#rest on irc.freenode.net 2012-01-25 (txt)

[00:05] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[00:06] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[00:23] mamund (mamund@nullrouted.info) left irc: Ping timeout: 248 seconds
[00:24] mamund (mamund@69.163.43.34) joined #rest.
[00:24] #rest: mode change '+o mamund' by ChanServ!ChanServ@services.
[00:33] mamund (mamund@69.163.43.34) left irc: Ping timeout: 245 seconds
[00:38] mamund (mamund@nullrouted.info) joined #rest.
[00:38] #rest: mode change '+o mamund' by ChanServ!ChanServ@services.
[00:45] mamund (mamund@nullrouted.info) left irc: Ping timeout: 252 seconds
[00:46] mamund (mamund@nullrouted.info) joined #rest.
[00:46] #rest: mode change '+o mamund' by ChanServ!ChanServ@services.
[01:13] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[01:24] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[01:32] mamund (mamund@nullrouted.info) left irc: Ping timeout: 255 seconds
[01:33] darrelmiller (~darrelmil@70.24.176.12) left irc: Read error: Connection reset by peer
[01:33] mamund (mamund@nullrouted.info) joined #rest.
[01:33] #rest: mode change '+o mamund' by ChanServ!ChanServ@services.
[01:34] darrelmiller (~darrelmil@70.24.176.12) joined #rest.
[01:41] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[02:17] Wombert (~Wombert@dslb-092-075-024-131.pools.arcor-ip.net) joined #rest.
[02:23] Wombert (~Wombert@dslb-092-075-024-131.pools.arcor-ip.net) left irc: Quit: Wombert
[02:50] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[02:52] mephju (~mephju@dslb-188-102-108-167.pools.arcor-ip.net) joined #rest.
[03:17] mhausenblas (~mhausenbl@wlan-nat.fwgal01.deri.ie) joined #rest.
[03:42] mhausenblas (~mhausenbl@wlan-nat.fwgal01.deri.ie) left irc: Quit: mhausenblas
[03:53] mhausenblas (~mhausenbl@wlan-nat.fwgal01.deri.ie) joined #rest.
[04:03] <Jarda> hmm.. in a job listing someone linked on IRC, I found from the keywords: "SOAP Rest" as one keyword
[04:03] <Jarda> excuse me buy wtg
[04:31] <mk_> sounds good.
[04:39] <Jarda> *but *wtf
[04:39] <Jarda> wha't SOAP Rest?
[04:39] <Jarda> wtf with my typing today
[04:39] <Jarda> must be this bt keyboard
[04:41] <mk_> to be fair it could be a project to create REST facade over SOAP endpoints
[05:06] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[05:10] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left irc: Ping timeout: 240 seconds
[05:11] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[05:15] bigbluehat1 (~bigblueha@nmd.sbx10523.greensc.wayport.net) joined #rest.
[05:16] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[05:17] bigbluehat1 (~bigblueha@nmd.sbx10523.greensc.wayport.net) left irc: Client Quit
[05:31] Hakon|mbp (~hakon1@193.137.202.84.customer.cdi.no) joined #rest.
[06:04] bigbluehat1 (~bigblueha@nmd.sbx10523.greensc.wayport.net) joined #rest.
[06:09] bigbluehat1 (~bigblueha@nmd.sbx10523.greensc.wayport.net) left irc: Read error: Connection reset by peer
[06:19] sbanwart (~sbanwart@66.6.175.18) joined #rest.
[06:41] pezra (~Adium@c-71-56-237-160.hsd1.co.comcast.net) joined #rest.
[06:45] Wombert (~Wombert@xdsl-78-34-218-164.netcologne.de) joined #rest.
[06:48] <darrelmiller> Keep your heads down, serialseb is on the warpath.
[07:05] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[07:09] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left irc: Ping timeout: 248 seconds
[07:14] mephju (~mephju@dslb-188-102-108-167.pools.arcor-ip.net) left irc: Quit: Ex-Chat
[07:22] <mk_> :|
[07:23] pezra (~Adium@c-71-56-237-160.hsd1.co.comcast.net) left irc: Quit: Leaving.
