2011-03-10T04:02:57 *** mtisi has quit IRC 2011-03-10T04:13:11 *** mjgiarlo has quit IRC 2011-03-10T04:15:17 *** jamesrf has quit IRC 2011-03-10T05:22:43 *** john____ has joined #evergreen 2011-03-10T05:22:56 hello 2011-03-10T05:25:38 *** john____ has quit IRC 2011-03-10T06:19:39 *** jamesrf has joined #evergreen 2011-03-10T06:28:23 *** mjgiarlo has joined #evergreen 2011-03-10T07:07:54 *** artunit has quit IRC 2011-03-10T07:08:24 *** artunit has joined #evergreen 2011-03-10T07:11:17 *** zigo has joined #evergreen 2011-03-10T07:19:53 *** rickd_ has joined #evergreen 2011-03-10T07:32:36 more memcache weirdness... we have two squeeze VMs, cloned from the same image. One of them works fine with the server defined as localhost:11211 and the other one kept coming up with the 'no authentication seed found' error until I changed it to 127.0.0.1:11211... 2011-03-10T07:32:55 (in /openils/conf/opensrf.xml, for clarity) 2011-03-10T07:38:09 *** zigo_ has joined #evergreen 2011-03-10T07:38:21 *** zigo_ is now known as zigo-_- 2011-03-10T07:38:26 *** zigo has quit IRC 2011-03-10T07:53:25 we found what might be a clue - the squeeze server that worked with localhost:11211 had the IPv6 settings commented out in /etc/hosts... don't know if that's related, but it was the only thing we could see that was different in the server configs 2011-03-10T07:54:48 for safety, we're having our opensrf.xml default to 127.0.0.1:11211 for memcache, so all should continue to be well 2011-03-10T07:58:30 *** zigo-_- has quit IRC 2011-03-10T08:07:37 *** rickd_ has quit IRC 2011-03-10T08:09:51 *** rickd_ has joined #evergreen 2011-03-10T08:11:53 csharp: I have seen oddities with IPv6 host entries "trumping" IPv4 ones, and if the service only listens on IPv4 for whatever reason.... 2011-03-10T08:14:06 *** rjackson-isl has joined #evergreen 2011-03-10T08:19:00 I see that there's some evidence for it being related to IPv6: http://www.mail-archive.com/users@sogo.nu/msg03844.html 2011-03-10T08:19:31 in that case, they explicitly tell memcached to listen on 'localhost' rather than the default 127.0.0.1 2011-03-10T08:20:52 Thus making it so that whatever "localhost" returns is being used on both ends 2011-03-10T08:24:22 I should note I have seen systems where "localhost" wasn't 127.0.0.1 even in IPv4 though 2011-03-10T08:25:29 *** kmlussier has joined #evergreen 2011-03-10T08:30:12 *** AaronZ-PLS has joined #evergreen 2011-03-10T08:30:45 *** AaronZ-PLS has joined #evergreen 2011-03-10T08:39:17 *** mtisi has joined #evergreen 2011-03-10T08:44:30 *** Shane-S has joined #evergreen 2011-03-10T08:48:39 I am getting ready to build Evergreen using the docuwiki, and I need to do open SRF first, but I have not found what I need the start with server wise. I am using the latest 10.04.2, and I am where I can pick DNS/LAMP/Etc. Should I choose anything or use manual and pick nothing? 2011-03-10T08:49:06 sorry Ubuntu 10.04.2 2011-03-10T08:49:09 Shane-S: probably best to just do a basic server install 2011-03-10T08:49:25 Shane-S: that is, just do the minimum 2011-03-10T08:49:55 thanks, I just skipped that section for now then. 2011-03-10T08:50:09 I will see how I fair :D 2011-03-10T08:50:38 Shane-S: I always choose openssh-server so I don't have to manually install it later, but otherwise, I just do "basic server" or just don't select anything 2011-03-10T08:50:56 * csharp can't remember the Ubuntu installer options from memory 2011-03-10T08:58:15 if I follow the srf wiki, I notice it uses 127.0.1.2 and 127.0.1.2 for private and public domains, should I (can I) use 127.0.0.1 for both if I am testing it on a single machine internally, or will it need at least a lan IP 2011-03-10T09:00:05 Shane-S: you'll want to follow the instructions as closely as possible 2011-03-10T09:00:45 Shane-S: even though they're each pointing to your own machine, ejabberd needs to see private and public as separate networks 2011-03-10T09:01:03 (and separate from 127.0.0.1) 2011-03-10T09:02:24 *** Dyrcona has joined #evergreen 2011-03-10T09:03:20 csharp: WHY does ejabberd need to see those individually? 2011-03-10T09:03:32 I run 8 ejabberd domains on the same IP privately. 2011-03-10T09:04:29 tsbere: sorry - I must be mistaken about why - I've always assumed that was the rationale behind the separate IPs 2011-03-10T09:04:48 * csharp didn't come up with that ;-) 2011-03-10T09:05:47 I figured that it might be for knowing what service a given process was connecting to, as netstat and friends can't tell you what username they used. But beyond that it isn't a requirement. 2011-03-10T09:06:34 I will use the directions verbatim and see how it goes :P 2011-03-10T09:06:51 it a virtual server so no harm done if it is messed up 2011-03-10T09:06:58 Shane-S: good call, in any case - start with the default and customize it later 2011-03-10T09:08:11 * tsbere just finished changing duration and fine rules to say minutes instead of days for testing purposes 2011-03-10T09:15:44 *** yboston has joined #evergreen 2011-03-10T09:19:10 *** gdunbar has joined #evergreen 2011-03-10T09:37:03 *** jenny has joined #evergreen 2011-03-10T09:39:31 re IPv6, we always turn it off 2011-03-10T09:42:26 and re private/public IPs for the domiains, the instructions are like that primarily to provide physical separation to match the conceptual separation 2011-03-10T09:42:29 IIRC 2011-03-10T09:43:49 I was about to edit a due date to say "I am tired of waiting for this item to be overdue" when I looked at the clock and said "oh, it is overdue". 