#rest on irc.freenode.net 2012-01-27 (txt)

[00:57] <mk_> talios, darrelmiller - definitely get involved in that discussion on the group
[00:57] <mk_> it's a pretty fundamental design to HAL so it's worth going through in detail
[01:18] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[01:18] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[01:43] DracoBlue1 (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[01:44] DracoBlue1 (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left irc: Client Quit
[01:48] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[02:37] talios (~textual@ip-118-90-67-172.xdsl.xnet.co.nz) joined #rest.
[02:43] talios (~textual@ip-118-90-67-172.xdsl.xnet.co.nz) left irc: Quit: Textual IRC Client: http://www.textualapp.com/
[02:43] talios (~textual@ip-118-90-67-172.xdsl.xnet.co.nz) joined #rest.
[02:55] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[03:01] Wombert (~Wombert@62.205.97.157) joined #rest.
[03:09] mr_yall (~mr_yall@scifo.mexico86.co.uk) joined #rest.
[03:35] <mk_> talios: dunno if you caught the above two lines I think you may have been disconnected
[03:35] <talios> oh?
[03:35] <mk_> doesn't really matter I just said if you should definitely contribute as it's fundamental part of HAL's design
[03:36] <talios> contribute what?
[03:36] <mk_> When I first responded to the guy I actually pen'd something about how I might make it a constraint that the self relation can only be a single relation on its own
[03:37] <mk_> talios: sorry contribute to the thread
[03:37] <talios> oh right.
[03:38] <mk_> anyone on your team should get involved too
[03:38] <mk_> I'm interested to hear the arguments for doing it
[03:38] <talios> Well, you -could- get around that "self relation" restriction, by having an additonal <link rel="http://minted"; href="same-as-self"/>
[03:38] <mk_> sure
[03:38] <mk_> tbh I think you're better off doing
[03:38] <talios> but then - you may as well just collapse that to rel="self http://minted"; which is effectively the same thing
[03:38] <mk_> <link rel="type" href="/types/widget" />
[03:39] <mk_> but I would seirously consider whether that is a Good Idea(tm) or not
[03:40] <talios> I'm not 100% in agreement with my coworkers ideas on how we shoud use HAL as it is tho :)
[03:40] <mk_> talios: what's the thinking behind needing to do that ?
[03:41] <talios> Mostly, that a resoure shoudl be self-describing - "Roy Says So" :)
[03:41] <mk_> i.e. what kind of situation would a resource need to assert its own significance
[03:41] <mk_> ah.
[03:41] <mk_> I think that's a perversion of the term self-descriptive
[03:41] <talios> Main 'valid' usecase - sending HAL representations via a message bus.
[03:42] <mk_> o.O
[03:42] <mk_> that sounds fun! :D
[03:42] <talios> Also, I had the thought earlier - if minting a rel on a top level resource is bad, why have the href either then.
[03:42] <mk_> is that something you are trying to do now or thinking about in the future ?
[03:42] <talios> you followed a link target - THATS THE HREF :P
[03:43] <talios> we want a message bus in the future, but its been part of my thinking of why you'd want a self-describing rep. you -might- even use SMTP as a transport rather than HTTP.
[03:43] Last message repeated 1 time(s).
[03:43] <mk_> talios: for protocols that don't have a Content-Location header
[03:43] <talios> doh
[03:44] <mk_> If you are using SMTP
[03:44] <talios> mk_: ahh but Content-Location is a header on a resource, you followed a link. Content-Location is part of the resource. things shoud be rel driven.
[03:44] <talios> so the argument is.
[03:44] <mk_> I meant the 'why is there an href on the root resource' thing
[03:44] <mk_> + debugging and shit like that
[03:45] <talios> true. but if everything is rel driven from a link - you shouldn't NEED it.
[03:45] <mk_> yeah right, it is a bit redundant
[03:45] <mk_> easier to calc relative URIs from a parser
[03:45] <talios> debugging was also my other reason for minting the <resource rel=""/> - its -added- documentation for -me-. for debugging, for whatever.