[07:23] <mk_> darrelmiller: apparently experts don't talk on rest-discuss
[07:23] bigbluehat1 (~bigblueha@nmd.sbx10523.greensc.wayport.net) joined #rest.
[07:24] <darrelmiller> ohhh, bigbluehat1 I see you. I saw there is some moving and shakin going on in your area.
[07:24] <mk_> :o
[07:26] pezra (~Adium@c-71-56-237-160.hsd1.co.comcast.net) joined #rest.
[07:27] <bigbluehat1> :)
[07:28] <bigbluehat1> guess my mac mini woke up to back up
[07:29] HighBit_ (~eqhmcow@adsl-98-69-182-207.rmo.bellsouth.net) joined #rest.
[07:29] SvenDowideit_ (~SvenDowid@203-206-171-38.perm.iinet.net.au) joined #rest.
[07:31] <darrelmiller> I guess I'm a bit slow. This was posted at the beginning of Jan but I only saw it today http://www.infoq.com/news/2012/01/Katz-CouchDB-Couchbase-Server
[07:32] <trygvis> you snoose, you loose!
[07:33] axisys_ (~axisys@ip68-98-189-233.dc.dc.cox.net) joined #rest.
[07:34] KevBurnsJr_ (~KevBurnsW@50.0.103.39) got netsplit.
[07:34] ssedano (~ssedano@unaffiliated/ssedano) got netsplit.
[07:34] chilversc (~chris@about/csharp/regular/KeeperOfTheSoul) got netsplit.
[07:34] mogsie (~mogsie@62.101.198.35) got netsplit.
[07:34] axisys (~axisys@ip68-98-189-233.dc.dc.cox.net) got netsplit.
[07:34] HighBit (~eqhmcow@adsl-98-69-182-207.rmo.bellsouth.net) got netsplit.
[07:34] SvenDowideit (~SvenDowid@foswiki/developer/SvenDowideit) got netsplit.
[07:34] kke (~kimmo@46.19.32.53) got netsplit.
[07:34] bigbluehat (u632@gateway/web/irccloud.com/x-wkusvqusacvtgvgo) got netsplit.
[07:34] Mike_L (~leonhard@shevek.tamale.net) got netsplit.
[07:34] pc1oad1etter (~pc1oad1et@ec2-174-129-99-58.compute-1.amazonaws.com) got netsplit.
[07:34] SvenDowideit_ -> SvenDowideit
[07:34] Possible future nick collision: SvenDowideit
[07:34] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) joined #rest.
[07:34] HighBit_ -> HighBit
[07:34] Possible future nick collision: HighBit
[07:40] <mk_> darrelmiller: rest-discuss isn't that scarey, is it ? :)
[07:40] <darrelmiller> hell yeah. Scares the crap out of me when I post there.
[07:40] <darrelmiller> Almost as scary as ietf-http-wg
[07:40] <mk_> I really believe what I said on twitter too - the biggest issue with rest-discuss (afaict) is "experts" posting their opinion as subjective fact
[07:41] <mk_> and using fancy terminology in circular arguments in order to try and seem like they are making sense
[07:41] <mk_> e.g.
[07:41] <darrelmiller> yeah, but are you just referring to some "troublesome elements" ?
[07:41] <mk_> well I'm pretty sure that is me
[07:42] <darrelmiller> hehe. I didn't think so.
[07:42] <mk_> it isn't that bad though
[07:42] <mk_> I think some people just need to grow a pair, really
[07:42] <darrelmiller> :-P
[07:44] <mk_> "Dear person-with-conflicting-viewpoint, I humbly recieve your wisdomful and articulately delivered points. However I must apologise, I believe I have found a flaw in one or two of your points and I would like, if I may, to attempt to analyse what these are and how I would otherwise approach the issue at hand"
[07:44] <mk_> vs.
[07:45] pc1oad1etter (~pc1oad1et@ec2-174-129-99-58.compute-1.amazonaws.com) got lost in the net-split.
[07:45] mogsie (~mogsie@62.101.198.35) got lost in the net-split.
