#rest on irc.freenode.net 2011-12-05 (txt)

[00:36] f00li5h (~f00li5h@unaffiliated/f00li5h) left irc: Read error: Connection reset by peer
[00:50] ssedano (~ssedano@unaffiliated/ssedano) joined #rest.
[01:11] quest88 (~quest88@c-98-207-205-137.hsd1.ca.comcast.net) left irc: Quit: quest88
[02:24] kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) left irc: Ping timeout: 244 seconds
[02:56] grove (~grove@aggw006.cappelendamm.no) left irc: Quit: grove
[03:01] grove (~grove@aggw006.cappelendamm.no) joined #rest.
[03:54] grove_ (~grove@aggw006.cappelendamm.no) joined #rest.
[03:55] grove (~grove@aggw006.cappelendamm.no) left irc: Ping timeout: 252 seconds
[03:55] grove_ -> grove
[05:07] grove (~grove@aggw006.cappelendamm.no) left irc: Quit: grove
[05:17] mephju (~mephju@dslb-094-222-022-096.pools.arcor-ip.net) joined #rest.
[05:25] mephju_ (~mephju@dslb-094-222-022-096.pools.arcor-ip.net) joined #rest.
[05:25] mephju_ (~mephju@dslb-094-222-022-096.pools.arcor-ip.net) left irc: Client Quit
[05:26] Wombert (~Wombert@dslb-092-075-013-091.pools.arcor-ip.net) joined #rest.
[05:36] Wombert_ (~Wombert@dslb-088-065-193-251.pools.arcor-ip.net) joined #rest.
[05:37] Wombert (~Wombert@dslb-092-075-013-091.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds
[05:39] Wombert (~Wombert@dslb-092-075-002-064.pools.arcor-ip.net) joined #rest.
[05:40] Wombert_ (~Wombert@dslb-088-065-193-251.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds
[05:43] Wombert (~Wombert@dslb-092-075-002-064.pools.arcor-ip.net) left irc: Ping timeout: 244 seconds
[05:45] grove (~grove@aggw007.cappelendamm.no) joined #rest.
[05:52] Wombert (~Wombert@dslb-088-066-206-217.pools.arcor-ip.net) joined #rest.
[05:57] grove (~grove@aggw007.cappelendamm.no) left irc: Quit: grove
[05:58] grove (~grove@aggw006.cappelendamm.no) joined #rest.
[05:59] Wombert (~Wombert@dslb-088-066-206-217.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds
[06:03] Wombert (~Wombert@dslb-092-075-019-241.pools.arcor-ip.net) joined #rest.
[06:06] Wombert_ (~Wombert@dslb-092-075-020-157.pools.arcor-ip.net) joined #rest.
[06:07] bigbluehat (~bigblueha@nmd.sbx08595.greensc.wayport.net) joined #rest.
[06:08] Wombert (~Wombert@dslb-092-075-019-241.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds
[06:09] Wombert (~Wombert@dslb-092-075-014-249.pools.arcor-ip.net) joined #rest.
[06:10] Wombert_ (~Wombert@dslb-092-075-020-157.pools.arcor-ip.net) left irc: Ping timeout: 255 seconds
[06:14] Wombert (~Wombert@dslb-092-075-014-249.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds
[06:14] Wombert (~Wombert@dslb-088-066-201-154.pools.arcor-ip.net) joined #rest.
[06:26] mr_yall (~mryall@host86-160-76-32.range86-160.btcentralplus.com) joined #rest.
[06:28] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[06:35] grove (~grove@aggw006.cappelendamm.no) left irc: Quit: grove
[06:40] fu-manchu (~fumanchu@adsl-99-30-180-185.dsl.sfldmi.sbcglobal.net) joined #rest.
[06:57] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[06:57] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[07:05] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Quit: Leaving...
[07:17] grove (~grove@193.201.9.46.customer.cdi.no) joined #rest.
[07:25] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[07:26] bigbluehat (~bigblueha@nmd.sbx08595.greensc.wayport.net) left irc: Quit: Leaving.
[07:26] bigbluehat (~bigblueha@nmd.sbx08595.greensc.wayport.net) joined #rest.
[08:03] grove_ (~grove@aggw006.cappelendamm.no) joined #rest.
[08:04] mamund shows up
[08:05] <mamund> steveklabnik: yay!
