[00:33:58:] agenda and stuff at http://esw.w3.org/topic/QaDev [00:34:35:] niq/xover sent regrets... [00:35:02:] * karlcow sent regrets too [00:41:58:] ======================= [00:42:04:] Action Items Review [00:42:06:] ======================= [00:42:25:] please note here if there is any notable change in your AI status [00:42:46:] ACTION: Nick to get public Web account -- done according to niq [00:42:50:] * yod completed AI on jpg version of header/footer [00:43:07:] ACTION: Nick to finish installing mod_validator on qa-dev -- done, see [00:43:09:] http://www.w3.org/mid/Pine.LNX.4.53.0405302321180.3554@hugin.webthing.com [00:43:17:] * yod started discussing with Dom on regenerating (virtually) all "valid foo" icons [00:44:15:] I think I've done my ACTION: Bjoern to modify priorities in CssValidator's bugzilla [00:44:30:] I've also incorporated the old BUGS.html page into Bugzilla [00:45:26:] * scop will also do the png->jpg footer/header change in checklink [00:46:16:] I've made some progress on using libcroco for a replacement... but nothing interesting to report yet. [00:47:17:] ACTION: Olivier to harmonize checklink's layout with MarkUp Validator -- this is done, no? [00:48:49:] * yod notes checklink harmonization done according to temporary plan i.e hardcoded links to instance on vwo [00:49:34:] yod: yep, I noticed, will add the h/c'd .jpg to the URLs in CSS [00:49:36:] * yod sends pointer to dom [00:50:07:] I am opposed to the legal footer on http://qa-dev.w3.org/wlc/ -- it is very misleading. The W3C privacy policy explicitly claims that the site does not use cookies, checklink hower does. [00:50:08:] I did not hear back from W3C when raising this issue six weeks or so ago. [00:50:41:] (http://lists.w3.org/Archives/Member/w3c-archive/2004Apr/0032.html member-only) [00:51:22:] bjoern_h: there has been some internal followup [00:51:37:] need to amend policy to include bugzilla and checklink [00:52:17:] ACTION olivier push [modification of privacy policy wrt cookies in bugzilla and checklink] forward [00:52:43:] <__Yves> I would note that the p3p description of jigsaw.w3.org leave the door to cookies open for the css validator [00:53:31:] Though there is a general problem that the footer talks about interaction with the entire site which for me includes jigsaw.w3.org... [00:57:46:] [17:35] maybe we should use some kind of process to make sure we know about regrets in advance [00:57:46:] [17:35] so we can reschedule meetings and stuff [00:57:47:] [17:36] Maybe in the wiki... [00:57:47:] [17:36] A new heading === Regrets for next meeting === [00:57:47:] [17:36] and everyone not present adds his name [00:58:09:] adding it now * press corner * [00:53:39:] off-topic Note: quite a few mentions of markup val and checklink in press [00:53:52:] ... including local press, which is an interesting sign [00:53:54:] offline press? [00:54:07:] <__Yves> dead tree press [00:54:14:] both dead tree and intarweb [00:54:15:] hehe :-) [00:54:33:] it would be good to have some place in the wiki to cover press coverage... [00:54:45:] there is http://esw.w3.org/topic/CssValidator which can be used for that kind of thing [00:54:56:] So feel encouraged to add links there [00:55:14:] Did not get around to do something similar for MarkupValidator yet [00:55:18:] http://www.w3.org/Press/ [00:55:37:] has them, although obviously among many others [00:55:59:] it does? [00:56:01:] err [00:56:07:] http://www.w3.org/Press/Articles-2004.html [00:57:36:] http://clipping.prod-pressindex.com/download4.asp?idmail=4974142&langue=1&idpart=1 [01:03:49:] ====================== [01:04:03:] CSS Validator [01:04:04:] ====================== [01:04:32:] * What to do with BUGS.html now that all bugs are in bugzilla? [01:04:43:] http://jigsaw.w3.org/css-validator/BUGS.html [01:05:05:] I think there are only two or three bugs left [01:05:12:] the page is more or less useless now [01:05:16:] <__Yves> a pain to maintain things in two places, a blurb and a link to bugzilla would be better [01:05:30:] Or a redirect? [01:05:49:] if we can't generate a page from bugzilla there (afraid we can't) then should not bother maintaining both [01:06:14:] pointer would be good enough to me [01:06:45:] I have some preference for a 301 redirect but no strong