2011-03-10T09:44:11 I've been wondering about the rationale behind the head/drone brick model... what if rather than 6 head + 2 drones bricks, we just have 18 "heads" (meaning self-contained "bricks" with no drones attached)? 2011-03-10T09:44:26 *** kmlussier has quit IRC 2011-03-10T09:44:32 csharp: Technically each head in that case is a drone 2011-03-10T09:44:37 *** sfortin has joined #evergreen 2011-03-10T09:44:55 I mean in a 1 head + 2 drones setup, if a drone goes out, it brings the brick down anyway, right? 2011-03-10T09:45:15 csharp: I've been wondering about that too, was thinking to break up a demo brick (head+drones) into 3 demo mini-bricks 2011-03-10T09:45:16 (at least until you edit .oils_brick.cfg) 2011-03-10T09:45:24 Assuming each brick has only one drone doing any given task 2011-03-10T09:45:24 csharp: no, the brick would still run in that case 2011-03-10T09:45:25 csharp: no, it just puts extra load on the other drone 2011-03-10T09:46:29 If your brick has one drone that does W and Y and a second drone that does X and Z the brick won't be able to function fully if either drone goes down. If both drones do XWYZ then a single drone going down won't take the brick with it (but will slow the brick down overall) 2011-03-10T09:46:40 the idea is that managing (say) 3 bricks as units from the config point of view is simpler than managing 9 entirely standalone systems 2011-03-10T09:47:05 both config on the bricks themselves, and on the load balancer etc 2011-03-10T09:47:36 more bricks means harder to track down crazy shit when one brick is misbehaving 2011-03-10T09:47:51 We are, for the time being, planning on a single head with possible drones config to start with. If we find we need more we can switch to load balancing multiple bricks later. 2011-03-10T09:48:30 so, you can add capacity by adding drones, which has minimal configuration overhead -- add a new host section to opensrf.xml, restart settings service (alone), fire up drone. no need to touch any "outside" bits 2011-03-10T09:48:51 * bradl slaps eeevil "stop touching my outside bits" 2011-03-10T09:49:04 bradl: eww 2011-03-10T09:49:13 tsbere: that will make rolling restarts impossible, unfortunately. which is a big benefit of multi-brick 2011-03-10T09:49:32 eeevil: True. 2011-03-10T09:49:49 eeevil: bradl: berick: thanks for the clarifications! 2011-03-10T09:50:57 it's really about layering of responsiblities .. if the outside IT dept is in charge of the load balancer then you have to get them to do work for you when you add another standalone brick 2011-03-10T09:51:16 Yeah, thanks all. :) 2011-03-10T09:51:19 well, it's not "really about" ... but that's another intended benefit 2011-03-10T09:52:34 unrelated ... objections to me pushing the grace period patch from last night into 2.0.3 2011-03-10T09:52:37 ? 2011-03-10T09:52:41 http://paste.lisp.org/display/120369 2011-03-10T09:54:47 I have no objections at this point. Seems sane to me. 2011-03-10T09:55:19 *** justin has joined #evergreen 2011-03-10T09:56:46 good morning folks. I have a client that we are trying to sell on Evergreen but all the test systems they have tried have uber slow webpacs. I've managed to find a few emails talking about a webpac rewrite, but wasn't able to turn up much to tell the client. 2011-03-10T09:57:07 Any idea about when we might see things like css/js aggregation, sprites, etc in the webpac? 2011-03-10T09:58:21 Or are their some sites that have particularly performant webpacs that we could borrow from/look at? 2011-03-10T10:00:35 *** justin__ has joined #evergreen 2011-03-10T10:02:37 justin: the rewrite is with kcls i believe, not sure if it is live 2011-03-10T10:02:40 *** justin has quit IRC 2011-03-10T10:02:59 I see, thanks eby 2011-03-10T10:03:40 their site still says search speed as an issue tho 2011-03-10T10:05:19 Wow, it looks nice though. I'll have to check out their performance later though... Drupalcon wifi isn't the best. 2011-03-10T10:06:14 if you are going to evergreen conf there is also a session on alternate opacs 2011-03-10T10:06:41 might interest you 2011-03-10T10:06:54 We're thinking about going, but I have a conflicting meeting. If anyone has any sense we can cancel the meeting... fingers crossed. 2011-03-10T10:09:02 So, in testing my grace periods patch I found a bug in fine generation. Went back to trunk without my patch the bug is still there. 2011-03-10T10:09:56 Anyone have any idea why having fines to generate on checkin would cause a "No active transaction to commit" message? I am thinking the fine generator is somehow closing the DB transaction. 2011-03-10T10:10:17 Note that fines are being generated, but the checkin isn't happening 2011-03-10T10:10:26 tsbere: a bug recently surfaced there... 