[03:46] <mk_> yeah I think that is a step too far for me
[03:46] <mk_> you're on the server side
[03:46] <mk_> you know how your URIs map to your resources
[03:47] <mk_> the trouble with the self+type rel is that it exposes something to the client that encourages them to behave in a Bad Way (tm)
[03:48] <mk_> if the only way for them to establish semantics is via previous rels.. they are *forced* into a discovery/traversal approach
[03:48] <mk_> as soon as you start exposing typed'ness you've leaked out information which means they are no-longer constrained
[03:50] <talios> A thought I did have earlier as well, if you followed a link for rel="x", one reason to have additional links would be to say "yes I am x, but I also contain y". Which may as I said above, be other explicit <link/> elements pointing to the same URI
[03:51] <talios> the href on a resource -is- a link in itself from what I can tell .
[03:52] <talios> (it certainly is when you view the current JSON representation).
[03:53] <mk_> yeah it's a link
[03:53] <mk_> btw <resource> used to be just a non-self-closing link element
[03:54] <mk_> <link>.. embedded content...</link>
[03:54] <mk_> and then I changed it to resource so it was clearer
[03:54] <talios> right.
[03:55] <mk_> 2 secs brb
[03:57] <talios> Ok, so letting my brain drift ( its 1am ). thats two <link rel="http://minted">;'s to the same URI, one somewhere else, and one embedded on the actual resource.
[03:59] <talios> https://gist.github.com/4b7d3922bf66d0d55547 <- I guess the <resource rel="self xxx"/> is ONLY relevant on the XML representation, where the rel="self" is an attribute.
[03:59] Fullmoon (~Fullmoon@dsl-stat-43-17.mmc.at) joined #rest.
[03:59] <talios> unlike JSON where it's an explicit "_links": entry
[04:02] Wombert (~Wombert@62.205.97.157) left irc: Quit: Wombert
[04:11] <talios> mk_ - thats a long 2secs :)
[04:11] <mk_> yeah sorry I'm back now
[04:11] <talios> heh
[04:12] <mk_> there's not really any difference between _links: { self: { ... and <resource href=".."
[04:12] <mk_> it's just because JSON isn't as expressive as XML so I had to find an elegant way to deal with it
[04:13] <talios> yeh. I kinda preferred the earlier _href attribute actually
[04:14] <talios> it does serve to hightlight clearly that it is a link tho. which is why I don't particular see an issue with having multiple links pointing to same href that self does.
[04:15] <mk_> honestly I can't remember why that was done away with, I think I was just trying to reduce the 'surface' area of HAL so it didn't "impose" as much on the JSON
[04:15] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[04:15] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[04:15] <Fullmoon> What is HAL?
[04:15] <mk_> much of a muchness
[04:16] <mk_> Fullmoon: http://blog.stateless.co/post/13296666138/json-linking-with-hal
[04:16] <mk_> the spec is here: http://stateless.co/hal_specification.html
[04:16] <talios> mk_ - that version is quite out of date from github now :(
[04:17] <mk_> talios: I think you might be hitting a cached version
[04:18] <talios> mm did you update it recently? as Isaac noticed that at work yesterday as well
[04:18] <mk_> pretty sure I pushed it to the server when I pushed the update to github
[04:18] <mk_> tbh the latest revision is not complete, I just needed to make a start on it
[04:19] <mk_> e.g. there's a Transclusion section that's completely empty
[04:20] <talios> One thing around this whole rel discussion, in your examples you have "customer", if the recommendation is to use a URI for minted rels, should the examples highlight that?
[04:22] <talios> interestingly tho - if they are URIs, how would _links.upsell[0].href work - if the upsell was a URI. _links["http://upsell";][0].href I guess?
[04:26] Wombert (~Wombert@62.205.105.58) joined #rest.
[04:31] <mk_> yeah
[04:31] <mk_> I don't do that in the examples cos it's less pretty :)
[04:32] <mk_> but you are right - the examples don't follow that recommendation
[04:33] mk_ shrugs
[04:33] <talios> heh - maybe at least ONE example then under the recommendations section of using URIs.