[07:45] Mike_L (~leonhard@shevek.tamale.net) got lost in the net-split.
[07:45] chilversc (~chris@about/csharp/regular/KeeperOfTheSoul) got lost in the net-split.
[07:45] ssedano (~ssedano@unaffiliated/ssedano) got lost in the net-split.
[07:45] kke (~kimmo@46.19.32.53) got lost in the net-split.
[07:45] KevBurnsJr_ (~KevBurnsW@50.0.103.39) got lost in the net-split.
[07:45] bigbluehat (u632@gateway/web/irccloud.com/x-wkusvqusacvtgvgo) got lost in the net-split.
[07:45] axisys (~axisys@ip68-98-189-233.dc.dc.cox.net) got lost in the net-split.
[07:45] <mk_> "Dear dickhead, you are wrong, here's why -"
[07:45] <mk_> personally I'd prefer people did the latter, as it saves me time
[07:46] <darrelmiller> That preference seems to contradict your previous objection to 01people posting their opinion as subjective fact01.
[07:48] <darrelmiller> I don't think there is a right and wrong in most cases. There is cause and effect. "If you do that then this will happen"
[07:50] <mk_> oh for sure I was being stupid there
[07:51] <mk_> the discussion about generic media types is a good example though
[07:51] <mk_> where you get people going "here's the thing about REST you are missing/getting wrong"
[07:52] <mk_> when really what we are talking about are design descisions that don't have a right or wrong answer
[07:53] <darrelmiller> That's the problem with an "architectural style". Different people design things different ways and they forget that different designs may still be valid wrt the style.
[07:53] <mk_> It would be less of a problem for me if those people were able to actually follow their opinion through with some rationale that explains what effect their particular design has vs another
[07:53] <mk_> but this rarely ever happens
[07:54] <mk_> we seem to have a lot of people who basically state an opinion.. and then start reeling off intra-REST buzzwords like self-descriptive and visibility
[07:54] <darrelmiller> yeah that really gets my goat. I hate the "because the spec says so" argument.
[07:54] KevBurnsJr_ (~KevBurnsW@50.0.103.39) joined #rest.
[07:54] ssedano (~ssedano@unaffiliated/ssedano) joined #rest.
[07:54] chilversc (~chris@about/csharp/regular/KeeperOfTheSoul) joined #rest.
[07:54] mogsie (~mogsie@62.101.198.35) joined #rest.
[07:55] <mk_> yes and there's the whole "that's the spec" thing
[07:55] <mk_> which makes sense on public-http, just not on rest-discuss
[07:55] <mk_> which is the main reason I didn't start a thread on public-http about partial PUT :)
[07:55] kke (~kimmo@kuuma.ho.ro) joined #rest.
[07:55] bigbluehat (u632@gateway/web/irccloud.com/x-wkusvqusacvtgvgo) joined #rest.
[07:55] Mike_L (~leonhard@shevek.tamale.net) joined #rest.
[07:55] pc1oad1etter (~pc1oad1et@ec2-174-129-99-58.compute-1.amazonaws.com) joined #rest.
[07:56] <mk_> I will do it after the re-charter though, fo sho!
[07:56] <darrelmiller> But there is a reason the "spec says so", even if it is "we had to make an arbitrary choice".
[07:57] <fu-manchu> yes, and the really sad part is how many of those reasons aren't clearly articulated anywhere
[07:58] <darrelmiller> That's one of the awesome things about httpbis, many of the threads are spent digging up the reasons why decisions were made a certain way.
[07:58] <fu-manchu> HTTP needs a "requirements" section
[07:58] <fu-manchu> yes!
[07:59] <fu-manchu> quick, before the grey beards all pass on
[07:59] pezra (~Adium@c-71-56-237-160.hsd1.co.comcast.net) left irc: Quit: Leaving.
[07:59] <fu-manchu> that problem is *far* worse in the database world--soon we'll have massive database implementations that nobody will know how to fix when they break
[08:00] <fu-manchu> or the fix will violate some unspoken constraint
[08:00] <darrelmiller> fu-manchu: whoa there. Fielding is not much older than me. I think we have a few years in us yet.
