Originally posted by lysbleu
View Post
Announcement
Collapse
No announcement yet.
Firefox 29.0 For Linux Is Now Available For Download
Collapse
X
-
Originally posted by Cyber Killer View PostOf course - because all interface changes after a piece of software is in production should be optional. Change whatever you want, but the default interface for the user should stay the same unless that user chooses to enable the new look.
1 - bloat software with plenty of UI options that would become impossible to maintain as time goes by
2 - never change anything concerning UI in a particular software, in doing so we would have the exact same UI as in Netscape Navigator 1.0 that was sooooooo superior, running on top of a GNOME 1.0 inspired interface.
3 - fork a piece of software every time we want to make a change in the UI, in order to have the older version for narrow minded people and the new one for others
4 - Stop developing software after version 1.0, apart from bug-fixing. Oh wait, a bug in a 1.0 software is unacceptable. Never mind.
Comment
-
there are bigger usability issues
Originally posted by Cyber Killer View PostYeah, it would be better if it used Qt (especially on my KDE desktop)...
- Everytime I turn on my computer, Firefox greets me with "This is embarrasing", asking if I want to restore the last open tabs. Perhaps not a usability problem for the average Phoronix reader, but try to explain this to your parents... I can tell you that the issue did no good to my father's already battered confidence in Linux.
And when Firefox has learnt to exit cleanly upon KDE logout:
- Stop fsyncing that sqlite db every mouse click! I'm totally okay with losing a few seconds of history (specifically, /proc/sys/vm/dirty_writeback_centisecs) in the unlikely event of power loss. Logging out of KDE is so drastically more likely, please optimize for that instead!
Originally posted by Adarion View Post>And disable not just Java but Flash by default if they are going to switch off plugins by default at all.
That is, on x86, I would -- just try to install Adobe Flash on ARM... Firefox needs to be more respectful of people who choose NOT to install Flash too!
Comment
-
Originally posted by omer666 View PostSo you think that we should either:
1 - bloat software with plenty of UI options that would become impossible to maintain as time goes by
2 - never change anything concerning UI in a particular software, in doing so we would have the exact same UI as in Netscape Navigator 1.0 that was sooooooo superior, running on top of a GNOME 1.0 inspired interface.
3 - fork a piece of software every time we want to make a change in the UI, in order to have the older version for narrow minded people and the new one for others
4 - Stop developing software after version 1.0, apart from bug-fixing. Oh wait, a bug in a 1.0 software is unacceptable. Never mind.
The way this stuff is handled nowadays (and I'm not talking only about Firefox, but generally) - stuffing new stuff down users throats whether they like them or not is definitely unacceptable.
Comment
-
Originally posted by omer666 View Postin order to have the older version for narrow minded people
Comment
-
Originally posted by omer666 View Post3 - fork a piece of software every time we want to make a change in the UI, in order to have the older version for narrow minded people and the new one for others
Originally posted by Cyber Killer View PostThe way this stuff is handled nowadays (and I'm not talking only about Firefox, but generally) - stuffing new stuff down users throats whether they like them or not is definitely unacceptable.Code:echo ">=www-client/firefox-29.0" >> /etc/portage/package.mask
What bothers me about Mozilla and now Firefox is the marketing. It smells funny now that lucid people would have to admit once again that the power of the web isn't *really* in our hands, it's in the hands of the developers whose code we run. It should be obvious but I'll say it again: the only truly 100% customizable, configurable, optimized, obedient, sexy, secure, powerful, standard-conforming, kick-ass {browser|editor|reader|converter|synthesizer|engin e|library|tool} is a halfway usable text editor, a C compiler, and a CPU. Oh, also you, your time, your attention span, and your power supply. But even then, the "obedient" part can still be compromised-- just read Reflections on Trusting Trust by Ken Thompson.
Comment
-
Originally posted by Vim_User View PostOf course, because not wanting to have your workflow broken makes you narrow minded. Sounds the same as "You don't like Gnome, you are unable to handle change!" Maybe your arguments would sound better if you wouldn't just insult people for not liking the same things as you.
But I've been using it for a while now, since I run the nightly builds.
After a while, you get used to it.
It's not so bad once you get used to it. It's pretty much same as before, not much has changed.
Comment
-
Originally posted by Vim_User View PostOf course, because not wanting to have your workflow broken makes you narrow minded. Sounds the same as "You don't like Gnome, you are unable to handle change!" Maybe your arguments would sound better if you wouldn't just insult people for not liking the same things as you.
Comment
-
Originally posted by intellivision View PostWhat you lose is freedom and performance.
Firefox's new Javascript interpreter outperforms Chromium, it also uses less memory and CPU per tab than Chromium does too.
ASM.js is a suitable replacement for NaCl too because, as its not running native code, there's less to exploit with malicious scripts.
It's not more free than Chromium in any sense. Both are open-source projects, although Firefox's artwork / branding is non-free so Debian ships it as Iceweasel while it ships Chromium unaltered.
PNaCl (the old NaCl isn't enabled by default) is not any less secure than a JavaScript virtual machine. In fact, since it relies on very low-level guarantees to provide security rather than the correctness of an entire JavaScript virtual machine, I would consider it more secure if anything. Anyway, it's a joke to be talking about Firefox in a security context since it runs everything (image decoding, video decoding, audio decoding, JavaScript, CSS/HTML, DOM, WebRTC, networking, [...]) entirely within the same fully trusted process. A single bug in a video decoder is enough to get full control of all your browser data and anything your user owns.
Unlike asm.js, PNaCl offers 64-bit integers, threads, shared memory and similar features expected from native code. It has negligible overhead (mostly due to the portability guarantees required, losing information about stuff like alignment) compared to the 50-100%+ hit for using asm.js since it's a proper compiler IR in SSA form rather than a hack. These features cannot be added to asm.js without destroying the entire argument behind it which is backwards compatibility.
I haven't seen any numbers showing either browser consumes less CPU or has faster JavaScript. I know Firefox causes more CPU wake ups than Chromium on my machine, but that's just anecdotal and may not be the same across all hardware / environments. Mozilla's own numbers have no clear winner.
Comment
Comment