See also: IRC log
Noah: First thing is to open the floor so that we can discuss what is on the agenda
one of the question is: do we need to continue to track issues or should we track products (like on the product page). We need a discussion but f2f time is already filled with technical discussions.
Ashok: are there important things buried in issues that might disappear if we move to products
Yves: how to track not-yet-products, create a new product? use a generic product?
Noah: can be done using actions.
... administrative stuff
please register for scribing slots
approval of minutes
<noah> http://www.w3.org/2001/tag/2011/12/22-minutes
RESOLUTION: minutes above approved
<noah> http://www.w3.org/2001/tag/products/
fragment identifiers & mime type: not scheduled as no progress were done
RESOLUTION: close the Web Application State (pending publication of the final Note)
DKA: wrt API minimization, we should leave it in a better state, but no discussion is scheduled
noah Is API minimization worth doing
DKA Absolutely
noahOK, to be considered after the election
<noah> ACTION: Ashok to draft product page on client-side storage focusing on specific goals and success criteria Due: 2012-01-17 [recorded in http://www.w3.org/2012/01/04-tagmem-irc]
<trackbot> Created ACTION-647 - Draft product page on client-side storage focusing on specific goals and success criteria Due: 2012-01-17 [on Ashok Malhotra - due 2012-01-11].
<glenn> possible minor typo under "Completed" table in products page: "Completion ... was announced on 30 December 2012"?
<Yves> indeed
http://www.w3.org/2001/tag/products/persistence.html
Noah: main question is "is there anything after the workshop report we need to do" ?
jar: Workshop took place on dec 8th in Bristol
Draft report: http://www.w3.org/2001/tag/2011/12/dnap-workshop/report.html
jar: We didn't have people saying that it was an impossible goal. No pure "cool URI" advocate stating that there was no problem to solve. No advocate of archives being sufficient. We didn't talk about what 'persistence' was. To me persistence is like trust.
Yves: trusting that it will be persistent, or that the representation is trustable (ie: not tampered with)
jar: There is a gradation of attacks, squatting being a threat to persistence, so it's both. People had a shared intuition of what persistence meant. Registry of media types or link relation is a good example of persistence. Even in the case of ISBN, it happened that some numbers were recycled, threatening the persistence of the identifier.
jar: There was the idea that in theory, persistent domain names were doable. One example is the use of the .arpa system This removes possibility to change owner, so create an immutable name (or mutable through a review process)
Larry: .arpa currently only IPV4-related?
<JeniT> http://www.iana.org/domains/arpa
jar: it's just the constituency that is interesting. Like .invalid
Noah: what is the update bandwidth of .arpa ?
jar: it's RFC-based; the outcome was not "move everything under .arpa". Options can be register a subdomain, a tld that offers persistence, etc...
noah A little confused: earlier you said there was not great interest in a new TLD, but now you're saying .arpa is just an existence proof. Isn't it an existence proof for a new TLD?
jar Not necessarily. What's important is that you've shown that a regime like this can be created. Could be realized as new TLD, could be realized as new domain under .arpa, or one could retroactively gold plate some existing domain(s)
(discussion about persistence of names vs persistence of representations and resolution)
<noah> Henry, do you need/want time to speak on this? We've got ~30 mins, but I want to devote the last 15 or so to TAG next steps
jar: Biological [species] names is a good example of a system where noone is reponsible for (it works by just publishing the name) but works.
Tim: ...but this is not a system under lots of stress, like the dns system is.
Noah: There are different communities there, one where this 'dns-like stress' is irrelevant, and others where it is the default
Yves: We are shifting from trust in the persistence to provenance issues of the representation, those are different.
<ht> http://www.w3.org/2001/tag/2011/12/dnap-workshop/report.html
<Zakim> ht, you wanted to note that I'm supposed to be reporting too!
<ht> no audio?
<masinter> (I said): when we do protocol design to design something to solve a problem, it's important to identify the scope of what problem is being solved, and waht isn't. In this case, we're picking at the edges of trust & security because of issues like SOPA take-down notices or someone disagreeing about biological names in some communities. The report on persistent domains hasn't been clear about waht the scope is, and that's why we're picking on it.
