Google chrome – Global marketshare by version • Hacker News Board

Re: Google chrome – Global marketshare by version

#31

Post by jonathansizz » Tue, Jan 28, 2014, 2:48 PM UTC

thirsteh wrote: post #10

nivla wrote: post #6

Earlier quoted context omitted.

Although I do agree with you that silent updates are a far better choice, its not perfect, sometimes it can introduce unexpected changes. Let's take for example the latest Chrome update (v32). It replaces the OS based scrollbar with a custom one. It also removes the ^v arrows around it. As someone who regularly uses them, it breaks my workflow :( .Sad but I can live with it. Had it ended there it would have been acce…

> There is a reason why silent updates/auto-updates aren't popular in the server/enterprise world because it could sometimes break things. Thankfully we're seeing a shift where people will accept this if it means serious security issues are fixed promptly. I frankly couldn't care less if my scrollbar acts a little strange if it means two remote code execution holes were fixed.

Remote execution holes in the scrollbar code? How about if they fix series security issues promptly without simultaneously breaking basic UI elements?

Re: Google chrome – Global marketshare by version

#32

Post by thirsteh » Tue, Jan 28, 2014, 3:27 PM UTC

jonathansizz wrote: post #31

thirsteh wrote: post #10

Earlier quoted context omitted.

> There is a reason why silent updates/auto-updates aren't popular in the server/enterprise world because it could sometimes break things. Thankfully we're seeing a shift where people will accept this if it means serious security issues are fixed promptly. I frankly couldn't care less if my scrollbar acts a little strange if it means two remote code execution holes were fixed.

Remote execution holes in the scrollbar code? How about if they fix series security issues promptly without simultaneously breaking basic UI elements?

> Remote execution holes in the scrollbar code?

You make it sound like that's not fathomable, but an astonishing number of security issues are caused by insecure font and UI rendering.

> How about if they fix series security issues promptly without simultaneously breaking basic UI elements?

While I agree that it's great if things don't break, I'm not sure what is really broken. I'm running Chrome right now and the scrollbar doesn't bother me. (It seems more like somebody not liking a change than it being "broken.")

If you really dislike the change so much, you can just disable the auto-updater--but then you're on your own. Running an old browser is like sailing in a shark-infested ocean in a wheel of swiss cheese.

Either way, you're suggesting they fix a hole in all previous versions of the software, and release new, fixed versions with no other changes. This is not feasible for a browser.

Re: Google chrome – Global marketshare by version

#33

Post by frik » Wed, Jan 29, 2014, 8:13 PM UTC

scholia wrote: post #30

Daiz wrote: post #26

Earlier quoted context omitted.

Chrome 32 really is a terrible update with some major regressions all over the place, all (apparently) for the sake of some custom scrollbars. To add onto the list of issues, popup windows ( including things like Chrome dev tools in their own window ) constantly pop up a "Page is not responding" error, despite working just fine. Same for any Chrome apps that you might be running in their own windows. I switched to Ca…

I went back to Firefox 7 months ago. That was when it became better than Chrome: faster, more stable, consumes fewer resources, supports more tabs, better UI.

I still think Chrome peaked around version 20, I have an portable version of it around. With it I can open hundreds of tabs without any problems and it is perfect. Chrome 32 is so bad that I question the QA department/process of Google. Three days ago I Canary crashed on start (Win7) had to wait til next silent update, but now it's a lot better than 32.

Firefox 26 is great, the only missing thing is multi-process support - otherwise a single website can hang the whole browser... not so good if you open dozens of tabs and keep them open for days.

Re: Google chrome – Global marketshare by version

#34

Post by scholia » Thu, Jan 30, 2014, 12:33 PM UTC

frik wrote: post #33

scholia wrote: post #30

Earlier quoted context omitted.

I went back to Firefox 7 months ago. That was when it became better than Chrome: faster, more stable, consumes fewer resources, supports more tabs, better UI.

I still think Chrome peaked around version 20, I have an portable version of it around. With it I can open hundreds of tabs without any problems and it is perfect. Chrome 32 is so bad that I question the QA department/process of Google. Three days ago I Canary crashed on start (Win7) had to wait til next silent update, but now it's a lot better than 32. Firefox 26 is great, the only missing thing is multi-process sup…

I think it was the Chrome 26/27 era that finally drove me away....

Understood re Firefox 26 multi-process support, but it makes a big difference to resource consumption. I've not had any problems but if necessary you could kill the browser thread and restart excluding the bad tab....