[08:01] <fu-manchu> just enough years to properly document what's already been built ;)
[08:02] <darrelmiller> Oh my, rest-discuss is getting infected with REST vs SOAP threads. Sigh.
[08:05] pezra (~Adium@173-164-36-90-colorado.hfc.comcastbusiness.net) joined #rest.
[08:08] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[08:09] <bigbluehat1> darrelmiller: yeah, it's based on his blog post here: http://damienkatz.net/2012/01/the_future_of_couchdb.html
[08:10] <bigbluehat1> take them both as Damien's personal choice
[08:10] <darrelmiller> bigbluehat1: So does this affect you?
[08:11] <bigbluehat1> eh, not really :)
[08:11] <bigbluehat1> I believe in the replication/HTTP part of CouchDB—which is still on the "roadmap" for Couchbase
[08:11] bigbluehat1 is looking up the Couchbase, Inc "response" to the subsequent fallout
[08:12] <bigbluehat1> darrelmiller: here 'tis: http://blog.couchbase.com/couchbase-commitment-to-open-source-and-couchdb
[08:12] <darrelmiller> bigbluehat1: thx
[08:13] <bigbluehat1> The Future of CouchDB is CouchDB :)
[08:13] <bigbluehat1> at least for me :)
[08:14] <mk_> I'm still on Riak
[08:14] <bigbluehat1> The Future of Membase is Couchbase which at some point will be even more like CouchDB…so wins all 'round :)
[08:14] <bigbluehat1> mk_: I forgive you ;)
[08:14] <bigbluehat1> there's no reason, imo, we can't all "win"
[08:14] <mk_> the thing about the local-replication thing is kind of cool
[08:14] <mk_> but like..
[08:14] <mk_> connectivity is getting more ubiquitous all the time
[08:14] <mk_> it's going to get to the point where 'offline' is going to become meaningless
[08:15] <bigbluehat1> connectivity is only as good as the thing between you and the thing you're connecting to
[08:15] <bigbluehat1> which in most cases is out of your immediate control
[08:15] <bigbluehat1> so it's not to be completely trusted
[08:15] <mk_> right, but there's caching stuff that deals with this
[08:15] <bigbluehat1> there will be ample times when you'll *want* to be offline
[08:15] <bigbluehat1> like when your government goes sour…for one
[08:15] <mk_> lol
[08:15] <bigbluehat1> :)
[08:15] <darrelmiller> mk_: Europeans have a different perspective on connectivity than north americans. Size is actually significant still.
[08:15] <bigbluehat1> it's a real problem for real people
[08:15] <mk_> CouchDB - the tinfoilhat of distributed databases
[08:16] <bigbluehat1> darrelmiller: and Canadians wrote the book on size
[08:16] <bigbluehat1> well the sequal to the Siberian Delima anyway :)
[08:16] <bigbluehat1> mk_: harhar
[08:16] <mk_> :P
[08:17] <bigbluehat1> I'm into hats, as you know ;)
[08:18] <bigbluehat1> odd
[08:18] <bigbluehat1> sorry…wrong window :-P
[08:20] <mk_> bigbluehat1: is it possible that the replication protocol couch uses could be abstracted ?
[08:21] <mk_> or does it really require that the persistence is couchdb based ?
[08:21] <mk_> i.e. could you emulate couchdb with localstorage and javascript ?
[08:22] <mk_> for an 'offline' app
[08:25] <bigbluehat1> you mean like this: http://www.dataprotocols.org/en/latest/index.html
[08:25] <bigbluehat1> there's also a bevy of (sort of) new DB's that are essentially "couch" but total re-creations of the concept
[08:25] <bigbluehat1> TouchDB, PouchDB, etc
[08:26] <mk_> oh ok
[08:26] <mk_> cool, thanks
[08:26] <bigbluehat1> that's a good site to watch for further development in that area
[08:26] <bigbluehat1> the core replication protocol has needed documentation
[08:26] <bigbluehat1> but for the most part it's a simple concept
[08:26] <bigbluehat1> that's quite portable
[08:27] <bigbluehat1> across db's…even RDBMS—with the right additions
[08:27] <bigbluehat1> in fact, TouchDB uses sqlite for storage
[08:27] <bigbluehat1> PouchDB I believe uses IndexedDB
[08:27] <bigbluehat1> https://github.com/mikeal/pouchdb
[08:27] <bigbluehat1> https://github.com/couchbaselabs/TouchDB-iOS
[08:27] <mk_> in your talk you said it's a deterministic conflict resolution mechanism ?