[08:05] grove (~grove@193.201.9.46.customer.cdi.no) left irc: Ping timeout: 240 seconds
[08:05] grove_ -> grove
[08:06] <mamund> mikekelly: only decent link i have on this is an old PPT from Fielding: [http://gbiv.com/protocols/waka/200211_fielding_apachecon.ppt
[08:09] grove_ (~grove@193.201.9.46.customer.cdi.no) joined #rest.
[08:13] grove (~grove@aggw006.cappelendamm.no) left irc: Ping timeout: 245 seconds
[08:13] grove_ -> grove
[08:14] technoweenie (~technowee@host-86-220-9-69.midco.net) joined #rest.
[08:37] nuclearsandwich (~nuclearsa@64-79-127-126.static.wiline.com) joined #rest.
[08:41] <whartung> Monday everyone. Seems to always happen around this time of the week...
[08:42] KevBurnsJr (~KevBurnsJ@50.0.103.39) joined #rest.
[08:47] bigbluehat (~bigblueha@nmd.sbx08595.greensc.wayport.net) left irc: Quit: Leaving.
[09:02] quest88 (~quest88@c-98-207-205-137.hsd1.ca.comcast.net) joined #rest.
[09:05] quest88 (~quest88@c-98-207-205-137.hsd1.ca.comcast.net) left irc: Client Quit
[09:12] <steveklabnik> mamund: i also think i'm starting to get your statement about 'programming media types'
[09:12] <steveklabnik> i'm re-doing that code test first and OO rather than procedural, it's interesting.
[09:13] peers (~ajp5103@128.118.235.29) joined #rest.
[09:15] mamund was away
[09:16] mamund waves to whartung
[09:17] grahamc (~gchristen@173-166-174-177-washingtondc.hfc.comcastbusiness.net) joined #rest.
[09:17] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Ping timeout: 244 seconds
[09:18] <mamund> steveklabnik: yeah, if you start to "think in media types" it provides a new POV for client coding.
[09:18] <mamund> you still have total freedom on _how_ you implement the client (OO, functional, event-loop, etc.)
[09:19] <steveklabnik> yeah.
[09:19] <mamund> when you get a version of the maze running, please send along a link (to code and client, if possible) and i'll add it to the list of implementations on the media type page
[09:19] <steveklabnik> this particular iteration is OO.
[09:19] <mamund> actually, i'd love to hear now diff coding styles (OO, etc.) "work" in this media-type POV
[09:19] <steveklabnik> sweet. yeah, i'll convert that procedural code into a maze solver, and show an OO one too.
[09:20] <mamund> IOW, it'd be interesting to see if some styles "work better" than others
[09:20] <mamund> (easier to map intentions to affordances, etc.)
[09:21] <steveklabnik> yeah
[09:21] <steveklabnik> in this particular case, I have a Maze class
[09:21] <steveklabnik> basically
[09:21] <steveklabnik> that can query the current response for the appropriate rels
[09:23] <mamund> yep
[09:27] <fu-manchu> one thing we've found with JSON-based media types is that Python consumers deserialize the representation to a native dictionary, and then find it's not usually useful to wrap up or transform that into custom objects--the ADT approach allows the programmer to be more efficient in designing the API and transport factors than they would if an OO layer were in between
[09:27] <mamund> hmmmm
[09:27] <mamund> fu-manchu: that's interesting
[09:28] <mamund> does this work "as well" when using a hypermedia-style JSON format (like collection+json or hal+json)?
[09:28] dmco (~dmco3@orchestra/user/dmco) joined #rest.
[09:29] <mamund> IOW, it is "easy" to find the links, links-rels, processing any tempaltes/forms?
[09:29] bigbluehat (~bigblueha@adsl-74-177-96-212.gsp.bellsouth.net) joined #rest.
[09:29] bigbluehat (~bigblueha@adsl-74-177-96-212.gsp.bellsouth.net) left irc: Client Quit
[09:32] dmco (~dmco3@orchestra/user/dmco) left #rest.
[09:32] <peers> Can anyone recommend any articles on "best practices" for designing a response envelope? Like where to stick URIs and how to label them, what to do with validation errors, etc.
[09:33] <peers> Or is that stuff pretty "as long as it works"
[09:33] <mamund> subbu's book "RESTful WEb Services Cookbook" has a number of recipes for designing responses.
[09:34] <mamund> google his web site and you'll see the TOC
[09:35] <peers> mamund Looks pretty thorough, thanks
[09:35] <mamund> np
[09:36] <mamund> ther are proly others sources, but that's the first one i reach for<g>
[09:36] <fu-manchu> mamund: yes, it works as well, as long as the media type or API docs define those constructs well enough
[09:36] <mamund> :)
[09:36] <mamund> cool
[09:36] <mamund> i know enough python to be a danger to self and others....