[04:33] <mk_> yeah I think that's probably the best bet
[04:33] talios mutters something about spec writers ;p
[04:33] <mk_> haha
[04:33] <talios> are you an "enterprise architect" by day? :)
[04:33] <mk_> no I'm a rails developer :<
[04:34] <mk_> I work for cloud computing and e-commerce companies mostly, building stuff
[04:35] <mk_> I first got into REST 3 and a bit years ago cos I had to build a big integration platform
[04:35] <talios> rails - thats like worse than enterprise ( coming from "the java guy" ) :)
[04:35] <mk_> oh man. I love rails :)
[04:36] <mk_> to be fair, I've only been commited to working with it since the big rails 3 re-write
[04:36] <mk_> which changed a tonne of stuff I really didn't like before
[04:36] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[04:37] <talios> about the same here then - i'd already started looking at REST 3.5ish years back, then started at $work and doing everything "resty" - tho that was mostly HTTP+JSON more than anythig "blessed by Roy" :)
[04:38] <talios> Making the transition to start using Link: headers was a big game changer for us, tho it's still only used in a few parts of our system. Now we're gong thru a much larger rework
[04:39] <mk_> Link headers are cool but I think they get used for stuff where it could just go in the body
[04:40] <mk_> super glad they got re-instated on the web though
[04:40] <mk_> hooray for mnot!
[04:40] <talios> true. one thing I like about them is the uniformity of processing regardless of Content-Type, i.e. same place to find links if you're looking at a text/plain application/pdf or text/csv representation
[04:40] <mk_> right, that's exactly where they shine
[04:41] <mk_> and also for other funky stuff like this link invalidation thing I worked on
[04:41] <talios> That reminds me - I've meaning to write my linking blog.
[04:41] <mk_> but that's the key - I really use them as a fallback for adding linking capabilities to legacy media types
[04:43] talios looks at the clock -shiz! 1:42am. I should really sleep - planning a 3 hour drive to a music festival tomorrow ( then the drive back in the evening ).
[04:44] <mk_> eek!
[04:44] <mk_> good talking to you
[04:44] <mk_> have a good one :()
[04:44] <talios> indeed - you to :)
[04:47] sbanwart_ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[04:48] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Remote host closed the connection
[04:49] Wombert (~Wombert@62.205.105.58) left irc: Quit: bai
[04:51] talios (~textual@ip-118-90-67-172.xdsl.xnet.co.nz) left irc: Quit: Textual IRC Client: http://www.textualapp.com/
[05:01] sbanwart_ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 252 seconds
[05:12] <mk_> darrelmiller: I think it's possible to make a decent case as to why that is not a good idea
[05:29] <Fullmoon> Besides Web Services With Rest, and the Rest Cookbook, any other good books that you could recommend?
[05:36] <mk_> no.
[05:36] <mk_> definitely not that dodgey one about Hypermedia with node and HTML
[05:36] <mk_> whoever wrote that didn't really put much thought into it
[05:37] <Jarda> mk_: this http://shop.oreilly.com/product/0636920020530.do ?
[05:37] <mk_> http://www.amazon.com/Building-Hypermedia-APIs-HTML5-Node/dp/1449306578
[05:38] <mk_> yeah that one
[05:38] <mk_> load of bollocks, right?!
[05:38] <mk_> :D
[05:38] <Jarda> ok, then you should maybe discuss a little with mamund ..