[08:27] <bigbluehat1> TouchDB is by #cocuhbase
[08:28] <bigbluehat1> right
[08:28] <bigbluehat1> it's essentially arbitrary
[08:28] <bigbluehat1> but consistently so :)
[08:28] <mk_> yeah I remember you saying that and it making no sense to me
[08:28] <mk_> :D
[08:28] <bigbluehat1> so every CouchDB (or deriviatives) should always pick the same winner
[08:28] <bigbluehat1> it's the consistency of the choice that's necessary
[08:28] <mk_> right I think it's the word arbitrary that throws me off
[08:29] <bigbluehat1> so that the replication history is consistent—you should always get the same conflict when replicating to different CouchDB instances
[08:29] <bigbluehat1> yeah :)
[08:29] <mk_> when you say arbitrary do you mean nieve ?
[08:29] <bigbluehat1> meaning, the choice isn't based on data, size, etc
[08:29] <mk_> naive*
[08:30] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[08:30] <mk_> but it's based on something, right ?
[08:30] <bigbluehat1> think of it as "relaxed" :)
[08:30] <bigbluehat1> it is based on something—hence its consistency
[08:30] <mk_> isn't that fundamental to the replication protocol ?
[08:30] <bigbluehat1> but it's not going to attempt to figure out how you want the conflict resolved—which is the road to messes
[08:31] <bigbluehat1> as any mobile phone's sync system proves
[08:31] <bigbluehat1> yeah, it's essential to the protocol
[08:31] <mk_> be cool if you could distribute conflict resolution algorithms in a document
[08:31] <bigbluehat1> when a conflict happens there should be no confusion (among instances) which doc would "win"
[08:31] <bigbluehat1> yeah, there's talk of doing that
[08:31] <mk_> like you can with the couch apps
[08:32] <bigbluehat1> however, there's much to be said for removing any blockades to replication
[08:32] <bigbluehat1> and doing any resolution at the human or app level
[08:32] <bigbluehat1> so that replication can just work
[08:32] <bigbluehat1> the conflicting documents are kept
[08:32] <mk_> well one thing I didn't understand about that
[08:32] <mk_> is if you have a conflict
[08:32] <bigbluehat1> so the app can pick a different winner—which is completely fine and intended
[08:32] <mk_> and it gets arbitrarily (but consistently resolved) and the world moves on from that resolution and further changes get made
[08:33] <mk_> how can you post-rationally resolve that wiht human intervention ?
[08:33] <bigbluehat1> diff :)
[08:33] <bigbluehat1> or whatever you'd like to show your user for resolving the issue
[08:33] <bigbluehat1> in many cases the conflict may not matter
[08:33] <bigbluehat1> and whoever edits last "wins"
[08:33] <mk_> "Dear user, your data is probably fucked up - here is the diff, good luck. Love CouchDB-using-douchebag"
[08:34] <bigbluehat1> and in the cases where it does matter, you've got the conflicting docs to display/parse/offer for resolution to the user
[08:34] <bigbluehat1> heh
[08:34] <mk_> :P
[08:34] <bigbluehat1> when has an app guessed correctly?
[08:34] <bigbluehat1> it certainly would be possible to write your client app to resolve conflicts for you based on your own needs
[08:34] <mk_> well some apps are trivially solved with timestamp of the commit, right ?
[08:35] <bigbluehat1> mk_'s changes should always beat darrelmiller's for instance ;)
[08:35] <bigbluehat1> but that's not for CouchDB to decide
[08:35] <mk_> that is correct.