[09:37] <fu-manchu> an OO layer tends to make the domain concepts easier at the expense of inefficient message exchange
[09:37] <fu-manchu> I guess that's my point
[09:37] <fu-manchu> hides the wrong things
[09:37] <mamund> is this kind of parsing done via the "list comprehension" feature of python?
[09:37] <fu-manchu> it could be, but not critical
[09:37] <mamund> ok
[09:38] <fu-manchu> for our media types, we do more walking of JSON object trees than anything else
[09:38] <mamund> ahhh
[09:38] <mamund> kinda like a JSON "dom"
[09:38] <fu-manchu> right
[09:38] <mamund> and the "walking" is customized for the media type?
[09:38] <fu-manchu> yes
[09:38] <mamund> the code "knows where to find stuff"
[09:38] <mamund> yep
[09:38] <fu-manchu> well
[09:39] <fu-manchu> in a generic way
[09:39] <mamund> i se
[09:39] <fu-manchu> not really "where to find stuff" but "what to do with each piece you find"
[09:39] <mamund> ok, understood
[09:39] <mamund> state-machine style
[09:40] <fu-manchu> there's no code doing "x = root.items[3].foo.bar" or other out-of-band knowledge; just recursive descent of child nodes defined by the media type
[09:41] <mamund> i see
[09:49] <steveklabnik> fu-manchu: i'd say that json comment is true for rubyists too
[09:49] <steveklabnik> it goes well with activerecord
[09:49] <steveklabnik> i mean, that's pretty much what activeresource is based on
[09:51] <fu-manchu> interesting. I would have thought the desire to add new methods to the native types would overwhelm any Ruby effort ;)
[09:51] mephju (~mephju@dslb-094-222-022-096.pools.arcor-ip.net) left irc: Quit: Verlassend
[09:52] <steveklabnik> hahah
[09:52] <mamund> ohhhhhhhh
[09:52] <steveklabnik> ?
[09:53] <mamund> just trying to goad you two into a cat fight over ruby<g>
[09:53] <steveklabnik> lol
[09:53] <fu-manchu> haha
[09:53] <mamund> :)
[09:53] <steveklabnik> python is a fine language, i just never actually use it. because i use ruby instead.
[09:54] <fu-manchu> now that ruby development is stabilizing, I'm trying to figure out where all the cool kids are going next so they can feel like they influenced a language
[09:55] <fu-manchu> maybe they're all working on node.js now instead
[09:55] <mamund> bingo
[09:55] <steveklabnik> many rubists are using node, yeah
[09:55] <steveklabnik> node is too specific though
[09:55] <steveklabnik> i dont think it's the future.
[09:55] <mamund> future of ????
[09:55] <whartung> mankind
[09:55] <steveklabnik> genearl web applcations
[09:56] <whartung> or web apps, that too
[09:56] <steveklabnik> it is not the next rails.
[10:02] <steveklabnik> the world doesnt neccesarily need another rails...
[10:03] <mamund> well, ruby is not "the next rails" either<g>
[10:03] <mamund> IOW, i think the comparision is better done between ruby and node, actually
[10:03] <mamund> maybe not
[10:04] <steveklabnik> i mean, you can do reactor in ruby, it's been there forever
[10:05] <mamund> :)
[10:05] <steveklabnik> MRI has been limiting in the past, but now that jruby and rbx are getting there...
[10:05] bigbluehat (~bigblueha@adsl-74-177-96-212.gsp.bellsouth.net) joined #rest.
[10:05] steveklabnik shrugs
[10:05] <mamund> IMO, rails is so appealing because ruby is so appealing
[10:05] <mamund> js? is that appealing?
[10:06] <steveklabnik> see, this is where i get a little bit language-ist.
[10:06] <mamund> :)
[10:06] <steveklabnik> which is why i try to stick to the async vs. standard line of argument
[10:06] <steveklabnik> ... which I do think is true.
[10:06] <mamund> ??
[10:07] <steveklabnik> like, node being all Reactor, non-blocking. vs 'standard webapp mvc'
[10:07] <mamund> hmmm
[10:07] <steveklabnik> i actually like javascript.
[10:07] <mamund> so node's async is what? unappealing?
[10:07] <steveklabnik> i dont think it's useful for normal apps.
[10:07] <steveklabnik> basically.
[10:07] <mamund> ahhh
[10:07] <steveklabnik> it is not a clear win.
[10:08] <mamund> hmmm
[10:08] <mamund> normal apps are what? web apps?
