RDB2RDF Working Group Teleconference -- 18 May 2010

<trackbot> Date: 18 May 2010

<mhausenblas> hhalpin, that is the funny thing - I didn't receive any error message

<mhausenblas> MacTed, can you scribe, please?

<MacTed> mhausenblas - I may be worse than no scribe. a personal failing.

<mhausenblas> oh, I thought you could help us improving the minutes quality ;)

<boris> Hi all

<mhausenblas> hi boris

<mhausenblas> thanks a lot, hhalpin for forwarding the agenda

I'll scribe!

<mhausenblas> scribenick: hhalpin

notes there was a spam attack

ok, let's go!

Admin

http://www.w3.org/2010/05/11-rdb2rdf-minutes.html

PROPOSAL: Accept the minutes of the 11 May 2010 telecon,

<mhausenblas> +1

ACCEPTED: Minutes from the 11th May 2010 telecon

Introduction of new members

Ahmed: We will discuss the use-cases more today
... and make sure we all agree before publication
... Let's introduce the new people!

Alexander: Alexander DeLeon, working at Technical University of Madrid
... worked with peer to peer reasoning, description logic, linked data

Ahmed: Welcome aboard!

Boris: Also working at ontology engineering group at Polytechnic University Madrid
... developed mappings between databases and ontology in past work

face-to-face SemTech 2010

maybe I'm muted!

but yes 6 weeks in advance

its OK if we miss it by a week...

ericP and Ivan will be at Semtech, same with me likely.

Orri: I will be there

Ahmed: I will be there.

Juan: Dan and I will be there.

PROPOSAL: Fix face-to-face dates at Semtech 2010.

Juan: We discussed earlier one day before one day after.
... i.e. one day before and one day after.

<scribe> ACTION: mhausenblas will send out the exact date (perhaps the day before and/or after) with a proposed concrete date to the list [recorded in http://www.w3.org/2010/05/18-rdb2rdf-minutes.html#action01]

<trackbot> Sorry, couldn't find user - mhausenblas

<scribe> ACTION: mhausenb will send out the exact date (perhaps the day before and/or after) with a proposed concrete date to the list [recorded in http://www.w3.org/2010/05/18-rdb2rdf-minutes.html#action02]

<trackbot> Created ACTION-55 - Will send out the exact date (perhaps the day before and/or after) with a proposed concrete date to the list [on Michael Hausenblas - due 2010-05-25].

Ahmed: I think it was the day before

Yes, RDF Next Steps is the week afterwards.

<Ashok> SemTech: http://semtech2010.semanticuniverse.com/

<scribe> ACTION: mhausenblas to create a wikipage for face-to-face [recorded in http://www.w3.org/2010/05/18-rdb2rdf-minutes.html#action03]

<trackbot> Sorry, couldn't find user - mhausenblas

<scribe> ACTION: mhausenb to create a wikipage for face-to-face [recorded in http://www.w3.org/2010/05/18-rdb2rdf-minutes.html#action04]

<trackbot> Created ACTION-56 - Create a wikipage for face-to-face [on Michael Hausenblas - due 2010-05-25].

Use-case document

Ahmed: Have people had a chance to read it?

+1, ahve read it.

Ahmed: I have spent a lot of time reading it
... there is a lot of terminology I am uncomfortable with.

Ashok: I had another look at it.
... its a few small things
... the one big item
... I would have liked to have the requirements to come from the use-cases
... reading it, requirements and use-cases are separate
... no visible connections.
... I would like to see those links.

Ahmed: I agree with Ashok on this.
... there is a lot of terminology
... things like micro-parsing that I am not sure what it means
... very confusing statement
... but if I can't understand it, then what about outside world?
... mapping to outside data-types?

I'm going to note there was going to be an action to make a glossary

but it appears EricP went on vacation a bit too quickly without making glossary

<Ashok> yes. glossary is incomplete

<mhausenblas> hhalpin, the glossary is there

<mhausenblas> right

<mhausenblas> just a placeholder ATM

maybe Ahmed could send an e-mail out with the vocabulary items he wants explained?