2011-03-10T10:10:49 tsbere: two db transactions fighting for the same DB row, one wins, the other times out 2011-03-10T10:11:16 tsbere: if yr logging sql, you should see an insert into money.billing that takes several seconds to complete 2011-03-10T10:11:38 * Dyrcona can't find a class that will take a blob of MARC21XML and create a MARC Record object, neither MARC::Record nor MARC4J look like they can do this. 2011-03-10T10:11:54 berick: Am I allowed to ignore this bug as far as my patch functionality goes, then? 2011-03-10T10:11:57 and the number of seconds the commit takes will be roughly equivalent to the cstore keepalive interval 2011-03-10T10:12:27 tsbere: maybe setting fine intervals to minutes wasn't such a good idea. :) 2011-03-10T10:12:40 tsbere: yes, it's a bug regardless of your work 2011-03-10T10:13:03 Dyrcona: I figured it was slower than seconds but faster than hours or days! :P 2011-03-10T10:13:54 eeevil: maybe it's too early for me but isn't your proposed grace period patch checking the opposite condition than the one you want? 2011-03-10T10:14:12 * Dyrcona runs off to write yet another SAXContentHandler subclass. 2011-03-10T10:14:22 Hmm, any ideas about this repair environment error I'm seeing in our logs: http://paste.lisp.org/display/120378 2011-03-10T10:15:01 bshum: I see "simpla_record" when I would expect "simple_record" 2011-03-10T10:15:10 That's what I'm wondering 2011-03-10T10:15:23 bshum: Also "transation" instead of "transaction" 2011-03-10T10:15:28 Would that be something wrong with our action triggers? Or something wrong in the code? 2011-03-10T10:15:34 I have no clue. 2011-03-10T10:15:37 * Dyrcona doesn't have to, he hopes. 2011-03-10T10:15:38 But it seems odd 2011-03-10T10:15:43 Indeed 2011-03-10T10:16:36 Aha 2011-03-10T10:16:50 It is something wrong with the event definition's environment 2011-03-10T10:17:10 Thanks tsbere for pushing me in the right direction 2011-03-10T10:17:57 bshum: so was it your action trigger or in the code? 2011-03-10T10:18:27 csharp: I just scanned through our environment table in A/T schema and saw the mispelled simpla instead of simple 2011-03-10T10:18:34 Not sure how it got there though 2011-03-10T10:18:41 But easy enough to correct 2011-03-10T10:18:50 don't forget your billable_transaction typo 2011-03-10T10:19:35 WEird 2011-03-10T10:19:44 There seem to be quite a few mispellings 2011-03-10T10:19:48 billabe_transcation.summary 2011-03-10T10:19:54 Should be billable? 2011-03-10T10:20:08 And transaction 2011-03-10T10:21:46 I don't know how these got mangled 2011-03-10T10:21:54 Most are cloned events 2011-03-10T10:22:08 Or remnants from our 1.6 to 2.0 2011-03-10T10:23:39 bshum: are your libraries able to create these? 2011-03-10T10:23:55 csharp: No, HQ runs the A/T show 2011-03-10T10:23:59 hmm 2011-03-10T10:24:05 I'm wondering if maybe we copied events badly 2011-03-10T10:24:11 Or introduced typos somehow 2011-03-10T10:24:26 Or we cloned from bad events initially. 2011-03-10T10:24:39 bshum: too much celebration when going to 2.0? :-) 2011-03-10T10:24:39 Granted, cloning hasn't worked quite right apparently 2011-03-10T10:24:54 eeevil: I annotated your paste with what I *think* you meant. 2011-03-10T10:32:15 *** justin__ has quit IRC 2011-03-10T10:34:08 *** jenny has left #evergreen 2011-03-10T10:34:24 *** jenny has joined #evergreen 2011-03-10T10:39:13 dbwells: yes ... % ==, good catch 2011-03-10T10:40:19 yay, one more mark in the 'not crazy' column 2011-03-10T10:41:53 What would I need to do to get a Wiki account? 2011-03-10T10:42:46 dbwells: :) 2011-03-10T10:42:49 *** parsr has joined #evergreen 2011-03-10T10:43:48 Authority record search/browse in 2.0 -- is there a public URL to access outside of the Staff Client? 2011-03-10T10:44:22 parsr: not yet 2011-03-10T10:45:00 ok -- thanks eeevil 2011-03-10T10:45:26 parsr: (stay tuned, though ...) 2011-03-10T10:46:13 Tunes is my middle name -- looney tunes that is! 2011-03-10T10:49:15 *** dbs has joined #evergreen 2011-03-10T10:49:15 *** dbs has joined #evergreen 2011-03-10T10:51:16 I'm disinclined to add the fine gen patch to trunk or 2.1, in anticipation of seeing tsbere's approach to replacing grace stuff 2011-03-10T10:51:30 unless there's screaming for it 2011-03-10T10:51:39 eeevil: All I have left is the upgrade script. And I am in the middle of figuring out the last piece of that. 2011-03-10T10:51:52 tsbere++ 2011-03-10T10:52:53 Trying to decide if it is worth it to try and do something other than "just set everything to 1 in action.circulation" 2011-03-10T10:53:12 compared to saying "set everything that isn't checked in" as an alternate 2011-03-10T10:54:14 AaronZ-PLS: check with phasefx re: getting a wiki account 2011-03-10T10:54:28 or me 2011-03-10T10:54:42 AaronZ-PLS: or dbs :-) 2011-03-10T10:54:50 AaronZ-PLS: privmsg me your name, desired wiki username, and email address 2011-03-10T10:55:13 dbs:Thanks, Willdo 2011-03-10T11:04:39 rebuilding both the 2.1 and 2.0.3 tarballs, both suffered the newpo issue 2011-03-10T11:05:16 So my upgrade script has to drop types, functions, and views, then recreate them all. But it appears to be working now. 2011-03-10T11:06:43 eeevil: If you want to pull from my git repo you can see it here: http://git.mvlcstaff.org/?p=tsbere/ILS.git;a=shortlog;h=refs/heads/graceperiods 2011-03-10T11:07:05 Otherwise I will be throwing stuff on launchpad shortly 2011-03-10T11:09:36 eeevil: looking at the backscroll, I have concerns about this changeset http://svn.open-ils.org/trac/ILS/changeset/19672 2011-03-10T11:10:31 I think that is going to fail for anyone who goes from 1.6.0 -> 1.6.1 -> 2.0 2011-03-10T11:11:31 dbwells: why will it fail? 2011-03-10T11:11:48 I mean, what's the path to failure 2011-03-10T11:13:01 Right, sorry. The 1.6.0 -> 1.6.1 adds that definition using NEXTVAL, so trying to add the same one at '20' will fail on the owner/name unique constraint. 