IRC log of dawg on 2004-03-25

Timestamps are in UTC.

15:29:03 [RRSAgent]
RRSAgent has joined #dawg
15:29:21 [DanC]
DanC has changed the topic to: RDF DAWG 25 Mar
15:29:30 [alberto]
alberto has joined #dawg
15:29:43 [Zakim]
+ +1.760.476.aaaa
15:29:57 [DanC]
agenda + convene, take role, review record, agenda, and misc actions
15:30:06 [Zakim]
15:30:10 [DanC]
agenda + Amsterdam meeting arrangements
15:30:17 [alberto]
alberto has joined #dawg
15:30:24 [DanC]
DanC has changed the topic to: RDF DAWG 25 Mar chair: DanC; scribe: ?VOLUNTEER_PLS
15:30:26 [Zakim]
15:30:34 [Zakim]
15:30:35 [Zakim]
+ +1.317.151.aabb
15:30:41 [DanC]
agenda + Use cases and Requirements
15:30:50 [KendallC]
argh, zakim always gets my name wrong
15:30:52 [Zakim]
15:31:01 [DanC]
Zakim, KendalC is KendallC
15:31:01 [Zakim]
+KendallC; got it
15:31:04 [KendallC]
15:31:10 [Zakim]
15:31:12 [dirkx]
dirkx has joined #dawg
15:31:25 [DanC]
Zakim, [ASemantics] holds DirkG
15:31:25 [Zakim]
+DirkG; got it
15:31:41 [Zakim]
15:31:50 [alberto]
alberto has joined #dawg
15:32:35 [DanC]
DanC has changed the topic to: RDF DAWG 25 Mar chair: DanC; scribe: KendallC
15:32:45 [DanC]
Zakim, who's on the phone?
15:32:45 [Zakim]
On the phone I see Tayeb, ??P13, +1.760.476.aaaa, KendallC, +1.317.151.aabb, EricP, [ASemantics], DanC
15:32:47 [Zakim]
[ASemantics] has DirkG
15:32:49 [Zakim]
15:32:54 [DaveB]
DaveB has joined #dawg
15:33:15 [Zakim]
15:33:31 [Zakim]
15:33:36 [DanC]
Zakim, ??P20 is DaveB
15:33:37 [Zakim]
+DaveB; got it
15:33:45 [AndyS]
zakim, ??P13 is AndyS
15:33:48 [alberto]
alberto has joined #dawg
15:33:49 [Zakim]
+AndyS; got it
15:33:55 [DanC]
Zakim, aaaa is RobS
15:33:55 [Zakim]
+RobS; got it
15:34:27 [Zakim]
15:34:35 [alberto]
alberto has joined #dawg
15:34:41 [DanC]
Zakim, who's on the phone?
15:34:41 [Zakim]
On the phone I see Tayeb, AndyS, RobS, KendallC, +1.317.151.aabb, EricP, [ASemantics], DanC, DanielK, DaveB, Patrick, Pat_Hayes
15:34:43 [Zakim]
[ASemantics] has DirkG
15:35:36 [alberto]
+39 is Alberto Reggiori - on the phone too
15:36:34 [dirkx]
list attendees
15:36:40 [dirkx]
zakim, list attendees
15:36:40 [Zakim]
As of this point the attendees have been Tayeb, +1.760.476.aaaa, +1.317.151.aabb, EricP, KendallC, DirkG, DanC, DanielK, Patrick, DaveB, AndyS, RobS, Pat_Hayes
15:37:09 [Zakim]
15:37:32 [DanC]
Zakim, take up item 1
15:37:32 [Zakim]
agendum 1. "convene, take role, review record, agenda, and misc actions" taken up [from DanC]
15:38:40 [DanC]
Zakim, next item
15:38:40 [Zakim]
agendum 2. "Amsterdam meeting arrangements" taken up [from DanC]
15:38:59 [alberto]
15:39:28 [dirkx]
Details - will be adding more details as we get closer.
15:39:31 [alberto]
1st f2f meeting in AMS (Leiden actually) - 22-23 April
15:39:33 [ericP]
ACTION ericP: set up meeting registration
15:41:18 [DanC]
Zakim, next item
15:41:18 [Zakim]
agendum 3. "Use cases and Requirements" taken up [from DanC]
15:41:48 [dirkx]
Dirk will follow hotel sitation (please email me if you find them full) and will try to keep up adding additional ones (the current ones where verified on thursday/friday as still having rooms ).
