00:07:41 dwalker has joined #echo 00:36:53 rubys has joined #echo 00:37:52 dajobe: any thoughts on 02:20:50 jcgregorio has joined #echo 02:56:44 jcgregorio has joined #echo 03:19:52 burtonator has joined #echo 03:44:33 burtonator has joined #echo 03:47:15 burtonator has joined #echo 03:51:48 burtonator_ has joined #echo 04:21:15 burtonator has joined #echo 04:24:06 Brendyn has joined #echo 04:24:12 Brendyn has left #echo 04:24:24 PhilOR has joined #echo 04:25:07 devnull56 has joined #echo 04:25:18 hey guys 04:26:36 hey 04:28:31 i would like to voice my suport for an offset parameter in any getRecentPosts type call, which the metaweblog api currently lacks 04:34:47 offset parameter, can you give me an exampe? 04:36:03 well i don't know the structure of the atom spec for these type of calls, but with the metaweblog api... 04:36:14 there's a getRecentPosts method that takes a certain number of posts 04:36:22 so 20 returns the last 20 posts 04:36:36 but there's no way to requests the last 20-30 posts 04:36:57 so if i wanted to create a client that let's you view the first 1-10 posts, then 11-20, then 21-30, etc 04:37:08 i would have to download posts 1-10, then 1-20, then 1-30 04:39:14 ah yes, the AtomAPI does have this, with the query parameters atom-begin-range and atom-end-range 04:39:28 great, thanks 04:39:51 also, i've contacted dave winer about adding this to the metaweblog spec that's currently being revised, but have received no response 04:40:01 i don't know if there's anyone else i could contact 04:40:58 http://bitworking.org/rfc/draft-gregorio-07.html#Model_Find 04:41:38 i can't wiat 04:41:43 *wait 04:41:44 to use this 04:42:01 thanks, that's good to hear. 04:44:13 OT historical note: the third reply when Ev announced blogger.getRecentPosts asked for an offset 04:44:30 (the second asked for it to be put in the documentation, which it still hasn't) 04:45:48 yeah, the lack of it made things inconvenient 04:46:11 it seems like for not-echo to work, we would have to make it painful to not use it 04:46:23 so it just doesn't become another standard 05:06:03 devnull56 has left #echo 06:27:49 PhilOR_ has joined #echo 06:57:44 Arien has joined #echo 07:42:43 burtonator_ has joined #echo 08:01:18 burtonator_ has joined #echo 08:15:52 d2m|off has joined #echo 09:31:50 danbri has joined #echo 09:32:03 hi all 09:32:10 * danbri only just heard about this channel 09:37:45 burtonator has joined #echo 10:30:38 burtonator has joined #echo 11:11:41 burtonator has joined #echo 12:03:11 dhaun has joined #echo 12:41:23 Sjoerd has joined #echo 12:47:08 rubys has joined #echo 13:01:13 http://tap.stanford.edu/xemantics.html 13:02:20 RemyDWD has joined #echo 13:06:38 replace xoir:xoid with rdf:about, and xoir:ref with rdf:resource, and I don't see much difference 13:09:47 for Atom at least, as it's structure is already so close to rdf 13:36:07 sjoerd_visscher has joined #echo 13:49:25 WernerVogels has joined #echo 14:07:04 WernerVogels has left #echo 14:29:44 rubys has joined #echo 14:34:05 jcgregorio has joined #echo 14:52:58 emanu has joined #echo 14:58:42 sjoerd_visscher has joined #echo 15:39:37 bitsko_ has joined #echo 15:45:16 sjoerd_visscher has joined #echo 15:48:09 dwalker has joined #echo 15:54:35 Hat has joined #echo 16:24:05 rubys has joined #echo 16:24:13 Sjoerd, are you there> 16:24:58 javax.xml.transform.TransformerException: xsl:copy-of is not allowed in this position in the stylesheet! 16:27:01 I converted atom2rdf.xml to .xsl... and then tried to use the result with Xalan (JDK 1.4) 16:30:08 sjoerd_visscher has joined #echo 16:30:29 Sam, I just saw your question on the logs 16:31:16 the irc connection is problematic 16:31:43 while I was trying to connect I fixed the bug 16:31:55 I'll try again... 16:33:03 I'm using XML::LinXSLT in perl, which is very forgiving 16:33:32 the first atom2rdf.xml file forgot to declare the rdf namespace, and it didn't give an error 16:33:45 Works! 16:33:53 does anyone know a better xslt library for Perl? 16:33:59 Do you have access to JDK 1.4? 16:34:24 I have no idea 16:34:57 I built the perl code by putting examples together 16:35:20 or do you mean using java directly? 16:35:27 I'm asking if you have JDK 1.4 installed on your machine. I'll take that answer as a 'no'. ;-) 16:36:02 ah, I think I can manage to install that 16:36:50 Here's what I'm playing with for tomorrow: http://intertwingly.net/stories/2003/08/17/ 16:37:36 libxslt? it's been a while... 