2011-03-10T11:13:55 line 55 of this: http://svn.open-ils.org/trac/ILS/browser/branches/rel_1_6_1/Open-ILS/src/sql/Pg/1.6.0.4-1.6.1.0-upgrade-db.sql 2011-03-10T11:14:31 dbwells: so, the question is ... when is the sequence pushed to 100 to "reserve" event def ids before that upgrade script, we should be good 2011-03-10T11:14:44 or before that use of NEXTVAL 2011-03-10T11:15:08 that ... was not a sentence ... :) 2011-03-10T11:15:15 :) 2011-03-10T11:15:37 "when is the sequence pushed to 100 to "reserve" event def ids?" "if it's before that use of NEXTVAL, we should be good" 2011-03-10T11:17:51 dbwells: it happens very early on. it's in the 1.4.0-1.6.0 upgrade script 2011-03-10T11:18:00 eeevil: And launchpad goodness https://bugs.launchpad.net/evergreen/+bug/732679 2011-03-10T11:18:09 tsbere: rad, thanks 2011-03-10T11:19:14 Perhaps another post-2.0 overhaul task is "reduce use of synthetic keys where good natural keys exist". Insert usual refrain about wishing we had an extensive test suite to avoid resulting breakage :) 2011-03-10T11:19:50 eeevil: so, in our case, we ran the 1.6.0.4->1.6.1 upgrade, and have that event def at ID 101. So the changeset's insert of the same definition at ID 20 is going fail, right? 2011-03-10T11:20:34 dbwells: yes ... you're still going to have to hack the upgrade script 2011-03-10T11:21:11 (now, I note that that's /always/ been the case, and the upgrade script still would have failed for you for basically the same reason before that changeset) 2011-03-10T11:21:49 eeevil: actually, the upgrade script worked fine, just ran it last week (?) 2011-03-10T11:21:51 but now, at least, it /can/ succeed. it could not have before even on a clean-1.6.1 to 2.0 upgrade 2011-03-10T11:22:28 dbwells: it should not have, unless you already had an event def with id=20 2011-03-10T11:23:27 It seems to me that we have made it work on a clean 1.6.1 but made it break on any 1.6.0->1.6.1->2.0. 2011-03-10T11:24:32 dbwells: I blame the 1.6.0-1.6.1 for that (for using nextval) 2011-03-10T11:24:41 And we don't have any event def 20 :( Getting confused... 2011-03-10T11:25:07 can we do this by action_trigger.event_definition.name instead, using a function like config.modify_metabib_field()? 2011-03-10T11:25:34 Will not having event def 20 in the future hurt us? 2011-03-10T11:26:07 bshum: the upgrade script takes event def 15 (if it exists) and moves it to 20 2011-03-10T11:26:30 bshum: and now, if it didn't exist, inserts a 20 2011-03-10T11:26:51 Find atevdef.name = 'Password reset request notification', check ID value; if 15, move it to 20 with the logic already in place (just embed it in the function); if it doesn't exist, insert it 2011-03-10T11:26:56 ? 2011-03-10T11:27:12 And if people have changed the name of the event, well, the world ends in 2012 anyway 2011-03-10T11:27:32 dbs: I think that's the right track 2011-03-10T11:28:02 if someone is inclined to code that up, I will hold off for another couple hours 2011-03-10T11:28:20 but there's a much bigger fix in the upgrade script that needs to get out there and supercede 2.0.2 2011-03-10T11:28:32 specifically, that all booking data will be lost 2011-03-10T11:28:55 that is certainly much more important 2011-03-10T11:29:25 So def 20 is supposed to be password reset (not wherever it ended up during our upgrade to 2.0). And we should probably move that back to 20 to get in line with the rest of the world? 2011-03-10T11:29:43 (trying to avoid headaches down the road too) 2011-03-10T11:29:58 bshum: ideally, yes, it's a reserved id 2011-03-10T11:31:59 eeevil: Okay, thanks for the confirmation. Sorry to interrupt the task at hand, just trying to make sure we don't slip up on our system moving forward. 2011-03-10T11:32:22 eeevil: I'll take a crack at it making it a little safer. Only fair since I brought it up, I suppose. 2011-03-10T11:32:30 dbwells++ 2011-03-10T11:33:32 In trying to decide what to work on next, I realize that I have a task at hand if my most recent patch is accepted. Namely adding it to my indb circ docs. 2011-03-10T11:38:59 *** parsr has quit IRC 2011-03-10T11:40:27 gmcharlt++ #note to the list about the booking issue 2011-03-10T11:53:53 eeevil: dbs: will something as simple as this work in the commonly conceivable cases? http://paste.lisp.org/display/120381 2011-03-10T11:56:46 dbwells: Pretty SQL, I was just figuring out how to write the second one that you listed there. 2011-03-10T11:57:07 dbwells++ # worked to move our password reset back to id 20 2011-03-10T11:57:21 we could search on the name instead, if that seems any more reliable 2011-03-10T11:58:02 I think hook is way more reliable 2011-03-10T11:58:42 I guess all of this assumes that there is no existing ID = 20 2011-03-10T12:01:41 right, I think all the solutions have made that assumption so far. 2011-03-10T12:08:09 Seems fair 2011-03-10T12:10:39 that assumption is fine, IMO. again, "reserved" ids 2011-03-10T12:11:34 dbwells: so, we have to further assume that there is, indeed a password.reset_request def, and that there are no changes between the 2.0 version of that def and the 1.6.1 version 2011-03-10T12:13:06 dbwells: to address those, would it be reasonable to use the INSERT ... SELECT ... WHERE (like I did with the hook) after the laste UPDATE to make sure that in the end there is a def 20 (and the hook) 2011-03-10T12:15:33 gmcharlt: I'm not sure how adding the translated versions of author names to authority records will help Christoph with searches on either version of the name, although at some point it would be nice to update metabib.