15:41:58 [KendallC]
attendees: alberto asemantics, andyS, jean-francois inria, dank, kc,
15:41:58 [KendallC]
robs, patricks, dajobe, ericp, dancon, jos de roo, pat hayes
15:41:58 [KendallC]
1 april next meeting
15:41:58 [KendallC]
dajobe scribe volunteer
15:41:58 [KendallC]
minutes approved (no opposed, no abstens)
15:42:11 [KendallC]
danc reserved bridge
15:42:11 [KendallC]
amsterdam meeting arrangements, dirk gives us
15:42:11 [KendallC]
ericp take over registration for f2f
15:42:11 [KendallC]
discuss telcon participation @ f2f
15:42:11 [KendallC]
danc working on f2f agenda
15:42:43 [KendallC]
discuss use cases and requirements
15:43:55 [KendallC]
attendee: dirk asemantics
15:44:48 [KendallC]
15:47:16 [KendallC]
15:47:19 [KendallC]
argh, sorry
15:47:55 [DaveB]
also attendee, josD - came in during roll call
15:48:04 [DaveB]
oh you caught that, sorry
15:48:08 [KendallC]
15:48:26 [DanC]
KendallC, no need to paste everything from emacs into this log. but if you could start taking notes here now, that might help.
15:49:50 [KendallC]
danc points out that our examples *are* the use cases (maybe clearer to call them user *stories* then?)
15:50:02 [DanC]
"story" is fine by me.
15:52:12 [alberto]
two possible applications/use-cases for "tell me about X": data/metadata browser, metadata crawler
15:52:56 [KendallC]
dirkx: split this use case (tell me about foo) into some specific domains. It's very genric.
15:52:59 [KendallC]
er, generic.
15:53:18 [KendallC]
the google harvesting case, follow the hierarchy case, tell me more case...
15:54:01 [KendallC]
use case: AndyS's "Find the email address of John Smith"
15:54:39 [KendallC]
robs: we should address precisely what the user wants to do with the results of this query.
15:54:55 [dirkx]
Making the use case more specific: "Browse" as in 'discover'; "Browse" as in selectively/interactively follow certain references; "Browse" as in 'discover' more information/refinements; "Browse" as in retrieve some alternative presentation to do something 'different' with.
15:55:12 [KendallC]
dirkx: thx
15:55:37 [KendallC]
andys: the use case *should* say, if it doesn't, that the query returns an email address, not some RDF.
15:55:43 [DanC]
(that's bryanT speaking?)
15:55:46 [dirkx]
I think andy meant that the Use case does -not- say that you get back RDF - rather some result. Such as an email. That is application specific. i read it that way.
15:55:48 [KendallC]
oh, sorry
15:55:53 [dirkx]
Sorry Andy - you just said that I think.
15:56:30 [KendallC]
andys: 2 revisions coming
15:57:11 [KendallC]
danc: does anyone think our tech won't solve this problem? No.
15:57:34 [dirkx]
DanC: it would not solve the whole problem - just be a key element/step in that process.
15:57:35 [rob]
that's white pages, not yellow...
15:57:52 [KendallC]
<strike>No.</strike> :>
15:58:42 [KendallC]
some hints of consensus forming around idea of solving this use case
15:59:05 [JosD]
JosD has joined #dawg
15:59:59 [KendallC]
use case: EricP's use case/action item
16:01:24 [dirkx]
Not sure if 'real-life' sizing our use cases are that useful. In actual reality 80% of any given 'solution' for such a case will have nothing to do with RDF and even less is DAWG scope concerned. So you are building use cases which are out of scope to whcih you may add a postfix listing just that part which is DAWG specific.
16:01:37 [DaveB]
EP-1 doesn't seem to do anything with ?context in the collect
16:01:49 [rob]
I think scope comes second. Use cases come first.
16:02:26 [DanC]
PatrickS tells a story of a vendor with complex parts... wants to put a catalog online...
16:03:26 [rob]
All these "tell me abouts" suddenly seem very interactive and less automated...
16:03:48 [KendallC]
most of our stories seem very human facing, which keeps surprising me
16:04:00 [dirkx]
rob: that is the result of a premisse; that it must be a story understandable to Aunt Mary.
16:04:16 [rob]
I feel that the open-ended "tell me about" is intrisically harder to deal with automatically.