<masinter> jar replied: it was interesting that at the workshop we didn't feel a need to discuss scope at length
<jar> (I said): The scope was relative: We want domain-name-based naming that will be perceived as as "persistent" as, say, MIME types or ISBNs (persistence standard imported from other domain)? and are actionable
<jar> Gavin Brown of CentralNIC
ht: The .arpa solution came from nowhere, nobody was expecting it. The idea was that gold plated domain named ought to be governed by community process. The idea was also to use .arpa to create new persistent domains (like .doi.arpa)
<jar> gold-plating needs a community process, IETF is a good example of such a process
<jar> the governing document for .arpa already sets out the requirements. only agreement needed would be with IAB - ICANN, IANA not in the picture .arpa is different that any other tld, no registrar, no contract.
<noah> Could a new TLD be created with the same characteristic?
ht: I am not at all sure that the IAB would be happy to create something that would look like a real domain under .arpa
<jar> But expecting resistance from IAB since this would imply lots of ordinary DNS traffic through .arpa - a new idea
the ideal approach would be to have parts of existing domains persistents
<noah> I need to cut this off in 2 mins. We are at time, and need 10+ minutes on TAG goals.
<jar> by registering dx.doi.org.arpa, side effect would be to make dx.doi.org 'persistent'
we could use the .arpa to record that dx.doi.org is persistent, but not binding resolution to the .arpa domain but leave it to dx.doi.org system (ie: normal dns behaviour)
<masinter> Doesn't ISOC run .org? Maybe just asking ISOC to offer some persistence guarantees for organizations that meet some persistence criteria?
<jar> PIR runs .org
Noah: Henry, do you have an email with what you just talked about?
<jar> yes, that's a promising option, larry
<ht> http://www.w3.org/2001/tag/2011/12/dnap-workshop/report.html
(discussion on the product page)
<noah> http://www.w3.org/2001/tag/products/persistence.html
Tim: we need to create a scenario document
<jar> timbl calls for a TAG document explaining the process we might like to see in the future for gold-plating
<jar> masinter: What doesn't work now that would work better if we succeed?
<masinter> it isn't foolish to try to obtain consensus
<ht> No, but it's foolish to give me an action to obtain it!
<masinter> criterial for me is that it be clear to a reader not already invested in this, "what will work better, if the TAG does this work?"
<noah> ACTION-528?
<trackbot> ACTION-528 -- Henry Thompson to create and get consensus on a product page and tracker product page for persistence of names -- due 2012-01-02 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/528
<ht> trackbot, ACTION-528 due 2012-01-06
<trackbot> ACTION-528 Create and get consensus on a product page and tracker product page for persistence of names due date now 2012-01-06
<masinter> somewhere, at least one of the success criteria has to be doing something
BREAK
slides: http://www.w3.org/2001/tag/2012/01/mimeweb.pdf
<noah> FWIW: Larry's use of the term language strikes me as sensible, if informal, but it's only vaguely related to the carefully negotiated definition the TAG agreed a few years ago
<noah> I think there's the core of something very good here...
analogy between mime type and persistent names
<noah> My intuition is that we'll do better to challenge ourselves to start by making this as focused and narrow as possible. If more general principles emerge, we'll find them.
<noah> I'm very nervous about the top down view of how languages evolve.
relation between persistent names and evolution on what names represent
See ACTION-595 Draft a report on Mime and the Web
<glenn> poor connectivity here at present, will attempt to rejoin tomorrow AM
different pace between what is defined for email, and what is defined for the web
email needs backward compatibility, web forward compatibility
[ndlr: For better readability, use cases are listed here]
success criteria is to address 50% of the use cases listed
<noah> http://www.w3.org/2001/tag/products/mimeweb-2011-12-24.html
<noah> AM: I heard versioning of languages versus versioning of references. Those are different things
<noah> AM: I also heard versioning of languages versus XML languages.
