Monday, June 23, 2003

Safari 1.0 was released for the Mac today, which marks the first General Announcement of an Apple-made web browser. Some visitors (thanks and thanks) had reported a major bug in the way Contact Sheet’s homepage was rendered in the previous Safari Beta — completely lacking a vertical scrollbar.

With the 1.0 announcement I thought I’d take the time to investigate the problem, seeing how my bug reports to Safari went unanswered (the bug still appears in 1.0). I looked at my stylesheets, then HTML, and found the problem with a bit of JavaScript, specifically a document.write method to detect the user’s screen size that caused the scrollbar to disappear — clearly a Safari bug. I altered the placement of the code slightly and the scrollbar now appears.

I didn’t hop on this problem when it was first brought to my attention because (a) Safari is a beta browser (b) since I now sport XHTML and CSS validity, it was likely to be a browser bug (see a) and (c) only 1% - 2% of the daily users use Safari — although this is likely to increase with the 1.0 release, and with my bug workaround in place.

Which begs the question, at what point should web designers care about site/browser compatibility? Is it purely about the percentage of users? Compliance with web standards? At most places I’ve worked it seems that folks use the sheer volumes when creating policy. But if “only” 1% of your users are viewing with Netscape, 1% of a million are a lot of people viewing a potentially imperfect site. With the browser market pie chart becoming more fragmented — a good thing, mind you — designers will have to consciously draw the line somewhere. And we already do with old versions.

So, what are your browser compliance criteria?


Comments


by Rob Foxx » Jun 25, 2003 5:14 AM

I concentrate on IE/O/Moz - then I take my chances unless I'm asked to specifically target a browser. Chances are if you cover IE/0/Moz your site should work well in any other browser worth having. Promised myself a long time ago that I wouldn't pander to the bleeding edge minorities, which Safari still is. Unless you target an audience full of die hard Macsturbators then I guess in the right here and now it's not essential to work on any "be nice to Safari" fixes.

Comments


by Keith » Jun 25, 2003 12:52 PM

Macsturbators!?! Hahaaaa. I resemble that remark. But hey it's looking great in Safari now....

Well with most of my projects, I try my best to take care of the latest versions IE, Opera, Mozilla and Safari on both Macs and PC and go from there. On my last project I couldn't get everything working in Opera for PC so I gave up. I don't worry about NS much any more.

At my day job, on our Intranet, I mainly concentrate on IE 5.5 and up, as that is our "Gold Standard" - but I do check things is other browsers (I work on a Mac) and if I have time may do some tweaking.

Comments


by ss » Jun 25, 2003 1:23 PM

I'm also interested in the criteria for choosing which browsers to support.

Yesterday Bob Sawyer alerted me to a major meltdown when viewing Contact Sheet in Mozilla 1.2.1 -- you couldn't click on anything in the contents area of the page. I should have things working again shortly. Mozilla 1.2.1 is not the latest release, but 1.3 is too buggy for his taste, and 1.4 hasn't been stabilized yet.

With many of the smaller browsers, they're releasing updates every month or two, creating far too many versions to easily test. If getting users to upgrade from NN 4.79 after two years, imagine getting people to upgrade from Firebird 0.5 to 0.6 after using it a month.

Comments


by Rob Foxx » Jun 26, 2003 7:26 AM

That's a valid point - as the minority browsers trample each other in their stampede to catch users falling off the ailing IE juggernaut it seems like the multi-threaded development/release strategy is more counter productive than anything.

Why don't those working on Firebird / Moz / etc pull down the shutters and get their collective heads working on a worthy milestone for release, rather than firefighting a deluge of bugs with every bleeding edge point.point release they drag screaming off of the production line?

I would definately strive to support a polished browser release first and foremost. Who gives a rats ass about pop-up killers and all the rest of it? Build a standards compliant browser that flies and then furnish it with gizmos.

Comments


by FriedFish » Oct 22, 2003 11:03 AM

I think you guys are missing the whole point of the mozilla project.. All these browsers (firebird,mozilla,netscape,galeon,etc) are using the same core rendering engine! That means that when one group comes across a bug and fixes it, they fix it for everyone else who bothers to update as well - each group is contributing to building that "polished browser". Yes, they each have their own little tweaks to the base code, but these are rarely more than gui changes and a few speed/efficiency hacks rising from what they chose to leave out.
As for coding your site FOR Safari - the whole point of standards compliance (which as far as I'm aware is something Safari is trying its darndest to keep up) is that you shouldn't have to code on a browser to browser basis. If you code according to standards, your page should pretty much look and work the same on ANY compliant browser.

Search

Syndication

RSS: .91 / 1.0 / 2.0