00:08:32 Hmm, still discussing this? 00:08:59 Apparently not :) 00:09:27 sorry, I was just replying to Mark Nottingham again :-) 00:10:08 Heh. Really *no problem*! 00:11:17 here: http://groups.yahoo.com/group/rss-dev/message/2255 00:12:26 Mark:- 00:12:28 > Yes, RDF is another layer of format on top of XML. However, 00:12:28 > it would be VERY difficult to get the IETF to approve another + 00:12:28 > extension to media types, IMHO. 00:12:33 sbp: which is a shame 00:13:30 Heh. You quote me :) 00:13:35 Quoting the modules spec :) 00:14:34 Hmm. Would you want RDF+RSS? Or RDF+XML? (Is there RDF+N3?) 00:14:40 yep :-) "why quote when you can quote a quote?", said Sean. 00:15:04 you'd want rdf+xml for XML RDF 00:15:19 rss+rdf+xml for RSS's weird thing 00:15:24 I would think, in general, that MIME types should focus harder on the encoding stuff. 00:15:26 and n3+rdf for N3 00:15:45 as, character encoding? yeah, that's also a big issue 00:15:54 No, I meant encoding in general. 00:16:23 what do you mean by "encoding"? it's quite an abstract term 00:16:25 I.e., if RDF means "data model" I'm not sure it's all *that* relevant to MIME type. 00:16:56 I could see, for example, rss+n3 for R(N3)SS1.0 00:17:01 yeah, you're probably right. But it's an instruction to the parser all the same 00:18:26 Hmm... odd note on GL 00:21:39 it's nice to see important news being circulated to the relevant working groups, though 00:27:40 DTRT is pretty cool for looking up RFCs... 00:42:17 mnot-laptop has joined #rdfig 00:42:45 hey, just the man! 00:42:54 Hi Mark 00:42:55 * mnot-laptop has evil, evil plans... 00:43:06 hello 00:43:14 * sbp wonders if he should start running now 00:43:44 re: foo+rdf+xml: 00:43:49 (from rfc3023) 00:43:53 A.14 What happens when an even better markup language (e.g., EBML) is 00:43:53 defined, or a new category of data? 00:43:53 In the ten years that MIME has existed, XML is the first generic data 00:43:53 format that has seemed to justify special treatment, so it is hoped 00:43:53 that no further suffixes will be necessary. However, if some are 00:43:54 later defined, and these documents were also XML, they would need to 00:43:56 specify that the '+xml' suffix is always the outermost suffix (e.g., 00:44:01 application/foo+ebml+xml not application/foo+xml+ebml). If they were 00:44:01 not XML, then they would use a regular suffix (e.g., 00:44:02 application/foo+ebml). 00:44:42 wow, mnot comes to argue in person! 00:45:18 * mnot-laptop should be working... IRC is the perfect antidote. 00:45:25 lol 00:45:32 antidote to what? 00:45:37 work, of course 00:45:40 Ah. 00:45:49 I thought you meant the antidote was working... which, I suppose it is 00:46:13 sbp, you do know that RSS 0.9 was specified as XML, not RDF too, right? 00:46:25 of course 00:46:58 AaronSw: infinite reference loop; standby for reboot. 00:47:10 AaronSw has left #rdfig 00:47:14 AaronSw has joined #rdfig 00:47:16 oops 00:47:31 mwahaha, I fixed that infinite loop bug last time this happened 00:47:59 * sbp wonders what that's all about 00:49:10 working is the antidote is working is the antidote... and so forth. Sorry, bad humour. 00:49:40 so... application/rss+xml ? 00:50:05 sure, if you must :-) 00:50:18 well, I'd really, really like to ;) 00:50:23 Do it! 00:50:28 Ignore the vile dissenter! 00:50:33 lol 00:50:43 I think you mean vile *heretic* dissenter. 00:50:54 yeah, bijan, come on 00:51:14 He's mired in facile utopean visions of RSS-as-it-may-be-in-his-grubby-paws not RSS-as-sanctified-on-the-pure-altar-of-pragmatic-usage! 00:51:24 pragamtists win again! 00:51:54 All I want to do is dispatch a feed URI to an aggregator without using cut-n-paste. 00:52:29 SethR has quit 00:52:48 Tough, you namby-pamby-mechanistic-automator-of-rabbits-swine-pig! 00:52:56 OK, I'll put together a draft and float it around Real Soon Now. 01:21:03 SethR has joined #rdfig 01:22:01 hey tim, did my diagram make any sense to you? 01:43:00 timbl-lap has joined #rdfig 01:43:00 tim has quit 01:46:08 [ServerNotice] Hi all. Carter will be disconnecting momentarily. 