16:05:50 [KendallC]
action item: patricks to write up his car-parts story
16:06:46 [KendallC]
(err, i assume that was a real action item, patrick. if not, lemme know.)
16:07:19 [KendallC]
use case: danc's geo story
16:07:37 [KendallC]
some liking of alberto's (?) proposed solution
16:07:39 [DanC]
"Re: Use case: tiger map/census data: have it your way" and thread
16:07:46 [KendallC]
16:08:26 [KendallC]
alberto: it can be seen as a case for a kind of extensibility requirement in the QL
16:08:45 [KendallC]
danc: didn't mean it as a story to motivate extensibility
16:09:04 [KendallC]
danc: rather, that some basic maths should be in the QL
16:10:58 [alberto]
exslt is a good example of that i.e. extend XSLT easly with namespace
16:12:08 [alberto] library like for GIS "operators" for example - or something along those lines
16:12:57 [KendallC]
16:14:49 [KendallC]
Zakim, mute me
16:14:49 [Zakim]
KendallC should now be muted
16:16:46 [rob]
One of our technologies here at NI involves viewing relational databases as RDF data.
16:18:14 [KendallC]
there's some appeal for danc's geo story
16:18:32 [KendallC]
dajobe: speaks in support of EP1 (is that right?)
16:19:02 [DaveB]
16:20:46 [ericP]
EP-4 aka FatAnnotationQuery:
16:21:20 [alberto]
Andys - isn't that about Annotea and Amaya ?
16:21:20 [ericP]
AndyS, cool use case!
16:22:02 [ericP]
Annotea curently shows what annotaitons there are on the "current" document
16:22:20 [alberto]
ericP: ok one step further - cool
16:22:26 [Zakim]
- +1.317.151.aabb
16:22:28 [ericP]
AndyS's sounds like it shows usefule stuff when hovering
16:23:26 [alberto]
AndyS UC sounds like "inline metadata" case - attach some external metadata to a link/image - interesting
16:23:52 [AndyS]
alberto - I guess so! trying to be a UC, not tech :-)
16:24:03 [rob]
I can stick around.
16:24:07 [alberto]
AndyS - +1 ! :-)
16:24:21 [KendallC]
Zakim, unmute me
16:24:21 [Zakim]
KendallC should no longer be muted
16:25:20 [DanC]
Zakim, who'se here?
16:25:20 [Zakim]
I don't understand your question, DanC.
16:25:24 [DanC]
Zakim, who's here?
16:25:24 [Zakim]
On the phone I see Tayeb, AndyS, RobS, KendallC, EricP, [ASemantics], DanC, DanielK, DaveB, Patrick, Pat_Hayes, JosD
16:25:26 [Zakim]
[ASemantics] has DirkG
16:25:27 [Zakim]
On IRC I see JosD, alberto, DaveB, dirkx, RRSAgent, Zakim, DanC, KendallC, AndyS, eikeon, rob, ericP
16:25:48 [patH]
patH has joined #dawg
16:25:52 [KendallC]
danc straw polls: do you prefer more free-form discussion or a discussion based on an outline or document.
16:26:36 [KendallC]
(ouch, i may have identified some jos contributions as alberto. mea culpa!)
16:28:28 [DaveB]
post a protocol use case AndyS :)
16:29:00 [KendallC]
it led to rob s suggesting we remove protocol from the charter. -wink-
16:29:12 [AndyS]
I haven't seen one that does not assume a common protocol if not local access
16:29:20 [KendallC]
(well, sorta suggesting it anyway...)
16:29:27 [alberto]
new uses cases will fit into the overview document - UC discussion should not stop at the moment - there is still a lot to discuss
16:29:31 [KendallC]
rob: :>
16:29:58 [KendallC]
mine didn't assume local access, but i'm not sure i know what you mean
16:30:44 [rob]
That introduction is already loading against self-contained RDF repositories...
16:31:35 [DaveB]
names please, speaker
16:31:44 [KendallC]
that's rob shearer
16:32:55 [rob]
Good point: we're querying RDF models; who parses is beyond our scope.
16:33:37 [KendallC]
danc's straw outline: a generic intro to the problem space (?); a list of use cases: email address, parts catalog, rss feeds; distilled technical requirements; relations to "related techs"
16:34:54 [Zakim]
16:35:13 [ericP]
i'm amused by the idea of it being a draft position
16:35:19 [KendallC]
danc: if yr interested in being an editor, write up a page-long outline of a first document and send to danc.