AM: versionning of XML vs versionning of HTML
<noah> LM: Yes, one of David Orchard's documents was about that
<noah> AM: I think there are good extracts from Dave's documents that might be useful
AM: what about sniffing?
<Ashok> Larry, you had a document on sniffing ... what about progressing that document?
<masinter> ashok, the document on sniffing turned into issues in the websec tracker on the sniffing document, which i have now volunteered to edit
Tim: mime type is key to the web architecture. There was also a model of versioning done by Jonathan
consistent ways of identifying vesion, or relationships
<masinter> guidelines for: when to use a new MIME type vs. registering a new one
<Zakim> noah, you wanted to talk about scoping this bottom up
<masinter> guidelines for: when to use a version indicator as a paramter of a MIME type
but it's better to be very crisp on little pieces
Noah: maybe one thing would be to say "let's take javascript, and figure out what people want from the mime type registration when javascript evolves", then same thing for one or two more languages
jeni: being able to take a larger theory and narrow it to a specific use case would be to test the theory and get the use cases to provide feedback
<Zakim> timbl_, you wanted to go back to the 8 use cases
<noah> 1?
<Zakim> noah, you wanted to focus on mime-registration related stuff
<masinter> https://plus.google.com/106838758956333672633/posts/BbiiK6C937V
noah: working on generic versionning for XML proved very time consuming. this effort should focus on registering media types rather than telling the world how to define a language
LM: in the preface, I made it clear that the goal was not to have general definition of terms, but local ones only
<noah> NM: Larry, are you mostly buying into my suggestion that we scope this effort mostly to the parts necessary to tell a story about MIME registration
<noah> LM: Yes, except in so far as we need to look at other bits to verify that they are out of scope.
Yves: 5 use cases are about language versions, one is about discrepancy between advertized metadata and "real" content
<noah> NM: I'd be much happier if the use cases said things like: ">MIME type registration of evolving versions of {HTML, JavaScript}"
versioning of specifications vs versioning of implementations
Noah: HTML specs were always careful not to do big incompatible change to the meaning of a tag. <table> will always roughly mean a table. HTML 1.0 didn't have <img> at some point <img> was introduced and still now it means image
this is one property of HTML that people rely on
Noah: that can be one principle that might be outlined
LM: happy to narrow down the issue, however there are always architectural issues behind
… I have an action item on websec to work on sniffing
[ndlr: added afterward]Discussion about work on use cases, should this be in a wiki to further expand on those use cases ? No consensus on the otcome yet, so no actions were created out of this discussion.
<noah> ACTION-531?
<trackbot> ACTION-531 -- Larry Masinter to draft document on architectural good practice relating to registries -- due 2011-12-26 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/531
<noah> ACTION-595?
<trackbot> ACTION-595 -- Larry Masinter to create a report on Mime and the Web -- due 2011-12-29 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/595
<noah> ACTION-636?
<trackbot> ACTION-636 -- Larry Masinter to update product page for Mime and the Web -- due 2011-12-08 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/636
AM: what is the ultimate goal, one large comprehensive document, or lots of small ones?
the big document may never get done
noah: we need to change our product page to be more incremental
ashok: I would like to say 'this is the big long-range goal, and these are the short-term steps'
<noah> Reword http://www.w3.org/2001/tag/products/mimeweb.html to:
<noah> 1) Make clear the main goal is mime registration of evolving languages -- only bring in broader issues as necessary
<noah> LM: The registry performs an important function of review which entries apply to which versions.
Noah: how about stating that work on that bigger product is done and split it in smaller products?
LM: might be a distraction, better to get this product
<noah> Current goal: The goal of this activity is to help guide the use of MIME protocol elements in Web specifications and implementations, and to analyze, document, and propose solutions to difficulties with current effective use of MIME in the Web.
<noah> Proposed goal: The goal of this activity is to help guide the use of MIME protocol elements and Mime registratoins in Web specifications and implementations, and to analyze, document, and propose solutions to difficulties with current effective use of MIME types and MIME registrations for languages that evolve.