*_field_entry with entries from linked authority records 2011-03-10T12:16:46 dbs: yes, I was anticipating functionality with that response, but hey ... giving you and me impetus to enrich the bib searches with authority keywords is a worthy goal ;) 2011-03-10T12:17:28 gmcharlt: that would actually make authority records sort-of useful 2011-03-10T12:17:33 can't have that 2011-03-10T12:18:32 at that point, I wonder whether we start teaching open-ils.search about solr so that we can cut down on the crazy database server RAM specs for large collections 2011-03-10T12:19:15 eeevil: well, it can't hurt. Safety first, someone once said to me. OTOH if someone manages to get to 1.6.1 without that definition, they might have bigger problems. 2011-03-10T12:19:20 dbs: indeed not. we would disappoint some catalogers by removing one source of their complaints about ILSs 2011-03-10T12:20:03 eeevil: also, let me verify that that definition hasn't changed since 1.6.1 2011-03-10T12:21:13 dbwells: or they remove it ... of course, there's also the risk of /mulitple/ defs using password.reset_request, but that's probably not a big deal 2011-03-10T12:21:30 or, rather, a likely case 2011-03-10T12:22:23 not a, that is 2011-03-10T12:22:56 eeevil: if they have removed it for whatever reason, is there another part of the script which depends on it being there? 2011-03-10T12:23:05 eeevil: actually, we almost did that - for libraries that operate entirely in French, I was going to bust out the notices at the system level instead of all consortial 2011-03-10T12:23:31 dbs: ahh, I see 2011-03-10T12:23:49 instead went the one-template-to-rule-them-all approach via stat_cat 2011-03-10T12:24:05 Upgrades suck. 2011-03-10T12:24:15 Indeed 2011-03-10T12:24:41 dbwells: no, removing won't break any RI, but the feature depends on it existing ... and removing the def is the wrong way to "disable" it 2011-03-10T12:24:59 *** frank_ has joined #evergreen 2011-03-10T12:25:05 hello 2011-03-10T12:25:07 dbs: rather than me rooting around, are you aware off-hand if that event definition changed between 1.6.1 and 2.0? 2011-03-10T12:26:04 i´d want to get some help, when i install staff client in a client machine, i get the error "TypeError: g.my_libs_tree is null", does someone know what do i have to do to fix it? 2011-03-10T12:27:22 eeevil: as long as we aren't the ones removing it, is it really our responsibility to put it back? Oh, I'll just stop wondering and do it anyway. 2011-03-10T12:28:56 frank_: you ran "autogen.sh -u" on the server? 2011-03-10T12:29:57 dbwells: It should be identical 2011-03-10T12:30:15 dbs: great, thanks 2011-03-10T12:31:10 Hmm, what's the "activity meter" showing in the staff client? Network activity during different actions? 2011-03-10T12:31:27 dbs: yes, i did it, but i guess i already know eahy i get that problem, i changed the locale to US and it worked but if i change it to other one, i get the error message 2011-03-10T12:33:02 dbwells: if someone removed the def, they broke the feature (same with the hook, which should be put back if missing). the only reason to do that is they were using a shotgun to "disable" it. it's unbreaking, and if they want it disabled they should do so the prescribed way ... so, in this case, I think we should put it back. or, dbs (since you wrote that feature), am I all wet? 2011-03-10T12:33:39 *** jennam has quit IRC 2011-03-10T12:33:44 eeevil: I'm with you, man. There's an "active" column on the event_def table for a reason, right? 2011-03-10T12:34:28 * dbs peers at 530K of auth records from NRCan 2011-03-10T12:36:47 *** AaronZ-PLS has quit IRC 2011-03-10T12:40:12 eeevil: dbs: I am not saying fixing stuff someone broke with a shotgun is a bad idea. It just seems like an odd job for the upgrade script, that's all. Nevertheless, the opportunity is here, so I'll get it in there. 2011-03-10T12:41:18 *** eeeevil has joined #evergreen 2011-03-10T12:53:29 *** eeeevil has quit IRC 2011-03-10T12:57:19 *** AaronZ-PLS has joined #evergreen 2011-03-10T12:59:27 *** kmlussier has joined #evergreen 2011-03-10T13:14:20 *** jeffdavis has joined #evergreen 2011-03-10T13:16:30 *** jenny has quit IRC 2011-03-10T13:25:41 *** dbs has quit IRC 2011-03-10T13:30:21 eeevil: patch committed to rel_2_0 and 2_0_3 2011-03-10T13:30:31 dbwells: rock, sir 2011-03-10T13:33:32 dbwells: looks great! thanks again! 2011-03-10T13:33:56 no problem 2011-03-10T13:40:35 wrapping now ... wheee! 2011-03-10T13:58:05 *** rjackson-isl has quit IRC 2011-03-10T13:59:34 *** tjvcarter has joined #evergreen 2011-03-10T14:05:06 apropos of nothing, here's to hoping that a binary xml contrib module does indeed get added to PG soon ... 