[10:08] <steveklabnik> yeah. the kind of thing Rails does. largely server-side, CRUDdy
[10:08] <mamund> ok
[10:08] <mamund> so, let me ask this....
[10:08] <mamund> if you coded in ruby/rails just as you do today...
[10:09] <mamund> but the _results_ we async, non-blocking, would you like ruby/rails less?
[10:09] <steveklabnik> nope. it's the jumping through hoops to get the asynch that makes it not worth it.
[10:10] <mamund> yes!
[10:10] <mamund> that's my POV
[10:10] <mamund> "jumping through hoops"
[10:10] <mamund> from my perspective, this is what rails did
[10:10] <mamund> we used to "jumpt through hoops" to do web apps (forms, state, etc.)
[10:11] <mamund> rails "fixed that"
[10:11] <steveklabnik> yeah.
[10:11] <mamund> by adopting a very dev-friendly lang and marrying it to a very easily-groked pattern (MVC)
[10:11] <mamund> so....
[10:11] <mamund> i think the probl for node is not the tech.
[10:11] <mamund> it's the dev expoerience.
[10:11] <steveklabnik> well, i'm not sure you can get around it.
[10:12] <mamund> and we're pretty early on the curve for that.
[10:12] <steveklabnik> that is quite true.
[10:12] <mamund> well, i think that was the common belief before rails.
[10:12] <mamund> anyway, it's all pure speculation right now
[10:13] <mamund> just my idel thoughts<g>
[10:13] <steveklabnik> :)
[10:13] <steveklabnik> agreed.
[10:29] <mikekelly> erm
[10:29] <mikekelly> mamund: that's kind of missing the point a bit
[10:29] <mamund> yes, i think so, too.
[10:29] <mamund> as soon as i saw the step about adding methods, i saw trouble
[10:30] <mikekelly> async functional code is not going to compete with oop in western world
[10:30] <mamund> LOL
[10:30] <mamund> ever?
[10:30] <mamund> :)
[10:30] <mikekelly> our whole western world is built on objectifying things
[10:30] <mamund> heee
[10:31] <fu-manchu> more importantly, most webapps are much more sensitive to latency (for their human users) than they are to throughput. async approaches tend to optimize for the latter
[10:31] <mikekelly> ? it's true
[10:31] <mikekelly> + javascript is a wack language
[10:31] <mamund> erlang
[10:31] <mikekelly> it's cool but it's not at all friendly
[10:32] <mikekelly> ruby is about as friendly as you can get
[10:32] <Jarda> haha
[10:32] <mamund> yep - designed to be so.
[10:32] <Jarda> language wars
[10:32] <Jarda> C#!
[10:32] <Jarda> Microsoft!
[10:32] <trygvis> scala!!
[10:32] <fu-manchu> F#!
[10:32] mamund runs to grab ptich fork and torch
[10:32] <trygvis> F##$%&#/(%
[10:32] <fu-manchu> :D
[10:32] <mikekelly> I don't think there's much contention that javascript is a shit langauge on the whole
[10:32] <Jarda> I love javascript
[10:33] <Jarda> it has it's stupid legacy shitty stuff
[10:33] <Jarda> but the core language is great
[10:33] <trygvis> meh
[10:33] <trygvis> haskell!
[10:33] <Jarda> most people saying "javascript sucks" are actually saying "browser dom model sucks"
[10:33] <mikekelly> I just don't see async stuff like node taking over the web world
[10:34] <mikekelly> pretty much the point of HTTP is that you don't really have to care too much about the backend
[10:34] <mikekelly> ok if you have auth that changes a bit
[10:34] <mikekelly> depending on how you do auth.
[10:34] <mikekelly> :)
[10:35] <mikekelly> the bit of the web world it can and probably will take over is the smart proxy stuff
[10:35] <mikekelly> which is what I use it for atm, becaue I'm too lazy to figure out how to write nginx modules
[10:36] <mikekelly> + even if I could write nginx modules there aren't that many people who could maintain it
[10:36] <Jarda> Yeah well I don't really care what language everyone uses
[10:37] <Jarda> you can do a fine http backend with almost any language
[10:37] <Jarda> some make it easier, some not..
[10:37] <trygvis> haskell!!
