Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Google Chrome, Google's Browser Project

blogoscoped.com

Post reply on HN
Print view View original

101–110 of 112 posts

Re: Google Chrome, Google's Browser Project

#101

Post by rcoder »

I'm curious why there aren't any platform partisans screaming "bloody murder" about Google's obvious flaunting of GUI tradition on this one. Putting tabs on top of the window titlebar may be the greatest UI innovation since tabs themselves, but it's going to make Chrome windows look and act differently from every other application on the user's computer.

That being said, a browser using WebKit + Gears + solic process sandbox model appeals to me a lot. Props to Google for explicitly referencing the Biba model, too -- if they can get web developers to start paying attention to the body of security theory published in the last 40 years, the browser itself could be a huge flop, and still leave us better off than we are currently.

Re: Google Chrome, Google's Browser Project

#102

Post by LogicHoleFlaw »

ars wrote: post #87
hooande wrote: post #86

Earlier quoted context omitted.

Other than the faster javascript, I don't see a lot of "game changing" innovations in this browser. These seem like mostly incremental improvements and small features. If the java engine is significantly faster this could really push things forward. Otherwise this is less than world beating.

You missed a massive one: multithreading. I've wanted this for ages with firefox. Each tab runs on it's own?! It's exactly what I've been wanting.

Just to be nitpicky... it sounds like they're using separate OS processes rather than threads within a single process. Hence, each tab would have its own entry in `top`...

edit: that'll teach me to not read the comic before commenting. It sounds like they're actually using lightweight processes.

Re: Google Chrome, Google's Browser Project

#103

Post by jmtulloss »

nailer wrote: post #81
kilowatt wrote: post #27

Have you guys SEEN the respective code bases for Mozilla and WebKit? Really, this was inevitable. Their whole tabs-as-process thing is too, when you think about it. Just imagine the potential for a browser-as-OS-with-processes-and-everything and how much they must be salivating to get it on devices and desktops NOT running Windows. It's a hell of a trojan horse in their winning battle to make Microsoft irrelevant...…

> Their whole tabs-as-process thing is too, when you think about it. Could you please explain this too? I like being able to identify and kill individual tab and plugin processes. But I don't understand why you can't garbage collect a thread like the presentation seems to say.

It's because there isn't a compacting garbage collector running. The core is C/C++, so it relies on the OS allocator to keep memory from becoming fragmented. The OS allocators are good, but they aren't that good. In a browser you're constantly allocating and freeing large amounts of memory, so you're going to end up with some holes. Its usage pattern is way different than most applications because it's used like Google finally designed it: as a window into a bunch of different processes.

Re: Google Chrome, Google's Browser Project

#104

Post by jmtulloss »

bOR_ wrote: post #94
mpk wrote: post #47

Earlier quoted context omitted.

You can bet they will. The V8 engine is unlikely to present much of an obstacle. We have a large JS codebase and the Opera, WebKit (Safari), Spidermonkey (Gecko/Firebird), Rhino (Spidermonkey port to Java) and JScript (IE6/7) engines are no hassle to support. The hard part is DOM interaction and styling. But as they're using WebKit and are putting a lot of effort into compatibility - this should be a breeze. The real…

I know the portion of IE6 on the web is still around 25%.. but do you know what portion of your customers is actually using IE6 still? I wonder at what point it is justified to return a 'please update your browser, preferrably to one of these: [list of reasonable browsers]', just because of the effort it takes you to keep IE6 compatibility.

At a small startup with a huge and complicated JS base, we've done just that. There's no way we can justify the man hours required to support IE6.

Re: Google Chrome, Google's Browser Project

#105

Post by jotto »

johnyzee wrote: post #79
krschultz wrote: post #26

I don't think this is an attack on Firefox over 50 million bucks. I think this is Google looking at the web and saying we need a faster, more stable, more rich, browser that has these 20 features to allow us to push our web apps forward. Their game is to make web apps equal to desktop apps so that they can sell advertising on them - and hammer Microsoft. This goes towards that goal. I'm looking forward to it. Firefox…

That seems to be what they are saying. However, Google does not have a good track record of delivering product for non-technical end users. GMail, their one succesful homegrown, mass-market, non-search product that I can think of, is a mess (haphazard user interface, no improvement for years, still beta). For such a strategic app, that is worrying. As for their platform vision, it also seems very scattershot. They ha…

GMail has been improved in ways other than UI changes.

They have rewritten the HTML, CSS and JS to make it more hackable for Greasemonkey - and in the process, optimized that HTML, CSS, and JS to reduce number of connections and subsequently load time.

The backend has seen massive scaling. They added POP and IMAP.

I don't know what I would change for GMail. It's a strong email client that gets the job done. I wouldn't consider it "worrying".

Re: Google Chrome, Google's Browser Project

#106

Post by yters »

bootload wrote: post #39
gaborcselle wrote: post #23

I love the comic book format of this - great idea!

"... I love the comic book format of this - great idea! ..." Thank Scott McCloud. McCloud is to the comic world what Stallman ~ http://www.scottmccloud.com/inventions/inventions.html (Creators bill of rights) and Linus is to the computing world by writing & thinking extensively how to innovate ~ http://www.amazon.com/exec/obidos/search-handle-url?_encodin... and push both the art and science of comics as hard as he c…

Thank you, those are indeed excellent comics. I think this one is especially good:

http://www.scottmccloud.com/comics/mi/mi-25/mi-25-1.html

Re: Google Chrome, Google's Browser Project

#107

Post by schtog »

I'm using Chrome right now and it is fine, a cleaned up Firefox, but nothing mindblowing.

What do I sign up for in terms of privacy? I didn't read through the whole deal(who does?) but saw some stuff about ads, does Google track my surfing?

I'm missing the top right Firefox searchbar where I can search wikipedia and other sites though.

Re: Google Chrome, Google's Browser Project

#108

Post by nailer »

jmtulloss wrote: post #103
nailer wrote: post #81

Earlier quoted context omitted.

> Their whole tabs-as-process thing is too, when you think about it. Could you please explain this too? I like being able to identify and kill individual tab and plugin processes. But I don't understand why you can't garbage collect a thread like the presentation seems to say.

It's because there isn't a compacting garbage collector running. The core is C/C++, so it relies on the OS allocator to keep memory from becoming fragmented. The OS allocators are good, but they aren't that good. In a browser you're constantly allocating and freeing large amounts of memory, so you're going to end up with some holes. Its usage pattern is way different than most applications because it's used like Goog…

Thanks.

Re: Google Chrome, Google's Browser Project

#109

Post by netcan »

nailer wrote: post #83
netcan wrote: post #65

Earlier quoted context omitted.

Where is the browser == eyeballs? BTW, I think that the big success of Google is in realising that's really true. Web Searches == money. But not inherently.

Owning the browser is a better way to target the eyeballs.

Are you talking about eyeballs for search?

Because that'd be included in the above estimate ($120m p/y for %100 of the market) assuming they're paying mozilla fairly for defaults.

Do you mean something else? Like eyeballs on the browser itself?

Re: Google Chrome, Google's Browser Project

#110

Post by nailer »

netcan wrote: post #109
nailer wrote: post #83

Earlier quoted context omitted.

Owning the browser is a better way to target the eyeballs.

Are you talking about eyeballs for search? Because that'd be included in the above estimate ($120m p/y for %100 of the market) assuming they're paying mozilla fairly for defaults. Do you mean something else? Like eyeballs on the browser itself?

I'm talking about targeting eyeballs for ads.

By owning the browser, Google (who a media sales company) know more about you, and can generate a better yield for their customers (ie, advertisers).

Post reply on HN