#rest on irc.freenode.net 2011-11-14 (txt)

[00:16] ivanfi (~ivanfi@62.159.77.167) joined #rest.
[00:36] grove (~grove@aggw006.cappelendamm.no) left irc: Quit: grove
[00:43] grove (~grove@aggw006.cappelendamm.no) joined #rest.
[02:05] grove (~grove@aggw006.cappelendamm.no) left irc: Ping timeout: 245 seconds
[02:11] darrelmiller_ (~darrelmil@bas1-montreal08-1279584872.dsl.bell.ca) joined #rest.
[02:13] darrelmiller (~darrelmil@70.49.83.206) left irc: Ping timeout: 252 seconds
[02:32] grove (~grove@aggw006.cappelendamm.no) joined #rest.
[03:48] ivanfi (~ivanfi@62.159.77.167) left irc: Read error: Connection reset by peer
[03:56] ivanfi (~ivanfi@62.159.77.167) joined #rest.
[04:31] Hakon|mbp (~hakon1@c51009CE2.inet.catch.no) joined #rest.
[04:36] grove_ (~grove@aggw006.cappelendamm.no) joined #rest.
[04:39] grove (~grove@aggw006.cappelendamm.no) left irc: Ping timeout: 276 seconds
[04:39] grove_ -> grove
[05:45] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[05:54] darrelmiller_ (~darrelmil@bas1-montreal08-1279584872.dsl.bell.ca) left irc: Quit: ~ Trillian Astra - www.trillian.im ~
[06:19] grove (~grove@aggw006.cappelendamm.no) left irc: Quit: grove
[06:20] bigbluehat (~bigblueha@adsl-98-71-147-122.gsp.bellsouth.net) joined #rest.
[06:23] wav1 (~Adium@cpe-70-112-49-11.austin.res.rr.com) left irc: Quit: Leaving.
[06:23] darrelmiller (~darrelmil@bas1-montreal08-1279584872.dsl.bell.ca) joined #rest.
[06:25] wav1 (~Adium@cpe-70-112-49-11.austin.res.rr.com) joined #rest.
[06:26] wav1 (~Adium@cpe-70-112-49-11.austin.res.rr.com) left irc: Client Quit
[06:36] mamund sez 'hi'
[06:42] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Quit: Leaving...
[07:18] grove (~grove@193.201.9.46.customer.cdi.no) joined #rest.
[07:20] <mikekelly> yo
[07:21] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[07:22] <mikekelly> hmm.
[07:23] <mikekelly> steveklabnik: what do you think about rails' routing dsl ?
[07:23] grahamc (~gchristen@173-166-174-177-washingtondc.hfc.comcastbusiness.net) joined #rest.
[07:26] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Ping timeout: 248 seconds
[07:45] grahamc (~gchristen@173-166-174-177-washingtondc.hfc.comcastbusiness.net) left irc: Quit: Leaving...
[07:56] kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) left irc: Quit: Textual IRC Client: http://www.textualapp.com/
[07:58] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[08:00] kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) joined #rest.
[08:22] bradley-holt (~bradley-h@65-183-135-35-dhcp.burlingtontelecom.net) joined #rest.
[08:23] ivanfi (~ivanfi@62.159.77.167) left #rest.
[08:24] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[08:33] <ssedano> hi
[08:34] <ssedano> if I have a collection of items, that I return in a list of elements: <elements><link ref="http://example.com/elements/1"; rel="self" /></elements>
[08:34] <ssedano> is that right? or the rel should be another?
[08:34] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Quit: Leaving...
[08:37] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[09:05] quest88 (~quest88@c-98-207-205-137.hsd1.ca.comcast.net) left irc: Quit: quest88
[09:14] mrryanjohnston (04358972@gateway/web/freenode/ip.4.53.137.114) joined #rest.
[09:21] <mrryanjohnston> would a search, such as http://server/person?fname=rob be a POST or a GET?
[09:22] <Hakon|mbp> get
[09:22] <mrryanjohnston> thought so. thank you!