Noah: Larry will write up a product page describing goals of previous topic.
<noah> ACTION-531?
<trackbot> ACTION-531 -- Larry Masinter to draft document on architectural good practice relating to registries -- due 2011-12-26 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/531
<noah> Leave ACTION-531 for Friday
<noah> ACTION-595?
<trackbot> ACTION-595 -- Larry Masinter to create a report on Mime and the Web -- due 2011-12-29 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/595
<noah> ACTION-595 Due 2012-01-24
<trackbot> ACTION-595 Create a report on Mime and the Web due date now 2012-01-24
<noah> ACTION-595?
<trackbot> ACTION-595 -- Larry Masinter to draft a report on Mime and the Web -- due 2012-01-24 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/595
<noah> ACTION-636?
<trackbot> ACTION-636 -- Larry Masinter to update product page for Mime and the Web -- due 2011-12-08 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/636
<noah> ACTION-636 Due 2012-01-17
<trackbot> ACTION-636 Update product page for Mime and the Web due date now 2012-01-17
<noah> Noah to help Larry with ACTION-636, capturing new directions from Wed 4 Jan 2012
jar: [going through product page: http://www.w3.org/2001/tag/products/defininguris.html
... [reviewing schedule]
... idea is to make a call fro change proposals 15 Jan...
Ashok: will you ask others beyond us?
jar: Yes - the idea is to get community consensus. We should post call to linked-data and [other communities].
… idea is to drive this issue to something.
Noah: Do you think this will make a real difference?
jar: I think if there is no consensus and we withdraw the resolution then that could [have an impact]. If we get w3c consensus then that could have some effect. I think the current situation is not tolerable.
tim: as a member of this community - I haven't reacted yet ...
… I think there is a technical issue here. What came out of range14 was the 303 recommendation - that [is not good].
… We need to work on more efficient alternatives to 303.
… I'd like to see a conclusion that we're going to underscore the resolution but temper it with some engineering that will make systems work in practice.
jar: that would be a great outcome.
… I think there will be an outcome. Anything that's not the current situation is going to be positive.
JeniT: how will we assess the change proposals?
jar: we need to decide what the change proposal process would be.
JeniT: it could be quite hard to be seen as fair in assessing them.
jar: it would still go through w3c consensus process - through rec track - [no matter what the TAG process is]
... even a statement from the TAG that "there appears to not be consensus on this issue" would be positive.
… I proposed the idea of a "town meeting" teleconference.
<DKA> +1 to a "town meeting"
noah: the goals look good to me.
draft call for change proposals: http://www.w3.org/2001/tag/2011/12/uddp/cp-call.txt
jar: I have included a plea to review the work done so far on this issue.
Jenit: How would a change proposal show adequate understanding of those?
jar: Good question — I imagine that if it makes a claim that's refuted in one of the referenced documents then it's a problem. If it just says "there's no way to X" and the issue-57 documents gives a way to do X…. I should make a list of points that ought to be addressed. E.g. "why not just use hash URIs"..
noah: in this round, there will be some non-objective evaluation.
tim: we should say this is not the time to argue terminology.
<ht> +1 to requesting terminology discussion happen elsewhere
Jenit: If I was coming to this and thinking of writing a change proposal - I might be concerned that I go to the effort and it is rejected based on obscure reasons...
noah: could we just say "questions are welcome on the www-tag mailing list".
jenit: having a deadline for drafting change proposals and then a period for discussion / revision and then a final deadline when we will make an assessment...
noah: ideally we should put these out for community review...
jenit: could say "We encourage people to work together to create change proposals that reflect community…"
<noah> NM: Suggest a period of community review and refinement for all proposals to net out a good set of alternatives.
<noah> JAR: Good idea. Not currently in plan, but I'll add it.
<noah> NM: Do you want to do that in the product pages?
<noah> JAR: Yes but later.
jar: I need to add a clause welcoming proposals from anyone and explain that we're going after consensus.
JAR: call for change proposals would be accompanied by two change proposals: no change and withdrawal of resolution
Ashok: Do you expect something completely novel to turn up?