[10:37] <mikekelly> the one thing ruby/rails has (and node too) are good ecosystems
[10:37] <Jarda> and it also depends a lot of the framework used
[10:38] <mikekelly> gems and bundler are pretty big win for ruby
[10:38] <Jarda> and IMO that is one thing that node.js is still missing, for serious web dev
[10:38] <Jarda> express etc aren't that mature
[10:38] <mikekelly> express is pretty good
[10:39] <mikekelly> the front-end for our datastore is written in express
[10:39] <mikekelly> serve a fair wack of volume per day atm, no probs
[10:39] <Jarda> well, it doesn't really matter
[10:40] <Jarda> we have good http caches and stuff
[10:40] <mikekelly> sure, we have to transform the data too as it comes in and out
[10:41] <mikekelly> but anyway it's baically a smart proxy
[10:41] <steveklabnik> mikekelly: lol at "objectifying things"
[10:41] <mikekelly> steveklabnik: that was kind of not a joke -_-
[10:41] <mamund> :)
[10:42] <mikekelly> our language is built to do that
[10:42] <steveklabnik> mikekelly: i'm not joking either
[10:42] <steveklabnik> i'm an anarcho-communist, i'm picking up what you're putting down.
[10:42] <steveklabnik> the only way to deal with it is laugh :(
[10:42] <mikekelly> oh ha :)
[10:43] <mikekelly> yeah well - our way of operating hasn't done us too bad
[10:44] <mikekelly> there's a lot of negativity around west atm but there's huge waves of new tech coming through that will do weird things to the planet
[10:44] <mikekelly> it's not that pretty under the hood but it works
[10:45] <mikekelly> am I talking about rails or society?
[10:45] <mikekelly> this is confusing.
[10:46] <mamund> :)
[10:46] <steveklabnik> meh.
[10:46] <mikekelly> steveklabnik: if you're anarcho-communist does that mean Ron Paul is the devil?
[10:46] <steveklabnik> not the devil. just misguided.
[10:46] <mikekelly> lol
[10:47] <steveklabnik> the small-state capitalists have their hearts in the right place, they just don't know what it's like to not be white ;)
[10:47] <steveklabnik> aaaaaaand now i'm off-topic.
[10:47] <mikekelly> I think you've just watched too many of those Zeitgeist films.. amirit?!?!?! *trollface*
[10:48] <mikekelly> project penus or whatever
[10:48] <steveklabnik> lol
[10:49] <mikekelly> i'm from the "I don't have a party but if you ask me what I think about something I'll tell you" party
[10:50] <mikekelly> i'm deeply religious though
[10:51] <mikekelly> I believe in Richard Dawkins
[10:51] <steveklabnik> haha.
[10:51] <mikekelly> I'm pretty militant about that
[10:51] <mikekelly> just like him.
[10:52] <mikekelly> how the hell did I end up blabbing about this? ah yes. node.js.
[10:52] <steveklabnik> lol.
[10:53] <mikekelly> mamund: btw i'm not sure I like the connect stuff
[10:53] <mamund> why?
[10:53] <mikekelly> because the pattern is all fucked up and not very javascripty
[10:53] <mamund> LOL
[10:53] <mikekelly> or callbacky
[10:54] <mikekelly> I dunno what it is about connect it just never clicked
[10:54] <mikekelly> and then I saw jsgi which made tonnes more sense
[10:54] <steveklabnik> what do you guys think of the new subbu thingy?
[10:54] <steveklabnik> ql.io or whatevs
[10:55] <mikekelly> looks like a lot of hard work has gone into it, it doesn'tseem to solve any problems I have though
[10:56] <mikekelly> mamund: have you done anything with it ?
[10:56] <mamund> tested a bit
[10:56] <mamund> not a fan
[10:57] <mamund> that's cuz it doesn't solve problems i have, tho
[10:58] <mamund> not really interested in a proxy that handles cro-site queries using a single query lang and converts all the results into a single data format.
[10:59] <steveklabnik> yeah.
[10:59] <steveklabnik> it took me a bit just to figure out what the hell it was
[11:00] <mikekelly> right, I can see a need for the snazzy proxy stuff but Im pretty sure node gives me enough to build the bits I need without too much extra work
[11:00] <mamund> well, i think the POV of ql.io is...
[11:00] <mikekelly> and I don't really want the query language
[11:00] <mamund> "i need all this data from other people who don't give me data in the way i want it."
[11:02] <mikekelly> mamund: but the idea is that you write an adapter for the service and then query it through the special language ?
[11:03] <mamund> not sure of where this is deployed
[11:03] <mamund> he has several examples listed
[11:03] <mikekelly> I worked on a cloud computing project about 3 years ago where we were tyring to integrate a handful of different services into one
[11:04] <mamund> YQL, ODATA, SPARQL, ql.io all seem to be attempting to solve the same problem.