01:46:26 chihchun has quit 01:46:26 dc_rdfig has quit 01:46:26 sandro has quit 01:46:26 kifbot has quit 01:46:26 sbp has quit 01:46:26 bijan has quit 01:46:26 em has quit 01:46:26 Iron_SpermWhale has quit 01:46:26 oierw` has quit 01:46:26 grove has quit 01:46:26 mnot-laptop has quit 01:46:26 tav has quit 01:46:26 SethR has quit 01:46:26 danbri has quit 01:46:26 GabeW has quit 01:46:26 timbl-lap has quit 01:46:26 xbill has quit 01:46:26 AaronSw has quit 01:46:26 deltab has quit 01:46:54 AaronSw has joined #rdfig 01:46:54 deltab has joined #rdfig 01:46:54 em has joined #rdfig 01:46:54 kifbot has joined #rdfig 01:46:54 sandro has joined #rdfig 01:46:54 Iron_SpermWhale has joined #rdfig 01:46:54 dc_rdfig has joined #rdfig 01:46:54 chihchun has joined #rdfig 01:46:54 tav has joined #rdfig 01:46:54 danbri has joined #rdfig 01:46:54 sbp has joined #rdfig 01:46:54 GabeW has joined #rdfig 01:46:54 xbill has joined #rdfig 01:46:54 grove has joined #rdfig 01:46:54 bijan has joined #rdfig 01:46:54 oierw` has joined #rdfig 01:46:54 mnot-laptop has joined #rdfig 01:46:54 SethR has joined #rdfig 01:46:54 timbl-lap has joined #rdfig 01:46:56 This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome 01:50:31 mnot-laptop has quit 01:51:06 chihchun has quit 01:51:06 dc_rdfig has quit 01:51:06 sandro has quit 01:51:06 kifbot has quit 01:51:06 sbp has quit 01:51:06 bijan has quit 01:51:06 em has quit 01:51:06 Iron_SpermWhale has quit 01:51:06 tav has quit 01:51:06 SethR has quit 01:51:20 dc_rdfig has joined #rdfig 01:51:20 Iron_SpermWhale has joined #rdfig 01:51:20 sandro has joined #rdfig 01:51:20 kifbot has joined #rdfig 01:51:20 em has joined #rdfig 01:51:20 chihchun has joined #rdfig 01:51:20 sbp has joined #rdfig 01:51:20 bijan has joined #rdfig 01:51:26 SethR has joined #rdfig 01:53:27 tav` has joined #rdfig 01:56:26 SethR has quit 01:56:53 timbl-lap has quit 01:56:54 tim has joined #rdfig 02:01:10 [GlobalNotice] Hi all. If you're not watching the wallops, some intentional rehubbing is going on. Thanks. 02:01:10 tav` has quit 02:01:10 tim has quit 02:01:17 tim has joined #rdfig 02:01:30 tav` has joined #rdfig 02:04:19 tim has quit 02:19:22 sbp has quit 02:19:39 sbp has joined #rdfig 02:29:15 tav has joined #rdfig 02:51:49 mnot-laptop has joined #rdfig 03:27:03 FYI: http://www.mnot.net/papers/draft-nottingham-rss-media-type-00.txt 03:27:04 Label FYI not found. 03:29:40 just got email 03:29:58 chump it, please 03:30:41 squeeze me? 03:32:00 http://www.mnot.net/papers/draft-nottingham-rss-media-type-00.txt 03:32:00 A: http://www.mnot.net/papers/draft-nottingham-rss-media-type-00.txt from AaronSw 03:32:12 A:|The application/rss+xml Media Type 03:32:12 titled item A 03:32:57 * mnot-laptop wonders where the description of this nifty thing is... 03:33:24 do a /msg dc_rdfig help 03:33:33 thx 03:33:39 and/or see http://usefulinc.com/chump/ 03:37:52 now, if there were a link on http://rdfig.xmlhack.com/ that led to a resource with a media type of *ahem* application/rss+xml, I'd know how to include this in my own aggregator... assuming it's available in RSS? 03:42:51 * AaronSw looks for the question in there... 03:43:03 accidental question mark? 03:47:08 is http://rdfig.xmlhack.com/ available as an RSS feed? 03:47:55 yep -- index.rss 03:48:00 and each day too 03:48:05 aha 03:48:37 what's your aggregator? 03:48:45 is it friends with syndic8.com? 03:50:04 no 03:50:14 bespoke for an audience of one - me 03:50:26 :) 03:54:49 GabeW has quit 04:20:26 sbp has quit 04:40:40 draft revised to give equal time to all RSS formats. 05:08:59 That's better. 05:26:41 deltab has quit 05:29:42 deltab has joined #rdfig 06:44:16 xbill has quit 06:44:44 xbill has joined #rdfig 09:42:20 danbri has quit 09:54:28 dajobe has joined #rdfig 10:19:25 danbri has joined #rdfig 10:25:34 * danbri -> boston (back tommorrow) 10:25:41 back = back online! 