JAR: People might come up with something new, though I don't expect it given we've been talking about it for 10 years
LM: I have started thinking about this in a new way: URIs as protocol elements, with this being a way of providing a definition of what the URI is for languages to use
... originally it sounded like discovery: "how do we discover what this URI means?"
... but the language provides the meaning to the URI
... and the language may choose to inherit from the definition that you get from resolving the URI
NM: When this started, people accepted that you could make URIs for images and for people
... and it wasn't obvious that you couldn't respond with a 200 for a URI that meant a person
LM: The HTTP protocol is defined by the HTTP RFCs, which don't say anything about any of this, and httpRange-14 didn't change those
NM: There are certain words such as "representation" that different people read in different ways
JAR: This is all water under the bridge
jar: we're using URIs in RDF and how do we use them?
larry: httpRange-14 did not affect any language that didn't cite it.
tim: no httprange14 is about URLs
larry: I didn't like httprange14 before because it tried to address a larger scope than it should have.
tim: it ended up with the RDF community using URLs consistently with the html community.
larry: httprange14 is in scope for RDF and not for html.
noah: my view is that this should be a comment on status codes for the httpbis effort.
jar: I only care about RDF in this discussion. What's at
stake is the ability to refer to RDF from the Web. [JAR 2012-01-15: should be
"refer to the Web from RDF"]
... [going over baseline proposal: http://www.w3.org/2001/tag/2011/12/uddp/]
<ht> +1 to documentation over definition
tim: URIs should be universal
larry: in a SOPA case the counterfeit of chanel perfume was made to change the DNS resolution. If I wanted to make the case that this perfume was counterfeit then I better not use the URI because they were forced to change the resolution.
… "uncool URIs must change"
jar: this and the persistence problem really go together. RDF also suffers from the persistence problem.
larry: if you use a URI for meaning something in a context then you want that meaning to be persistent. If you use DNS names which you can't guarantee their persistence then ...
... another example- I make assertions about texaco and then chevron and texaco merger and they change all their uris to chevrontexaco.com...
… you have to accept the consequence.
… it may be that RDF2 will have a way of supplying a date context - please interpret these statements as being in context...
jar: [continues through document]
... http://www.w3.org/2001/tag/2011/12/uddp/#idp409552
… I've called out places where I've generalised.
jar: this enumerates the critical parts of the TAG resolution.
jenit: I don't agree that the 4xx response is not a critical part.
jar: OK I'll put it back in.
... it seems to me that 200 http is too much of a special case - if you're talking about web architecture then you're talking about rfc3986 retrieval - of which http 200 responses are a class.
larry: my change proposal is to avoid the phrase "http resource."
jar: I've already done this except where I'm quoting Roy Fielding.
... I give examples of ftp and data as being retrieval-oriented URIs.
... I will clarify further in the draft.
larry: in the web, there is an ambiguity between the touch point and the scope of the thing referenced?
jar: this is the reason you have documentation - to explain things.
larry: you cannot eliminate ambiguity.
jar: absolutely.
larry: the resources are not what are named by the URI . I have a problem with "the resource in question:...
jar: the ambiguity thing is a red herring.
[further wordsmithing]
jar: I'll clarify the second sentence.
... this is aimed at people using RDF.
ashok: "tis is a proposal for use with RDF."
<ht> That's a possible change proposal
<ht> but not a change to this document
larry: we could be happier when we say "this is RDF architecture.
... When you have linked data it is the function of linking data that you use the URI both for presentation and for meaning...
noah: let's say there's a solution adopted for the RDF community. It must be the case that the "document" community does nothing to conflict eo.
LM: We had URIs, now we have IRIs -- you don't try to impose a non-backwards-compatible meaning
JAR: If the RDF community accepts this, that's the end of the story
NM: If there was something that was backwards-incompatible for the document community, then that would be a problem
JAR: HTML doesn't have a stake in how this comes out
LM: Do load balancers and proxies have to start respecting this?
NM: Does Squid have to be aware of this? Is there RDF-Squid and Other-Squid?
