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

Chrome Release Cycle

docs.google.com

Post reply on HN
Print view View original

11–20 of 30 posts

Re: Chrome Release Cycle

#11

Post by ghshephard »

d_r wrote: post #5
angrycoder wrote: post #2

I've used chrome for about a year now. I've gotten so used to the seamless and silent updates that every other piece of software seems like nagware. I'm sure I'd be singing a different tune if it broke my browser during a silent update.

Lack of silent updates was one of the key reasons I switched from Firefox to Chrome (speed was another.) Seemed like every time I opened the browser to view a webpage I needed to see that very moment , Firefox would pop up an annoying update prompt that I'd keep clicking out of each time.

+1 - Interesting - that's actually why I've switched over to Chrome as well. Basically, every time I started up Firefox on my System at work on XP, Firefox would Freeze with a "Hey, we're going to upgrade you now - why don't you just go grab some coffee!" - So, slowly but surely, my goto browser started to become Chrome. The downside of this, is that, over time, I started firefox even less frequently, and, as a result, it was _always_ waiting for an upgrade. making it even worse.

Chrome just starts for me. I'm happy with it.

Re: Chrome Release Cycle

#12

Post by boredguy8 »

MrJagil wrote: post #4

One 'problem' with this modus operandi is that some companies use the big releases as selling points. It is interesting that the Chrome team does not feel the pressure to do this. Especially considering how heated the browser war is these days. Everyone seems to think Firefox 4 is going to be the salvation, for example. On the other hand, the silent updates are a blessing, and good features are another way of attract…

As the presentation explains: market features, not versions.

Re: Chrome Release Cycle

#13

Post by ximeng »

d_r wrote: post #5
angrycoder wrote: post #2

I've used chrome for about a year now. I've gotten so used to the seamless and silent updates that every other piece of software seems like nagware. I'm sure I'd be singing a different tune if it broke my browser during a silent update.

Lack of silent updates was one of the key reasons I switched from Firefox to Chrome (speed was another.) Seemed like every time I opened the browser to view a webpage I needed to see that very moment , Firefox would pop up an annoying update prompt that I'd keep clicking out of each time.

Hah! At the moment I read "at that very moment", I got a popup from Chrome's PDF reader in the background tab I loaded the article in:

"Some slides have failed to load and may appear blank. Click OK to try reloading the presentation, or cancel to continue viewing."

So Chrome hasn't completely ridded itself of modal popups just yet, though it's happens infrequently enough that I was pretty surprised.

Re: Chrome Release Cycle

#14

Post by ximeng »

I like the idea of hiding features behind flags. You can see evidence of this from the number of command line options in the source:

http://src.chromium.org/svn/trunk/src/chrome/common/chrome_s...

They do this even for fundamental features of the browser. For example, there's four different process models available:

http://dev.chromium.org/developers/design-documents/process-...

For planning software releases, I take inspiration from Chrome, Ubuntu, and Microsoft. They all have different release cycles, it's interesting to compare them.

Chrome: agile, 12 weekly releases, subtle upgrades, parallel versions with different stability levels.

Ubuntu: fixed six monthly schedule, well advertised schedule (e.g. http://www.youtube.com/watch?v=107E4RXNxYc) means difficult for them not to release on time, it's a selling point. Every 24 months long term support release.

Microsoft: major products released approximately every three years. Heavy marketing efforts around releases rather than features. Service pack releases and patches released on an ongoing basis.

I guess Chrome has less of an enterprise install base, which means they are more able to release on an ongoing basis. Enterprise IT teams like to install known good releases and not let other the vendors manage the upgrades.

Re: Chrome Release Cycle

#15

Post by lallysingh »

d_r wrote: post #5
angrycoder wrote: post #2

I've used chrome for about a year now. I've gotten so used to the seamless and silent updates that every other piece of software seems like nagware. I'm sure I'd be singing a different tune if it broke my browser during a silent update.

Lack of silent updates was one of the key reasons I switched from Firefox to Chrome (speed was another.) Seemed like every time I opened the browser to view a webpage I needed to see that very moment , Firefox would pop up an annoying update prompt that I'd keep clicking out of each time.

I also appreciate that when I close a tab, its memory is returned to the OS. My safari process tends to need killing once in a while, after it's been mean to my VM.

Re: Chrome Release Cycle

#16

Post by miniatureape »

"Market features, not versions," but to be sure this is Chrome's most important feature. It also mirrors what's happening in the development of HTML. http://www.mail-archive.com/whatwg@lists.whatwg.org/msg23160...

If IE had been engineered in the same way, we never would have gotten into the IE 6 trap and when all browser manufacturers adopt this strategy, web technology will move much more quickly.

Re: Chrome Release Cycle

#17

Post by IChrisI »

lawfulfalafel wrote: post #7
celticjames wrote: post #3

One thing that helps Chrome release often is that the api for building extensions is all javascript. By comparison, Firefox extensions can have binary components. This means there needs to be a certain amount of platform stability or extensions will break every six weeks. I think that in the future we'll see faster release cycles for Firefox as the api for extension building becomes more js based and less dependent o…

I don't know about that. I would love to have an extension that made the omnibar a lot more like firefox's awesomebar (good indexing, can search for key phrases for title of article and the such) but to my knowledge that's pretty much impossible because of the javascript limitation (I hope I'm wrong here, not really sure). I just wish google was a bit more subtle about wishing people search on google again when looki…

Funny, I wish Firefox's awesome bar let me press enter to go to the first result without pressing down first.

Re: Chrome Release Cycle

#18

Post by hokkos »

celticjames wrote: post #3

One thing that helps Chrome release often is that the api for building extensions is all javascript. By comparison, Firefox extensions can have binary components. This means there needs to be a certain amount of platform stability or extensions will break every six weeks. I think that in the future we'll see faster release cycles for Firefox as the api for extension building becomes more js based and less dependent o…

Chrome extensions can have binary componants with NPAPI : http://code.google.com/chrome/extensions/npapi.html

Re: Chrome Release Cycle

#19

Post by cookiecaper »

dman wrote: post #8
w1ntermute wrote: post #6

Earlier quoted context omitted.

The funny thing is that from a Linux user's perspective, Google was simply providing a workaround for what should have been the OS's job. On Linux, Firefox's auto-updates are disabled so that the OS can handle that instead. This is probably the best solution, as it gives you the most control over updates, and applies to all software system-wide. You're not forced to install updates at any time, and you can disable th…

The update popup for extensions is still annoying though - even on linux.

Many distributions' version of Firefox comes with that disabled out of the box.

Re: Chrome Release Cycle

#20

Post by cookiecaper »

celticjames wrote: post #3

One thing that helps Chrome release often is that the api for building extensions is all javascript. By comparison, Firefox extensions can have binary components. This means there needs to be a certain amount of platform stability or extensions will break every six weeks. I think that in the future we'll see faster release cycles for Firefox as the api for extension building becomes more js based and less dependent o…

Almost every extension I've tried has worked fine on Firefox 4, and certainly between the 3.x releases. While you might be right that this should be considered occasionally, it doesn't seem to be a big deal.
Post reply on HN