[08:35] <bigbluehat1> correct, timestamp stuff could be used now
[08:35] <bigbluehat1> and it's fairly trivial to setup a system to post-process conflicts
[08:35] <bigbluehat1> on document update—which is notified during replication as well as edits
[08:35] <mk_> is that built in ?
[08:35] <darrelmiller> heh!!
[08:35] <bigbluehat1> no, but there are hooks for it
[08:35] <darrelmiller> Age before beauty.
[08:36] <bigbluehat1> and there's talk about adding something like that to _design docs
[08:36] <bigbluehat1> :)
[08:36] <mk_> nice
[08:36] <bigbluehat1> like "resolver" or some such
[08:36] <mk_> afaik that's built into Riak
[08:36] <mk_> but I think you have to write the resolver in Erlang
[08:36] <mk_> which is balls
[08:36] <mk_> who does that ?
[08:36] <bigbluehat1> eh, Erlang's not so bad
[08:36] <bigbluehat1> it's ugly
[08:36] <bigbluehat1> but not so bad
[08:36] <mk_> oh cmon
[08:36] <bigbluehat1> :)
[08:36] <mk_> it looks like ass
[08:37] <bigbluehat1> learnyousomeerlang.com
[08:37] <darrelmiller> Erlang is awesome.
[08:37] Wombert_ (~Wombert@xdsl-78-34-218-164.netcologne.de) joined #rest.
[08:37] <mk_> darrel you code in C#
[08:37] <darrelmiller> I used it for at least an hour so I know.
[08:37] <mk_> everything is pretty to you.
[08:37] <bigbluehat1> :)
[08:38] <Hakon|mbp> :P
[08:38] <bigbluehat1> in CouchDB you'd be able to use any lang that's setup as a view server
[08:38] <bigbluehat1> so JavaScript, CoffeeScript, and (at your option) Erlang
[08:38] <darrelmiller> tbh, JSON is the ugliest thing I see these days. How messed up am I?
[08:38] <bigbluehat1> or with a bit more work, Python, Ruby, PHP (even)
[08:38] <mk_> CoffeeScript isn't a language!
[08:38] <bigbluehat1> darrelmiller: what else do you see? :)
[08:38] <bigbluehat1> mk_: meh
[08:39] <bigbluehat1> conlangs are languages too
[08:39] <mk_> I kid.
[08:39] <bigbluehat1> ask any Klingon
[08:39] <mk_> bigbluehat1: any plan to have linking stuff like Riak ?
[08:39] Wombert (~Wombert@xdsl-78-34-218-164.netcologne.de) left irc: Ping timeout: 240 seconds
[08:39] Wombert_ -> Wombert
[08:39] <bigbluehat1> like MapReduce chaining? or something different?
[08:40] <bigbluehat1> you mean where docs reference other docs?
[08:40] <mk_> yeah it's mapreduce under the hood
[08:40] <bigbluehat1> and can drag each other into fights?
[08:40] <bigbluehat1> yeah, you can do something like that now with MapReduce + ?include_docs=true (and/or a _list function)
[08:40] <bigbluehat1> there's also some view merging work that's been done
[08:40] <bigbluehat1> but not sure that really does chaining yet
[08:41] <bigbluehat1> but I hope to see that happen—and am going to push for it soon
[08:41] <bigbluehat1> as it'd be super handy for the BlueInk arch
[08:41] <mk_> I heard some Bad Things(tm) about couch's compacting blowing up on big datasets
[08:41] <mk_> FUD?
[08:41] <bigbluehat1> eh, mostly
[08:41] <bigbluehat1> depends on how big your data set is and how much space your disk has
[08:41] <bigbluehat1> and how often you compact
[08:41] <bigbluehat1> it won't loose data
[08:41] <bigbluehat1> but it might crash, etc depending on what you ask it to do where
[08:42] <bigbluehat1> compaction is essentially copying out live data into a new .couch db file
[08:42] <mk_> doens't it store a copy for each version of the doc rather than deltas ?
[08:42] <bigbluehat1> if you've not got room for that, it'll fail
[08:42] <bigbluehat1> right, but when you compact it only keeps the active version
[08:42] <bigbluehat1> and a revision tree--iirc
[08:42] <mk_> alright
[08:42] <mk_> sold
[08:42] <bigbluehat1> but no contents, just references that they existed for later replication attempts
[08:42] <mk_> where can I buy ?
