Spotting, swatting and not-ing
December 16, 2002
Feedback Farm
Have something to say about this article? Let us know below and your post might be the Post of the Month! Please read our Official Rules and Sponsor List.
Forums
Want to dig even deeper? Post to the new MacEdition Forums!
Recently Peter-Paul Koch wrote an excellent article for Apple on CSS bugs in IE5 for the Mac. While not wishing to toot my own horn, he noted that the most comprehensive bug guide around for that browser is the MacEdition Guide that I maintain – but I’ll be the first to admit that his explanations of the bugs he did cover were clearer and more comprehensive than mine. He also covered the fixes and workarounds to these bugs more thoroughly.
Bug charts and explanations are essential to Web development, which is still plagued by buggy and incomplete browsers manifesting brain-dead implementations of the specs. As Web authors become a bit more confident in their use of CSS and a bit more ambitious, they come up with things that haven’t really been tried before – and they find new bugs. The MacEdition bug guides and guides to CSS support in various browsers are always evolving as new information comes in. If you check out the MacEdition Guide to CSS Bugs in IE5/Mac now, you’ll see a bunch of new bugs that have only recently been identified and reported to me. The “huge text bug” is a gem.
Unfortunately, the focus on browser bugs has some negative consequences. We all know that browsers are buggy, so when we see something that doesn’t accord with what we expect, we blame the browser, especially when every browser does it differently. I’ve previously pointed out that sometimes it’s the fault of the designer, what with widths that don’t add up, browser sniffers that choke, or just plain sloppy markup (unfortunately my Brain Upgrade Initiative never really got off the ground, but there is still time.) Still, when only one browser is behaving differently from all the others, it’s very tempting to blame the odd browser out, rather than check the specs.
It may surprise many to know that when the odd browser out is IE5/Mac, it’s just as likely that it’s IE5/Mac that is doing the strictly correct thing, and all the other browsers are being overly accommodating of (technically) incorrect markup. Certainly I have received “bug reports” from well-meaning readers for the MacEdition Guide to CSS Bugs in IE5/Mac that turn out not to be bugs at all, just stricter applications of the rules than everybody is used to.
So what I’ve done is compiled them into a mini-guide as an adjunct to the bug guide – a “not-bug” guide, if you like. Now, obviously, it’s much harder to catch and classify a “not-bug”. Something is a “not-bug” if the browser in question is behaving correctly according to the specs, but all the other browsers are more forgiving, so people are used to the incorrect syntax working. Basically, it’s a “not-bug” if somebody thought it was a bug, but it’s actually correct behavior; but this also depends on individual interpretations of correct behavior. For now, there’s not many of these puppies; but I will be keeping an eye on the usual suspect discussion lists for people noticing singular behavior in particular browsers that happens to be correct. Our readers should still feel free to give me a heads-up about such things, even if you know I subscribe to a particular email list. Being on the other side of the world from most of you, most of the traffic on these lists happens when I’m asleep, so I subscribe to digests and sometimes miss things.
I’ve learned a lot from these not-bug reports. I never realized
that coordinates in image maps should always be zero or positive. We
got used to the incorrect interpretation because so many browsers
incorrectly treat negative coordinates as relative to the previous
point instead of relative to the top left corner of the map. Likewise,
it was only a notion in the back of my mind that if XHTML has
case-sensitive elements, stylesheets for XHTML documents should use
case-sensitive selectors (e.g. H1 in the stylesheet
won’t match <h1> in the document). IE5/Mac
does the strictly correct thing in these instances, but it’s easy
to forget what the correct thing is.
Why single out IE5/Mac? Surely there are other browsers that do the right thing, even though it looks like the wrong thing to authors who are used to whatever IE for Windows decides to throw up. The answer to this is simple. I’m a bit of an egomaniac (really!) and I like to check out MacEdition’s referrer logs, so when someone links to the MacEdition Guide to CSS Bugs in IE5/Mac to show how bad and buggy this browser supposedly is – often in contexts completely unrelated to CSS – I see those links. It annoys me that the many hours of hard work I and many other web developers have put in trying to isolate these bugs are being used to dismiss a browser undeservedly.
I’ve seen personal sites and discussion boards where people say, “oh IE5/Mac is buggy and here’s a link that says so” with no context and no comparison of the equally huge list of bugs in most any other browser. Some of IE’s bugs are egregious, but many of them are tricky, obscure things in areas of the specifications that few designers normally venture into. Sure, I’d like all browsers to implement floats the same way, but the bottom line is that the vast majority of sites will never find these bugs even if they use advanced CSS for laying out their pages.
No, it’s not the Windows IE that so many are used to (and there are still some things IE5/Mac does more correctly than IE6/Windows). And it doesn’t have the CSS Level 2 support that Mozilla and its relatives have. It’s not even my main browser for surfing anymore (yes, I switched to Chimera, too). But its support for CSS Level 1 is excellent, as is its support for many other important Web standards. IE5/Mac might be getting a bit long in the tooth, having been released back in 2000. It might be from Microsoft. It might be time for a new major version. But even with the extraordinary (and welcome!) march of OS X users to Chimera (more on that in a future column), if MacEdition’s logs are to be believed, IE5 remains the predominant browser of Mac users, and it deserves to be treated as an equal Web citizen. It has its bugs, but it’s pretty good overall. And I’d much rather you use it than Netscape 4 or OmniWeb.