- From: Ian B. Jacobs <ij@w3.org>
- Date: Mon, 14 Jan 2002 09:10:24 -0800
- To: www-archive@w3.org
IRC log of 14 Jan 2002 TAG teleconference Note: This IRC log has been edited, primarily to remove interactions prior to discussion beginning in earnest, and after the meeting was adjourned. Timestamps are US Pacific Standard Time. [07:36:03] <Ian> Regrets: DanC [07:36:10] <Ian> zakim, who's here? [07:36:12] <Zakim> I see NWalsh, TimBL, Stuart, PaulCotton, TBray, IJacobs-CA [07:37:01] <Ian> Where is Chris, Roy, David? [07:37:12] <Zakim> +DOrchard [07:37:50] * Ian general sentiment that 7:30 PT is early.... [07:37:55] * Ian is in PT today. :) [07:37:58] <DanC> not to mention the agenda was late ;-) [07:38:14] <Ian> Action item review: [07:38:33] <Ian> PC: I"d like to see agenda sent Friday. [07:38:39] <Ian> -------------- [07:38:47] <Ian> On the previous minutes: [07:38:59] * TimBL will try to get agenda out on Friday [07:39:00] <Ian> IJ: Thanks to DanC for help. [07:39:10] <Ian> (No comments on previous minutes) [07:39:21] <Ian> TimBL: Requests for agenda items? [07:39:29] <Ian> * Face-to-face meeting [07:39:55] <Ian> TimBray: We received a formal request for input. [07:40:28] <Ian> Media types (Mark Baker) [07:40:29] <Ian> http://lists.w3.org/Archives/Public/www-tag/2002Jan/0063.html [07:40:31] <Ian> --------------------- [07:40:34] <Ian> IPR requirements [07:40:55] * DanC wanders off to WebOnt ftf, where the break is over... [07:41:06] <Ian> TimBL: Patent disclosures necessary to TAG by TAG. Also, people who are not employees of Member companies need to sign collaborators' agreement. [07:41:09] <Ian> ...this means Roy. [07:41:15] <Ian> ...and Tim Bray. [07:41:16] <Zakim> + + [07:41:52] <Ian> +Roy. [07:42:02] <TimBL> Zakim, + is Roy [07:42:03] <Zakim> sorry, TimBL, I do not recognize a party named '+' [07:42:07] <TimBL> Zakim, who is here? [07:42:08] <Zakim> I see NWalsh, TimBL, Stuart, PaulCotton, TBray, IJacobs-CA, DOrchard, ??P11 [07:42:31] <TimBL> Zakim, ??P11 is Roy [07:42:32] <Zakim> +Roy; got it [07:42:33] <TimBL> Zakim, who is here? [07:42:34] <Zakim> I see NWalsh, TimBL, Stuart, PaulCotton, TBray, IJacobs-CA, DOrchard, Roy [07:42:41] <Ian> Invited expert agreement: [07:42:49] <Ian> http://www.w3.org/Consortium/Legal/collaborators-agreement [07:42:55] <Ian> Tech reports release form: [07:43:02] <Ian> http://www.w3.org/TR/Release [07:43:57] <Ian> IJ: Do you have to print and fax? [07:44:04] <Ian> TimBL: I think an email is ok, but we need to check. [07:44:12] <Ian> Action IJ: Check out what's sufficient; contact RF and TB. [07:44:15] <Ian> ------------------------- [07:44:21] <Ian> First ftf meeting [07:45:05] --> roy (~fielding@cx2098322-a.alsv1.occa.home.com) has joined #tagmem [07:45:13] * Ian welcomes Roy. [07:45:48] <Ian> Proposed day: Tuesday 29 January in D.C. [07:46:02] <Ian> TBL: Limitation on that - we're having a Team dinner that evening. [07:46:07] <roy> I can't make that date either. [07:46:12] <Ian> DO: Can't either. [07:46:33] <Ian> CL, TB, PC are speaking during the conf. [07:46:44] <Ian> 30 Jan: W3C steering committee meeting in BOS. [07:46:56] <Ian> PC: TBL, do you have any alternate dates? [07:47:13] <Ian> What was wrong with 1 Feb? [07:47:51] <Ian> DO, TB, PC: Can't make the 28th. [07:48:13] <Ian> CL: Can't make 1 Feb - I18N Workshop [07:48:23] <Ian> So that means: 28, 29, 1 out. [07:49:23] <Ian> TBL: Do people have a preference for meeting mid-week or end week? [07:49:40] <Ian> TB: I think we'll have to take what we can get.... [07:49:53] <Ian> Proposals: 4 Feb, 7 Feb, 8 Feb. [07:50:01] <Ian> Norm: Monday 4 Feb no good for me. [07:50:13] <Ian> IJ: I'll be in Chicago on the 8th. [07:50:33] <Ian> TB: That week bad for me (4-8 Feb) [07:50:56] <Ian> Week of 25-28 Feb out (tech plenary) [07:51:02] <Ian> 11 Feb: AB meeitng [07:51:23] <Ian> TB, DO: 18-22 works well [07:51:36] <Ian> NW: People going to plenary likely to leave on Friday of that week. [07:52:00] <Ian> DO: 14 Feb is V-day. [07:52:24] <Ian> 13 