TimBL: There are things that screw you up: Firefox transparently follows redirects, which is fine if it's 301s or 302s, but if it does it with 303s then it screws you up
JAR: 303 is already documented compatibly in HTTPbis
NM: Let's talk about making a resolution for JAR to take this forward
TimBL: I think it needs to recognise the problems with 303
JAR: the ISSUE-57 report has as its purpose to do just that
... this [UDDP] is just a baseline
... for change proposals
... I might be able to refer to the ISSUE-57 document in here
... there is a reference to that document in the call for change proposals
HT: it's worth saying in the call that you will find in the Issue 57 document a list of problems with the Fielding resolution
... different people have had different problems with it
JAR: I like that solution
... There's a lot in here about fragment identifiers, even though it's not really part of httpRange-14
<ht> I'm happy for the call to go out, with the minor changes proposed to JAR's baseliine doc't
JAR: I just need to know whether there's anything else I need to do before sending out the call
<noah> PROPOSED RESOLUTION: The TAG will circulate a call for proposals to (re)resolve issue httpRange-14. The call will be based on Jonathan Rees' "d proposal for a call for change proposals", which is based on the baseline draft at http://www.w3.org/2001/tag/2011/12/uddp/. Both are to be updated based on suggestions at 4 January 2012 F2F.
<noah> PROPOSED RESOLUTION: The TAG will circulate a call for proposals to amend issue httpRange-14. The call will be based on Jonathan Rees' "d proposal for a call for change proposals", which is based on the baseline draft at http://www.w3.org/2001/tag/2011/12/uddp/. Both are to be updated based on suggestions at 4 January 2012 F2F.
<noah> PROPOSED RESOLUTION: The TAG will circulate a call for proposals to amend issue httpRange-14. The call will be based on Jonathan Rees' "proposal for a call for change proposals", which is based on the baseline draft at http://www.w3.org/2001/tag/2011/12/uddp/. Both are to be updated based on suggestions at 4 January 2012 F2F.
<timbl_> PROPOSED RESOLUTION: The TAG will circulate a call for proposals to amend the resolution to issue httpRange-14. The call will be based on Jonathan Rees' "proposal for a call for change proposals", which is based on the baseline draft at http://www.w3.org/2001/tag/2011/12/uddp/. Both are to be updated based on suggestions at 4 January 2012 F2F.
<noah> RESOLUTION: The TAG will circulate a call for proposals to amend the resolution to issue httpRange-14. The call will be based on Jonathan Rees' "proposal for a call for change proposals", which is based on the baseline draft at http://www.w3.org/2001/tag/2011/12/uddp/. Both are to be updated based on suggestions at 4 January 2012 F2F.
<noah> Passes unanimously.
RESOLUTION: The TAG will circulate a call for proposals to amend the resolution to issue httpRange-14. The call will be based on Jonathan Rees' "proposal for a call for change proposals", which is based on the baseline draft at http://www.w3.org/2001/tag/2011/12/uddp/. Both are to be updated based on suggestions at 4 January 2012 F2F.
<noah> ACTION-624?
<trackbot> ACTION-624 -- Jonathan Rees to draft for TAG consideration a call for httpRange-14 change proposals -- due 2011-12-31 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/624
<noah> close ACTION-624
<trackbot> ACTION-624 Draft for TAG consideration a call for httpRange-14 change proposals closed
<noah> ACTION-625?
<trackbot> ACTION-625 -- Noah Mendelsohn to schedule followup discussion of http://www.w3.org/wiki/HttpRange14Options (per agreement in Santa Clara) -- due 2011-12-21 -- CLOSED
<trackbot> http://www.w3.org/2001/tag/group/track/actions/625
<noah> ACTION-589?
<trackbot> ACTION-589 -- Noah Mendelsohn to work with Jonathan to update URI definition discovery product page Due: 2011-08-18 -- due 2011-12-23 -- CLOSED
<trackbot> http://www.w3.org/2001/tag/group/track/actions/589
<noah> ACTION-201?
