#rest on irc.freenode.net 2012-02-05 (txt)

[00:22] Wombert (~Wombert@dslb-092-074-112-186.pools.arcor-ip.net) joined #rest.
[02:12] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[04:18] mhausenblas (~mhausenbl@188.141.67.15) left irc: Quit: mhausenblas
[04:26] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[05:16] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[05:57] mhausenblas (~mhausenbl@188.141.67.15) left irc: Quit: mhausenblas
[07:08] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Remote host closed the connection
[07:09] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[07:14] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Read error: Operation timed out
[07:38] grove (~grove@aggw006.cappelendamm.no) joined #rest.
[10:05] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[10:07] <Wombert> http://tools.ietf.org/html/draft-pbryan-http-json-resource-01
[10:07] <Wombert> le sigh
[10:15] <trygvis> that's a lovely mix of transport layer and data layer
[10:16] <trygvis> it's not claiming to be RESTful so it's not that bad
[10:17] <trygvis> if all the references to json was dropped it would just summarize a typical RPC way of doing http
[10:17] <Wombert> The client MUST include a query component in the Request-URI to distinguish it from the "read" operation.
[10:17] <Wombert> awesome
[10:17] <Wombert> fuck
[10:17] <trygvis> it's skipping a couple of hard parts like patching and how to make the query
[10:17] <trygvis> hehe
[10:17] <Wombert> this JSON hype is
[10:17] <Wombert> gnnnnh
[10:18] <Wombert> see the problem is
[10:18] <Wombert> that all the fucking nodejs hipsters
[10:18] <Wombert> will adopt something like this
[10:18] <Wombert> and then we're stuck with it for ten fucking years again
[10:18] <Wombert> gnnnh
[10:18] <Wombert> MUST NOT RAGE
[10:18] <trygvis> it'll be like ruby on rails all over again <o>
[10:18] <Wombert> \o/
[10:19] <trygvis> heh, there's a nice typo to. 5.7; The Request-URI MUST NOT contain a query component, in order to distinguish it from a "create" operation.
[10:19] <trygvis> that NOT shouldn't be there
[10:21] <trygvis> but it's not bringing much improvement on the world
[10:22] <trygvis> it has an error response thing, which is nice but no content type for it which is boo
[10:23] <Wombert> it also uses ETags for "resource version"
[10:23] <Wombert> Y U NO READ HTTP SPEC
[10:25] <trygvis> doesn't the http spec say that it's supposed to be opaque?
[10:25] <Wombert> nah it's the entity tag
[10:26] <Wombert> "The entity tag MAY be used for comparison with other entities from the same resource"
[10:28] <trygvis> it really would be nice to put all of these http/rest patterns on a wiki
[10:28] <trygvis> makes it easier to flog people publicly
[10:28] <trygvis> without being personal
[10:29] <Wombert> yeah true
[10:29] <Wombert> although honestly, someone who puts up an RFC like that deserves a flogging
[10:30] <trygvis> dunno, it really is about time that someone starts to put down all the patterns in writing
[10:30] <Wombert> true
[10:31] <trygvis> even they are "bad", just so that their errors can be documented
[10:31] <trygvis> and honestly, it's not that bad. it's just not REST
[10:32] <Wombert> It's a useless, half-arsed, half RPC half resource-oriented RFC that redefines several aspects of the HTTP spec
[10:32] <Wombert> and he did not even bother discussing the general idea first
[10:32] <Wombert> (I've searched for anything, the only stuff that comes up is his JSON-Patch RFC)
[10:32] <Wombert> s/anything/anything related/
[10:33] <trygvis> what does he redefine?
[10:34] <trygvis> the thing is that you can't expect people to read the entire http spec before they start making applications, so there is a need to define patterns/conventions
[10:34] <Wombert> entity tags for instance
[10:34] <Wombert> "An entity tag MUST be unique across all versions of all entities associated with a particular resource."
[10:35] <trygvis> I'm not arguing that it's a good document, but at least he's putting in effort in documenting practices
[10:35] <Wombert> right
[10:35] <trygvis> is that a violation? it's only restricting the values, not adding
[10:37] <Wombert> yeah, as he implicitly takes out the uniqueness aspect of it
[10:37] <Wombert> doesn't he?
[10:37] <Wombert> if he says "the entity tag of the entity is the version of the *resource*"
[10:38] <trygvis> doesn't seem like it, it feels like he's expecting a sequence number of revisions
[10:38] <Wombert> then all entities (representations) of the resource will have the same entity tag for the same version/revision
[10:39] <trygvis> is an entity the same as an representation?
[10:41] <Wombert> does that matter?
[10:41] <Wombert> A "strong entity tag" MAY be shared by two entities of a resource only if they are equivalent by octet equality.
[10:41] <Wombert> his are strong, as they have no W/ prefix
[10:41] <Wombert> A "weak entity tag," indicated by the "W/" prefix, MAY be shared by two entities of a resource only if the entities are equivalent and could be substituted for each other with no significant change in semantics. A weak entity tag can only be used for weak comparison.
[10:41] <Wombert> An entity tag MUST be unique across all versions of all entities associated with a particular resource. A given entity tag value MAY be used for entities obtained by requests on different URIs. The use of the same entity tag value in conjunction with entities obtained by requests on different URIs does not imply the equivalence of those entities.
[10:42] <trygvis> ah, ok
[10:42] <trygvis> I didn't know that they had to be byte-by-byte equal
[10:42] <Wombert> :)
[10:42] <trygvis> ok, then I agree that he's violating the spec
[10:54] edsu (~ed@li144-162.members.linode.com) joined #rest.
[11:26] mhausenblas (~mhausenbl@188.141.67.15) left irc: Quit: mhausenblas
[11:44] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[11:46] <trygvis> what would stop that ID to go through the process
[11:46] <trygvis> hopefully "someone" will discover the flaws
[12:34] Wombert (~Wombert@dslb-092-074-112-186.pools.arcor-ip.net) left irc: Quit: Wombert
[12:58] KevBurnsJr (~KevBurnsJ@c-67-164-98-121.hsd1.ca.comcast.net) left irc:
[13:49] Wombert (~Wombert@dslb-092-074-112-186.pools.arcor-ip.net) joined #rest.
[14:01] grove_ (~grove@84.115.45.31.customer.cdi.no) joined #rest.
[14:04] grove (~grove@aggw006.cappelendamm.no) left irc: Ping timeout: 252 seconds
[14:05] grove_ (~grove@84.115.45.31.customer.cdi.no) left irc: Ping timeout: 240 seconds
[14:45] mhausenblas (~mhausenbl@188.141.67.15) left irc: Quit: mhausenblas
[16:09] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[16:38] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Remote host closed the connection
[16:40] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[19:19] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 240 seconds
[21:00] Wombert (~Wombert@dslb-092-074-112-186.pools.arcor-ip.net) left irc: Quit: Wombert
[21:30] grove (~grove@84.115.45.31.customer.cdi.no) joined #rest.
[21:30] grove (~grove@84.115.45.31.customer.cdi.no) left irc: Client Quit
[22:09] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[22:14] grove (~grove@aggw006.cappelendamm.no) joined #rest.
[23:50] mhausenblas (~mhausenbl@188.141.67.15) left irc: Quit: mhausenblas
[00:00] --- Mon Feb 6 2012