ruote log_2010-06-10

2010-06-10 18:26:22 utc lbt finding time to edit the wiki some more...
2010-06-10 18:50:50 utc lbt kennethkalmer: got a sec?
2010-06-10 20:26:05 utc kennethkalmer hey lbt
2010-06-10 20:26:06 utc kennethkalmer sure, some coffee ?
2010-06-10 20:26:53 utc lbt mmm guatemala elephant?
2010-06-10 20:46:31 utc kennethkalmer why not
2010-06-10 20:46:44 utc kennethkalmer ACTION emits guatamalan elephant ;)
2010-06-10 20:47:19 utc lbt ACTION wakes up
2010-06-10 20:47:39 utc lbt anyhow.... I was trying to get a handle on it all and put this together... john's made a couple of comments (italic) and I just need to grok the "there is no big store" statement.
2010-06-10 20:50:01 utc kennethkalmer digesting information quickly
2010-06-10 20:54:49 utc kennethkalmer ok, i think i know where john is coming from
2010-06-10 20:55:28 utc lbt cool
2010-06-10 20:55:53 utc kennethkalmer just like he says, when an expression is applied they get a copy of the workitem
2010-06-10 20:55:55 utc lbt and obviously I'm referring to the couchdb/fs backing store
2010-06-10 20:56:11 utc kennethkalmer not like a "pointer" to the stored workitem
2010-06-10 20:56:15 utc kennethkalmer yeah
2010-06-10 20:56:31 utc kennethkalmer the mechanics look like it, but in principle it is a copy
2010-06-10 20:56:51 utc kennethkalmer which is updated by the expression, and handed back to storage for updating
2010-06-10 20:57:29 utc kennethkalmer hmmm, how to phrase this in my tiredness
2010-06-10 20:57:51 utc lbt aren't items passed via the store?
2010-06-10 20:58:10 utc kennethkalmer yeah, which means the storage also acts as a transport mechanism
2010-06-10 20:58:16 utc kennethkalmer that is pure convenience
2010-06-10 20:58:17 utc lbt excellent
2010-06-10 20:58:26 utc lbt I thought so
2010-06-10 20:58:35 utc kennethkalmer in the case of remote participants, the store itself is not at play
2010-06-10 20:58:41 utc lbt no, indeed
2010-06-10 20:58:52 utc lbt so in that case it could/should be transactioned
2010-06-10 20:58:57 utc lbt hence amqp
2010-06-10 20:59:02 utc kennethkalmer well in both cases really
2010-06-10 20:59:24 utc kennethkalmer john can fill you in on the fine details, i'm not nearly as up to date with the internals as i should be
2010-06-10 20:59:30 utc lbt and I'm trying to make it as light an introduction as possible too
2010-06-10 20:59:45 utc lbt to give a picture without too much detail all at once
2010-06-10 21:00:03 utc kennethkalmer then for that reason i think it can be stated that the storage "doubles up" as a transport mechanism
2010-06-10 21:00:25 utc kennethkalmer the introduction is very nice btw
2010-06-10 21:00:59 utc lbt "Another key point is that workitems are not passed directly from Participant to ruote; by default they are transported via a shared persistent store"
2010-06-10 21:01:45 utc lbt again... not a reference document; more giving a flavour of resilience capabilities
2010-06-10 21:01:53 utc lbt (thanks)
2010-06-10 21:02:01 utc kennethkalmer ... from local participants to ruote ...
2010-06-10 21:02:01 utc kennethkalmer +1
2010-06-10 21:02:22 utc kennethkalmer from a "sales point of view" i agree with you 100%
2010-06-10 21:02:38 utc kennethkalmer makes for great understanding during project evaluation
2010-06-10 21:02:54 utc kennethkalmer if the reader likes or dislikes it, they can explore it further
2010-06-10 21:03:02 utc lbt part 2 of the document will have the clarification and more technically correct wording
2010-06-10 21:03:14 utc kennethkalmer agreed
2010-06-10 21:03:51 utc kennethkalmer hmmm
2010-06-10 21:03:51 utc kennethkalmer just had an awesome idea
2010-06-10 21:04:00 utc lbt "Another key point is that workitems are passed in a reliable way from Participants to ruote; by default they are transported via a shared persistent store."