10:25:43 bye 10:39:50 danbri has quit 10:40:20 dajobe has changed the topic to: Semantic Web Chat - http://www.w3.org/RDF/Interest/#irc 11:55:39 larsbot has joined #rdfig 12:07:47 chihchun has quit 12:10:15 chihchun has joined #rdfig 12:15:16 chihchun has quit 12:17:14 chihchun has joined #rdfig 12:53:34 dajobe has quit 13:13:05 logger has joined #rdfig 13:13:05 topic is: Semantic Web Chat - http://www.w3.org/RDF/Interest/#irc 13:13:05 Users on #rdfig: logger chihchun larsbot xbill deltab mnot-laptop tav tav` dc_rdfig Iron_SpermWhale sandro kifbot em bijan jang lasDesk @AaronSw oierw` grove 13:13:05 This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome 13:37:53 tim-gone has joined #rdfig 14:15:57 bwm has joined #rdfig 14:24:09 * em waves to bwm 14:32:53 tim-gone is now known as tim-lurk 14:37:10 Hey, Tim-Lurk. :-) Do you think the definitive/authoritative semantics for an XML doc should be retreived from the namespace address (which means we need to settle what goes there) or from some other address available via some attribute like "sw:grammarAnnotatedWithSemantics" ? 14:43:56 I think the definitive/authoritative semantics for any XML doc I publish should be retreived from the namespace address, but this doe snOT mean we have to settle what goes there. 14:44:09 Theer are parts of this it is importnat no to muddle. 14:44:43 The HTTP spec says that if you have dereferenced an http: URI then what you get is authoritative. 14:44:58 This doesn't mean you will always succed. 14:45:09 It doesn't mean you can't get the information from other souyrces. 14:45:58 A 3rd party can alwys make a document which makes information it wants to publish about some namesapce, too. 14:45:58 mysql 14:46:16 I think I understand that -- but if you do get something, what do you get? (we can use the media-type to made several possibilities, but still.) 14:46:48 If you ask for text- 14:47:25 If you ask for text/xml should you get the XML Schema or a human-readable XHTML page about the format? (maybe that's not the best example.) 14:50:49 Another thing that's been bugging me: is . ]]> an abomination or a reasonable way to let people and machines know how to interpret some text? 14:51:49 (It would give blindfold a nice way to know the authoritative grammar for not-really-XML text.) 14:54:07 Sandro, the URI spec is the one which defines the relationship between a URI and its meaning. 14:54:28 For URIs starting with "http:", it hands off to the HTTP spec. 14:54:50 The HTTP spec allows format negotaiation, and then hands off to the MIME type registry. 14:55:10 The MIME type registry fopr application/xlm hands of to the XML spec. 14:55:28 The XML spec hands off to the namespace URI. 14:55:35 Goto 1 14:56:15 This loop terminates with some well-known URIs of course which your machine knows about because the programmer read the spec which quotyed the URI. 14:57:08 (re your ne example above, that is I think wrong in that XML has notations for doingthat) 15:00:05 (NOTATION) 15:00:48 When you go non XML, then you have to fall back to MIME type. 15:02:35 dajobe has joined #rdfig 15:06:44 I don't see how that definition process answers the quesiton of whether a server should provide an XML Schema or XHTML Documentation (or RDDL, etc) at the namespace URI. 15:08:16 tav` has quit 15:08:22 And my n3 example *is* XML. In fact, it's XML in the traditional "markup" sense, although rather trivially. It's certainly not XML in the object-encoding sense, where the relationship information is supposed to be conveyed in the markup instead of the text. 15:16:32 My problem with using the media type is that there's no good way for a computer to retrieve the formal meaning of a media type. I suppose we could encode URIs into media types or something strange. 