W3C | TAG | Previous: 22 Mar teleconference | Next: 5 April

Minutes of 29 March 2004 TAG teleconference

Nearby: IRC log | Teleconference details · issues list (handling new issueswww-tag archive

1. Administrative (15min)

  1. Roll call. SW, TBL, DC, NW, RF, MJ, CL, IJ. Regrets: PC
  2. Resolved to accept the minutes of the 22 Mar teleconference

    Action IJ: Remove a side comment per SW's suggestion

  3. Accepted this agenda
  4. Next meeting: 5 April. No meeting 12 April.

1.1 May ftf meeting

  1. See meeting page
  2. Action IJ: Look for more advice from W3C Team on suggested hotels.

2. Technical (75min)

See also open actions by owner and open issues.

2.1 Top Level Domains used as filters (.xxx, .mobile, etc.)

See email from Danny Weitzner and email about timing expectations.

.xxx and .mobile
SW: Why did DJW bring this to us?
DC: He suspects this is a bad idea and he thinks that if the TAG says it's a bad idea, that might have an impact.
  1. Economic arguments
  2. Architectural problems - putting info into a URL
  3. Social manageability
DC: There's a comment period open right now (so some time-sensitivity)
no - blocking .sex does not block all sex sites. finding them is no easier.
DC: I think that the market will cause this to fail.
.tv is actually a country code
DC: E.g., only a few hundred ".tv" domains.
More on the topic: http://www.chron.com/cs/CDA/ssistory.mpl/front/2461340
but yes it is sold as a novelty for television
Persistence decreased by arbitrary boundaries
Device Independence reduced
DC: Subdividing internet doesn't scale. If you have .mobile and .sex, what do you about sex for mobile devices?
However there is no requirement to have only television stuff under .tv for example
TBL: If you have .travel, then assume that you'll have a "dot" for every rubrique in the yellow pages. If you follow Kant, you end up dividing up the Web. You are imposing a centralized ontology. We could write a book on why that won't work. We could document the fact that the Web only worked because it DOES NOT have that. There are a number of systems pre-dating the Web that didn't work because people didn't agree with the ontologies of others. Some arch problems: persistence degrades. E.g., as a resource becomes more general, information will be moved out of a ".mobile" domain. Similarly, some information may be labeled adult and then later non-adult (e.g., "articles" in Playboy <wink>). There are philosophical arguments - who defines what is "mobile"? what is "porn"?: We've been through a huge discussion on this. PICS is the result of this discussion. Some people are happy to be labeled xxx, but others (larger group) that don't want to be labeled xxx but others want to label them as xxx - trying to manage this internationally is very difficult. E.g., dewey decimal system does not scale to the size of the Web.
TBL: (1) persistence (2) social unmanageability (3) economic argument. People have paid a lot for domain real-estate. By creating a new TLD you are changing the zoning. You may have to buy hundreds of domain names all of a sudden. Thus, these proposals make it tough to maintain a trademark, and add cost to businesses as well as destabilize the DNS.
Chris, you wanted to talk about .mobile, and about the cultural relativity of 'obscene'
CL: I agree about the cultural relativity argument. There are lots of mobile companies pushing for this. Designed for profit.
CL: Their proposal says that "The Web is for desktops." Which is very dangerous.
ah; that's a comment I could endorse: "the .mobile proposal is misinformed"
DC: Yes, let's please send that comment.
+1 to problems with "the web is for desktops" argument!
mario, you wanted to note that also TLDs like .ag for public owned companies are not accepted that much by the marked. Additionally, people are not awaiting finding company related information at e.g. sap.ag the first place. Normally, these domains are just seconding a .com one.
It flies in the face of device independence
similar .assoc.fr for example
and .tm.fr
MJ: Market shows reluctance to accept special domains for first domain name in the market. Something like .ag is often used as a secondary domain; usually people use .com anyway.
(I'm confused about how for-profit TLDs ever got on the map)
exact same as http://www.pepsi.fr/
although getting a raw .fr is very hard because you have to show you are a french company
DC: I'd like CL to pursue the mobile question, whatever we do. E.g., CL writes three paras, the TAG endorses. I'd like CL to send this to the official comments list forthwith.
SW: At end of RFC3675, Donald recommends rating services (PICS)
(wow... it's not easy at all to find the comment target email address)
ah... http://forum.icann.org/mtg-cmts/stld-rfp-comments/general/index.html
If i have an action, can it be crisply stated so that it can be closed when i have done it?
draft to www-tag and send to (wherever) is ok as an action
DC: "Public comment period begins 31 March" (ends 30 April)
"A public comment period will begin 31 March 2004, 23:59 UTC and will last for one month. Comments will be posted in a web-based forum. Instructions will be published when the comment period opens." -- http://www.icann.org/announcements/announcement1-26mar04.htm
Action CL: Send a draft to www-tag explaining why the .mobile proposal is misinformed. If the TAG supports the proposal, send to ICANN on the official mailing list.
CL: Should we follow up on other topics like .sex?
TBL: I think it would be appropriate for the TAG to get into the technology and society area. For instance, to get into the arguments discussed when PICS was developed - e.g., centralized categories are socially unacceptable. This was the argument behind distributed whitelist/blacklists. Do we think it's out of scope for the TAG because it's too societal?
DC: I think in scope. Just not inspired.
TBL: Could we endorse something that DJW has written up.
DC: Yes, though I would have to see it first.
TBL: Can we ask DJW to address the filtering issue if we address the device-independence issue?
SW: Yes, the TAG could, after review, endorse something that Danny has proposed.
TBL: We could publish something small that points to text (or take it over).
Action IJ: Talk to DJW about sending a proposal to the TAG (focusing on social issues) that the TAG could review and possibly endorse.

