2010-06-12
post/691580154
Jim Lynch, in Safari Reader: Apple’s Weapon of Mass Destruction (via, via)
This chap can’t have been around when Load Images was a browser setting. Or have read the CSS spec, and the assumption that users would be able to apply styles.
Of course, given he’s forced his rant over three pages (each of which are incredibly short and smothered with ads), and because I haven’t installed Safari 5 yet, I’ve only read the first third of his article, so perhaps he covers all of that. Somehow, though, I doubt it. And I’m not going to give him the satisfaction of page views to find out.
2009-12-08
A short list of Chrome issues (beta 1)
… most of which are actually due to me being stuck liking the way Safari does things, or a consequence of it being a beta.
- You can’t invert open behind - on Safari command shift click opens in a new window behind the current one with command click opening in a new window in front/focus, whereas Chrome is hard-wired to do the opposite (and with tabs to boot)
- Tab moves between all link and form elements, not just form elements. (I can’t remember if Safari picks this up from the system-level preferences or not, but it does what I want, and Chrome doesn’t.)
- There are no command key shortcuts to open bookmarks in the toolbar¹
- The combined text/title/URL completion in the “Omnibox” means you have to go too far into a URL to easily work around the lack of command key shortcuts
- You can’t set a default font size, so sites that honour relative fonts (like delicious, and (in places) Twitter) now have Huge Idiot Typefaces
- While Chrome sensibly uses the keychain (so, like Camino, it can share usernames and passwords set up in Safari) it doesn’t offer to complete the username part, meaning more typing.²
It’s been a nice evening, but despite being rock solid, Chrome goes back in the “for emergency use” box for a while. Sadly, given the team’s approach to customisation, it might be quite a long while indeed.
¹ Camino also gets this wrong, in a very odd (yet explicable) way: if the bookmark bar is visible, it works, but if it’s not, it doesn’t. The developers seem to believe that an invisible option is confusing. I believe that twenty vertical pixels on a laptop display are more precious than feedback. Sorry.
² Probably this is sensible from a security point of view but it’s annoying me, so I’m listing it anyway. I’m sure you’re sensible enough to make your own decision.
2009-07-27
Recovering From Safari Crashes
Every now and again, I’ll see a friend complain online that Safari has crashed and that they’ve lost a vitally important window. Usually I chip in that there is a session-recovery feature, even though it seems that hardly anyone knows about it.
If you look in the History menu, the last option in the third block is the one you want: Reopen All Windows From Last Session. This does what it says, restoring your state (although I’m not sure I’d swear that it’ll have saved the contents of that textarea containing that world-changing blog post.
Of course, Firefox handles this far better: the state isn’t just automatically saved, it’s automatically restored. There are ways to get this for Safari, like Glims, but they’re input manager hacks and therefore prone to breakage. (Glims also does a lot more than just session recovery, but that might be a downside for some users.) Maybe come Safari 5 someone at Apple will have been convinced to make this a feature, but I’m not holding my breath.
In the meantime, remember History > Reopen All Windows From Last Session. If you need to.
2008-09-19
JavaScript Engines and the IE Hegemony
Another day, another JavaScript performance increase. This time, it’s the WebKit team, with the somewhat ludicrously named SquirrelFish Extreme, which manages a tenfold speed increase on Safari 3’s JSKit engine, apparently.
Oddly, it seems as if most of the work involved has been done under the auspices of the Summer of Code, Google’s programme to get college students involved in open source projects. Of course, the other newcomer on the JS engine block is their own V8.
However, good as all the competition between these two (and Mozilla’s TraceMonkey, due to be released in Firefox 3.1 and, like the others, available in pre-release form) is, it ignores the elephant in the room. As I commented on a PC Pro blog post about Chrome and Firefox (after the V8 talk at Google Developer Day), this progress is pretty much passing IE by.
John Resig’s recent benchmarks of JavaScript engines show that Internet Explorer 7 is awful, and IE8, although better, is nowhere near the performance of the released competing browsers, let alone the improved versions currently under development. Indeed, after one chart he has to note:
No results for IE were provided as the browser crashes when running the tests, unfortunately.
(To be fair, he does go on to say that Safari nightlies also failed on Windows.)
So, what’s the outlook? Clearly JS-intensive applications are here to stay, and I think we’ll increasingly see interfaces to them on the server side and possibly the desktop too. However, the developers of the most cutting-edge of these apps will either be held back by, or have to explicitly exclude, the most common browser in the world. Maybe that’ll be enough to keep Microsoft focussed on improving, and from the notes to a recent interview, it looks like they know it. Here’s hoping it works out.