[05:38] <Jarda> I haven't read it
[05:38] <mk_> I know it's mike's I'm kidding
[05:38] <Jarda> oh
[05:38] <Jarda> :)
[05:38] <mk_> Fullmoon: I'm kidding btw that book is excellent
[05:38] <Jarda> :D
[05:39] <mk_> probably the most important book on 'REST' you can get
[05:39] <Jarda> got me there :)
[05:39] <mk_> the other stuff you get about 'REST' is really all just stuff you could get from reading the HTTP spec :|
[05:39] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[05:43] <mk_> REST cookbook is a pretty good reference though, to be fiar
[05:43] <mk_> oh REST in Practice is pretty good too, actually
[05:43] <mk_> that's written by the guys who came up with REST Bucks
[05:44] <mk_> mamund: I was trying to come up with a reference application we can work on (ala maze) but with write as well as read
[05:45] <mk_> I thought about something like a forum/discussion application
[05:45] <mk_> but that might be a bit boring
[05:46] <mk_> then I thought about maybe something like Battleships
[05:46] <Fullmoon> mk_: Cool, I'll have a look
[05:47] <mk_> Fullmoon: mamund wrote hypermedia APIs with HTML5 and node.js
[05:47] <mk_> he's a self-professed spiritual Guru in the REST world
[05:47] <Fullmoon> Like node.js node?
[05:47] <Fullmoon> Ah
[05:48] <mk_> yeah the node bit is just the server side tech he picked for the examples
[05:48] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[05:48] <Fullmoon> Uh very cheap for kindle..
[05:48] <mk_> even if you aren't interested in node there's a tonne of great stuff anyway
[05:48] <mk_> on the hypermedia side
[06:19] bigbluehat1 (~bigblueha@adsl-98-71-132-42.gsp.bellsouth.net) joined #rest.
[06:20] sbanwart_ (~sbanwart@66.6.175.18) joined #rest.
[06:24] sbanwart_ -> sbanwart
[07:06] <darrelmiller> mk_: yeah, tagging on the node and Html5 was a brilliant way of getting past the O'Reilly marketters to get a book out on Hypermedia design :-)
[07:07] <mk_> mamund: is an evil genius
[07:07] <darrelmiller> I should write a book on designing restful clients using coffeescript and CSS3.
[07:07] <mk_> ... :D
[07:20] <bigbluehat1> :)
[07:20] <bigbluehat1> do it!
[07:22] <mk_> what does everyone think about a reference hypertext application for battleships
[07:22] <bigbluehat1> do it!
[07:22] bigbluehat1 is in a very positive, encouraging mood
[07:22] <mk_> lol
[07:23] <mk_> should I shoot myself in the face ?
[07:23] <bigbluehat1> well...
[07:23] <bigbluehat1> we all draw the line somewhere :)
[07:23] <mk_> ...
[07:23] <bigbluehat1> so...
[07:23] <bigbluehat1> no
[07:23] <mk_> pwned.
[07:23] <mk_> ¬_¬
[07:23] <bigbluehat1> :(
[07:23] bigbluehat1 will be more careful
[07:23] <bigbluehat1> I miss him already....
[07:26] <bigbluehat1> ….mk_...?
[07:27] <mk_> X_X
[07:27] <bigbluehat1> he's gone to rest-discuss purgatory for sure...
[07:27] <bigbluehat1> this time unwillingly
[07:27] <mk_> haha
[07:27] <bigbluehat1> poor chap
[07:27] <bigbluehat1> :)
[07:27] <bigbluehat1> can anyone tell I'm trying to debug in IE8 right now?
[07:27] <mk_> ohh damn
[07:28] <mk_> don't support that shit man!!!
[07:28] <bigbluehat1> heh :)
[07:28] <bigbluehat1> I'm doing everything I can *not* to have to
[07:28] <mk_> fight powa!!!!
[07:28] <bigbluehat1> :)
[07:28] <mk_> fight the powa*
[07:28] <mk_> we are the 99%
[07:28] <bigbluehat1> heck yeah! ;)
[07:28] <mk_> end the fed!
[07:29] <bigbluehat1> I want to by a galopi and put a "IR1PRCNT" on it
[07:29] <mk_> hahahaha
[07:29] <bigbluehat1> oh…super…the bug's just magically gone!