Feb? [07:52:31] <Ian> PC: Weds are always bad for me. [07:52:40] <Ian> Proposed 12 Feb. [07:52:52] * Ian will be in Boston 11 Feb. [07:53:18] <Norm> Boston works for me [07:53:35] <Stuart> Boston 12th Feb would work for me [07:53:49] <Ian> TB: I think you should all come to Vancouver! [07:54:08] <Ian> TBL: I can manage the 12th in Boston. [07:54:23] <Ian> PROPOSED: 12 Feb in Boston. [07:54:26] <Ian> Stuart: Fine for me.' [07:54:28] <Ian> PC: Not sure yet. [07:54:45] <roy> Are we responsible for our own travel costs? [07:55:05] <Ian> PC: Can we have video conf from Vancouver? [07:55:20] <Ian> TB: I don't know about logistics, but DO and TB can ask around. [07:56:21] * DanC offers tentative regrets for any ftf meeting in Jan, Feb, Mar, or May [07:56:33] <Ian> TB: Any objections to checkout out video conf as a possibility? [07:56:40] <Stuart> no [07:56:46] <Ian> TimBL: I am in favor of video conf. [07:56:51] <Ian> Plan A: Everyone flies to MIT. [07:56:56] <Ian> Plan B: Find videoconf. [07:57:18] <DanC> video is all well and good, but don't expect it to substitute for a ftf meeting [07:58:36] <TimBL> We were considering a 2-site, DanC [07:58:57] <TimBL> Maybe we should think about 3. [07:59:26] <Ian> TBL: The AB handled a 3-site video conf pretty well, but Ann Bassetti on her own. Not so great. [07:59:43] <Ian> --On meeting at WWW2002--- [08:00:06] <Ian> PC: I had suggested Sunday morning 5 May. [08:00:12] <Ian> ...DanC pointed out Team meeting. [08:00:20] <Ian> PC: Another alternative is Friday 3 May. [08:00:27] <Ian> " Currently the AC meeting is planned for [08:00:27] <Ian> May 5-7 with the main part of the meeting being on Mon-Tue May 6-7 and the [08:00:27] <Ian> introductory sessions on Sun May 5. The AB is planning to meet on Sat May 4. " [08:01:09] <Ian> TBL: Team rehearsal has less emphasis than in previous years due to change in AC meeting format. [08:01:42] <Ian> PC: The WWW conf is from 7-11 May. [08:02:09] <Ian> TBL: I need to be present for some of new AC rep meeting. [08:02:25] <Ian> PC: All-day meeting 5 May with people walking in and out? [08:02:32] <Ian> TB: Not sure this is a good idea. [08:03:09] <Ian> PC: We are expect to present to the AC on M or T. [08:03:20] <Ian> ...(We should have ftf time before that). [08:03:36] <Ian> (No conflicts to meeting on Sunday) [08:03:52] <Ian> TBL: By the time we get to this point, we may have subgroups working on projects. Could report to main group on Sunday. [08:04:04] <Ian> [Several people intend to scuba, including PC, TB, IJ] [08:04:23] <Norm> Alas, no scuba for me, but I'll snorkle along side [08:04:33] <Ian> PROPOSED: all-day 5 May [08:04:50] <Ian> (with some people possibly dropping out periodically) [08:05:05] <TimBL> DanC? [08:05:07] <roy> 5 May is okay for me [08:05:32] <Ian> PC: You might ask Steve Bratt (COO) to take over some of TBL's role at the new AC rep meeting. [08:05:40] <Ian> --------------------------------------- [08:05:59] <Ian> Charter issues. [08:06:09] <Ian> - Setting up a process for hearing appeal. See charter. [08:06:09] <Ian> We have to elaborate this. Discuss briefly, possible subgroup [08:06:33] <Ian> TBL: A requirement to provide a process to hear an appeal. Don't know whether all will be formal appeals. Will get requests for input as well. [08:06:52] <Ian> ....question of accountability on www-tag. [08:07:20] <Ian> TB: Example of email signed XML Protocol WG [08:07:21] <Ian> http://lists.w3.org/Archives/Public/www-tag/2002Jan/0063.html [08:07:29] <Ian> TBL: We need an internal and an external process. [08:07:48] <Ian> - Need to keep an issues list (Ian?) [08:08:12] <Ian> - Have a representative of the issue in the TAG present to the TAG at a meeting. [08:08:50] <Ian> DO: I suggest that, since we will have a large number of issues over time. People who bring issues should also bring possible resolutions. [08:09:55] <Ian> DO: Choices that we recommend will be based on different use cases. [08:10:41] <Ian> TBL: We need a page that says "If you're asking