2010-06-10 21:04:05 utc lbt ACTION listens
2010-06-10 21:04:20 utc kennethkalmer that is great wording
2010-06-10 21:04:23 utc kennethkalmer idea:
2010-06-10 21:04:56 utc kennethkalmer couchdb itself can be a remote participant, with an interface provided by a couch-app
2010-06-10 21:05:26 utc kennethkalmer replication can handle the consume/reply mechanism
2010-06-10 21:05:33 utc kennethkalmer imagine couch/couch-app on maemo
2010-06-10 21:06:05 utc lbt listening... not that up on couch though. I've been focussing on rabbitmq
2010-06-10 21:07:18 utc kennethkalmer
2010-06-10 21:08:10 utc lbt ACTION reads
2010-06-10 21:08:20 utc lbt btw...
2010-06-10 21:08:58 utc lbt BOSS will be written around Ruote barring some major blowup next week
2010-06-10 21:10:14 utc kennethkalmer that looks awesome
2010-06-10 21:11:13 utc lbt well, it will be... :) That's not my writing :)
2010-06-10 21:12:08 utc kennethkalmer i'm seriously considering getting an N900 permantly
2010-06-10 21:12:33 utc kennethkalmer *permanently
2010-06-10 21:13:55 utc lbt they are nice
2010-06-10 21:14:22 utc kennethkalmer i'm enjoying my demo model very much
2010-06-10 21:14:33 utc lbt so long as you want a linux hack device that happens to make phone calls
2010-06-10 21:14:53 utc kennethkalmer my thinking exactly
2010-06-10 21:14:56 utc lbt we'll get some form of MeeGo on there
2010-06-10 21:15:11 utc kennethkalmer just curious if the OVI app store will ever perform
2010-06-10 21:15:21 utc lbt yes
2010-06-10 21:15:30 utc kennethkalmer would like to experiment in generating some kinda money from my devise too
2010-06-10 21:15:36 utc kennethkalmer even if it is $1
2010-06-10 21:15:38 utc lbt not sure if it'll do it in N900 times though
2010-06-10 21:15:45 utc lbt yeah...
2010-06-10 21:15:47 utc lbt hey...
2010-06-10 21:16:04 utc lbt
2010-06-10 21:16:15 utc lbt ACTION spots another beta-tester....
2010-06-10 21:16:55 utc lbt I need votes...
2010-06-10 21:17:30 utc lbt (the install link on that first page won't work until it gets to stable)
2010-06-10 21:18:01 utc lbt We're also not sure how to handle GPL apps on OVI
2010-06-10 21:18:11 utc lbt no reason why not though
2010-06-10 21:21:10 utc kennethkalmer back to ruote, is ruote-amqp working fine for you ?
2010-06-10 21:21:24 utc lbt no :)
2010-06-10 21:21:31 utc lbt I meant to hack on it tonight
2010-06-10 21:21:49 utc lbt sec... let me save and get the code up
2010-06-10 21:22:08 utc kennethkalmer cool, thanks
2010-06-10 21:26:03 utc lbt the : amqp = :default_queue => 'test' )
2010-06-10 21:26:08 utc lbt approach....
2010-06-10 21:26:16 utc lbt as per
2010-06-10 21:26:36 utc lbt doesn't seem to work
2010-06-10 21:26:55 utc lbt so none of the map_participant tricks work either
2010-06-10 21:30:22 utc kennethkalmer hmmm... will have to extend the tests then
2010-06-10 21:31:07 utc lbt I've had to have a couple of weeks on python so I'm a bit rusty.... but john mentioned that we should be registering classes
2010-06-10 21:31:47 utc lbt wondering if that part of r-a needs updating?
2010-06-10 21:32:50 utc lbt I am trying to make a distributed setup too - just to ensure I get the ipc
2010-06-10 21:34:11 utc lbt so I start with .... mmmm pastie died
2010-06-10 21:35:07 utc lbt
2010-06-10 21:36:24 utc lbt then
2010-06-10 21:37:42 utc kennethkalmer i'll have to spend some time with that code to see what is going on
2010-06-10 21:38:04 utc lbt I'll be looking at it again on saturday
2010-06-10 21:38:59 utc kennethkalmer great, apologies for the lag from my side
2010-06-10 21:39:10 utc kennethkalmer struggling with my FOSS balancing act as of late
2010-06-10 21:39:46 utc lbt np at all
2010-06-10 21:40:14 utc lbt I just need to get my head into it and "get it" at a lower level
2010-06-10 21:49:08 utc kennethkalmer hey weyus
2010-06-10 21:49:11 utc kennethkalmer how is the bus ride ?
2010-06-10 21:49:29 utc kennethkalmer and how is/was railsconf ?