilos de los sistemas operativos Microsoft Windows, MacOS y GNU/Linux GTK.

Organizador de lugares: ahora puedes visualizar, organizar y buscar en todos tus marcadores, tags e historial de búsqueda a través de múltiples vistas y carpetas inteligentes para almacenar tus búsquedas frecuentes. También puedes crear y restaurar copias de seguridad cuando tu quieras.

Velocidad: mejoras en el motor JavaScript así como las dirigidas a las optimizaciones de perfil siguen mejorando la perfonmance. Comparado con Firefox 2, las aplicaciones web como Google Mail y Zoho Office funcionan el doble de rápido en Firefox 3 Beta 5, y el popular SunSpider test de Apple mustra mejoras con respecto a versiones anteriores.

Para ver todas las notas del release originales vea la página oficial en inglés.

Una vez expuesta las mejoras expuestas según Mozilla, que si bien son importantes no siempre son las más correctas, vallamos un poco más lejos y veamos que podemos ver en esta versión, y a ampliar un poco las mejoras que tiene esta nueva versión.

Si me preguntan a mí si noto un incremento de velocidad, la verdad es que sí, sobretodo como dice en las release notes en aplicaciones web, o al ver videos flash, por ejemplo. GMail funciona com una fluidez jamás vista, y los servicios de video como YouTube y MegaVideo parecen funcionar mejor.

Aunque como dije anteriormente, si bien esta versión es más veloz, tiene un alto consumo de memoria RAM, que para algunos equipos es una contra muy grande, y para otros realmente no es muy importante.

Ahora, ¿qué tan elevado es este consumo con respecto a otros navegadores?

Pues, voy a intentar ser lo más gráfico posible:

Para el que no tiene ganas de contar, le resumo que en esa imagen hay 25 ventantas abiertas de Firefox y otras 25 de Opera.

Ahora veamos los consumos de recursos de cada uno de los procesos:

Firefox

Opera

No hay que saber mucho inglés para entender lo que hay en las imágenes. Definitivamente Firefox consume tres veces más memoria que Opera, y consume más recursos del microprocesador. Cabe destacar que cada ventana de Internet Explorer consume aproximadamente 25 MB, por lo cual si hacen el cálculo sabrán que 25 ventanas consumirían unos 625 MB, la verdad no hay palabras para esta cifra.

También quiero dejar claro, que el consumo de memoria no es proporcional a la velocidad de la aplicación, es decir que consuma más memoria no quiere decir que sea más lento. Por eso ahora viene la segunda pregunta.

¿Es más rápido Firefox 3 que otros navegadores?

Realmente es una pregunta la cual respuesta es difícil de comprobar, aunque tenemos varios tests para verificarlo. Entre los más destacados, podemos mencionar el SunSpider JavaScript Benchmark, el cual nos permite testear el rendimiento (tiempo de proceso) de JavaScript en los navegadores. Esta es la comparativa con respecto a los resultados de cada navegador:

Como ven, Firefox se muestra como el ganador absoluto, y se muestra una leve mejoría entra la BETA 4 y la BETA 5, seguidos muy de cerca por Safari.  Opera e Internet Explorer duplican la velocidad de procesado.

Otro punto muy importante en los navegadores, es si cumplen o no con los estándares, algo muy importante si quieres un navegador que te permita visualizar todas las páginas y correr todas las aplicaciones correctamente. Para verificar esto, existe el famoso test Acid 3, de la página Web Standards Project. Aquí tenemos la comparativa de este test:

Si bien Safari se lleva el trofeo, es seguido por muy poco por Firefox, superando notablemente a Opera.

Esto confirma lo antedicho, que el consumo sea más alto no necesariamente significa que la velocidad sea menor.

¿Qué tan seguro es Firefox?

La pregunta que sí tiene respuesta, Firefox es el navegador más seguro. Firefox 3 incluye una larga lista de sitios maliciosos en su lista negra, por lo que si intentas acceder a ellos te encontrarás con una página como esta, además de ser el navegador del cual se resuelven más cantidad bugs en menos cantidad de tiempo, lo que para mí es lo más importante. Si quieres un navegador seguro, Firefox es la mejor elección sin duda alguna.