15:18:46 Well, here W3C could set up a very useful repository. 15:19:28 That seems architecturally okay to you? I guess since there's already an IANA registration bottleneck, another one is okay..... 15:20:42 Hmmm. he your example being XML .... hmmm. Does it actually help? I suppose it solves the problem of getting the URI.... then you just need an RDF Property explaing that it is a psedo XML namespace and pointing to the BFG instead of the schema. 15:21:17 If IANA went down the road of persistent on-line repositories, I am sure we could me it on that road. 15:22:12 Right now, ietf.org has no persitence gurantee. The RFCs have no URI offially. 15:22:20 yeah. 15:22:52 We could run an annotation-server style service -- "W3c, what do you know about application/xml ?". 15:23:16 Yeah. 15:23:45 There is a long-standing work item for "someone" to organize a transition from MIME type string to URI. 15:24:42 There's also the angle that media-types can be seen as relative URIs. We could (try to) define them as being relative to http://www.w3.org/media-type/ and define the behavior for certain media-type GETs of those addresses. 15:25:06 Yeah. 15:26:37 sandro: that'd be incredibly useful 15:26:56 I take it the work item is "longstanding" because of the incredible opportunity for it to bikeshed? 15:27:20 sandro, there's also the IETF's HTTP media-type storage 15:27:29 Yes, exactly - the transition would happen like that. 15:27:47 bikeshed? 15:28:24 "paint it blue!" "no, paint it blue!" 15:28:41 Ah .... exactly, do we need fire that can be fitted nasally? 15:29:05 - HHGTTG, Douglas Adams 15:29:19 Heheh. 15:29:30 There was an I-D a little while back that had a proposed encoding of URIs into media types, and RDDL has been making noise about new HTTP headers that would replace mime types with URIs. 15:31:21 Sandro, if you get more than one pience of useful information about emdia types, I would encourage you as a first step to code it up in RDF and stick it on our web site. You could start by scraping the IANA registry and then merge your own observations, and so on, including blindfold gammars. 15:32:04 http://www.w3.org/2001/mime/types (.rdf, n3) for example. 15:32:04 B: http://www.w3.org/2001/mime/types from tim-lurk 15:32:11 Oooops 15:32:36 logger, ignore that 15:32:36 I'm logging. I don't understand 'ignore that', tim-lurk. Try /msg logger help 15:33:09 * logger is already logging 15:38:29 sbp has joined #rdfig 15:51:15 That sounds pretty good, Tim. You prefer one big file ("types") instead of directory structure matching the type names themselves "type/text/html"? The latter seems like it could evolve in the right direction better. In either case it would be (.rdf, .n3, ...) files. In the latter case, they'd only have a few things in them. 15:52:26 sbp has quit 15:59:39 Also, the RFC's call it "media type" not "mime type". I assume we should use .../2001/media/type... ? 15:59:59 my vote is for separate files 16:01:06 Some FTP servers will zip up directories for you. I can imagine asking for the whole collection or just a part.... Of course, I'd prefer a real RDF query interface there anyway. 16:05:32 So Aaron, how would you get from the XML namespace to the blindfold grammar which tells you when the document means? 16:05:48 I'd probably go with RDDL 16:06:20 Yeah. Who's objecting most to RDDL/ 16:08:41 I don't know of anyone objecting to it... it allows humans to read it, XML folks to read it, and can parse into RDF, as well as let you embed other stuff in it. 16:08:52 dajobe has quit 16:09:05 That's what I did for the RSS 1.0 namespace: http://purl.org/rss/1.0/ 16:09:21 And I'm trying to convince Dublin Core to use it, but they're worried they won't be able to keep the RDDL up-to-date. 16:12:08 larsbot has quit 16:12:14 GabeW has joined #rdfig 16:15:11 Hrm. I can't quite mesh my world-view with that of http://www.openhealth.org/RDDL/purposes/, but I guess I can just use my own for this. 16:16:43 They're just uris. 16:16:51 which makes them perfect for RDF translation 16:17:02 rddl has all the hallmarks of an excellent spec: the details are almost shorter than the acronym 16:17:10 Heheh! 