[08:42] <bigbluehat1> :)
[08:42] <bigbluehat1> http://couchdb.apache.org/ (and friends) :)
[08:43] <bigbluehat1> Couchbase uses the same tech/approaches, but spreads it across multiple servers via the Membase vBuckets stuff
[08:43] <mk_> oh ok
[08:43] <Hakon|mbp> i always though you couldn't rely on the versioning of couchdb because it could compact at any time?
[08:43] <mk_> so like Riak
[08:43] <mk_> just bolted on as an after-thought?
[08:44] mk_ *poke* *poke* *poke*
[08:44] <mk_> I had too much coffee, sorry
[08:44] <bigbluehat1> Hakon|mbp: it's not for document versioning
[08:44] <Hakon|mbp> and that you lost the version info when it compacted
[08:44] mhausenblas (~mhausenbl@wlan-nat.fwgal01.deri.ie) left irc: Quit: mhausenblas
[08:44] <bigbluehat1> it's MVCC
[08:44] <Hakon|mbp> ah
[08:44] <bigbluehat1> don't rely on the MVCC PostgreSQL uses either
[08:44] <bigbluehat1> it's not for that
[08:44] <bigbluehat1> you can, however, build a versioining system easily enough
[08:44] <Hakon|mbp> then i'm with you :)
[08:45] <bigbluehat1> but it's not part of CouchDB by default
[08:45] <mk_> sounds suspicious.
[08:45] <bigbluehat1> :)
[08:45] <mk_> :D
[08:45] <bigbluehat1> meh…I'll fight this one later ;)
[08:45] bigbluehat1 has to pick up his son from school
[08:45] <bigbluehat1> see you gents post-lunch
[08:45] <mk_> I'm kidding
[08:45] <mk_> cheers bud ;)
[08:45] <Hakon|mbp> you get no fight from me ;)
[08:46] bigbluehat1 (~bigblueha@nmd.sbx10523.greensc.wayport.net) left irc: Quit: Leaving.
[08:47] pezra (~Adium@173-164-36-90-colorado.hfc.comcastbusiness.net) left irc: Quit: Leaving.
[08:50] pezra (~Adium@173-164-36-90-colorado.hfc.comcastbusiness.net) joined #rest.
[09:19] <whartung> o/
[09:22] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[09:30] Wombert (~Wombert@xdsl-78-34-218-164.netcologne.de) left irc: Quit: Wombert
[09:34] <mk_> http://lists.w3.org/Archives/Public/www-archive/2012Jan/0027.html
[09:34] Wombert (~Wombert@xdsl-78-34-218-164.netcologne.de) joined #rest.
[09:34] <mk_> Ian Hickson ftw?
[09:34] <mk_> hahaha
[09:35] <fu-manchu> wow
[09:38] <mk_> I think he's going for douchebag-rock-star but just not quite achieving the rock-star status
[09:42] <fu-manchu> and here we find Roy's long-standing admonition to Hixie that "browsers are by far not the majority of traffic" arriving at some concrete consequences
[09:51] <whartung> wth is this?
[10:01] KarlHungus -> jssjr
[10:01] jssjr -> KarlHungus
[10:03] bigbluehat1 (~bigblueha@adsl-98-71-132-42.gsp.bellsouth.net) joined #rest.
[10:04] Wombert (~Wombert@xdsl-78-34-218-164.netcologne.de) left irc: Quit: Wombert
[10:09] <fu-manchu> dangit darrelmiller, always beating me to the easy #rest questions on SO
[10:10] fu-manchu upvotes
[10:12] <whartung> What? Another question rest tag I need to delete from SO?
[10:12] <whartung> I've been slacking off lately on my crusade...but the windmills, they keep popping up.
[10:17] mhausenblas (~mhausenbl@188.141.67.15) left irc: Quit: mhausenblas
[10:32] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[10:57] mhausenblas (~mhausenbl@188.141.67.15) left irc: Quit: mhausenblas
[11:20] bigbluehat1 (~bigblueha@adsl-98-71-132-42.gsp.bellsouth.net) left irc: Quit: Leaving.