2011-03-10T14:09:43 ok, 2.0.3-forReal tarball up 2011-03-10T14:12:53 and the 2.1-prealpha1 is up too 2011-03-10T14:13:43 *** jenny has joined #evergreen 2011-03-10T14:15:08 Building 2.0.3 SC 2011-03-10T14:15:42 bshum: thanks 2011-03-10T14:17:53 phasefx, you around? 2011-03-10T14:17:59 tsbere: I am 2011-03-10T14:18:12 Just to double check, but the 2.0.3 is the correctly named tar.gz right? Not really 2.0.3-forReal ? :) 2011-03-10T14:18:24 (making sure I didn't grab the wrong one) 2011-03-10T14:18:27 bshum: no, not forReal ;) 2011-03-10T14:18:30 so, yes 2011-03-10T14:19:04 there's no -forReal source tarball (I hope!) 2011-03-10T14:19:26 phasefx: Whipped up a quick "no empty workstation registration" patch. Any opinion? http://git.mvlcstaff.org/?p=tsbere/ILS.git;a=shortlog;h=refs/heads/emptyws 2011-03-10T14:21:31 tsbere: looks good to me. one thing worth re-considering (for that interface in general, not related to that patch) is the auth type used.. it times out very quickly, so currently, one doesn't want to lolligag while registering a workstation 2011-03-10T14:22:32 eeevil: Cool, cool :) 2.0.3 SC is here: http://ur1.ca/2yjyo 2011-03-10T14:22:43 phasefx: not wanting to lolligag is no excuse for not typing a name ;) 2011-03-10T14:22:44 ur1 is a bit slow though, if that link doesn't work I'll get another one 2011-03-10T14:22:55 tsbere: true dat :) 2011-03-10T14:23:04 would you like me to throw that on launchpad or something? Or at least tell my local git to make a patch file? 2011-03-10T14:23:35 tsbere: I'm going to need to learn how to pull from your git one day, but yeah, if I see something in email (which launchpad accomplishes), I'll eventually remember/get to it. tsbere++ 2011-03-10T14:23:52 phasefx: It is really easy. Want a set of git commands? 2011-03-10T14:24:32 90% of it is adding the remote. 2011-03-10T14:26:00 sure. context, I have a git-svn repository, and I have a remote for git.esilibrary.com. I think I can manage adding yours as a remote; I'd just want to sanity check the workflow for getting that specific commit into my master so I can push to svn 2011-03-10T14:26:56 phasefx: boils down to creating a topic branch for testing the commit, then either merging tsbere's branch or cherry-picking the specific commit 2011-03-10T14:27:01 so if this is all in master 2011-03-10T14:27:13 phasefx: git remote add tsbere git://git.mvlcstaff.org/tsbere/ILS < That will get you the new remote. Then just "git fetch tsbere" to update and tsbere/blah become valid merge targets. 2011-03-10T14:27:17 git checkout master 2011-03-10T14:27:19 bshum: uploading 2011-03-10T14:27:19 git svn rebase 2011-03-10T14:27:28 git checkout -b test-tsbere-stuff master 2011-03-10T14:27:31 add the remove 2011-03-10T14:27:34 *remote 2011-03-10T14:27:48 and either git merge or git cherry-pick 2011-03-10T14:29:15 eeevil: With all the fiasco yesterday, you didn't also cut a 1.6.1.8 yet right? (was going to help make that but it doesn't seem to exist yet) 2011-03-10T14:30:12 bshum: not yet, no 2011-03-10T14:31:04 phasefx: If still needed later I will throw a patch somewhere. But for the moment time to head out. 2011-03-10T14:31:22 tsbere: this is working so far, trying to remember how to do sign-off :) 2011-03-10T14:31:55 I see it for cherry-pick, but not merge 2011-03-10T14:32:20 phasefx: I think you have to create the branch with sign-off? 2011-03-10T14:32:30 no 2011-03-10T14:32:39 git cherry-pick -s to sign off 2011-03-10T14:32:40 or not ;) 2011-03-10T14:32:50 while cherry-picking 2011-03-10T14:33:01 or git commit --amend -s 2011-03-10T14:33:40 alright, let's see how that looks in trac 2011-03-10T14:34:12 no sign of git's Author: field 2011-03-10T14:34:27 yes, that's a limitation of git-svn 2011-03-10T14:34:43 * phasefx will give tsbere credit when backporting 2011-03-10T14:35:02 this could be backported far, do we have a string freeze in any of our branches? 2011-03-10T14:35:32 I'll just do 2.1 and not worry too much 2011-03-10T14:38:43 eeevil: I assume that with the changes made to the 1.6.1-2.0 upgrade script that's now packaged with 2.0.3, should update the note we have in the 2.0 upgrade wiki page to indicate that now people should be using the 2.0.3 upgrade script instead of the 2.0.2 2011-03-10T14:38:45 tsbere: alright, it's done, thanks! You're now fast-tracked with me for patches courtesy of git :D 2011-03-10T14:39:24 gmcharlt: thanks for sanity 2011-03-10T14:40:22 is there any field documentation UI anywhere other than the patron registration screen? 2011-03-10T14:40:30 *** dbs has joined #evergreen 2011-03-10T14:41:08 like anywhere field documentation is currently used on anything other than "ILS User" 2011-03-10T14:41:54 tsbere: just read your grace patch; it looks like you're setting a boolean for grace = true or false, so keeping the previous behaviour vs. adding an explicit "grace_period" definition (which was the direction I was going to head, until you said you were already going there) 2011-03-10T14:42:55 tsbere: difference being that we want to define grace periods of 15 minutes for a 4 hour loan, not 1 hour or 1 minute (if you define the duration as '240 minutes' instead of '4 hours') 2011-03-10T14:43:08 jamesrf: hrmm, I bet we could do something funky like add it as tooltips to autogrid and xul list headers (for xul lists using the IDL for column definitions) 2011-03-10T14:50:48 dbs: not sure if I understand your meaning, but doesn't the 'grace' setting (currently and in the patch) equal a certain number of fine_intervals of grace? 