IRC log of dig on 2006-07-25

Timestamps are in UTC.

00:18:56 [lkagal]

lkagal (n=lkagal1@234.68.171.66.subscriber.vzavenue.net) has joined #dig

02:17:59 [EliasT]

EliasT (n=EliasT_@c-71-232-132-142.hsd1.ma.comcast.net) has joined #dig

03:13:48 [DanC_lap]

DanC_lap (n=connolly@64-126-89-30-dhcp-kc.everestkc.net) has joined #dig

03:56:06 [DanC_lap]

DanC_lap has quit (Read error: 110 (Connection timed out))

04:13:15 [EliasT]

EliasT has quit ("Trillian (http://www.ceruleanstudios.com")

04:52:02 [crowell]

crowell (n=crowell@EASTCAMPUS-NINE-FORTY-EIGHT.MIT.EDU) has joined #dig

05:02:48 [crowell]

crowell has quit ()

06:26:23 [sheets]

sheets (n=dsheets@EASTCAMPUS-EIGHT-SEVENTY-NINE.MIT.EDU) has joined #dig

07:01:12 [sheets]

sheets has quit (Read error: 110 (Connection timed out))

11:54:24 [RalphS]

RalphS (n=swick@30-7-207.wireless.csail.mit.edu) has joined #dig

12:55:00 [timbl]

timbl (n=timbl@30-7-145.wireless.csail.mit.edu) has joined #dig

13:25:51 [claudiof]

claudiof (i=claudiof@31-35-17.wireless.csail.mit.edu) has joined #dig

13:28:39 [sheets]

sheets (n=dsheets@EASTCAMPUS-EIGHT-SEVENTY-NINE.MIT.EDU) has joined #dig

13:56:32 [evoli]

evoli (n=yuhsin@DEPARTMENT-OF-ALCHEMY.MIT.EDU) has joined #dig

14:50:37 [yosi_s]

yosi_s (n=chatzill@static-71-243-122-114.bos.east.verizon.net) has joined #dig

15:09:26 [DanC]

timbl, somehow you mailed that FOXHOLES thing from your calendar, but I don't see it there now.

15:09:32 [DanC]

do you plan to attend?

15:10:10 [DanC]

I do see a tabulator meeting in your calendar on Thursday. I'm at risk; considering taking the day off.

15:12:23 [timbl]

webcal://www.csail.mit.edu/events/eventcalendar/ical/csail.ics

15:12:35 [DanC]

aha

15:12:54 [timbl]

We have Dan Atkinds/NSF here till 3.

15:13:26 [timbl]

s/d//

15:13:34 [timbl]

I might try to rtuish down after

15:13:36 [DanC]

does that "We" include me? I don't recall inviting him. ;-)

15:14:04 [DanC]

oh... and webcal: ... wash your mouth with soap! ;-)

15:15:52 [dsheets]

dsheets (n=dsheets@30-6-40.wireless.csail.mit.edu) has joined #dig

15:16:09 [sheets]

sheets has quit (Nick collision from services.)

15:16:27 [dsheets]

dsheets is now known as sheets

15:16:59 [sheets]

timbl or DanC, should triples in the store ever change or simply be amended? For instance, if I'm logging the source retrieval and parsing in the store, should I just change the relevant status triples or should I timestamp each event and then when I need the data look for the new event?

15:18:07 [DanC]

"status triple" is something I'm not familiar with. I don't really know how the tabulator works in any detail.

15:21:25 [ryanlee]

ryanlee (n=ryanlee@30-6-75.wireless.csail.mit.edu) has joined #dig

15:21:51 [sheets]

"status triple" is just a statement about the internal state of the machine. For instance, the status of loading resource X will be represented in a triple.

15:22:22 [sheets]

This is more of a best practice question than anything.

15:47:13 [DanC]

has the status been a triple for a long time? or are you making it into a triple? my initial inclination is: don't.

15:48:15 [sheets]

The status is being made into a triple

15:48:22 [sheets]

Don't what? Change it in the store?

15:48:31 [DanC]

don't make it into a triple

15:48:43 [sheets]

erm...

15:48:59 [sheets]

This is http://dig.csail.mit.edu/issues/tabulator/issue24

15:49:51 [sheets]

Do you know of any ontologies that represent things in this domain? Right now I'm writing our own when I need new semantics.

15:49:54 [DanC]

I guess my picture of the target is different. My picture is: whatever data tabulator picks up from the world, it should pass on as RDF. But its internal state is not something I'm terribly interested in.

15:50:25 [DanC]

I know ericp did some work on an http ontology, and it was used in one or two places.

15:50:46 [sheets]

DanC, so instead of "I failed to retrieve this document" make it "This document retrieval returns this error code"?

15:51:12 [sheets]

I mean, I'm planning on including both right now

15:51:22 [sheets]

But you think the former is unnecessary?

15:51:35 [DanC]

I'd say "there is an http response whose code is 404"

15:51:54 [DanC]

I'd say "there is an http response about <doc1> whose code is 404"

15:52:33 [sheets]

DanC, sure, but what about "I've sent out a request for <doc1>" or "I've received headers for <doc1>"?

15:52:59 [sheets]

or "I've parsed 1000 triples in <doc1>"

15:53:21 [sheets]

This information will need to be consumed and displayed in gui elements eventually but should it be in the store?

15:53:27 [DanC]

ah... found it: http://www.w3.org/1999/xx/http

15:53:39 [sheets]

ah great

15:53:47 [DanC]

I'm afraid it doesn't have much in it

15:54:07 [sheets]

yes... that might change

15:55:33 [DanC]

Perhaps "there is a request for <doc1>" is worth capturing. I wouldn't bother with "I've parsed 1000 triples". In general, be careful about putting stuff in RDF when you don't have a clear use case for consuming it.

15:56:12 [sheets]

ok, sounds good

15:57:06 [DanC]

if you put "there is a request for <doc1>" then link to it when you put "there is a reply" in

15:57:50 [DanC]

http://www.w3.org/2000/10/swap/pim/email is reasonably complete, I think...

15:57:52 [sheets]

I'm thinking something like a chain of actions might be good, yes

15:57:56 [sheets]

ok

15:58:25 [DanC]

ah... I use http://www.w3.org/2000/10/swap/pim/email#references rather than in-reply-to

16:00:45 [DanC]

hmm... http://www.w3.org/2000/10/swap/pim/email should have a link to some of the things that use it, like aboutMsg.py and MASE, and perhaps to http://esw.w3.org/topic/EmailVocabulary

16:25:57 [yuhsin]

yuhsin (n=yuhsin@NO-KNIFE.MIT.EDU) has joined #dig

16:26:14 [evoli]

evoli has quit (Nick collision from services.)

16:26:15 [yuhsin]

yuhsin is now known as evoli

16:26:16 [evoli_]

evoli_ (n=yuhsin@DEPARTMENT-OF-ALCHEMY.MIT.EDU) has joined #dig

16:29:31 [evoli]

evoli has quit (Client Quit)

16:30:01 [evoli_]

evoli_ has quit (Client Quit)

16:31:41 [evoli]

evoli (n=yuhsin@DEPARTMENT-OF-ALCHEMY.MIT.EDU) has joined #dig

16:31:58 [eikeon_]

eikeon_ (n=eikeon@gandalf.mindlab.umd.edu) has joined #dig

16:34:39 [eikeon_]

eikeon_ has quit (Excess Flood)

16:36:23 [eikeon_]

eikeon_ (n=eikeon@gandalf.mindlab.umd.edu) has joined #dig

16:56:17 [lkagal]

lkagal has quit ()

17:07:56 [harveyj]

harveyj (n=harveyj@30-5-253.wireless.csail.mit.edu) has joined #dig

17:08:42 [sheets]

timbl, http://www.niem.gov/

17:09:30 [sheets]

They even have a pretty little java class viewer applet: http://niem.gtri.gatech.edu/iepd-ssgt/DataModelViewer.do

17:17:43 [harveyj]

harveyj has quit ()

17:52:44 [eikeon_]

eikeon_ has quit ()

17:52:59 [lkagal]

lkagal (n=lkagal1@30-5-242.wireless.csail.mit.edu) has joined #dig

17:59:46 [eikeon_]

eikeon_ (n=eikeon@gandalf.mindlab.umd.edu) has joined #dig

18:13:51 [claudiof]

claudiof has quit ()

18:25:58 [sheets]

So jsUnit doesn't play nicely with anything asynchronous

18:26:49 [sheets]

and the spec for XHR doesn't mention it but events don't get dispatched when XHR is called synchronously

18:31:58 [sheets]

DanC, http://www.openqa.org/selenium/ looks like it can do asynchronous testing but there would be activation energy. Thoughts?

18:33:02 [DanC]

activiation energy... meaning we'd have to learn to use it?

18:34:23 [DanC]

sorry, my stack is about 6 deep just now. don't have bandwidth for this.

18:35:22 [sheets]

DanC, activation energy = learning curve + lost time; no problem about your stack, don't worry about it; selenium looks like it can also test ajax UI

18:40:13 [harveyj]

harveyj (n=harveyj@pool-71-126-228-200.bstnma.east.verizon.net) has joined #dig

19:06:18 [hmslydia]

hmslydia (i=Lydia@30-5-13.wireless.csail.mit.edu) has joined #dig

19:18:20 [sheets]

So far for acceptable tabulatable mediatypes, I have: 'application/rdf+xml','application/owl+xml','application/xml','text/xml','text/html' Others?

19:19:45 [sheets]

what about n3, turtle, and triples?

19:19:52 [sheets]

erm... n-triples

19:20:30 [claudiof]

claudiof (i=claudiof@31-35-17.wireless.csail.mit.edu) has joined #dig

19:20:58 [timbl]

If you have text/html, give it q=0.5 as you want RDF by practice

19:21:09 [timbl]

n3 family: when we can parse them

19:21:26 [timbl]

text/n3+rdf

19:21:40 [sheets]

ah, i have application/xhtml+xml

19:21:58 [sheets]

hmm... the precedence is by q= not ordering?

19:22:11 [sheets]

also, how respected is 406?

19:22:55 [timbl]

by q=

19:23:04 [sheets]

eep, ok

19:23:05 [timbl]

How do you man, how respected?

19:23:28 [sheets]

Most servers will just send me whatever and not send a 406 if they can't satisfy my Accept:

19:23:58 [timbl]

Yes. I don't know the rules offf the top of my head, i am embarassed to say

19:24:26 [sheets]

the spec claims it should send a 406 but I've never seen that happen

19:24:32 [timbl]

Obviously if we get somethng we didn't want, we just ignore it.

19:24:33 [sheets]

I'll proceed with caution, thanks

19:24:37 [sheets]

right

19:24:37 [timbl]

Oh, it does happen.

19:24:47 [sheets]

ok, great

19:25:18 [timbl]

If we get stg we can handle, ten we process it. Note that anything containing "+xml" can be expected to be XML and therefore GRDDLable

19:26:28 [sheets]

erm... can I express that with Accept wildcards or do I have to accept all application/* and text/* and then do string processing for +xml at the end?

19:27:54 [timbl]

I think you just say accept application/xml

19:28:04 [sheets]

oh, all right

19:28:09 [timbl]

on th ereceiving end, yes you parse as XML anything with "+xml"

19:28:30 [sheets]

ok

19:28:42 [sheets]

so application/xml, text/xml, and text/html should cover me

19:28:54 [sheets]

until we start dealing with turtle and n3 and such

19:29:54 [timbl]

say explicitly you accept application/rdf+xml as that is what RDF servers key off. To say you grok RDF

19:30:26 [sheets]

yes... not at all

19:30:46 [sheets]

don't i have to tell them application/owl+xml and application/xhtml+xml then, too?

19:31:26 [claudiof]

claudiof has quit ()

19:32:54 [sheets]

Also, we should be able to handle rss1.0 feeds but they aren't valid XML -- what gives?

19:33:31 [sheets]

ah, never mind, firefox was being too user friendly

19:33:59 [timbl]

OWL should be served as application/rdf+xml

19:34:22 [sheets]

ok

19:35:19 [claudiof]

claudiof (i=claudiof@31-35-17.wireless.csail.mit.edu) has joined #dig

19:57:24 [eikeon_]

Does cwm depend on n3 being served up as a particular mime type when fetching via log:semantics?

19:59:18 [eikeon_]

... yosi_s, DanC, timbl?

19:59:32 [timbl]

Well, it will also guess text/plain as a seconf best, but the proper way yes it it gets a content type.

19:59:33 [DanC]

good question, eikeon

19:59:45 [DanC]

eikeon, why do you ask?

20:01:04 [timbl]

timbl has quit ()

20:02:07 [eikeon_]

Trying to debug why end to end demo is not working

20:02:17 [eikeon_]

Which content type?

20:02:35 [eikeon_]

application/n3?

20:02:51 [eikeon_]

text/rdf+n3?

20:03:34 [DanC]

z.add_header('Accept', 'text/rdf+n3, application/rdf+xml')

20:03:43 [DanC]

in swap/webAccess.py

20:04:02 [eikeon_]

Thank you... will try that.

20:13:27 [RalphS]

RalphS has quit ("bye for today")

20:26:27 [ryanlee]

http://simile.mit.edu/repository/longwell-demos/

20:38:49 [timbl]

timbl (n=timbl@30-7-145.wireless.csail.mit.edu) has joined #dig

20:45:05 [claudiof]

claudiof has quit ()

21:18:12 [peace-keep]

peace-keep (i=alerer@30-6-168.wireless.csail.mit.edu) has joined #dig

21:19:19 [peace-keep]

peace-keep has quit (Client Quit)

21:23:17 [peace-keep]

peace-keep (i=markus@30-6-168.wireless.csail.mit.edu) has joined #dig

21:25:51 [peace-keep]

peace-keep has quit (Client Quit)

21:26:03 [ryanlee]

ryanlee has quit ()

21:30:21 [eikeon_]

eikeon_ has quit (Read error: 104 (Connection reset by peer))

21:48:54 [evoli]

evoli has quit ("Lost terminal")

22:34:16 [timbl]

timbl has quit ()

22:34:58 [crowell_work]

crowell_work has quit ("leaving")

23:03:16 [hmslydia]

hmslydia has quit (Read error: 110 (Connection timed out))

23:29:38 [dsheets]

dsheets (n=dsheets@EASTCAMPUS-EIGHT-SEVENTY-NINE.MIT.EDU) has joined #dig

23:32:07 [sheets]

sheets has quit (Read error: 110 (Connection timed out))

23:33:30 [yosi_s]

yosi_s has quit ("Chatzilla 0.9.73 [Firefox 1.5.0.4/2006060817]")

23:40:40 [dsheets]

dsheets is now known as sheets

23:54:51 [sheets]

sheets has quit (Remote closed the connection)