the TAG a question, here's what the TAG needs to know." [08:10:59] <Ian> ...one point would be "Are there a set of apparent alternatives? If not, could you write them up?" [08:11:18] <Ian> TB: A couple of obvious points: [08:11:30] <Ian> - TAG first decides whether they will address the issue. [08:11:51] <Ian> TBL: We may address it on the spot as a local issue, but put into a long-term agenda. [08:11:55] <Ian> TB continuing: [08:12:05] <Ian> - We need a system for numbering and referring to issues. [08:12:14] * TimBL apologizes for butting on TimB [08:12:20] * TimBL apologizes for butting in on TimB [08:12:54] <Ian> IJ: Any bug-tracking tool suggestions? [08:13:20] <Norm> I'm quite happy with the SourceForge trackers; if there was any plan to run a W3C Forge [08:13:26] <Ian> TB: Issue number can be opaque string useful in email headers. Our primary form of records is mailing list. As long as you use issue numbers there, you're ok. [08:13:48] <TimBL> [open] [08:13:51] <TimBL> [open][close] [08:15:01] <Ian> DO: I suggested for SAML WG - not just an issue number or opaque string, but a transparent string: rough categorization, longer string (short version of the issue name), the issue number itself. [08:15:42] <Ian> TB: I'd like to avoid getting into an extended process. [08:15:56] <Ian> ...key for me is gating (yes/no we will handle) and public record with issue numbers. [08:16:21] * Ian notes NW's comments. [08:16:53] <Ian> Action IJ: Write up summary: (1) based on email (2) will use mnemonics (3) will be public and we will be accountable to it. [08:17:23] <TimBL> and (4) it will aloow reject as well as accept [08:17:30] <Ian> IJ: I think an index is also necessary. [08:17:50] <Ian> TB: Chairs need an internal system for keeping track. But mailing list subject lines are the root of everything. [08:18:53] <Ian> TB: Policy is only that the email has the unique id in the subject line. [08:19:10] <Ian> PC: Issue numbers aren't reusable. [08:19:57] <Ian> PC: How do we address the XLM Protocols issue in particular? [08:20:07] <Ian> - Do we take this on? [08:20:24] <Ian> ------------------ [08:20:27] <Ian> XML Protocols issue [08:20:34] <Ian> http://lists.w3.org/Archives/Public/www-tag/2002Jan/0063.html [08:20:43] <Ian> TB: Propose we take this on and call it issue 1. [08:20:49] <Ian> TBL: I'd like someone to present it to the WG now. [08:21:03] <Ian> Q. 1; What are the general guidelines or policies (if any) for W3C [08:21:03] <Ian> working groups in defining their own media types? Should they be [08:21:03] <Ian> defining them at all? [08:21:20] <Ian> Q. 2; If custom media types are required, should there be any [08:21:20] <Ian> commonality between them? [08:21:20] <Ian> Q. 3; What is, or what should be, the relationship between a media type [08:21:20] <Ian> and an XML namespace? [08:21:44] <Ian> TB: Some people responded that didn't understand Q2. [08:22:16] <Ian> DanC's reply: [08:23:44] <Ian> TBL: Q3 first came up with MathML WG- mixing math with xhtml. [08:24:10] <Ian> ...if you serve mixed doc as xml to netscape, netscape will recognize math through namespace and will display the math. [08:24:19] <Norm> Mark's Q: http://lists.w3.org/Archives/Public/www-tag/2002Jan/0063.html [08:24:32] <Ian> TBL: ...but if you send to IE, IE will feed it only to generic HTML machine, which would display as raw source. [08:24:35] <Norm> Dan's reply: http://lists.w3.org/Archives/Public/www-tag/2002Jan/0069.html [08:24:42] <Ian> TBL: ...you might be able to fix with a style sheet. [08:25:07] <Ian> TBL: Question is: should you be able to select your application based on outermost namespace in a document. My opinion is "yes", otherwise you can't create XML applications. [08:25:46] <Ian> ...we can talk about mixed-content documents afterwards. [08:26:13] <Ian> TB: If we say it's good for software to dispatch on namespaces (seems not to controversial), I don't think media types will help you much for highly mixed docs. [08:26:47] <Ian> DO: Manifest: constituents of XML "package". [08:27:09] <Ian> ....don't think we should duplicate namespace info outside the content. [08:27:32] <Ian> ....we should say that media type should say that it's an xml document. And that processes will have to unpickle at least the first namespace in the document. [08:27:49] <Ian> TB: It's more complicated than that - to deal with things properly, you have to dispatch as you work through the document. [08:27:53] * TimBL happy for people to use IRC to get on the q to speak [08:28:03] <Ian> TB: ...not sure much can be usefully said in the media type [08:28:09] <roy> A media type should reflect the purpose of the message. Sending a document as XML serves no useful purpose. [08:28:17] <Ian> TB: I don't think the media type architecture is going to help. [08:28:40] <Ian> DO: There's no point in munging the manifest into the MIME declaration. [08:29:24] <Ian> RF: The media type has more to do with visibility (and selection of handlers) than about the structure of bits in the content. If you send a generic XML, you are expecting a generic engine to process the namespaces. In practice, that's never the case. [08:30:20] <Ian> TB: Rules of game - spec must define namespace and media type. But below top level media type doesn't help. [08:30:44] <Ian> TBL: You can create an RDF document that has an HTML document inside it, but the HTML document is not mean to be displayed - RDF is talking about it. [08:31:17] <Ian> ....but the top-level semantics are important (e.g., "This is an HTML document first and foremost" or "This is SVG first and foremost") [08:31:32] <roy> [agrees with TBL] [08:31:47] <Ian> ...the outermost xml determines the meaning of the innermost (e.g., it may quote innermost). [08:32:33] <Ian> TBL Summarizing: [08:32:39] <Ian> Q3: There is some work to be done. [08:32:49] <Ian> Q2: Didn't quite understand. [08:33:03] <Ian> Q1: People seemed happy, but having discussed Q3, perhaps not as clear. [08:33:22] * Ian asks RF to summarize his point on IRC. [08:33:44] <Ian> TBL: Having some visibility at the toplevel is useful, even if it doesn't give you all the information you need. [08:34:04] <Ian> TB: I think we should agree to provide an answer to Q1 and Q3. And assign those to tag issues 1, 2. [08:34:55] <Ian> IJ: I'd like to put "Validation of mixed document types" on the issues list.... is this related to Q3? [08:35:55] <Ian> DO: My organization doesn't believe in this RFC. [08:36:10] <Ian> ...plus sign doesn't help that much. What version information are you talking about? [08:36:16] <Ian> ...you're still going to have to do more work.... [08:36:48] <Ian> TB: Do we want to take a position on that RFC? [08:36:58] <Ian> TBL: Then add Q2 to issues list as well. [08:37:05] <roy> RFC 3023 [08:37:40] <Ian> TB: I think we should discuss on www-tag list. [08:38:09] <Ian> TB: If we agree that our first step is to accept/reject. There should be a point in the process where we say what we are going to do (e.g., update a document, publish an arch rec, etc.). [08:38:31] <Ian> TB: I think we formally close the issue with an email. [08:38:52] <Ian> s/TB/TBL last comment. [08:39:06] <Ian> TB: I think that having something in the email archive will do. [08:39:47] <Ian> DO: I think that it can be problematic if issue resolved by email and nothing else happens. Better when there's a larger context, and appropriate resolution text put in that larger context (e.g., document). [08:40:11] <Ian> TimBL seques into next agenda item.... [08:40:13] <Ian> --------------------------------------------- [08:41:31] <Ian> TBL: Ideally, outline view of arch is available. Appended to it are descriptions in detail that link to TAG replies. [08:41:55] <Ian> TBL: I think that a glossary is important. [08:42:40] <Ian> DO: How will tag deal with multiple uses of same term. [08:43:16] <Ian> or two terms for same idea (e.g., namespace name and namespace URI both used) [08:43:26] <Ian> TBL: DanC kept a list of this type of