[11:04] <mikekelly> we created indiividual 'facades' basically for each service to bring them in line
[11:04] <mikekelly> that seemed to work ok
[11:05] <mikekelly> but there was no query language in there, it was just ad-hoc since all the services were so different
[11:06] <mikekelly> maybe it's apples/oranges
[11:06] <mikekelly> probably genius and going over my head
[11:06] mikekelly shrugs
[11:07] <mikekelly> I actually really like node btw, I don't want to sound like a node hater cos I'm not :)
[11:07] <mamund> LOL
[11:09] <mamund> i often setup HTTP "facades" for other services
[11:09] <mamund> sometimes just putting HTTP in front of SOAP
[11:10] <mikekelly> yeah we had to do that for performance reasons :)
[11:10] <mamund> bingo
[11:11] <steveklabnik> :)
[11:11] <mamund> have done this for CSV-style services, too.
[11:11] <mamund> even did some connectors to IBM360 quite a while back
[11:12] <mikekelly> random tangent.. I see you saw the ben carlyle bit on http evolution
[11:12] <mamund> yep
[11:12] <mikekelly> seemed pretty comprehensive but basically same stuff as before right?
[11:12] <mamund> yeah, imo
[11:12] <mikekelly> tbh I just skimmed it cos I'm lazy -_-
[11:12] <steveklabnik> link?
[11:13] <mikekelly> http://soundadvice.id.au/blog/2011/12/06/
[11:13] <steveklabnik> gracias
[11:13] <mikekelly> mamund: but I was guessing your tweet implied what I replied
[11:14] <mikekelly> "yeah ok cool but where's the rest?"
[11:14] <mikekelly> I almost posted that but didn't in the end cos it was too cheesey
[11:15] <mikekelly> plus the title didn't actually mention rest, to be fair
[11:15] <mamund> yes, that's what i said to myself
[11:15] <mikekelly> the opening paragraph does though.. ;)
[11:15] <mamund> yep
[11:16] fu-manchu plugs cherrypy as a great tool for such facades :)
[11:24] bob_sage (~chatzilla@outbound4.ebay.com) left irc: Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243]
[11:29] bigbluehat (~bigblueha@adsl-74-177-96-212.gsp.bellsouth.net) left irc: Quit: Leaving.
[11:39] grahamc (~gchristen@173-166-174-177-washingtondc.hfc.comcastbusiness.net) left irc: Quit: Leaving...
[11:47] <mikekelly> mamund: I emailed Jim about that Auth thing, so be interesting to see what he thinks
[11:47] Wombert (~Wombert@dslb-088-066-201-154.pools.arcor-ip.net) left irc: Quit: Wombert
[11:48] <mikekelly> I'm tempted to just blog about it and see what happens
[11:48] <mamund> yeah, that's a good start, i think
[11:48] <mikekelly> I'll see what Jim says first, if he says I'm retarded than I think I'll skip the blog post
[11:49] <mamund> LOL
[11:49] <mikekelly> Jim ?
[11:49] <mikekelly> I meant Ian
[11:49] <mamund> sounds better
[11:49] <mikekelly> whatever. same person, basically
[11:49] <mamund> heeee
[11:49] <mikekelly> they can read it together while they snuggle
[11:49] <mamund> the jim-ian hybrid?
[11:50] <mikekelly> yeah I used Jian(tm) in a rest-discuss post
[11:50] mamund struggles to get mental image out of his mind
[11:50] <mikekelly> did anyone spot that?
[11:50] <mikekelly> I don't even know Ian or Jim that well I really shouldn't go around making jokes like that :D
[11:51] <mikekelly> they're cool dudes though I'm sure they're fine with it..
[11:51] <mikekelly> maybe. hopefully.
[11:51] <mamund> i know ian pretty well, and he's pretty cool
[11:52] <mikekelly> yeah, apart from his machine-form delusion
[11:52] <mamund> LOL
[11:52] <mikekelly> nobody's perfect.
[11:52] <mamund> he's closer to your POV than mine, you know.
[11:52] <mikekelly> he wasn't at that rest circle jerk thing we went to
[11:52] <mikekelly> what was that called?
[11:53] <mamund> WS-REST?