[09:26] KevBurnsJr (~KevBurnsJ@50.0.103.39) joined #rest.
[10:00] <steveklabnik> mikekelly: i like it, mostly
[10:00] <steveklabnik> mamund: hi! can you check out the backlog from last night, and help clarify my thinking on sjtgraham's assertio
[10:00] <steveklabnik> assertion
[10:00] <steveklabnik> about PUT vs PATCH?
[10:00] <mikekelly> steveklabnik: there's no clear benefits to having PUT be a full replacement
[10:01] <mikekelly> but that's basically how it's defined so
[10:01] <steveklabnik> sure.
[10:01] <mikekelly> we don't have idempotent partial updates
[10:01] <steveklabnik> the question is one of definition
[10:01] mikekelly shrugs
[10:01] <steveklabnik> there's a line in the HTTP 1.1 spec that says 'this spec doesnt' define how put changes the server state'
[10:01] <mikekelly> or.. just ignore the spec like rails does :D
[10:01] <steveklabnik> and it's confusing to me.
[10:01] <steveklabnik> heh.
[10:01] <mikekelly> same here
[10:01] <steveklabnik> to be fair, it's in discussion for rails 3.2
[10:01] <steveklabnik> actually
[10:01] <mikekelly> plus there's no real mechanisms that relyh on 'full pPUT smenatics'
[10:02] <mikekelly> i.e. no caches actually respond with the PUT body
[10:02] <mikekelly> that's about as basic an intermeediary mecahnism as you can get and it doesn't happen
[10:02] <mikekelly> so you have to ask why that is useful at an application level
[10:02] <mikekelly> layer
[10:02] <mikekelly> i.e. layer 7 (http)
[10:02] <steveklabnik> mm
[10:03] <mikekelly> does rail's use of 'resources' or 'resource' not bother you btw ?
[10:04] <steveklabnik> not particularly. why would it? i mean, i just never thought about it, i guess
[10:08] <technoweenie> it kind of blows my mind people are still figuring out how PUT requests should work. i blogged about it in april and got a tweet saying Roy just redefined PUT a few days earlier to clarify
[10:09] <technoweenie> http://tools.ietf.org/html/draft-ietf-httpbis-p2-semantics-14#section-7.6
[10:10] <steveklabnik> interesting.
[10:11] <technoweenie> i try to follow that (which is easy having written an API from scratch this year), but i was also part of the Rails team when we made PUT the rails default
[10:17] Hakon|mbp (~hakon1@c51009CE2.inet.catch.no) left irc: Quit: Leaving...
[10:23] steveklabnik nods
[10:23] <steveklabnik> oh, technoweenie, i'm working on an experimental gem... steveklabnik/hateoas. Though I'm a bad git and merged in a 'remove ALL THE CODE' commit into master
[10:24] <steveklabnik> sorta like a cross between capybara and mechanize
[10:24] <technoweenie> steveklabnik: yea i've got one called Sawyer i'm hacking on
[10:24] <steveklabnik> i'm not sure it's going to be useful/interesting yet, but since you're working on similar stuff
[10:24] <technoweenie> mine wont be dependent on http i dont think. i wrote a zeromq adapter for faraday
[10:25] <steveklabnik> :)
[10:26] <technoweenie> sawyer is ready though i think. i'm going to use it in my next internal API
[10:26] <steveklabnik> cool. i remember checking it out a while ago
[10:26] <technoweenie> yea i've rewritten it 3 times, i think i have it now
[10:31] Hakon|mbp (~hakon1@158.222.202.84.customer.cdi.no) joined #rest.
[10:38] technoweenie (~technowee@host-86-220-9-69.midco.net) left irc: Remote host closed the connection
[10:57] KevBurnsJr_ (~KevBurnsJ@50.0.103.39) joined #rest.