2011-03-10T14:56:37 * dbs needs to jet but will reconsider 2011-03-10T14:56:57 dbs, others: if it's not already in, looks like the Google Summer of Code mentor organization deadline is in <24h: 11 March at 23:00 UTC 2011-03-10T14:57:37 (if you weren't already aware -- otherwise, disregard my heads up) 2011-03-10T14:58:48 jeff: thanks 2011-03-10T14:59:18 bshum: I'm not sure I follow ... anyone on 1.6.1 or before that wants to come to 2.0 should download 2.0.3, use that 1.6.1-2.0 (which gets them to 2.0.0) and then use the 2.0.0-2.0.1, etc, scripts 2011-03-10T14:59:48 and, ideally, they'd upgrade to 1.6.1.7 first, in the normal way 2011-03-10T14:59:58 (at least at the db level 2011-03-10T15:00:38 bonus (aside): once on 2.0.3, upgrading to postgres 9.0 should be possible 2011-03-10T15:00:51 eeevil: Right, guess we should update the 1.6.1-2.0 upgrade steps on the wiki (and other places, including DIG drafts) accordingly then. 2011-03-10T15:01:05 They still read 2.0.0 2011-03-10T15:01:34 Sent the patch for the downloads page to dev list as reply to your initial release team message 2011-03-10T15:03:06 *** tjvcarter has quit IRC 2011-03-10T15:06:45 @later tell dbs The "grace" variable, pretty much everywhere in the db, is intended to be multiplied by the fine interval for the total grace period. Thus, 0 is off, > 0 is on. 1 is "one fine interval" and recommended default from some point in IRC, 3 would be "three fine intervals", etc. 2011-03-10T15:06:45 tsbere: The operation succeeded. 2011-03-10T15:07:46 *** jeffdavis has quit IRC 2011-03-10T15:07:50 *** jeffdavis has joined #evergreen 2011-03-10T15:08:08 * tsbere will need to start doing the sign-off himself so that phasefx has less work to do when pulling from his git ;) 2011-03-10T15:08:43 Not that the emptyws branch was a lot of code this time around 2011-03-10T15:11:11 heh 2011-03-10T15:11:45 though signing off on one's own patches is not a bad idea, used a signal that a particular version of a patch is ready to submit 2011-03-10T15:12:10 true 2011-03-10T15:12:22 just means that a patch review process would be looking for more than one sign-off as a desideratum 2011-03-10T15:12:32 *** gdunbar has quit IRC 2011-03-10T15:12:50 *** jeffdavis has left #evergreen 2011-03-10T15:12:53 I am just glad today that our libraries won't be cluttering the db with "shortname-" entries ;) 2011-03-10T15:13:39 Our current "demo" system has several, last I checked. 2011-03-10T15:14:55 dbwells, tsbere: right, sorry - it's been two years since I set up most of our circ rules 2011-03-10T15:15:31 still, I'd rather see a grace_interval that is defined in and of itself rather than in terms of fine intervals 2011-03-10T15:16:16 The example I was using (4 hour loan, 15 minute grace period) was a real one from last week, accompanied by a request for a $15/hour fine 2011-03-10T15:17:16 It forces you to divvy up the fine intervals in weird ways, no? $3.75 / 15 minutes instead of the straightforward $15/hour 2011-03-10T15:17:43 I didn't change how the fines are calculated, just where the "grace" value comes from. I figure changing how the grace period is calculated is a different task. 2011-03-10T15:18:02 Shouldn't be too hard, though 2011-03-10T15:18:15 tsbere: right, I understand that. But given that you're moving to per-rule grace values, why not go the extra step? 2011-03-10T15:18:15 dbs: would NUMERIC instead of INT work? then you can specify a portion of a fine interval 2011-03-10T15:19:16 eeevil: that seems like an unnecessary complication on the configuration side. I would like to be able to put this in the hands of librarians, and math is hard! 2011-03-10T15:20:22 jeff: thanks for the reminder. yes, the clock is ticking :) 2011-03-10T15:21:03 If we think a grace_interval setting is potentially desirable I can see about adding that too, either as a secondary patch or an update to the current one. Doesn't look too bad for implementation right now (beyond the DB side, anyway). 2011-03-10T15:21:56 dbs: sure 2011-03-10T15:22:22 tsbere: I was thinking instead of - if you have a grace_interval setting, you don't need a grace INT setting, right? 2011-03-10T15:22:34 I would think if are adding a grace_interval, we don't need the grace setting. 2011-03-10T15:22:36 tsbere: this may benefit from a RFC to the general and dev lists, IMO 2011-03-10T15:22:46 yeah, what dbs said. 2011-03-10T15:22:57 (and I agree with dbs, one or the other) 2011-03-10T15:24:03 Either way the db change is the hardest part ;) 2011-03-10T15:24:24 though grace_period may be a better name, perhaps 2011-03-10T15:24:44 +1 to switching to grace to an interval as part of this patch. (and bonus points for finishing the part of the code in the "if(0)" block) :) 2011-03-10T15:25:05 been meaning to look at that for a long time 2011-03-10T15:25:07 dbwells: heh ... I second the bonus 2011-03-10T15:25:39 I was ignoring the if(0) due to it being, well, an if(0) :P 2011-03-10T15:27:42 tsbere++ 2011-03-10T15:28:24 * tsbere branches graceperiods to graceintervals locally, but will use grace_period instead of grace_interval code-side 2011-03-10T15:46:11 btw, csharp, nice detective work on the squeeze / ipv6 angle 2011-03-10T15:47:48 dbs: thanks! 