[11:53] <mikekelly> nah the one in munich
[11:53] <mamund> ahh
[11:53] <mamund> i missed that one
[11:53] <mikekelly> rest something
[11:53] <mikekelly> dunno
[11:53] <mamund> restunconf
[11:53] <mikekelly> that's it thank you
[11:53] <mikekelly> yeah I remember talking to Ian in a bar about xforms and link relations
[11:54] <mikekelly> it came up again on Stefan's blog yesterday again
[11:54] <mamund> ok
[11:54] <mikekelly> and he said pretty much the same thing
[11:54] <mikekelly> on there
[11:54] <mikekelly> 'your overloading rels'
[11:55] <mikekelly> but he made a post that basically was the thinking behind hal
[11:56] <mikekelly> infact, did you read that thread on stefan's blog ?
[11:56] <mikekelly> innoq.com/blog/stilkov/blog/2011/12/media-types/
[11:56] <mikekelly> http://innoq.com/blog/stilkov/blog/2011/12/media-types/
[11:56] <mamund> yep, saw that
[11:56] <mikekelly> http://www.innoq.com/blog/stilkov/2011/12/media-types/#comment-378264821
[11:56] <mamund> very happy to see this kind of talk going on
[11:57] <mikekelly> "HAL looks like a nice way to decorate stuff with links, but is perhaps too general for what we sought to achieve. I may be mistaken, but as I understand it, HAL doesn't tell me anything about what sits inside a resource element - which is nicely domain agnostic, but just shuffles the issue of determining what syntax or structure I can expect to have to deal with somewhere else. (One way of dealing with this with something like HAL would be by having the link rela
[11:57] <mamund> :)
[11:57] <mikekelly> but "just shuffles" is loaded language
[11:58] <mikekelly> which makes a design descision sound like something negative
[11:58] <mamund> hand-wavy, unspecific
[11:58] <mikekelly> it has 'just moved it' - but it's moved it somewhere important
[11:58] <mikekelly> away from the resource itself
[11:58] <mamund> :)
[11:59] <mikekelly> and into the place where the application is driven (i.e. the link preceding it)
[11:59] <mikekelly> I actually think a media type which enforces that constraint is a Good Thing
[11:59] <mikekelly> it's good because it's natural for server side to expose resources that way
[11:59] <mamund> that is the job of a media type (IMO)
[12:00] <mikekelly> and it's good for clients because all they can rely on is the rel
[12:00] <mikekelly> i.e. you know where you stand
[12:00] <mamund> to make these kinds of "forcing decsions" and apply uniformity/rigor to the representations.
[12:00] <mamund> yepo
[12:00] <mikekelly> I don't know how to say that in blog'y words
[12:00] <mikekelly> please can someone do that
[12:00] <mamund> :)
[12:01] <mikekelly> and finish the post with 'so mike sez just use hal and stfu about media types plz thx'
[12:01] <mamund> haha
[12:01] <mikekelly> we really just need everyone to get a boner for minting link relations
[12:02] <mikekelly> jus take the boner they have for media types and trasnfer it to link relations
[12:02] <mikekelly> and we're golden.
[12:02] <mamund> i think some major headway has been made recently in that area
[12:02] <mikekelly> headway
[12:02] <mikekelly> boner
[12:02] <mikekelly> there's definitely a joke in there
[12:02] <mamund> what i think needs to be done now is to "mint" lots of public client apps that take advantage of this feature
[12:03] mamund ignores that bathroom humor from the front row
[12:03] <mikekelly> :D
[12:03] <mikekelly> the apis I work on atm are all private
[12:03] <mamund> srsly, tho.
[12:03] <mikekelly> even the docs :|
[12:03] <mamund> yeah, that is most of my work, too.
[12:03] <mikekelly> which reminds me
[12:03] <mamund> i have to spend "off hours" doing stuff i can show publicly
[12:04] <mikekelly> I need to stop having interesting conversations with you and build something I can show people
[12:04] <mikekelly> basically what I'm saying is it's all your fault mike
[12:04] <mikekelly> as usual
[12:05] <mikekelly> .. you don't even care enough to respond.
[12:05] <mikekelly> shameful
[12:06] <mikekelly> mamund: I followed up w/ Ian but he hasn't responded
[12:06] <mamund> lol
[12:06] <mikekelly> "you're absolutely right - HAL does push everything into the link relation. How is this distinct from your original comment "following a typed link (e.g. a link with link relation "order"), the client's expectations are set with regard to what the response will represent" ? How does the specific media type you are describing handle this differently to HAL?"
[12:06] bigbluehat (~bigblueha@adsl-74-177-96-212.gsp.bellsouth.net) joined #rest.
[12:07] <mikekelly> fair or ?
[12:07] <mikekelly> how would you answer that?