[07:29] <bigbluehat1> amazing
[07:29] <mk_> I'm taking some credit for that
[07:30] <fu-manchu> I did that with my mind
[07:30] <bigbluehat1> oh…I fixed it X-P
[07:30] <bigbluehat1> heh
[07:30] <bigbluehat1> you both get the credit
[07:30] <bigbluehat1> I just did the typing…I guess
[07:31] bigbluehat1 refills his coffee
[07:45] pezra (~Adium@webmail.openlogic.com) joined #rest.
[08:34] whartung (~whartung@wsip-98-189-78-118.oc.oc.cox.net) left irc: Read error: Operation timed out
[08:38] whartung (~whartung@wsip-98-189-78-118.oc.oc.cox.net) joined #rest.
[08:38] bigbluehat1 (~bigblueha@adsl-98-71-132-42.gsp.bellsouth.net) left irc: Quit: Leaving.
[08:46] <whartung> o/
[09:05] <darrelmiller> \0
[09:06] <fu-manchu> \o/ (break it up)
[09:06] <whartung> <o> waaaaahhh
[09:12] hdave (~hdave@static-71-245-233-130.bstnma.fios.verizon.net) joined #rest.
[09:16] KevBurnsJr (~KevBurnsW@50.0.103.39) joined #rest.
[09:17] <mk_> am I missing something
[09:18] <mk_> why is it a crime to use URI templates in links ?
[09:19] <whartung> Everything you do is a crime mk_, it's just your character.
[09:19] <mk_> I know right
[09:19] <mk_> check this out
[09:20] <mk_> https://twitter.com/#!/serialseb/status/162179818434207745
[09:20] <mk_> that was yesterday, I think.
[09:20] <mk_> .. go figure.
[09:23] <darrelmiller> mk_: Just think, soon you and serialseb will be living just a short drive from each other. You could go have a beer and clear this stuff up ;-)
[09:23] <mk_> yeah for sure
[09:24] <mk_> I've met him already, he was at restunconf
[09:25] <mk_> he really doesn't get hypertext
[09:25] <mk_> I tried to explain it to him
[09:25] <darrelmiller> I know. I keep tabs on everyone.
[09:25] <mk_> I think we discussed it for about an hour and a half
[09:25] <mk_> alas, I failed
[09:27] <whartung> I judts haven't had time to follow rest-discuss
[09:28] <darrelmiller> Plant the seed, then wear them down over time. Eventually they will have a brilliant idea that will sound familiar.
[09:28] <fu-manchu> the best jokes are the ones you make other people tell
[09:29] <whartung> "restunconf" is that different from "unrestconf"?
[09:29] <whartung> The people that don't get hypertext simply have their heads buried in their own little implementations.
[09:30] <darrelmiller> That was the "us Europeans are jealous of RESTfest" conference.
[09:30] <whartung> oh
[09:44] axisys_ (~axisys@ip68-98-189-233.dc.dc.cox.net) left #rest.
[09:45] <mk_> well I'm still none the wiser what his issue is
[09:46] hobodave (~hobodave@pdpc/supporter/professional/hobodave) joined #rest.
[10:01] pezra (~Adium@webmail.openlogic.com) left irc: Quit: Leaving.
[10:02] pezra (~Adium@webmail.openlogic.com) joined #rest.
[10:11] pezra (~Adium@webmail.openlogic.com) left irc: Quit: Leaving.
[10:11] pezra (~Adium@webmail.openlogic.com) joined #rest.
[10:27] hdave (~hdave@static-71-245-233-130.bstnma.fios.verizon.net) got netsplit.
[10:28] Fullmoon (~Fullmoon@dsl-stat-43-17.mmc.at) got netsplit.
[10:28] chilversc (~chris@about/csharp/regular/KeeperOfTheSoul) got netsplit.
[10:28] mogsie (~mogsie@62.101.198.35) got netsplit.
[10:28] chilversc (~chris@about/csharp/regular/KeeperOfTheSoul) returned to #rest.
[10:28] hdave (~hdave@static-71-245-233-130.bstnma.fios.verizon.net) returned to #rest.
[10:28] mogsie (~mogsie@62.101.198.35) returned to #rest.