16:35:30 [Zakim]
16:35:41 [KendallC]
dajobe: asks for more use cases
16:36:05 [DaveB]
16:36:24 [KendallC]
kendallc: i asked for more protocol-oriented use cases
16:36:38 [rob]
Specific use cases are easier to shoot down!
16:36:48 [KendallC]
16:37:01 [KendallC]
rob: you *are* a trouble maker ;>
16:37:22 [DaveB]
I'm expecting uses cases to be discarded; it's good when we find them out of scope, not for this round of stuff.
16:37:23 [KendallC]
andys: would like to increase our general pool of UCs before *too much* sorting of them (err. I think?)
16:38:03 [KendallC]
the proposed document may do some of the work of sifting UCs
16:38:45 [rob]
I worry that we might "ignore" boring but relevent use cases.
16:38:50 [KendallC]
rob: yes
16:39:00 [KendallC]
editors shall have to guard against that
16:39:09 [KendallC]
(well IMO)
16:39:46 [KendallC]
alberto: would like the document to have a kind of use case "taxonomy" so that new UCs can be slotted into the right places.
16:40:10 [KendallC]
err, "taxonomy" too heavyweight. alberto wants a list of UC types, it seems.
16:40:33 [KendallC]
a lightweight categorization of UCs
16:41:13 [KendallC]
sorry, "taxonomy" my mishearing of alberto.
16:41:31 [rob]
I think we'll need 20-30 use cases in order to boil commonalities down to 5-7; let's not get complacent and focus too much on any one.
16:42:15 [AndyS]
Is this a good example of a UC section?
16:42:21 [DanC]
yes, starting with 20 to 30 and boiling to 5 to 20 is what I have in mind
16:42:40 [KendallC]
depends on what "boiling down" means
16:42:45 [rob]
It would be nice to get an idea of the full space, and then formally map the group to the 5-7 which represent them.
16:42:51 [DaveB]
(talking over each other on phone)
16:42:54 [rob]
(so that we know we haven't ignored the others)
16:43:27 [alberto]
requirements sound better - taxonomy is odd - I did not meant that - just some "organisation" of UCs
16:43:29 [KendallC]
andys: any inputs to the f2f
16:44:03 [KendallC]
danc: put f2f inputs on next week's agenda, danc will send mail
16:44:06 [DanC]
ACTION DanC: send mail to start discussion of ftf agenda
16:44:24 [Zakim]
16:44:25 [alberto]
16:44:25 [Zakim]
16:44:26 [Zakim]
16:44:27 [Zakim]
16:44:28 [Zakim]
16:44:29 [Zakim]
16:44:29 [Zakim]
16:44:30 [Zakim]
16:44:31 [Zakim]
16:44:33 [Zakim]
16:44:35 [Zakim]
16:44:37 [Zakim]
SW_DAWG()10:30AM has ended
16:44:39 [Zakim]
Attendees were Tayeb, +1.760.476.aaaa, +1.317.151.aabb, EricP, KendallC, DirkG, DanC, DanielK, Patrick, DaveB, AndyS, RobS, Pat_Hayes, JosD
16:44:42 [DanC]
ok, kendall, so you'll send proposed minutes in email?
16:44:48 [DanC]
RRSAgent, make logs world-readable
16:44:52 [DanC]
RRSAgent, make logs world-access
16:45:09 [KendallC]
danc: i shall do if you'll tell me where to find them on the web :>
16:45:15 [DanC]
RRSAgent, pointer?
16:45:15 [RRSAgent]
16:45:20 [KendallC]
16:46:42 [alberto]
alberto has left #dawg
16:47:13 [AndyS]
AndyS has left #dawg
19:04:33 [Zakim]
Zakim has left #dawg
19:06:07 [DanC]
RRSAgent, bye
19:06:07 [RRSAgent]
I see 3 open action items:
19:06:07 [RRSAgent]
ACTION: ericP to set up meeting registration [1]
19:06:07 [RRSAgent]
recorded in
19:06:08 [RRSAgent]
ACTION: item to patricks to write up his car-parts story [2]
19:06:08 [RRSAgent]
recorded in
19:06:08 [RRSAgent]
ACTION: DanC to send mail to start discussion of ftf agenda [3]
19:06:08 [RRSAgent]
recorded in