thing. [08:44:10] <DanC> http://www.w3.org/Architecture/Terms [08:44:14] * Ian thanks Dan. [08:44:26] <Ian> TBL: Interest in using annotation server? [08:44:37] <Ian> TB: Probably not. I think we should use generally available public tools. [08:44:47] <Ian> TBL: Can be used in Amaya or Netscape. [08:45:08] <Stuart> http://www.w3.org/1999/05/WCA-terms/ [08:45:09] <Ian> TBL: Another possibility - we keep a list of these annotations. [08:45:14] * DanC wonders if anybody present has actually used Annotea. [08:45:28] * Stuart I haven't [08:45:37] <Ian> TB: I think for most of us, TAG work is going to place another load on our time. I don't have time to debug software. [08:45:41] <Ian> PC, DO: Agreed. [08:45:55] <Ian> TBL: So we should stick to email and CVS. [08:46:12] * DanC regrets not setting up a wiki last week [08:46:14] <Norm> Is Annotea the Amaya one or the IE one? Scott Boag's been saying how nice the IE one is, but... [08:46:20] <Ian> (TBL notes that per his action item, he put in a request re cvs to systems tema) [08:46:56] <Ian> TBL: Perhaps IJ's primary job will be to create a tree (Table of Contents) for the architecture of the Web. [08:47:10] <Ian> ...this would be a reference for us. We write documents to fit into the spaces. [08:47:18] <Ian> ...and document relations between these things. [08:47:32] <Ian> TB: Are we going to be more reactive or proactive? [08:47:38] <Ian> TBL: Charter is that we do both. [08:48:53] <Ian> TBL: When I find that I'm answering the same arch question 3 times, I write down my thoughts (clear or not). [08:49:21] <Ian> ...so I've created a tree first, and filling it in non-linearly. [08:49:43] <Ian> So, is first task to write up the tree? [08:50:01] <Ian> TBL: Brainstorming with yellow stickies. [08:50:26] <Ian> TBL: A tree will never be perfect. [08:50:36] <Ian> TBL: For next time, we can brainstorm on top-level cut. [08:51:03] <Ian> DO: We're going to have to think about the typical taxonomy problem - what's the toplevel decider. [08:51:14] <Ian> TBL: We have to realize that it's arbitrary, but we have to do it for our own work. [08:51:26] <Ian> TBL: We can design it as a Web (document dependences). [08:51:36] <Ian> ..that's what we've done for sem web advanced development. [08:51:53] <Ian> ...we generate graphs with graph viz. [08:52:02] <DanC> contribution the TOC: formats/protocols/naming. maybe provider-view/consumer-view too. [08:52:08] <Ian> DO: I was referring to TB's software for navigating.... [08:52:55] <Ian> TB: Propose answering Mark Baker, and put issues on issues list. [08:53:14] <Ian> Action IJ: Put together a page on doing this. Publish first few issues. [08:53:46] <Ian> IJ: Length of ids? [08:53:50] <Ian> TB: Short is good. [08:53:58] <TimBL> mnonic [08:54:00] <Ian> TBL: Would like to see mnemonic. [08:54:16] <TimBL> Resolved - Action Ian to accept issues 1-3 [08:54:48] <Ian> Action IJ: Change name of conf on zakim from monthly to week. [08:55:09] <Ian> Action item review [08:55:14] <Ian> TBL: Find out what kind of editing access to the W3C Web site [08:55:14] <Ian> will be available to TAG participants. [08:55:14] <Ian> PC/IJ: Summarize issues suggested on www-tag (including [08:55:14] <Ian> technical comments, liaison request). (An initial [08:55:14] <Ian> categorization of input may be found in the IRC log.) [08:55:14] <Ian> DO: As part of preparation for TAG panel at W3C's Technical [08:55:14] <Ian> Plenary 2002, solicit input from chairs on what issues the [08:55:14] <Ian> TAG should address, and which documents the TAG should [08:55:14] <Ian> produce. [08:55:30] <Ian> TBL: Looks like CVS will be available [pending] [08:55:52] <Ian> PC/IJ: PC did his part, IJ [pending] [08:55:56] <Ian> DO: Not done. [08:56:03] <Ian> ...should be done by next week. [08:56:13] <Ian> -------------------- [08:56:17] <Ian> Next meeting 21 Jan, same time. [08:56:20] <Ian> ADJOURNED -- Ian Jacobs (ij@w3.org) http://www.w3.org/People/Jacobs Tel: +1 718 260-9447
Received on Monday, 14 January 2002 12:10:24 UTC