[10:39] Fullmoon (~Fullmoon@dsl-stat-43-17.mmc.at) got lost in the net-split.
[10:53] pezra (~Adium@webmail.openlogic.com) left irc: Quit: Leaving.
[10:56] pezra (~Adium@webmail.openlogic.com) joined #rest.
[10:56] pezra (~Adium@webmail.openlogic.com) left irc: Client Quit
[10:57] pezra (~Adium@webmail.openlogic.com) joined #rest.
[12:32] KevBurnsJr_ (~KevBurnsJ@c-67-164-98-121.hsd1.ca.comcast.net) joined #rest.
[12:33] hdave (~hdave@static-71-245-233-130.bstnma.fios.verizon.net) left #rest.
[12:47] hobodave (~hobodave@pdpc/supporter/professional/hobodave) left irc: Quit: Textual IRC Client: http://www.textualapp.com/
[13:23] pezra (~Adium@webmail.openlogic.com) left irc: Quit: Leaving.
[13:27] pezra (~Adium@webmail.openlogic.com) joined #rest.
[13:29] pezra (~Adium@webmail.openlogic.com) left irc: Client Quit
[13:34] pezra (~Adium@webmail.openlogic.com) joined #rest.
[13:38] pezra (~Adium@webmail.openlogic.com) left irc: Client Quit
[13:40] pezra (~Adium@webmail.openlogic.com) joined #rest.
[13:40] pezra (~Adium@webmail.openlogic.com) left irc: Client Quit
[13:43] pezra (~Adium@webmail.openlogic.com) joined #rest.
[14:04] sbanwart (~sbanwart@66.6.175.18) left irc: Ping timeout: 240 seconds
[14:10] pezra (~Adium@webmail.openlogic.com) left irc: Ping timeout: 276 seconds
[14:15] pezra (~Adium@webmail.openlogic.com) joined #rest.
[15:06] Hakon|mbp (~hakon1@72.135.202.84.customer.cdi.no) joined #rest.
[15:18] <darrelmiller> oooh there is a new URI Template spec available http://tools.ietf.org/html/draft-gregorio-uritemplate-08
[15:22] <darrelmiller> mk_: "Within client-side implementations, a URI template has many of the same properties as HTML forms"
[15:37] pezra (~Adium@webmail.openlogic.com) left irc: Ping timeout: 276 seconds
[15:41] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[15:43] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Read error: Operation timed out
[15:48] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[15:53] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Read error: Operation timed out
[15:56] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[15:59] <KevBurnsJr> wow 08
[15:59] <darrelmiller> yeah, I completely missed the 07 in sept.
[16:00] <fu-manchu> I still wish there was a "uri templates for rest" that didn't have all the complicated crap to support *every* possible permutation of URI
[16:00] <fu-manchu> all I need is {/foo}, {?bar}, and maybe {;baz}
[16:00] <darrelmiller> They have levels. Level 1 is pretty simple.
[16:01] <darrelmiller> It's level 2,3 and 4 that get a bit more funky :-)
[16:02] <fu-manchu> gracias, wasn't aware they had clumped them into levels
[16:02] <fu-manchu> it's a little bit more sane now :)
[16:03] <darrelmiller> I haven't quite figured out how level 1 supports query params though.
[16:04] Hakon|mbp (~hakon1@72.135.202.84.customer.cdi.no) left irc: Read error: Connection reset by peer
[16:04] <fu-manchu> foo?bar={value}
[16:04] <darrelmiller> yeah that's what I was thinking.
[16:04] <fu-manchu> for a very expansive definition of "supports" :)
[16:05] <darrelmiller> but that format is not nearly as nice as the foo{?bar} syntax which I think is only introduced in level 3
[16:06] <fu-manchu> I still don't see why they have level 2 at all. You should be able to infer reserved expansion if you're using "/" for example
[16:07] <darrelmiller> dunno.
[16:07] <fu-manchu> meh
[16:07] <fu-manchu> I wrote my own into Shoji 'cause I couldn't wait ;)
[16:08] <darrelmiller> is there a mailing list for this working group?