Ahmed: I'd like to send you an e-mail
... to make sure we can work through the rest of reservations
... we should really work through these
... once people publish it, if people don't understand it, it's a problem.

Ashok: Ahmed, it would look better with more explanation
... lots of small things like that.
... the question is how to improve it

Michael is an editor!

<MacTed> ericP - are you out there?

<mhausenblas> no, ericP is on holiday

Notes that you need to send exact changes in text.

<MacTed> ah.

<MacTed> forgot that.

Why not e-mail a list of changes to the list-serv

Basically, we need to fix it up this week

its supposed to be published next Tuesday

<MacTed> refresh URL?

michael: within the cell there is structure
... and I want to define something that maps that to RDF

Ashok: its a mapping from a value

http://www.w3.org/2001/sw/rdb2rdf/use-cases/

Souri, latest URI above (it should always be the same)

Ahmed: I just don't understand this

Ashok: this requirement and the following one should have some further explanation

<LeeF> WGs are supposed to publish a draft every *3* months, I believe. (heartbeat requirement)

<Ashok> Yes, the date is old :-)

last update is $Id: Overview.xml,v 1.39 2010/05/11 17:21:15 eric Exp $

notes that we need to publish

Ashok: Can we stop the publication process

it would be better if actual textual changes were suggested rather than just vague "I don't understand" comments

Ahmed: I'm not on board with publication
... it's in an embarrassing state.

MacTed: I agree that it's not that great

Generally governance is by vote.

You can raise an objection

<LeeF> I don't understand why we can't go ahead and publish and record any objections to publishing.

Ahmed: SPARQL took 5 months
... We shouldn't rush

<Ashok> Harry. The W3C works by consensus ... we rarely vote

Notes that we do vote when there is real difficulty with consensus

Ahmed: A month is not enough

Notes that very few people have commented, and the feedback is too vague. You have to address exact changes.

Ahmed: We should publish something we should be proud of.

<LeeF> Ashok, note that consensus does not require unanimity.

Ahmed: its worse to publish something that is confusing.

MacTed: Should we publish with only headers

<LeeF> I could support the proposal from MacTed.

thinks that proposal is a bit crazy

we should just sit down and actually send it comments, happy to sit on phone with Ahmed and Ashok tomorrow.

But I would really prefer if people sent in actual comments with *actual* suggested text.

Ahmed: Some of us were not aware of the size of it
... I'm a very busy guy
... we are where we are.
... the section headers are confusing.
... we need a better transition

I'm happy to delay the publication more

<mhausenblas> http://www.w3.org/2001/sw/rdb2rdf/wiki/Use_Cases_and_Requirements/Reviews

but we need to actually have a discussion that is not vague and meta-level

notes that ericP and michael have not addressed many of the comments, i.e. any of the grammatical notes made by me.

Ahmed: I did not approve it
... My comments have not been addressed.

Michael: We discussed them and reached consensus

<LeeF> Can we please stick to the queue?

<LeeF> This is not professional behavior.

Ahmed, calm down :)

+1 LeeF

[lots of discussion, got a bit heated!]

Ashok: Could we follow up with Harry and Michael and work through it line by line
... it may take 3 hours
... but then we'll have a better document.

+1 Ashok

Ahmed: I may not have 3 hours
... i'd prefer if anyone can participate

<Zakim> LeeF, you wanted to note that WG members and chairs have a time commitment to the group

I'm happy to book an extra 2 hours for our next telecon, and can even do it now...

that's fine with me.