Conclusiones

Firefox consume mucha memoria, eso no se puede negar, aunque es un precio que hay que pagar por una navegador tan completo, con tantos plugins, con una seguridad excelente, y una perfonmance insuperable. Todavía falta para la liberación de la versión final, aunque las espectativas son grandes. Si tu equipo te lo permite, te recomiendo este navegador.

Se espera que la próxima versión sea la final, por lo cual intentaré analizarla nuevamente, y quizás haga una comparación con Opera, el cual para mí es otro de los mejores navegadores.

Agradezco cualquier crítica y comentario, como siempre.

Descarga Firefox 3 BETA 5 - Mozilla.com

Salu2

WebKit is available for Mac OS X And Windows

http://webkit.org/

Forse per la maggior parte delle persone questo screen non dirà assolutamente nulla...ebbene, significa che il browser purtroppo ancora poco diffuso, che prende il nome di Opera, aderisce al 100% agli standard imposti dal consorzio W3C.

Per chi non lo sapesse l'Acid3 è un insieme di test con cui è possibile verificare l'aderenza di un browser agli standard web.
L'obiettivo principale dell'Acid3 è spostato sulle nuove tecnologie Web2.0 come ECMAScript e DOM Level 2. Qualche test si occupa anche di Scalable Vector Graphics, XML e data: URIs.

Potete effettuare da voi il test Acid3 da questo link

Attualmente, nessun browser passa completamente il test, aggirandosi di media attorno al 70%, quindi potete ben capire come questo traguardo sia molto importante per il team di Opera, essendo i primi ad averlo raggiunto.

Per ora il team di sviluppo ha annunciato la notizia solamente proponendo uno screen, perchè ci sono ancora qualche piccole modifiche da fare al codice, e sperano di poter rilasciare una technical preview su labs.opera.com entro la prossima settimana circa.

Se volete provare l'ultima release beta di Opera 9.5 potete scaricarla da questo link (è una versione beta, quindi ancora molto instabile)

Fonte della news

[AGGIORNAMENTO]

Nel giro di poche ore, anche Webkit (Safari) ha raggiunto il 100% , tuttavia però rilasciando anche una build al pubblico.

Con tutte queste corse, sembra di essere davvero in un GranPremio, chi sarà il vincitore secondo voi, Internet Explorer, Firefox 3, Opera 9.5 oppure Safari?

Que coisa não?

Yksikään nykyisistä Internet-selaimista ei noudata www-julkaisussa käytettäviä standerdeja (standardeihin voit tutustua W3C:n sivustolla)riittävästi läpäistäkseen uuden ACID-3 -testin. Tiettävästi tällä hetkellä yleisimmistä selaimista parhaiten ACID-3-testistä on selviytynyt Opera Opera 9.50 beta build 9815 versiollaan. Nykyiset uusimmat viralliset versiot Firefoxista tai Internet Explorerista eivät läpäisseet edes vanhempaa ACID-2 -testiä, mutta Operan uusin versio (9.26 build 8835) läpäisee. Kannattaa siis käyttää Operaa, jos haluatte mahdollisimman laaja-alaista toimivuutta, mutta Firefoxia, jos haluatte selaimen lähes äärimmäistä mukailtavuutta. Asiasta kertoi suomeksi Digitoday ja englanniksi DailyTech.

Standardit eivät ole vain selainvalmistajien ongelma, sillä jokaisen kotisivujaan väsäävän tulee huolehtia, että omat sivunsa täyttävät www-julkaisun standardit. Sivujensa standardien mukaisuutta voi käydä testaamassa W3C:n validaattoreilla ilmaiseksi. Varsinaisten standardien lisäksi kannattaisi sivujen suunnittelijan ajatella myös erityisryhmiä ja heidän laitteidensa tarpeita, joten erilaiset näytöt ja mahdolliset puhesyntetisaattorit tulee myös huomioida värisokeita ihmisiä unohtamatta heitäkään. Ja sitten, kun sivut ovat standardien mukaisia ja niitä katsellaan standardien mukaisilla selaimilla, on kaikilla kivaa.

Valitettavasti kaikki yleisetkään palvelut eivät halua noudattaa standardeja ja siksi esimerkiksi tämä minun wordpressissä sijaitseva blogini ei ole standardien mukaista materiaalia.