[16:09] <KevBurnsJr> I think they're piggypacking on the uri list
[16:10] <KevBurnsJr> http://lists.w3.org/Archives/Public/uri/
[16:10] <darrelmiller> thank you, my google foo was failing me.
[16:11] <KevBurnsJr> I only recall cause I posted on there after writing my PHP uri tempale -04 parser
[16:11] <KevBurnsJr> whiiiich I should prooobably update
[16:11] <KevBurnsJr> http://lab.kevburnsjr.com/php-uri-template-parser
[16:13] <darrelmiller> I suppose I should think about writing a C# one at some point.
[16:14] sbanwart_ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[16:14] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Remote host closed the connection
[16:40] Hakon|mbp (~hakon1@226.86.202.84.customer.cdi.no) joined #rest.
[16:43] sbanwart_ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 260 seconds
[16:47] steveklabnik (~steve@c-67-165-87-244.hsd1.pa.comcast.net) joined #rest.
[16:47] <steveklabnik> mamund: congrats on your book man. I just finished it. Great job.
[16:50] <darrelmiller> steveklabnik: mamund is mia, haven't seen him here all day. Must be having fun in SF.
[16:53] <steveklabnik> :)
[16:53] <steveklabnik> it's hard not to!
[16:53] <steveklabnik> how have things been around here?
[16:53] <darrelmiller> Just the usual bitching and griping...
[16:53] <steveklabnik> heh
[16:54] <steveklabnik> i'm giving the first of my hypermedia talks at ruby conferences next week.
[16:54] <darrelmiller> Although there is excitement about the URI template spec getting approved.
[16:54] <steveklabnik> still putting it together.
[16:55] <darrelmiller> Cool.
[16:55] <steveklabnik> oh yeah? did it, or is it close?
[16:55] steveklabnik hasnt heard
[16:55] <darrelmiller> https://datatracker.ietf.org/doc/draft-gregorio-uritemplate/history/
[16:56] <steveklabnik> wooo
[16:56] <darrelmiller> steveklabnik: Are you familiar with restfulie ?
[16:56] <steveklabnik> six years
[16:56] <steveklabnik> not bad
[16:56] <steveklabnik> yes, i have a patch in for some documentation on the ruby one
[16:57] <steveklabnik> why do you ask?
[16:57] <darrelmiller> ok. I was just wondering. I haven't heard much about it recently.
[16:57] KevBurnsJr (~KevBurnsW@50.0.103.39) left irc: Quit: tgimff
[16:57] sbanwart_ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[16:58] Hakon|mbp (~hakon1@226.86.202.84.customer.cdi.no) left irc: Read error: Operation timed out
[17:02] <steveklabnik> yeah, i havent either.
[17:02] <steveklabnik> i dont use it for anything
[17:24] bigbluehat (u632@gateway/web/irccloud.com/x-wkusvqusacvtgvgo) left irc: Quit: Connection closed for inactivity
[17:31] sbanwart_ -> sbanwart
[17:37] steveklabnik (~steve@c-67-165-87-244.hsd1.pa.comcast.net) left irc: Ping timeout: 244 seconds
[17:46] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Remote host closed the connection
[17:54] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[18:01] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Quit: Konversation terminated!
[18:04] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[18:36] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 252 seconds
[18:42] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[18:55] rpaddock (~rpaddock@rrcs-76-79-208-202.west.biz.rr.com) joined #rest.
[18:57] <rpaddock> If a entity has an additional requirement that must be met in order for a PUT to process correctly, should that throw a 400 or a 409 on failure?
[19:05] <fu-manchu> depends on the requirement, and whether the user can resolve it, but often 409
[20:04] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 260 seconds
[20:37] quest88 (~quest88@c-98-207-205-137.hsd1.ca.comcast.net) joined #rest.
[23:29] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[23:29] mhausenblas (~mhausenbl@188.141.67.15) left irc: Client Quit
[00:00] --- Sat Jan 28 2012