<trackbot> ACTION-201 -- Jonathan Rees to report on status of AWWSW discussions -- due 2011-12-28 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/201
<noah> ACTION-201 Due 2012-03-31
<trackbot> ACTION-201 Report on status of AWWSW discussions due date now 2012-03-31
<noah> ACTION-282?
<trackbot> ACTION-282 -- Jonathan Rees to draft a finding on metadata architecture. -- due 2012-01-31 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/282
<noah> Jonathan will bump 282 date later.
<noah> ACTION: Jonathan to post call for change proposals to amend the resolution to httpRange-14 per 4 January 2012 TAG Resolution Due: 2012-01-17 [recorded in http://www.w3.org/2012/01/04-tagmem-irc]
<trackbot> Created ACTION-648 - Post call for change proposals to amend the resolution to httpRange-14 per 4 January 2012 TAG Resolution Due: 2012-01-17 [on Jonathan Rees - due 2012-01-11].
DKA Product page: http://www.w3.org/2001/tag/products/PublishingLinking-2011-12-27.html
JeniT We've been kicking this draft around - Dan I have revised it recently.
http://www.w3.org/2001/tag/doc/publishingAndLinkingOnTheWeb-2012-01-04.html
JeniT following discussion with Rigo - the direction he recommended was to have this doc focused on technical aspects of publishing and linking and to have some other mechanism for having stronger statements that we want to make - e.g. right to link, flagging issues with transforming proxies, etc - questions with legal questions associated with them.
…. we want to answer 3 main questions today. First, is that a reasonable way forward; 2nd if so what is the best mechanism for making those opinion statements?; 3rd given that we make those statements, what statements should we make?
… so first question - is that a reasonable way to structure this work?
Larry There are different kinds of opinions. There are opinions about the technical impact are and opinions about legislation. I wanted you to separate out the opinions about legislations from the technical opinions.
JeniT yes - that's what we've done for the latest draft.
Noah: it strikes me as the right direction to try.
Dan: this is based on some additional feedback from Rigo.
Jenit: what should the opinion statements be?
<masinter> You can remove opinions about legal without removing opinions about control
Dan: Bullet points on messages we want to convey. We have taken these out of the document.
Jeni writes on board "hosting != possesion"
Larry: Can we keep this in the document rephrased as technical point?
<masinter> ""It is impossible to control dissemination of content-based unwanted material, merely by imposing restrictions on service providers offering transformation services, because such services are not able to differentiate wanted from unwanted content. The result would be severely limited services, instead.""
<masinter> http://lists.w3.org/Archives/Public/www-tag/2011Dec/0120.html
Dan: Rigo would feel this would be better left out
... this is opinion
Larry: If you cannot distinguish what you can publish and what you cannot, you cannot publish anything
... distinguish mechanically at reasonable cost
<masinter> "common carrier"
Noah discusses wording re automated algorithms to distinguish ...
scribe: what you can publish and what you cannot
... copyrighted vs. non-copyrighted material
Peter: A lawyer you not care whether it is automated or not
<masinter> in the "legal opinion" view, I'd want to make web hosting and transformation services to be like "common carriers"
Jar: I think you can consider making the point without using legal
... you can talk about requirements
<jar> System requirements come from a variety of sources. Laws and contracts are just particular examples of requirement sources. So talk about requirements, and avoid any hint that some legal assessment is being made.
Jeni: We want to separate the two documents: legal and technical
Larry: Goverments do not have to make things illegal in order to prohibit them
Jeni writes on board "filtering content can be hard".
Larry: We should point about other means of control other than criminalization
Dan: We can work in parallel on these two things ... first publish the technical work
Jeni writes "copying is needed for proxying /archiving"
Jeni writes "rewriting links is necessary for archiving"
Jeni writes "transforamation is needed for search engines"
Jeni writes "users don't know where they are going when they click on a link/prefetch"
Jeni writes "deep linking is necessary -- right to link
<jar> greasemonkey is a terrific innovation… but may be incompatible with requirements
scribe: linking is a speech act
... network effects
Dan: A speech act is something that is protected under UN Resolution ...