LeeF: Notes that we have time committements, which I think very people have put in.
... publishing this document is an internal step
... and it's ok.
... if the i's are not dotted and t's not crossed

MacTed: We have had many discussions
... I think the editors have not reflected all the discussions
... but just some of them.
... however, I'm willing to do it.
... and we've committed to partipate in other decisions

<LeeF> +1 to MacTed's frustration

PROPOSAL: We delay publication one week, and then we book up to two extra hours to go through all the comments
... All comments must be sent it this week before the meeting, and comments must be precise about what sentences they don't understand/like and ideally suggest textual changes.

<Souri> If needed, I won't mind meeting for 2+ hrs any time (even) tomorrow or day after (after a careful reading tonight of the latest version) in helping revise the document

Ahmed: How about for things where I don't have

Michael: Either a concrete proposal or removal

+1 Souri

Michael: it's difficult to guess or to read people's mind when being editors
... we need concrete proposals

Ashok: Documents have lots of editors, have you actually gone through and fixed them?

Michael: The editor notes are both for the editors and to communicate things under discussion to readers as well.
... happy to remove editor notes or turn them concrete

Does next week work for an extra hour or two?

Ahmed: You need to send comments to editor list

<LeeF> http://www.w3.org/2001/sw/rdb2rdf/wiki/Use_Cases_and_Requirements/Reviews ?

Ahmed: to whole list not just editors

Michael: Note that I have been tracking these comments
... see wiki page

Ahmed: I only read e-mails

(notes we need to go forward with this proposal and get consensus!)

<Souri> I agree with the suggestion that, esp. if you cannot meet for some extra time, please send *very specific* review comments and suggestions via email

PROPOSAL: We delay publication one week, and then we book up to two extra hours to go through all the comments

<Souri> +1

<alexander> +1

<mhausenblas> +1

<Seema> +1

RESOLUTION: We delay publication one week, and then we book up to two extra hours next meeting to go through all the comments

PROPOSAL: All comments must be sent it this week before the meeting, and comments must be precise about what sentences they don't understand/like and ideally suggest textual changes

i.e. comments must be sent it by FRIDAY

<LeeF> I think this is good advice, but I don't see how it makes sense as a WG resolution :)

PROPOSAL: All comments must be sent it this week (by Friday) and comments must be precise about what sentences they don't understand/like and ideally suggest textual changes

<Souri> Can we do Saturday end-of-day instead of Friday?

Saturday end of day works for me

<Souri> Gives me an extra day

PROPOSAL: All comments must be sent it this week (by end-of-Saturday) and comments must be precise about what sentences they don't understand/like and ideally suggest textual changes

<Souri> +1

<mhausenblas> +1

(sorry, I may need Sunday to help put the comments in)

(Monday is very busy with other WGs for me)

RESOLUTION: All comments must be sent it this week (by end-of-Saturday) and comments must be precise about what sentences they don't understand/like and ideally suggest textual changes

PROPOSAL: If Ahmed can not make next meeting for extra time, he schedule the time this week with editors and Team contact, and sends the times in e-mail to listserv.

Ashok: Nervous about doing it differently
... let's just stay longer on Tuesday

<mhausenblas> oh, am I a co-chair? didn't know that :D

Ashok: If we start scheduling differently, it will be very hard.

let me mention something!!

<mhausenblas> LeeF, not that I'm aware of

so we're definitely doing an extra long meeting on Tuesday

but if Ahmed has scheduling trouble, then we spend the extra time this week

and then go through

<Souri> So, are we meeting 11am-1pm EDT next Tues?

another meeting with him and send the comments to him.

Yes Souri.

<MacTed> I'll look forward to doing my own major rererereview upon my return....

Although hopefully we will not need all that time!

trackbot, meeting adjourned

<trackbot> Sorry, hhalpin, I don't understand 'trackbot, meeting adjourned'. Please refer to http://www.w3.org/2005/06/tracker/irc for help

RSSAgent, draft minutes

- DRAFT -

RDB2RDF Working Group Teleconference

18 May 2010

Attendees

Contents

Admin

Introduction of new members

face-to-face SemTech 2010

Use-case document

Summary of Action Items

Scribe.perl diagnostic output