opinion. [01:07:12:] <__Yves> fine either way [01:07:36:] Where would a link / redirect go? [01:07:37:] <__Yves> generating by scraping bugzilla seems hell, so 301 or page + link [01:07:38:] http://www.w3.org/Bugs/Public/buglist.cgi?product=CSSValidator&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED [01:07:46:] would be all open bugs / feature requests [01:07:56:] <__Yves> maybe a page with fixed bugs and open bugs would be better [01:08:06:] http://www.w3.org/Bugs/Public/buglist.cgi?product=CSSValidator would be all bugs [01:08:10:] Yves: depends on how much success I will have in cleaning up templates (so far - not much) [01:08:50:] btw. http://bugs.gnome.org/ is nice (-r than what we have)... [01:09:12:] <__Yves> yod: good luck [01:09:12:] in my view if redirect then redirect to open bugs. people should (?) be able to search for other bugs from there [01:09:43:] That would probably be http://www.w3.org/Bugs/Public/ or http://www.w3.org/Bugs/ [01:10:46:] We could have multiple links, search for bugs, show all bugs, show open bugs... [01:11:01:] <__Yves> +1 [01:11:40:] So someone should replace the BUGS.html with some Bugzilla intro and pointers to specific CSSValidator Bugzilla entry points [01:12:06:] can do it, but not before 10+days [01:12:12:] <__Yves> in general all the static pages should have a revamped CSS as well [01:12:13:] anyone else wants to do it? [01:12:17:] see also bugs.kde.org frontpage for link ideas if you run out :) [01:12:37:] I can look at it, do it if it is not done when you return :-) [01:13:00:] sounds acceptable [01:13:50:] ACTION: (bjoern or) olivier to replace the BUGS.html with some Bugzilla intro and pointers to specific CSSValidator Bugzilla entry points [01:14:03:] next thing would be [01:14:05:] * What to do with TODO? (http://jigsaw.w3.org/css-validator/TODO) [01:14:14:] I could not make much sense out of that page [01:14:20:] *** JibberJim has quit IRC ("Bye!") [01:14:26:] <__Yves> TODO -> redirect to open bugs or feature requests? [01:14:38:] or to BUGS.html? [01:14:51:] http://dev.w3.org/cvsweb/2002/css-validator/TODO.html [01:15:14:] TODO could have a more general, strategical overview of what we plan to do [01:15:23:] ... but only if we're keen on maintaining that [01:15:40:] (e.g mention our experiments with libcroco, etc) [01:15:43:] * __Yves needs to run, will be online in ~20mn (going back home for a teleconference night starting at 19h) [01:15:50:] I am not. Neither that page nor the current code :-) [01:15:53:] later, Yves [01:16:09:] <__Yves> hum, not sure to mention that in a TODO [01:16:51:] ok then [01:17:07:] Seems like we should do something similar with it as with BUGS.html [01:17:16:] more focussed on feature requests... [01:17:37:] or just keep it as is and remove the links to it [01:17:51:] "This page no longer maintained, see whatever" [01:18:13:] the latter I think [01:18:33:] bugzilla does not separate bugs from feature requests well enough for it to make sense [01:18:43:] yeah... [01:19:01:] so make that an AI similar to the BUGS.html AI? [01:19:11:] yes [01:19:47:] ACTION =~ s/BUGS/TODO and BUGS/ [01:20:13:] * How to fix line endings? -- http://www.w3.org/mid/3f36c357.166471943@smtp.bjoern.hoehrmann.de [01:20:40:] I've still no idea where this is coming from [01:20:59:] but it makes it rather difficult to maintain these things [01:21:06:] broken editor surely [01:21:15:] I could convert the line endings and commit, but that would be rather ugly [01:21:36:] could edit repository in place, but risky :) [01:21:58:] Make a backup and do :) [01:24:09:] I'll see whether that can be done graciously [01:24:28:] if not, then we should fall back to convert and commit [01:24:35:] find . -name "*,v" | xargs dos2unix [01:25:28:] in theory yes [01:25:39:] ok, I can do the latter if everything else fails [01:26:06:] ACTION olivier see if line endings can be fixed directly in repository [01:26:21:] I suggest that's it for the CssValidator for today [01:26:33:] the remaining item is not that important currently [01:27:00:] and answers to it are conditioned to my contacting PLH/Sijtsche, no? [01:27:37:] to some extend, yes. [01:28:06:] ========================= [01:28:15:] interlude (Log Validator) [01:28:18:] ========================= [01:28:29:] will