On Jun 19, 10:22 am, smorgan <[EMAIL PROTECTED]> wrote: > Gerv wrote: > > Sorry, but that's rubbish. If all we have are 29 sites which are broken, > > then the fact is that the vast majority of web developers _are_. > > We have a lot more than that. I've been answering reading Camino > feedback, forums, etc. for a very long time. I always tell users to > complain to the sites' admins, but I don't file TE bugs on every > single on of those sites, because I don't have the time to spend > filing bugs that will either sit and rot with no action taken, or in > some cases be closed as (essentially) "we don't care about sites that > are sniffing".
Just to reiterate what Stuart said here, bug 334967 is just the tip of the iceberg. Its deps are either TE bugs that were originally filed against Camino and which I added to the tracking bug when I learned of it, other TE bugs I've happened upon randomly when searching for something someone's reported and which matched the issue, and a few bugs added by other (non-Camino) triagers who happen to know about the tracking bug. To my knowledge, no-one's made any concerted effort to go through open TE bugs and pull out all the sniffing-related ones that have been filed since TE was opened. Moreover, every site that sniffs and gets filed in Bugzilla doesn't get kicked to TE; rather than flood TE with so many bugs that the major sites get lost in the noise (or, lost more than than they are now in the morass that is TE), many just get closed INVALID. By contrast, I think we see about 1 site per week in the Camino forum (and I don't read the feedback list, but I'd guess it's at least that there, too). As Stuart notes, we always request people write to webmasters and point them at http://www.geckoisgecko.org (started by a Camino user, btw). I'm delighted to see interest in actually doing something about the problem of sniffing, but it's going to need some serious commitment from those who care about Gecko and have resources and clout to leverage. For example, I would have loved to see outrage (and education efforts) from the organization that shepherds Gecko when http://developer.yahoo.com/yui/articles/gbs/ first came out in early 2006; it was a squandered opportunity to push "rendering engines" instead of "browsers" (that page illustrates everything that is wrong with people who *think* they're actually performing best-practice detection these days, and if you send different code to different browsers [based on rendering engines], what code do you send to X-grade browsers? Gecko code? WebKit code? Random generic code?). Instead, we're left with a situation where Yahoo properties work correctly with Firefox but not other Gecko browsers, and each different property has to be evangelized separately (and, really, for each browser on each property, since they tend to ignore "sniff for rendering engine" and just add each new complaining browser, when they deign to add at all). As I recall, the BBC announced a very similar policy at the same time. We can complain, but since we're not Firefox (or Safari, or IE), not A-Grade browsers, it doesn't really matter. And, as I mentioned in bug 334967 when I introduced http://wiki.mozilla.org/User:Sardisson/Gecko_is_Gecko , the existing "when and how to sniff" documentation in http://developer.mozilla.org/en/docs/Browser_Detection_and_Cross_Browser_Support is in need of a serious upgrade. Too much time and effort is spent on differentiating Mozilla from Netscape 4, and the amount of time the document spends on describing sniffing for different UA string components seems to run counter to the goals of no sniffing/feature sniffing/rendering engine and engine version sniffing trio (which aren't, in my opinion, adequately stressed in that document). As I said in the bug, I'm not the right person from a technical perspective to write or revise the doc, but I'm happy to help, and rather than simply complaining, I did draft http://wiki.mozilla.org/User:Sardisson/Gecko_is_Gecko as an idea of what I think would be better (also, thanks to Boris for the edits he made). Also, once the existing sniffing document is fixed (or replaced), it (or its successor) needs its visibility raised; you can only find that page in DevMo if you're specifically looking for it, which seems less- than-helpful if we want to promote "no sniffing" as our first ideal. But all of this takes time, both in person-hours and, worse, in lifespans of old ideas. In the absence of any meaningful efforts on the sniffing-education front now (or, really, in the last four years) by those with leverage, we need to do what's best for Camino users right now. That we have to, and what we need to do to do that, sucks. My hope is that it is only a temporary hack.... Smokey _______________________________________________ dev-tech-layout mailing list [email protected] https://lists.mozilla.org/listinfo/dev-tech-layout