[12:31] nuclearsandwich (~nuclearsa@64-79-127-126.static.wiline.com) left irc: Remote host closed the connection
[12:32] DracoBlue (~Adium@dslb-088-075-066-254.pools.arcor-ip.net) joined #rest.
[12:32] DracoBlue (~Adium@dslb-088-075-066-254.pools.arcor-ip.net) left #rest.
[12:36] Wombert (~Wombert@dslb-088-066-201-154.pools.arcor-ip.net) joined #rest.
[12:55] technoweenie (~technowee@host-86-220-9-69.midco.net) left irc: Remote host closed the connection
[13:10] mamund was away'
[13:11] <mamund> mikekelly: you still around?
[13:26] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[13:42] talios (~amrk@203.84.135.120) joined #rest.
[14:07] peers (~ajp5103@128.118.235.29) left irc:
[14:33] nuclearsandwich (~nuclearsa@sjs-mh-wifi-1-1-lc-int.sjsu.edu) joined #rest.
[14:34] bigbluehat (~bigblueha@adsl-74-177-96-212.gsp.bellsouth.net) left irc: Quit: Leaving.
[14:53] Ngarthl (erlenha@colargol.tihlde.org) got netsplit.
[14:57] grove (~grove@193.201.9.46.customer.cdi.no) left irc: Quit: grove
[15:04] Ngarthl (erlenha@colargol.tihlde.org) got lost in the net-split.
[15:08] mamund is all done for the day, ta-ta!
[15:08] <steveklabnik> bye!
[15:34] Wombert (~Wombert@dslb-088-066-201-154.pools.arcor-ip.net) left irc: Quit: Wombert
[15:36] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Quit: Leaving...
[15:44] mr_yall (~mryall@host86-160-76-32.range86-160.btcentralplus.com) left irc: Ping timeout: 240 seconds
[15:54] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[16:43] dreinull (~dreieins@217.18.70.225) left irc: Ping timeout: 248 seconds
[17:23] nuclearsandwich (~nuclearsa@sjs-mh-wifi-1-1-lc-int.sjsu.edu) left irc: Remote host closed the connection
[17:40] dreinull (~dreieins@217.18.70.225) joined #rest.
[17:55] <steveklabnik> mamund: for when you return: https://gist.github.com/1431783
[18:01] sbanwart__ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[18:05] sbanwart__ -> sbanwart
[18:06] Wombert (~Wombert@dslb-088-066-201-154.pools.arcor-ip.net) joined #rest.
[18:15] KevBurnsJr (~KevBurnsJ@50.0.103.39) left irc:
[18:33] sbanwart_ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[18:34] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 244 seconds
[19:00] sbanwart__ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[19:01] sbanwart_ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 268 seconds
[19:01] <steveklabnik> mamund: actually, I decided to give it a full repo: https://github.com/steveklabnik/web_maze
[20:11] Wombert (~Wombert@dslb-088-066-201-154.pools.arcor-ip.net) left irc: Quit: Wombert
[20:19] kennethreitz (~kennethre@adsl-99-191-194-225.dsl.pltn13.sbcglobal.net) joined #rest.
[20:28] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Quit: Leaving...
[20:29] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[20:35] f00li5h (~f00li5h@unaffiliated/f00li5h) joined #rest.
[20:40] sbanwart__ (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 240 seconds
[20:48] talios (~amrk@203.84.135.120) left irc: Quit: Bye!
[21:11] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Quit: Leaving...
[21:11] fu-manchu (~fumanchu@adsl-99-30-180-185.dsl.sfldmi.sbcglobal.net) left irc: Quit: Lost terminal
[21:13] fu-manchu (~fumanchu@adsl-99-30-180-185.dsl.sfldmi.sbcglobal.net) joined #rest.
[21:42] kennethreitz (~kennethre@adsl-99-191-194-225.dsl.pltn13.sbcglobal.net) left irc: Quit: Computer has gone to sleep.
[22:11] kennethreitz (~kennethre@ip65-46-252-62.z252-46-65.customer.algx.net) joined #rest.
[22:24] kennethreitz (~kennethre@ip65-46-252-62.z252-46-65.customer.algx.net) left irc: Quit: Computer has gone to sleep.
[22:26] grove (~grove@aggw006.cappelendamm.no) joined #rest.
[23:15] nuclearsandwich (~nuclearsa@74-93-3-241-SFBA.hfc.comcastbusiness.net) joined #rest.
[23:28] nuclearsandwich (~nuclearsa@74-93-3-241-SFBA.hfc.comcastbusiness.net) left irc: Remote host closed the connection
[00:00] --- Tue Dec 6 2011