release log validator 0.3 tomorrow [01:28:47:] with a CSS validation support thanks to bjoern's SOAPy module [01:28:52:] just FYI [01:29:17:] * bjoern_h pre-announced that in the wiki :-) [01:29:22:] cool [01:29:33:] Note that it will consider all www.w3.org/QA/ documents invalid... [01:29:44:] as the style sheet uses a non-css2 feature [01:29:51:] display:inline-block [01:31:01:] ah [01:31:09:] well, some chips may fall [01:31:30:] not an awfully big deal, but good to know [01:32:06:] indeed :) [01:32:04:] ========================= [01:32:10:] Link Checker [01:32:10:] ========================= [01:32:34:] we don't have an entry in the agenda, but could be good to have an idea of the timeline [01:32:43:] 3.9.3 almost there, need a couple of CSS tweaks [01:33:18:] and decision whether to fix (== add) the "#skip" fragments or remove the link, opinions? [01:33:47:] could you elaborate on this? [01:34:17:] just a sec... [01:35:00:] http://qa-dev.w3.org/wlc/checklink?uri=http%3A%2F%2Fqa-dev.w3.org%2Fwlc%2Fchecklink%3Furi%3Dfoo&hide_type=all&depth=&check=Check [01:35:39:] IOW, there's a "hidden" "#skip navigation" link in the header, but not all responses have the ID [01:36:00:] how about replacing with #results1 ? [01:36:13:] hmm not really skipping _navigation_ [01:36:47:] ...and there's no #results1 either in eg http://qa-dev.w3.org/wlc/checklink?uri=foo [01:37:04:] indeed [01:37:26:] IMO skip not terribly useful in checklink, but having it would be consistent with rest of v.w.o [01:37:59:] * bjoern_h can live without such links... [01:38:19:] * scop too [01:38:20:] agreed it's not very needed, though adding the a name would not be terrible to do [01:38:38:] remove? [01:38:55:] *** SHS` (SHS@213-152-38-112.dsl.eclipse.net.uk) has joined the channel [01:39:03:] remove is much easier :) [01:39:09:] (and fine with me) [01:39:10:] remove then. [01:39:13:] ok [01:39:52:] apart from those, I think we're (finally) ready for a quick beta period [01:42:41:] Who is going to announce the beta on www-validator? [01:43:04:] (and the install stuff etc.?) [01:43:15:] I think we can start the beta before thursday, and continue until yod's back [01:43:44:] I can install the beta and send announcement (my) thursday morning [01:44:05:] good. I'll commit the remaining bits tonight [01:45:04:] that leaves about-two-weeks before our next meeting and potential decision to release [01:45:39:] (there may be some gripes about robots stuff, need feedback on that) [01:46:12:] could you draft some text on what exactly you want feedback for the announcement and send that to yod/list? [01:46:38:] will do tonight [01:47:42:] ========================== [01:47:46:] Markup Validator [01:47:47:] ========================== [01:48:19:] shocking! check's output and why.html now valid [01:48:42:] has W3C lost all sense of irony? [01:48:52:] no - we didn't have one from the start [01:49:29:] * bjoern_h not sure about that, I see much evidence... [01:49:45:] but I am not going to mention names here :-) [01:49:55:] other than the choice of a banana for a footer image? [01:50:07:] * yod answered that mail in private, BTW [01:50:47:] pondered copying the list, but was concerned about potential "basic" interpretation of my message [01:50:48:] how unfortunate :( and yes, many other things suggest a sense of irony [01:51:10:] There is always the option to cc interested parties like me. [01:51:13:] the irony of talking about lack of irony , I guess... [01:51:51:] I think, even though that is unfortunate, we should defer decisions on this particular issue until the next meeting [01:52:11:] this is too important to do anything in haste in response to it [01:53:09:] exactly [01:53:16:] I suggest we fix in in CVS though [01:53:30:] then we can take our time [01:54:36:] * bjoern_h not in favour of that [01:54:56:] let's revisit when you are back [01:55:06:] ok, sure, whatever [01:55:19:] * what to do with documents without encoding information [01:55:27:] detail? [01:55:36:] does not make much sense to discuss without xover [01:55:40:] I had a conversation with Martin about it [01:55:53:] he said the current behaviour was likely to be the best we could do [01:55:53:] well, it currently defaults to UTF-8 which is rather suboptimal [01:56:15:] Martin argued that UTF-8 was generally safer [01:56:35:] We can test for UTF-8 first and if that fails use something else, for example [01:57:08:] but this is nothing that fits well into this meeting [01:57:49:] likewise for the 'referer' and 'm12n' items on the agenda [01:58:39:] EXPN on "/check?uri=referer vs. /check/referer" ? [01:59:28:] There are caching issues with MSIE and it's breaking a long tradition and the same look/feel for CssValidator and MarkUp Validator [01:59:38:] CssValidator still uses /check/referer... [02:00:12:] ok [02:02:28:] don't see much point on discussing it again without xover around [02:02:37:] indeed [02:02:43:] all the arguments for and against having been stated already [02:03:06:] although *not at the same time* [02:03:27:] it might make more sense to discuss this on www-validator first... [02:03:39:] perhaps [02:03:50:] are you willing to orchestrate such a discussion? [02:03:55:] <_Yves> I am not on www-validator, so cc www-validator-css if it involves css validator [02:04:11:] any known issues why reverting to /check/referer wouldn't be trivial? [02:04:40:] (technically) [02:04:46:] reverting would not be hard AFAICT [02:04:47:] Not that I know of, there might be parameter parsing issues, not sure [02:05:23:] it's more a question of whether the issues with check?uri=referer are worse than the ones with check/referer, really [02:05:45:] ok, let's wait for xover's insight [02:05:52:] (I would generally suggest to drop that thing as more and more people use filtering software that breaks links to it; just maintain it for backwards compat...) [02:06:11:] yod, yes, I can do that at some point [02:06:45:] thanks [02:06:53:] ... [02:07:31:] one more agenda item worth noting - the QAWG is likely to have a meeting in London in October [02:08:17:] (probably 25-29) [02:08:45:] provided the QAWG survives its meeting next week, or that the QAIG meets instead [02:09:08:] anyway... there is the idea of having a qa-dev meeting in europe around or at these dates [02:09:35:] not necessarily feasible, but worth investigating [02:09:52:] * _Yves can host [02:10:35:] In Nice? [02:11:13:] <_Yves> sophia antipolis [02:11:17:] <_Yves> quite cloes [02:11:50:] yod, the question is who would be generally interested? [02:12:01:] yes [02:12:41:] * scop is generally interested in that [02:12:41:] I would, but the financial pressure would not be on me [02:15:07:] I am interested... and somewhere around london would be much cheaper than Sophia-Antipolis for most of us. [02:16:10:] * yod can get in touch with QA?G meeting host and see whether they could host one more day [02:16:13:] Kopenhagen would also be an option, depending on whether we want to meet with these QA folks. [02:16:37:] or if qa-dev could be included and invited as part of that meeting [02:17:00:] That would certainly make some sense [02:17:24:] based on that info, we could plan further [02:17:40:] * yod notes level of interest, in principle at least [02:17:41:] The sooner the better [02:19:56:] what's up with "making sure the suggestion link actually is used for suggestions"? [02:20:25:] I suggest we edit the style sheet in place with p.helpwanted { display: none } and defer anything else until our next meeting... [02:20:43:] * yod wants people with strong opinions to discuss with other people with strong opinions [02:20:43:] I don't really want to have these "suggestions" for another two weeks... [02:21:24:] use mail thread, give deadline for reply(?) [02:21:46:] hmm... ok... [02:22:55:] that's not immensely satisfying but I would really prefer a discussion than people stating their opinions to me and have me decide [02:23:18:] this is not such an emergency that it requires an autoritarian decision [02:23:56:] well, xover suggested to make these things almost invisible in 0.6.7, display:none is not much different [02:24:47:] for the users, no, in terms of decision, I feel it is [02:25:39:] I am afraid that lets-do-something-now is not compatible with lets-look-at-this-later... [02:26:02:] but I'll send mail to the list [02:26:06:] we'll see [02:26:10:] thanks [02:26:26:] * yod hope we can look-at-it-now [02:26:36:] I may be naive, but as you say, we'll see [02:28:57:] ADJOURNED Log file closed at: 2004-06-09 02:34:35