16:17:22 Yeah -- but presumably some applications will care about specific values. So I need to understand if my link makes sense for those applications. 16:17:40 I'm not being facetious (well, not bvery) - it's a sign of strength imho 16:17:47 I agree. 16:22:08 And of course there's the forking of http://www.w3.org/2001/rddl/ (which XMLSchema did, I guess, since rddl.org has no persistence guarantee.) 16:23:04 Hmm, you consider making a backup of the DTD forking? 16:23:27 They still use the RDDL Public ID. 16:24:04 Oh, okay..... 16:25:02 So no one is pointing to http://www.w3.org/2001/rddl/ instead of rddl.org for their rddl:resource links? 16:25:21 I don't think so. 16:25:32 okay. 16:26:12 XMLSchema certainly doesn't. 16:27:34 good. 16:28:00 If it's good enough for XMLSchema and AaronSw, it's good enough for me. :-) 16:28:11 Heh, heh. :-) 16:29:04 Wow, there's an awful lot of RSS feeds. 525k feeds, 4388 Live, 300 dead and 568 unparsable. 16:29:47 Is there some directory of all of them? 16:29:55 Yep, www.syndic8.com/ 16:30:01 Numbers from http://www.syndic8.com/polllist.php 16:30:14 It started just recently, so we've been filling it up with data the past few weeks. 16:30:42 It's incredibly effective -- it makes me excited to see how many people are willing to contribute to ensuring people publish well-formed/valid XML. 16:31:09 173 users. 16:31:49 What's the best (linux or web) UI for RSS feeds? 16:32:14 Hmm. 16:32:38 Perhaps AmphetaDesk -- http://disobey.com/amphetadesk/ 16:32:43 developer is a friend of mine. 16:33:29 I'm also a developer on peerkat, http://peerkat.sourceforge.net/ 16:35:47 Do either of them do good things with keeping track of what you've seen, or is that too 1980's? 16:36:07 Heheh. 16:36:20 No, I keep trying to convince them to do that but they haven't yet. 16:37:46 I've spent a long time understand how to merge different information and media sources, and I can't quite figure out where RSS fits in. But I haven't really used it, except peripherally as a use on sites which use it (slashdot and its ilk). 16:39:07 do you mean merge as in, "see it all on the same page" or merge as in "intermix the feeds into one big list"? 16:39:39 mnot-laptop has a nice rss setup 16:41:05 thx. Unfortunately, it's not ready for prime time (but what is?) 16:41:22 neither, really -- I mean that (circa 1988) it was pretty obvious that e-mail and instant messages and bulletin boards were all on the same spectrum, and the question was how to flesh out the spectrum so we could share tools and work in other points on it. That's why I'm into RDF. :-) 16:42:06 Ah. 16:42:37 sandro, have you seen threadsML? 16:42:39 .google threadsml 16:42:42 .google threadml 16:42:51 xena has joined #rdfig 16:42:56 .google threadml 16:42:56 no results found. 16:42:59 From that perspective slashdot (etc) were a huge step backwards. Now I see it more as a step sideways... Similarly RSS. 16:43:15 ThreadsML: http://www.quicktopic.com/7/H/rhSrjkWgjnvRq 16:43:55 Yeah, whatever -- I think we need to solve some SemWeb infrastructure problems first. 16:44:15 At least I got them to use RSS 1.0 ;) 16:44:18 (or I do.... obviously other people will proceed as they like.) 16:44:21 RSS just restricts RDF in a way that allows more people to understand (or perhaps ignore) what's going on, I think 16:46:36 mnot, you're on the list of people I really should talk to about XML-RDF-Convergence. But I haven't made sense of your e-mailed proposal yet, so I may not be able to have an intelligent conversation about it. 16:47:25 Are you going to Nice? 16:48:42 Unfortunately, no. Not because of any terrorist threat or budget problem; these things are minor annoyances to the truely dedicated AC rep. The thing that *will* give me pause, however, is a pregnant wife with a certain look in her eye. *sigh* 16:49:34 I'm on a "personal" travel ban until at least the end of December. 16:50:04 congrats mnot-laptop on upcomming family addition! 16:50:14 * em joins and reads logs 16:50:27 Heh heh. Congrats. My pregnant wife isn't due till May, so she's coming to Nice. 16:50:31 Thanks! I understand there's reason to return them. 16:50:47 :) 16:50:54 Ooh, mnot has the opportunity to take the record for youngest child reading up on RDF! 16:51:22 sandro: cool; congrats. 16:51:30 Ah, congrats Sandro! 16:51:31 We'll see... ;) 16:51:32 * em would happy relinquish the record 16:51:46 Just paste some nodes and arcs into these photos... http://www.mnot.net/photo/people/charlie/baby.html 16:52:33 Aaron, you're a goof. :-) 16:52:55 This statement has already been asserted several times, methinks. ;-) 16:53:19 sorry... you need more definative proof of active particpation... ala http://www.w3.org/2001/09/06-ecdl/slide4-0.html 16:53:58 * em will be happy to not that mnot-laptop and sandro should have more books to choose from by the time their new additional arive 16:53:59 sandro: so, my proposal is really just a bit of glue that allows you to extract statements from XML. I think it's quite similar to blindfold, except that it leverages the structure in an XML schema. 16:54:14 ah, that's beautiful! 16:55:30 * mnot-laptop thinks a background in caching could be more useful to start with... in terms of keeping food down, diapers, etc. caching is much more useful technology than being able to model the product. 16:55:55 Yeah -- I just didn't understand some of you terms and structures well enough to udnerstand possible differences. But I'm fairly unschooled in Schema. My hope is to write a blindfold grammar for XML-schema with annotations. That will allow either approach to work, with my code. When I'm actually trying that, I'll have to grapple the issues you've presumably solved. (and then I'll see if I agree :-) 16:56:48 * sandro never thought of parenting as cache-management before. 16:56:49 mnot-laptop... does The Schema Adjunct Framework e.g. http://www.extensibility.com/resources/saf.htm and the like show up on your radar screen at all? 16:57:14 I'm pretty new to schema as well. I don't know if there are any unique technical problems being solved; it's more trying to address the social problems of getting people to use RDF and giving them easy hooks to get into it. 16:57:22 the notion of adding semantics to structure, etc 16:57:31 em: yes, had a look at SAF. 16:58:19 * em applauds mnot-laptop "social problems of getting people to use RDF and giving them easy hooks to get into it." 16:58:53 * mnot-laptop bows graciously 16:59:18 * em wonders if anyone took away an action item from the previous IMT URI discussion on todays log? 16:59:24 SAF is good, however I needed to be more tightly tied into the structure to be able to exploit it, so attributes in another namespace and appinfo seemed the way to go. 17:00:01 mnot-laptop, agreed 17:00:38 mnot-laptop is now known as mnot 17:02:12 em, what sort of action? 17:03:34 "There is a long-standing work item for "someone" to organize a transition from MIME type string to URI." 17:04:25 (a) either pursuing the social side of this with IANA or (b) offering additional trusted services "e.g. library of congress... what can you tell me about text/xml" 17:04:43 s/or/and\/or 17:04:45 There's an I-D about that out there somewhere, IIRC. The question is, did you want to be able to control what's at the end of them? 17:05:38 dajobe has joined #rdfig 17:06:53 mnot (if your answering my question/request?) can you dig up the ID? For starters I'd like a URI. Requesting whats at the end of them at that point is gravey... but of course, I'd like some machine processible information ala RDF 17:07:09 they already have URIs... 17:07:28 http://logicerror.com/mediatype 17:07:36 -> http://logicerror.com/mediaType-uri 17:07:53 -> ftp://ftp.isi.edu/in-notes/iana/assignments/media-types/ 17:08:17 As documented in http://rfc2048.x42.com/ 17:09:32 AaronSw, see point about either pursuing the social side of this with IANA ... more specifically wrt persistent naming policy and connical uri declaration. From there we work on what might be resolved when one dereferences ftp:// http:// IANA 17:09:54 What is more persistant than being spelled out in an RFC? 17:10:06 RFCs are mirrored across the world. 17:10:07 I think the idea is to have authoritative information at the location. For current-style media types that probably just means RDF pointers to definition documents, and maybe some non-normative pointers to other information. I'd like it to include normative poitners to formal semantics provided in an RFC, but I'm not sure that will happen. Maybe we shouldn't worry about non-XML media types. 17:10:15 and it seems unlikely those URIs would be reused. 17:11:25 Hmm, Donald E. Eastlake 3rd used to work at Cybercash. ftp://ftp.isi.edu/in-notes/iana/assignments/media-types/application/cybercash 17:11:26 Didn't know that. 17:11:39 agreed... the point is *someone* needs to contact, educate, help explain, etc. we're been talking about this for years in Dublin Core but no one really has followed up on this... thats why i was asking about action items. 17:11:58 contact who? educate what? 17:12:15 Anyway, those URIs are documented in in plain old ASCII with all the authoritative information sent to IANA. 17:12:19 we're coming up with good ideas that i wish we would someone formalize for others (e.g. Semantic Web Challanges) or have individuals own 17:12:19 * sandro 's son is asking about em's son, having seen the web picture now. 17:12:30 * em waves to sandro's son 17:12:38 * em proxies for ales 17:12:57 Heheh, /2001/sw/ should have an Ask Alex section. 17:13:09 lol 17:14:08 * sandro 's son is very confused about emotes. What does it MEAN when he SAYS he waves to me? Is he actually waving? 17:15:20 AaronSw.. btw, do you have somehting that builds an RDF schema from ftp://ftp.isi.edu/in-notes/iana/assignments/media-types/media-types ? 17:15:39 Or just instance data 17:15:43 No... what would be in this RDF schema? 17:15:51 s/schema// 17:16:34 No, but i don't think it'd be too hard. 17:16:35 * AaronSw ducks 17:16:43 http://search.ietf.org/internet-drafts/draft-mealling-iana-urn-01.txt 17:16:44 C: http://search.ietf.org/internet-drafts/draft-mealling-iana-urn-01.txt from mnot 17:16:49 oh, bugger 17:16:59 C:|An IETF URN Sub-namespace for Registered Protocol Parameters 17:16:59 titled item C 17:17:12 thx 17:17:35 So, it's a URN, but there are people busily working away at allowing us to dereference those, so no problem, right? 17:18:05 DanC is against it, airc 17:18:10 hmm, as I recall 17:18:11 the draft? 17:18:33 using URNs instead of plain-old HTTP uris 17:18:43 ah. 17:19:07 Well, I'd think this would be a perfect use case for URNs; if he's against this, is he against URNs in general? 17:19:28 Well, yes... 17:19:34 ah again. 17:19:54 I'd think you'd have found this out by now being on the URI Cabal and all 17:19:57 ;-) 17:20:05 Oh, I only play one on TV. 17:20:06 Aaron, are you claiming the ftp.isi.edu URIs for media types are "the identifiers" for those types, or simple "some" identifiers? Fetching from them isn't very useful. 17:20:16 Not useful? 17:20:18 What do you mean? 17:20:51 Well, you don't get anything machine-understandable, do you? :-) 17:21:04 ASCII! Everyone understands ASCII. 17:21:11 Sure, it's low-tech, but boy does it last. 17:21:21 back in my day... 17:22:12 How could they be anything but 'some'? Sure, they're well-known, but they aren't unique. IIRC there's a long-lived transition away from isi going on... 17:22:22 * sandro knows Aaron knows the difference between ASCII and RDF in terms of machine understanding. 17:22:31 :-) 17:22:43 It's low-tech, but that's how the IETF does things 17:22:51 run an annotation service if you want to say more 17:23:03 I want to get back a formal grammar, with semantic annotations, for the media type. 17:23:53 It would be nice to have an authoritative service, where the annotated grammar is approved by the IETF process. 17:24:20 Well you just go convince the IETF of that... heh! 17:24:22 But I guess there's no reason one can't run an amorphous service, and bits in it can be marked authoritative as appropriate. 17:26:21 Yeah, that'd be useful. 17:26:30 I mean, if they do go with URNs that's really all you can do. 17:27:22 Unless URN resolution starts working. :-) But maybe that's just an annotation service. 17:27:57 right 17:28:03 c18n or whatever they call it 17:29:24 what's c18n ? 17:30:16 http://images.google.com/images?q=tbn:3amebJIUJ2Q:kaladarshan.arts.ohio-state.edu/China/yungang/jpgs/c18/c18n/C7259Yungang18NWaESiBodhsHd.jpg 17:30:18 D: http://images.google.com/images?q=tbn:3amebJIUJ2Q:kaladarshan.arts.ohio-state.edu/China/yungang/jpgs/c18/c18n/C7259Yungang18NWaESiBodhsHd.jpg from sandro 17:30:18 contextualization 17:30:24 dajobe: silly chump 17:30:30 dajobe: silly chump 17:30:32 heheheh! 17:30:38 D:silly chump 17:30:38 commented item D 17:30:38 take out the space 17:31:04 c15n 17:31:06 sorry 17:31:50 Contextualization 17:32:11 Contextualization of Resolution 17:32:25 from the let's-invent-long-words dept. 17:33:18 antidisestablishmentarianism = a26m 17:35:24 pneumonoultramicroscopicsilicovolcanoconiosis = p44s 17:35:30 that's my favorite word 17:37:07 lets not chump http://www.dictionary.com/doctor/faq/p/pneumono.html 17:37:47 heheh 17:38:46 tav` has joined #rdfig 17:42:32 mnot returns and reads... 17:42:57 ah, c15n - I sat in on that. Was interesting in a very theoreticaly way; didn't go *anywhere* 18:08:33 tav has quit 18:13:49 tav`` has joined #rdfig 18:14:35 dajobe: you around and tracking rdf-interest? 18:15:00 yes and maybe 18:15:23 specfically Vassilis email wrt RDF Syntax 18:15:41 subject " RDF/XML Containers: Syntax & Semantics" ? 18:15:45 yes please 18:15:50 is all about test cases from an issue brian wrote 18:16:11 we need to prob. revisit that mess again 18:16:15 Vassilis (et.al) has large scale running code based on original RDF specs 18:16:32 I've asked them to review the new rounds of docs to make sure nothing hurts 18:17:18 they (sigh) made comments on mapping to triples again, despite it not actually being in the refactoring doc. 18:51:38 SethR has joined #rdfig 18:52:34 http://robustai.net/mentography/reificationContext.gif 18:52:34 E: http://robustai.net/mentography/reificationContext.gif from SethR 18:52:52 E:| The mentography of reification 18:52:53 titled item E 18:54:03 E: [http://lists.w3.org/Archives/Public/www-rdf-logic/2001Oct/0185.html|also see posting in RDF-LOGIC] 18:54:03 commented item E 18:54:56 E: comments as always welcome 18:54:57 commented item E 19:03:01 mnot has quit 19:13:16 bwm has quit 19:14:08 tav` has quit 19:14:08 tav`` has quit 19:20:19 SethR has quit 19:35:55 tav has joined #rdfig 20:42:27 look has joined #rdfig 20:45:04 * sandro gets lucky. He decides "grammar" is not the right term for what blindfold works with. He decides "formal language definition" is probably the most-correct term. Then he notices the blindFoLD name still works well. :-) 20:45:16 heheh 20:45:26 That's quite lucky 20:46:13 I thought so. :) 21:09:47 tav`` has joined #rdfig 21:17:40 mnot has joined #rdfig 21:49:22 dajobe has quit 21:57:52 sbp has joined #rdfig 22:20:20 sbp has quit 22:21:25 Has anyone written a classified ad server that outputs RDF? 22:23:08 Hmm... don't think so. 22:23:51 OK, that's one of my pipe dream projects