2.2 Web Architecture Document Last Call


  1. Last Call issues list (sorted by section)
  2. Annotated version of WebArch
  3. Archive of public-webarch-comments
  4. List of actions by TAG participant
  5. Additional actions
    1. Action IJ 2004/02/09: Incorporate editorial suggestions (see minutes of that meeting for details).

Actions 2004/03/15 (due 25 March?) to review sections: No progress today


  1. If people have sent comments per above actions, we will review those comments.
  2. We will look at comments with broad impact, such as those from Pat Hayes
  3. We will continue our backwards walk through the document.
Pat Hayes comments
[Meaning of key terms like identifier, resource, representation]
SW: I agree that we use "identify" in both sense. I was sympathetic to his comments.
DC: This is the "myth of names and addresses" Get used to it.
RF: If you look in the dictionary, the term has both definitions. We are using in the descriptive sense (in which both meanings are legit).
NW: I was sympathetic to PH's arguments for many years then decided that's not the way the world is, so time to move on.
(I haven't read his comments very closely, but based on Stuart's summary, I've heard similar arguments many^7 times)
TBL: I think PH has hit on description of the semantic Web. By confusing his meanings C and D, we create the Web. That's the Web. That's what makes the Web the Web.
RF: I agree with TBL on that.
TBL: PH's first sentence includes the word "conflict". I think he's binding it to disjoint sets in his mind (in his own ontology).
SW: Some assertions that we make in our good practice notes, he asserts, make good sense in one world, but are crazy/meaningless in the other world.
Ian, you wanted to mention another issue about people/agents
(people/agents: asked and answered, no? action timbl)
I think most text on agents was written without considering that agent might include person
IJ: My point is not to define "agent" in another way, only that people have commented that some GPNs don't make sense for all types.
Issues manola6, parsia7, parsia14, manola17, manola23, i18nwg16
manola6 is asking about "user agent", not "agent"
DC: I thought we said "Do so at your own risk."
DC: I suggest changing to "do so at their own risk."
TBL: What about "have no license to"
DC: I think this makes sense for people and software; could use some editorial fixing.

DC Proposal: Adjourn and work on actions during next 30 minutes.

[Discussions of action items]

IJ: I expect to do editorial work, tracking.
CL: [IJ missed]
SW: I shall be reviewing section 2.3 issues
NW: I shall be looking over section 3
MJ: I will be helping CL with section 4. [And sending comments to public-webarch-comments]
DC: I need to mull over PH comments, respond to Kopecky
TBL: Follow up with David Booth
RF: Section 2, finish draft of URI spec this week.

The TAG did not discuss issues below this line.

2.3 Review of open action items related to issues

The TAG expects to review the list of open actions by owner and to close any that are obvious to close. TAG participants are encouraged to review this list before the meeting, as well as other action items listed in this agenda.

3. Status report on these findings

See also TAG findings

4. Other action items

Ian Jacobs for Stuart Williams and TimBL
Last modified: $Date: 2004/03/29 22:39:40 $