16:38:05 access to the java files is not allowed 16:38:29 the xml files not either 16:38:44 only the xsl file is accessible 16:38:47 fixed 16:40:06 background: the RDF validator is based on Jena's ARP which depends on Xerces 2 and SAX. 16:40:36 First, I took a SAX writer sample from Apache which simply outputs SAX events into XML. 16:41:05 Then I created a small driver program which uses Xalan to look for a stylesheet and transform the source appropriately. 16:41:19 Whether or not the stylesheet is present, what is produced is a series of SAX events. 16:41:35 The hope is that something like this could be easily integrated into ARP. 16:42:22 ah, that's a very cool idea! 16:45:17 the stylesheet PI is really missing a target mime-type 16:46:00 I'm already using it on my atom feed to create my rss feed 16:46:02 what should the PI look like? 16:46:38 media, title, charset? 16:46:45 just xml-stylesheet with an extra target-type="..." 16:48:06 I want to transform the Atom file to xhtml, rdf and rss, but there's no way to say that 16:49:28 closest thing I can imagine is to declare that they are separate media 16:49:46 xhtml=display? 16:52:19 I know that you have the input xr for atom2rdf on your machine as a static file... do you have the output xslt as a static file someplace? 16:52:43 no 16:53:27 on media: I vaguely remember the possibility to add restrictions to media, like width < 800 16:54:36 ah, here it is: http://www.w3.org/TR/css3-mediaqueries/ 17:16:38 Enhanced has joined #echo 17:18:52 Enhanced has left #echo 17:20:14 yining has joined #echo 17:36:56 rubys has joined #echo 18:05:30 xian has joined #echo 18:43:51 crysflame has joined #echo 18:43:55 heh 18:44:20 KevinMarks has joined #echo 18:44:29 greetings! 18:44:49 add this to my lurk list 18:45:09 * crysflame nods. 18:55:55 Sjoerd has joined #echo 19:08:17 burtonator has joined #echo 19:09:26 jcgregorio has joined #echo 19:20:11 hi all 19:20:23 * danbri wonders how long this channel has been around... 19:21:14 hi 19:22:08 a week, at least 19:22:42 ah ok, haven't missed quite so much 19:23:46 * bear points to the topic 19:23:52 bitsko has a logging bot 19:24:30 cool 19:39:04 Sjoerd has left #echo 19:39:15 Sjoerd has joined #echo 20:17:53 jcgregorio has joined #echo 20:37:01 burtonator_ has joined #echo 21:00:13 wkearney99 has joined #echo 21:01:15 wkearney99 has left #echo 21:25:42 burtonator_ has joined #echo 21:41:15 hay has joined #echo 21:57:42 dwalker has joined #echo 22:51:45 rubys has joined #echo 22:52:57 danbri! 23:18:49 sbp has joined #echo 23:26:42 half-baked idea:- 23:26:53 treat Atom as an alternate serialization of RDF 23:27:22 use some standard non-namespaced attributes to point out a couple of bits (actually, just thre would do it, I think) 23:27:40 document the structure so that others can use it and have it redeployed 23:28:02 that way, extensions would have to follow the framework, and you would lose the inextensibility-of-transformations problem 23:28:12 hmm 23:33:54 it'd look something like:- 23:33:55 23:33:55 type="http://purl.org/alternatethingy/ns#" > 23:33:55 23:33:55 dive into mark 23:33:57 23:33:59 2003-08-05T18:30:02Z 23:34:01 23:34:03 Mark Pilgrim 23:34:05 23:34:07 23:34:09 Atom 0.2 snapshot 23:34:11 23:34:13 2003-08-05T08:29:29-04:00 23:34:15 2003-08-05T18:30:02Z 23:34:22 23:34:22 application/xhtml+xml 23:34:22 23:34:23
23:34:25 23:34:27 23:34:29 23:34:29 emanu has joined #echo 23:34:31 23:34:33 23:35:33 basically, you'd treat atom:* as typed nodes, atom:*/* as propElts, and atom:*/* with children as having rdf:parseType="Resource" on it 23:35:45 unless they had type="literal" in there 23:35:54 convert id to rdf:about and uri to rdf:resource 23:37:25 hey, you could even call it the atom model or something. RDF wouldn't be the only necessary output 23:37:46 hmm 23:38:50 * sbp wonders a) how reusable it'd be, not that it matters, b) whether the non-namespaced attr trick is solid from an architectural standpoint 23:40:57 hmm. it'd be easier to mix the alternate-serialization stuff and atom, but that'd be confusing 23:41:23 and you'd have to qualify the attrs anyway... 23:42:13 oh. different namespace for and . that'd be perfect 23:42:50 23:42:50 23:42:51 [...] 23:43:01 but people aren't going to "get" that 23:43:17 erk 23:44:24 still, it's better than 23:44:52 like people are even going to bother to put that in their feeds anyway unless it's required or heavily recommended by the specification 23:45:19 and like I really want to have to do an(other) HTTP GET for every nearly-RDF document I come across 23:47:18 time to write code, I guess 23:51:32 * sbp should probably be diplomatic about the namespace