<noah> NM: I agree. I'm happy to see the TAG talk about the importance of network effects & Metcalfe's law. I'm happy to see >the W3C< make the connection to UN Resolutions. I don't see the technical content that makes the connection to the UN part of the TAG's remit at W3C.
Larry: Protocols should be explicit whether links imply automatic behaviour without additional action
Jeni writes " linking vs. inclusion"
Larry: Links could be speech acts or automatic actions
Noah disagrees ... points to separation of concerns. Links can be processsed in different ways
Larry: Protocols could annotate links to indicate whether link should be followed
Jeni: Topics on board are examples we can pull out to talk about legislation or contracts, etc.
Ashok: These are good starting statements
Most people in room think this is a good direction
Tim: "It would be reasonable for a Goverment to conclude that ..."
Larry: Larry: (reading aloud)In telecom field this is the legal Common Carrier issue ... does this translate to web hosting etc. ?
... used in Common Law Countries
Dan: Article 19 on Human Rights ... fundamental right
Jeni: Moving to vehicle ... how to disseminate these points
Noah: We need a base technical document with good, simple, non-inflammatory examples
... We can decided how to disseminate on a case-by-case basis
... two documents technical document and a document with examples.
... perhaps combine into one document
Larry: Perhaps ask ISOC for some help
Noah: We should publish our document first
Discussion about updating the product page
Ashok: We need a new mechanism to put W3C positions front and center
Tim: Perhaps Web Foundation could pick that up
Larry: This seems the main thing that ISOC does ... we could get them the technical background
jar: But this project is about what we think ... voice of the technical community
Yves: Most impt thing is to publish the technical document
<jar> +1
Dan: Let's publish the technical document and then debate how to disseminate the other stuff
Larry: Should we review your latest documemt?
jar: Technical document could have some compelling examples
<jar> The main (tech) document can contain plenty of compelling examples, based on general system requirements (not on legal considerations)
<jar> (which is similar to what I think Larry was saying earlier, about administrative control)
<noah> http://www.w3.org/2001/tag/2012/01/CopyrightLinkingTopics.jpg
<noah> ACTION-627?
<trackbot> ACTION-627 -- Noah Mendelsohn to schedule very detailed line-by-line review of Pub&Linking draft at January F2F -- due 2011-12-23 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/627
<noah> need to reopen 627 until draft is ready
<noah> Never mind
<noah> ACTION-627 Due 2012-01-10
<trackbot> ACTION-627 Schedule very detailed line-by-line review of Pub&Linking draft at January F2F due date now 2012-01-10
<noah> ACTION-627 Due 2012-01-17
<trackbot> ACTION-627 Schedule very detailed line-by-line review of Pub&Linking draft at January F2F due date now 2012-01-17
<noah> ACTION-629?
<trackbot> ACTION-629 -- Daniel Appelquist to with help from Jeni to propose changes to goals, success criteria etc. for publishing/linking product page -- due 2011-11-11 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/629
<noah> ACTION-629 Due 2012-01-17
<trackbot> ACTION-629 With help from Jeni to propose changes to goals, success criteria etc. for publishing/linking product page due date now 2012-01-17
<noah> ACTION-541?
<trackbot> ACTION-541 -- Jeni Tennison to helped by DKA to produce a first draft of terminology about (deep-)linking etc. -- due 2011-12-20 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/541
<noah> Changing description
<noah> ACTION-541?
<trackbot> ACTION-541 -- Jeni Tennison to helped by DKA to produce draft on technical issues relating to copyright/linking -- due 2012-01-31 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/541
<noah> ACTION: Jeni to produce a document with examples motivating the technical points in the Copyright/Linking document Due: 2012-03-20 [recorded in http://www.w3.org/2012/01/04-tagmem-irc]
<trackbot> Created ACTION-649 - Produce a document with examples motivating the technical points in the Copyright/Linking document Due: 2012-03-20 [on Jeni Tennison - due 2012-01-11].
The following notes were taken on the whiteboard during this session:
