IRC log of dig on 2011-02-21

Timestamps are in UTC.

00:00:43 [danbri]
danbri has quit (Remote host closed the connection)
01:47:04 [oshani]
oshani (~oshani@c-71-233-151-72.hsd1.ma.comcast.net) has joined #dig
02:01:48 [oshani]
oshani has quit (Quit: Mama nidi!)
02:12:29 [oshani]
oshani (~oshani@c-71-233-151-72.hsd1.ma.comcast.net) has joined #dig
03:17:56 [melvster]
melvster has quit (Ping timeout: 276 seconds)
04:43:14 [webr3]
webr3 has quit (Ping timeout: 260 seconds)
04:47:03 [webr3]
webr3 (~nathan@host86-133-148-238.range86-133.btcentralplus.com) has joined #dig
05:37:54 [lkagal]
lkagal has quit (Quit: lkagal)
06:00:51 [melvster]
melvster (~melvster@p5797F1BD.dip.t-dialin.net) has joined #dig
06:30:03 [kennyluck]
kennyluck (~kennyluck@EM111-188-36-146.pool.e-mobile.ne.jp) has joined #dig
07:38:32 [kennyluck]
kennyluck has quit (Quit: kennyluck)
08:23:35 [kennyluck]
kennyluck (~kennyluck@EM111-188-142-78.pool.e-mobile.ne.jp) has joined #dig
08:32:22 [melvster]
melvster has quit (Ping timeout: 246 seconds)
08:43:30 [oshani]
oshani has quit (Quit: Mama nidi!)
09:11:41 [timbl]
timbl (~timbl@212.70.34.22) has joined #dig
09:26:42 [timbl]
timbl has quit (Quit: timbl)
09:53:00 [danbri]
danbri (~danbri@ip176-48-210-87.adsl2.static.versatel.nl) has joined #dig
10:25:02 [bblfish]
bblfish has quit (Remote host closed the connection)
10:26:47 [Ipaparrizos]
Ipaparrizos has quit (Remote host closed the connection)
10:28:26 [melvster]
melvster (~melvster@p579F90FA.dip.t-dialin.net) has joined #dig
10:41:45 [kennyluck]
kennyluck has quit (Quit: kennyluck)
11:45:34 [kennyluck]
kennyluck (~kennyluck@netDHCP-169.keio.w3.org) has joined #dig
11:53:22 [danbri]
danbri has quit (Remote host closed the connection)
12:12:09 [danbri]
danbri (~danbri@ip176-48-210-87.adsl2.static.versatel.nl) has joined #dig
12:18:05 [IvanHerman]
IvanHerman (~ivan@83.160.85.48) has joined #dig
12:30:37 [timbl]
timbl (~timbl@212.70.34.22) has joined #dig
12:44:50 [ericP]
timbl, ears on?
12:45:41 [ericP]
more specifically, ears on for RW dirs in SWObjects?
12:45:56 [ericP]
(sorry, "file-system-backed" is more precise)
12:46:04 [timbl]
Hi
12:46:06 [timbl]
yup
12:46:31 [ericP]
i just sent you an email with usage details
12:47:37 [ericP]
it will 404 on data/newFile.ttl, but i think we should check with chimezie and DAWG on what they're doing there
12:48:12 [timbl]
They are probably not making a data wiki
12:48:17 [timbl]
(or are they?)
12:49:12 [timbl]
So currently can edit existing files but not make new ones
12:49:13 [ericP]
well, they're making a correspondance for HTTP {GET,PUT,DELETE} /foo and INSERT DATA { </foo> { ... } }
12:50:04 [timbl]
The tabulator stuff and the existing servers which support it all give an empty resource when you accesss a new URL.
12:50:15 [ericP]
if i understand, the datawiki protocol doesn't distinguish between a non-existent resource and an empty one, right?
12:50:47 [timbl]
If you really don't want to then you'd have to put some header in to indicate that the resource could be inserted to
12:50:48 [ericP]
what if it read 404, plus the "i'm editable" header as an invitation to create?
12:50:59 [timbl]
right it doens't distinguish
12:51:00 [ericP]
soft minds run together
12:51:10 [ericP]
(re: "some header")
12:51:32 [ericP]
so there is an "i'm editable" header, isn't there?
12:52:18 [ericP]
MS- something, iirc
12:52:29 [ericP]
rep.addHeader("MS-Author-Via", "SPARQL");
12:52:48 [timbl]
Logiucally though to say a nonexistent document is editable is a stretch.
12:53:22 [timbl]
The user of tabulator needs to know the diff between a empty doc in the wiki an a real 404 on another server which is nota woiki
12:53:49 [timbl]
So the tabualtor would have to show it as an empty doc
12:53:58 [ericP]
i'd not expect the endpoint to return MS-Author-Via if it were not editable
12:53:58 [timbl]
with success not a red dot
12:54:41 [timbl]
Would the assumption be then that you can do an INSERT DATA with n previous PUT?
12:54:53 [ericP]
i believe so
12:55:05 [ericP]
not having thought about it for long, mind you
12:55:17 [timbl]
I think it bastardizes HTTP less to just say that the erver has an infinit enumber f empty files
12:55:58 [ericP]
hmm, i prefer preserving the distinction between non-existence and empty
12:56:34 [ericP]
has it worked yet?
12:56:47 [ericP]
oh, i overhauled the logging entirely
12:56:54 [timbl]
has what worked yet?
12:57:10 [ericP]
i was wondering if you'd poked at the new .dmg
12:57:19 [timbl]
I have to get or make a new version, right
12:57:19 [timbl]
new .dmg
12:57:26 [timbl]
Where is this things home page?
12:57:41 [ericP]
huh
12:57:46 [ericP]
good question
12:57:52 [ericP]
do people use the web for things like that?
12:58:29 [ericP]
presuming you mean the datawiki bits of SWObjects, i guess it would go in the sourceforge SWObjects wiki
13:00:26 [ericP]
it's still pretty bleeding edge (less than a day old), and almost entirely untested, so i'm not sure i'm bold enough to admit its existence
13:01:16 [ericP]
btw, snowing again here. might be good cc skiing
13:01:39 [timbl]
so where shoul di get a dmg from?
13:05:25 [ericP]
[delivered out of band]
13:06:22 [ericP]
the new mods don't change the old 404 behavior; they just back it with the filesystem
13:11:00 [timbl]
so any write goes back to disk immediately?
13:14:34 [timbl]
"rewrites even when you query a graph (doesn't know when it's
13:14:34 [timbl]
referenced for R vs. W" from one extreme to another
13:21:40 [IvanHerman]
IvanHerman has quit (Quit: bye guys)
13:25:34 [timbl_]
timbl_ (~timbl@212.70.34.22) has joined #dig
13:25:34 [timbl]
timbl has quit (Read error: Connection reset by peer)
13:25:35 [timbl_]
timbl_ is now known as timbl
13:41:48 [ericP]
there's a mutex on the DB around SPARQL operation execution (e.g. SELECT, INSERT...)
13:42:52 [ericP]
as each graph is referenced, the db checks to see if there's a filesystem map for that graph
13:43:10 [ericP]
if there is, it (re-)reads that graph from the disk
13:43:32 [ericP]
(leaving the graph empty if there's no corresponding file)
13:44:14 [ericP]
all these graphs are added to a dirty list, which is flushed after the op finishes and before the mutex closes
13:44:17 [ericP]
.
14:10:34 [oshani]
oshani (~oshani@c-71-233-151-72.hsd1.ma.comcast.net) has joined #dig
14:44:05 [oshani]
oshani has quit (Quit: Mama nidi!)
14:44:34 [oshani]
oshani (~oshani@c-71-233-151-72.hsd1.ma.comcast.net) has joined #dig
14:46:41 [oshani]
oshani has quit (Client Quit)
14:47:19 [oshani]
oshani (~oshani@c-71-233-151-72.hsd1.ma.comcast.net) has joined #dig
15:04:15 [oshani]
oshani has quit (Quit: Mama nidi!)
15:50:09 [lkagal]
lkagal (~lkagal@pool-96-237-240-136.bstnma.fios.verizon.net) has joined #dig
16:25:41 [amy]
amy (~amy@65-78-21-147.c3-0.wrx-ubr6.sbo-wrx.ma.cable.rcn.com) has joined #dig
18:58:14 [oshani]
oshani (~oshani@30-7-126.wireless.csail.mit.edu) has joined #dig
19:04:24 [timbl]
timbl has quit (Quit: timbl)
19:10:19 [presbrey]
ericP, timbl, I think it does distinguish
19:10:30 [presbrey]
200/empty vs. 404/empty
19:35:14 [oshani_]
oshani_ (~oshani@30-7-126.wireless.csail.mit.edu) has joined #dig
19:35:14 [oshani]
oshani has quit (Read error: Connection reset by peer)
19:35:14 [oshani_]
oshani_ is now known as oshani
20:07:47 [timbl]
timbl (~timbl@212.70.34.22) has joined #dig
20:51:51 [ericP]
2011-02-21T19:04:24Z -!- timbl [~timbl@212.70.34.22] has quit [Quit: timbl]
20:51:51 [ericP]
2011-02-21T19:10:19Z <presbrey> ericP, timbl, I think it does distinguish
20:51:51 [ericP]
2011-02-21T19:10:30Z <presbrey> 200/empty vs. 404/empty
20:52:06 [ericP]
presbrey, did you mean 404/non-existent?
20:53:29 [manu1]
manu1 has quit (Remote host closed the connection)
20:54:02 [manu1]
manu1 (~chatzilla@pool-96-240-178-55.ronkva.east.verizon.net) has joined #dig
22:29:49 [betehess]
betehess (~betehess@c-24-128-51-56.hsd1.ma.comcast.net) has joined #dig
22:42:05 [presbrey]
an empty RDF document is always produced
22:42:43 [presbrey]
404 indicates it hasn't ever been written
22:44:08 [presbrey]
it varies whether PUTing an empty RDF document results in a future 200
22:45:26 [presbrey]
by storage type and implementation year
22:47:20 [presbrey]
PUTing invalid RDF to an empty URI always remains a 404 document
22:47:54 [presbrey]
(afaik)
22:48:00 [presbrey]
you're porting this into swobjects?
22:48:21 [presbrey]
er implementing
23:03:45 [ericP]
presbrey, i think i might have accidentally done it this AM
23:04:18 [ericP]
(some time before dawn, was a bit hazy)
23:14:39 [ericP]
presbrey, actually, the conceptual functionality was there before, but created a new RdfDB which backs certain graph names with locally accessable files
23:15:01 [ericP]
("certain" = those that match 1 of n supplied regex transformations)
23:26:26 [presbrey]
cool
23:27:18 [presbrey]
are you saving them as .n3?
23:50:56 [ericP]
presbrey, .ttl -- just s p o