[10:59] KevBurnsJr_ (~KevBurnsJ@50.0.103.39) left irc: Client Quit
[11:00] <KevBurnsJr> Just submitted a PO for Hypermedia APIs through Beatport education budget =D
[11:00] <KevBurnsJr> http://www.amazon.com/dp/1449306578
[11:00] <steveklabnik> :)
[11:03] <mamund> :)
[11:35] grove (~grove@193.201.9.46.customer.cdi.no) left irc: Quit: grove
[11:38] mrryanjohnston (04358972@gateway/web/freenode/ip.4.53.137.114) left irc: Quit: Page closed
[11:39] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Ping timeout: 260 seconds
[11:50] Hakon|mbp (~hakon1@158.222.202.84.customer.cdi.no) left irc: Quit: Leaving...
[11:55] mamund notices it's kina quiet here
[11:56] <darrelmiller> That's 'cos whartung is not here.
[11:56] Hakon|mbp (~hakon1@158.222.202.84.customer.cdi.no) joined #rest.
[12:20] <mamund> LOL
[12:21] <mamund> i'm "heads-down" trying to get some end-of-year stuff closed out
[12:21] <mamund> haven't been very "chatty" lately
[13:42] Nicolas_Leonidas (~chatzilla@cmr-208-97-116-42.cr.net.cable.rogers.com) joined #rest.
[13:42] <Nicolas_Leonidas> hi
[13:42] <Nicolas_Leonidas> can someone help me with this question I have here http://stackoverflow.com/questions/8110540/rest-vs-xmlrpc-vs-my-own-json-outputting-script
[13:42] mamund waves to Nicolas_Leonidas
[13:43] <Nicolas_Leonidas> I'm not sure whether I should move to rest or xmlrpc or keep doing what I'm doing
[13:43] Nicolas_Leonidas waves back at mamund
[13:43] mamund backs up, SO is not his area<g>
[13:43] <mamund> darrelmiller: this is yours<g>
[13:43] <Nicolas_Leonidas> mamund: it's a REST related question
[13:43] <Nicolas_Leonidas> it doesn't have much to do with SO
[13:44] <mamund> darrelmiller has eperience w/ SO and RESTy bits there
[13:48] <mamund> Nicolas_Leonidas: yeah, i scanned the Q, not at all sure what this is about.
[13:48] <Nicolas_Leonidas> mamund: so see I've already made a server like thing for my JSON calls do you know what JSON is?
[13:49] <mamund> yes
[13:49] <Nicolas_Leonidas> so it has some commands and stuff, it works with my pages
[13:49] <Nicolas_Leonidas> I'm wondering if I should have use REST instead
[13:49] <mamund> instead of waht?
[13:49] <darrelmiller> mamund: Oh, you're not stepping on my toes, feel free ;-)
[13:49] <Nicolas_Leonidas> instead of building my own json server
[13:50] <mamund> no darrelmiller hop right in here
[13:50] mamund swoops his arm in a circle, palm up, pointing to his pal darrelmiller
[13:52] <darrelmiller> Nicolas_Leonidas: An application built based on the principles of REST would be quite an effective way of serving up JSON responses.
[13:55] <darrelmiller> Nicolas_Leonidas: In fact it has the added benefit of being just as capable of returning pretty much any media type.
[13:55] <Nicolas_Leonidas> darrelmiller: so you think I should move to REST right?
[13:55] <Nicolas_Leonidas> what is the main difference between REST and XMLRPC?:
[13:55] <Nicolas_Leonidas> and SOAP?
[13:56] <darrelmiller> XMLRPC is limited to XML and only takes advantage of a small subset of the capabilities of HTTP.
[13:57] <darrelmiller> REST is more flexible. However, it is often more work to get going. I can't say whether it is the right choice for you.
[13:57] <darrelmiller> I would need much more information about your application.
[13:58] <Nicolas_Leonidas> darrelmiller: well see, this part of application only needs to receive some data from the server, and send some data to it
[13:58] <Nicolas_Leonidas> for example I authenticate using that json server I have made, I read the time of the server, I ask the server to send the client sorted data and stuff like that
[13:59] <darrelmiller> You may find that some type of JSONRPC is sufficient for you. The principles of REST are there to allow different clients to access your service as it continues to evolve over a long period of time.