[11:37] Wombert (~Wombert@80.187.152.101) joined #rest.
[11:40] pezra (~Adium@173-164-36-90-colorado.hfc.comcastbusiness.net) left irc: Quit: Leaving.
[11:42] Wombert (~Wombert@80.187.152.101) left irc: Quit: Wombert
[11:47] KevBurnsJr_ (~KevBurnsW@50.0.103.39) left irc: Quit: restart
[11:49] KevBurnsJr_ (~KevBurnsW@50.0.103.39) joined #rest.
[11:58] bigbluehat1 (~bigblueha@adsl-98-71-132-42.gsp.bellsouth.net) joined #rest.
[14:02] pezra (~Adium@webmail.openlogic.com) joined #rest.
[14:20] bigbluehat1 (~bigblueha@adsl-98-71-132-42.gsp.bellsouth.net) left irc: Quit: Leaving.
[14:26] Wombert (~Wombert@dslb-092-075-024-131.pools.arcor-ip.net) joined #rest.
[14:30] sbanwart (~sbanwart@66.6.175.18) left irc: Read error: Operation timed out
[14:58] Wombert (~Wombert@dslb-092-075-024-131.pools.arcor-ip.net) left irc: Quit: Wombert
[15:27] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) left irc: Read error: Connection reset by peer
[15:27] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) joined #rest.
[15:46] KevBurnsJr_ (~KevBurnsW@50.0.103.39) got netsplit.
[15:54] KevBurnsJr_ (~KevBurnsW@50.0.103.39) returned to #rest.
[16:10] pezra (~Adium@webmail.openlogic.com) left irc: Quit: Leaving.
[16:35] Wombert (~Wombert@dslb-092-075-024-131.pools.arcor-ip.net) joined #rest.
[16:38] Wombert_ (~Wombert@dslb-092-074-121-149.pools.arcor-ip.net) joined #rest.
[16:38] Hakon|mbp (~hakon1@193.137.202.84.customer.cdi.no) left irc: Remote host closed the connection
[16:40] Hakon|mbp (~hakon1@193.137.202.84.customer.cdi.no) joined #rest.
[16:40] Wombert (~Wombert@dslb-092-075-024-131.pools.arcor-ip.net) left irc: Ping timeout: 276 seconds
[16:40] Wombert_ -> Wombert
[16:48] <darrelmiller> whartung: BTW, I have my own private internet that does not allow intermediaries that drop bodies on GETs.
[16:48] <whartung> private idaho...
[16:49] <whartung> yea... I mean, it's a valid question, is it a realistic concern -- but the point is do you have a plan when the unrealistic happens.
[16:50] <darrelmiller> I plan on whining on the Http-wg mailing list to all the intermediary developers while they are specing out HTTP/2.0 to see if they will ensure GET bodies are passed through :-)
[16:51] <whartung> dd you see that blurb about SPDY being incorporated?
[16:52] <darrelmiller> I've seen a number of comments ranging from "let's start from scratch and redesign the internet" to "let's just rubber stamp SPDY".
[16:52] <darrelmiller> I think it is way too early to tell which way it will go.
[16:53] <darrelmiller> But I'm hoping that a working implementation of WAKA will appear at some point in the near future.
[16:53] fu-manchu crosses his fingers but won't hold his breath
[16:54] <whartung> in lieu of SPDY?
[16:55] <darrelmiller> Why not?
[16:55] <darrelmiller> Apparently there are a lot of similarities.
[16:55] <whartung> yea I'm not familiar enough with either
[18:56] KevBurnsJr_ (~KevBurnsW@50.0.103.39) left irc: Quit: donezo
[22:14] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) left irc: Ping timeout: 252 seconds
[22:24] mhausenblas (~mhausenbl@188.141.67.15) joined #rest.
[22:48] mhausenblas (~mhausenbl@188.141.67.15) left irc: Quit: mhausenblas
[00:00] --- Thu Jan 26 2012