[13:59] <Nicolas_Leonidas> darrelmiller: yes, this application does NOT need other clients to use it's server at this point
[14:00] <darrelmiller> Think of it this way, and I may get flamed for this, but REST is a bunch of hoops that you can jump through to avoid having to version your service in the future.
[14:00] mamund gets a torch
[14:00] <darrelmiller> If you don't mind versioning your service and potentially breaking your client then REST may be more effort than it is worth.
[14:01] <darrelmiller> mamund: What? You didn't like that description :-)
[14:01] <mamund> LOL
[14:01] <mamund> just tossing bon mots from the balcony
[14:02] <darrelmiller> I thnk am going to tweet that and see how much abuse I get.
[14:02] grove (~grove@193.201.9.46.customer.cdi.no) joined #rest.
[14:02] <mamund> heeee
[14:03] <darrelmiller> Apparently someone is trying to save me as twitter is currently down.
[14:03] <mamund> karma is sweet
[14:03] <mamund> Nicolas_Leonidas:
[14:03] <darrelmiller> Yeah, but I'm stupidly persistent.
[14:03] <mamund> you getting what you need here?
[14:06] <Nicolas_Leonidas> REST vs RPC shouldn't be that different in my environment, I'm using Zend Framework, very similar setup
[14:07] <mamund> Nicolas_Leonidas: not sure that you mean here.
[14:07] <mamund> you mean you think REST is the same as RPC?
[14:07] <mamund> or you mean somethning else?
[14:07] <mamund> finally, what is it you want to do anyway?
[14:08] <Nicolas_Leonidas> mamund: well no, but can you tell me what separates the two?
[14:08] <mamund> ahh
[14:08] <mamund> ok
[14:08] <mamund> R
[14:08] <Nicolas_Leonidas> in simple baby language?
[14:08] <mamund> RPC is a style of arch where client and server share a list of procedures
[14:09] <mamund> REST is a style of arch where client and server share a media type
[14:09] <mamund> in REST the list of procedures is not public and is not of interest to clients.
[14:09] <mamund> rEST clients are interested in the hypemedia controls in a message (links and forms) and the data that comes back in responses.
[14:09] <mamund> does that help?
[14:10] <Nicolas_Leonidas> so you are saying in REST, we only request say some sort of data and get it
[14:11] <Nicolas_Leonidas> we can't tell the server to do something specific to the data?
[14:11] <mamund> not saying that at all
[14:11] <Nicolas_Leonidas> can you give me an example where REST would be better than RPC and vise versa?
[14:11] <mamund> saying that the _way_ in which this is done in a REST arch is very diff than in an RPC arch
[14:11] <Nicolas_Leonidas> I'm sorry, I was born a slow person
[14:11] <mamund> np
[14:11] <Nicolas_Leonidas> oh, so the outcome is the same, they can both be used for the same reasons?
[14:11] <mamund> slow and steady wins the race, my friend.
[14:11] <mamund> you're doin' fine
[14:12] <mamund> REST is an advantage in systems that..
[14:12] <mamund> 1) client and server are not coded by the same group
[14:12] <mamund> 2) the system (app) must last a long time (years) w/o breaking
[14:13] <mamund> 3) clients are focused on "solving problems" that the servers might not know about (but are a party to)
[14:13] <Nicolas_Leonidas> 2, is where the versioning thing darrelmiller was referring to right?
[14:13] <Nicolas_Leonidas> right, like google map should be a server
[14:13] <mamund> yeah, that's a bit part of it
[14:14] <mamund> REST arch makes it possible to change several parts of the system w/o "breaking things"
[14:14] <mamund> adding caching servers
[14:14] <mamund> changing the origin server to a new domain
[14:15] <mamund> updating the origin server from PHP to JSP, etc.
[14:15] <mamund> supporting new clients that only know HTML, not JSON (or vice versa)
[14:15] <mamund> and so forth
[14:15] <mamund> does that help?
[14:15] <Nicolas_Leonidas> yes, it does help a lot
[14:15] <Nicolas_Leonidas> so google map would be a good example of where REST would work good right?
[14:15] <mamund> it could yes
[14:16] <Nicolas_Leonidas> how about RPC now?
[14:16] <mamund> IIRC, the "API" they expose is not very REST-like, tno.
[14:16] <Nicolas_Leonidas> where does that come into play?
[14:16] <mamund> that's their choice, of course
[14:16] <mamund> RPC owrks well when all those things i meaniotned earlier don't matter enough to do the work needed to mitigate them<g>
[14:17] <mamund> for example
[14:17] DracoBlue (~Adium@dslb-088-075-066-254.pools.arcor-ip.net) joined #rest.
[14:17] <mamund> 1) the list of procedures will not change (no arg changes, etc.)
[14:17] <mamund> 2) the server will not change addreses, domains
[14:18] <mamund> 3) the server will not support clients that need other formats (JSON, XML, CSV, HTML, etc.)
[14:18] <Nicolas_Leonidas> I see, that's very interesting
[14:18] <mamund> 4) the clients will always be hard-coded to know what URIs and bodies to send, the server will not change them.
[14:18] <mamund> cool
[14:18] <mamund> these are chioces in _style_
[14:18] <mamund> folks make them based onwhat they need to get done and teh effort involved
[14:18] <Nicolas_Leonidas> I'll print these and read them more carefully, thank you very much mamund
[14:18] <mamund> LOL
[14:18] <mamund> you're welcome here any time
[14:19] <mamund> lots of smart folks here to help
[14:19] mamund is just kinda mouthy today
[14:19] <Nicolas_Leonidas> thanks have a nice time
[14:21] bigbluehat (~bigblueha@adsl-98-71-147-122.gsp.bellsouth.net) left irc: Quit: Leaving.
[14:22] DracoBlue (~Adium@dslb-088-075-066-254.pools.arcor-ip.net) left irc: Quit: Leaving.
[14:23] Nicolas_Leonidas (~chatzilla@cmr-208-97-116-42.cr.net.cable.rogers.com) left irc: Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243]
[14:28] grove (~grove@193.201.9.46.customer.cdi.no) left irc: Quit: grove
[14:44] mamund is outta-here!
[16:29] twilliams (~twilliams@apache/committer/twilliams) joined #rest.
[17:09] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
[18:07] KevBurnsJr (~KevBurnsJ@50.0.103.39) left irc:
[18:08] gchristensen (~gchristen@unaffiliated/grahamc) joined #rest.
[18:18] twilliams (~twilliams@apache/committer/twilliams) left irc: Quit: Computer went to sleep
[18:20] Hakon|mbp (~hakon1@158.222.202.84.customer.cdi.no) left irc: Quit: Leaving...
[19:15] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Ping timeout: 245 seconds
[20:01] KevBurnsJr (~kevburnsj@c-76-126-10-63.hsd1.ca.comcast.net) joined #rest.
[21:10] KevBurnsJr (~kevburnsj@c-76-126-10-63.hsd1.ca.comcast.net) left irc:
[21:39] grove (~grove@193.201.9.46.customer.cdi.no) joined #rest.
[21:44] grove (~grove@193.201.9.46.customer.cdi.no) left irc: Quit: grove
[21:49] technoweenie (~technowee@host-86-220-9-69.midco.net) joined #rest.
[22:13] darrelmiller (~darrelmil@bas1-montreal08-1279584872.dsl.bell.ca) left irc: Ping timeout: 252 seconds
[22:26] grove (~grove@aggw006.cappelendamm.no) joined #rest.
[22:59] gchristensen (~gchristen@unaffiliated/grahamc) left irc: Quit: Leaving...
[23:53] ivanfi (~ivanfi@62.159.77.167) joined #rest.
[00:00] --- Tue Nov 15 2011