2011-03-10T15:48:58 Huh. So far amazingly painless. 2011-03-10T15:49:04 Assuming no typos 2011-03-10T15:49:06 >_> 2011-03-10T15:59:55 *** rsinger has quit IRC 2011-03-10T16:04:03 *** sfortin has quit IRC 2011-03-10T16:07:56 *** kmlussier has quit IRC 2011-03-10T16:08:57 dbs: for the GSoC, any objections to assigning the $500 stipend that the mentoring organization would get to either the SFC (if EG joins) or to the pool for the next Evergreen conference? 2011-03-10T16:14:05 *** bshum has quit IRC 2011-03-10T16:27:02 gmcharlt: no way man. I'm in it for the big bucks 2011-03-10T16:27:19 *** rsinger has joined #evergreen 2011-03-10T16:28:19 dbs: all is revealed! ;) 2011-03-10T16:28:39 (seriously though - makes sense to me) 2011-03-10T16:28:47 anyway, since the application is an HTML form, I'm filling it out 2011-03-10T16:28:52 and will pastebin each piece 2011-03-10T16:28:56 So, I think I have everything except for the "bonus" done 2011-03-10T16:35:23 gmcharlt: awesome, thanks man 2011-03-10T16:35:59 I was planning on adding a brief intro to the "ideas" document, pointing to some developer resources that would assist a new dev poking around 2011-03-10T16:37:04 dbs: sounds good 2011-03-10T16:38:28 tsbere: to answer your LP question, the autoload stuff is to add fields to the Circulator object being defined here, not the circulation Fieldmapper object which is automagically defined elsewhere. So you only need 'grace' in the AUTOLOAD_FIELDS if you intend to say something like '$self->grace' at some point. 2011-03-10T16:41:09 (or wherever you use a Circulator object, like in run_method) 2011-03-10T16:45:24 dbs: just added an idea to the ideas page 2011-03-10T16:51:54 dbs: do you know how to bring the pastebot back? 2011-03-10T16:52:25 gmcharlt: I was the contact in the past, but for now you paste then go back to paste.lisp.org and copy/paste 2011-03-10T16:52:33 http://paste.lisp.org/+2KWA 2011-03-10T16:52:36 * dbs will be gone for another half hour or so 2011-03-10T16:53:01 dbs: OK - when you get back, check the backscroll 2011-03-10T16:56:46 dbs: also, please establish a link ID on the GSoC so that I can list you as a backup admin on the applicaiton 2011-03-10T17:00:12 http://paste.lisp.org/+2KWA/1 # answer to how mentors were selected 2011-03-10T17:02:51 dbwells: As I do indeed do the ->grace (and in the new version ->grace_period) bit I believe I was correct in wanting that in the autoload, then. (I was pretty sure, one of my early tests failed until I added it anyway) 2011-03-10T17:02:58 but thanks for the confirmation 2011-03-10T17:03:52 Or at least I was doing that. I may have --amended that line out. 2011-03-10T17:03:53 Hmmm 2011-03-10T17:03:56 * tsbere will test that again 2011-03-10T17:05:09 Oh, right, I switched from storing it outside the recurring fine rule to just using it inside 2011-03-10T17:06:19 *** tsbere has quit IRC 2011-03-10T17:07:02 *** rickd_ has quit IRC 2011-03-10T17:07:15 *** tsbere has joined #evergreen 2011-03-10T17:07:39 *** rickd_ has joined #evergreen 2011-03-10T17:13:14 http://paste.lisp.org/+2KWA/2 # plan for disappearing mentors 2011-03-10T17:18:11 gmcharlt++ 2011-03-10T17:20:31 *** AaronZ-PLS has quit IRC 2011-03-10T17:21:23 http://paste.lisp.org/+2KWA/3 # how we paln to encourage students to participate 2011-03-10T17:22:15 gmcharlt: I thought we switched to every other week again on the meetings? 2011-03-10T17:22:29 Or would we go back to weekly for the gsoc period? 2011-03-10T17:22:39 tsbere: thanks, I had forgotten 2011-03-10T17:23:15 doing it weekly during the gsoc period might be nice for the students, but I'll just put down every other week on the application 2011-03-10T17:23:59 * tsbere has gone back to loan duration, fine interval, and now grace period durations measured in minutes for testing 2011-03-10T17:24:19 Amazed it took my marc record, though. 2011-03-10T17:28:34 http://paste.lisp.org/+2KWA/4 # plan for disappearing students 2011-03-10T17:28:41 *** Dyrcona has quit IRC 2011-03-10T17:34:17 gmcharlt++ 2011-03-10T17:35:51 dbs: do you know of a past GSoC organization that would be willing to vouch for us? 2011-03-10T17:35:59 I noticed that hte SFC is one 2011-03-10T17:42:15 gmcharlt: Maybe also ask on #code4lib? 2011-03-10T17:43:28 moodaepo: thanks, I've done so 2011-03-10T17:45:52 http://paste.lisp.org/+2KWA/5 # project description 2011-03-10T17:51:57 gmcharlt: yeah, I was going to suggest SFC - bkuhn has assisted in the past with GSoC and various member projects 2011-03-10T17:52:10 sorry, I'm primarily in parental mode atm 2011-03-10T17:52:38 dbs: no worries, although I expect to see patches from the kid as soon as possible ;) 2011-03-10T17:52:44 *** jenny has left #evergreen 2011-03-10T17:53:56 gmcharlt: for #1, perhaps also "* past history of demonstrated willingness to provide guidance to new contributors to Evergreen" 2011-03-10T17:54:34 dbs: added 2011-03-10T17:56:50 gmcharlt: not sure if evidence like http://markmail.org/search/?q=evergreen#query:evergreen%20list%3Aorg.georgialibraries.list.open-ils-general+page:1+state:facets would be useful or not - at least 2/3 mentors are on top there :) 2011-03-10T17:57:32 s/past history of demonstrated/demonstrated/ perhaps? 2011-03-10T17:57:41 and 1/3 needs to write more ;) 2011-03-10T17:57:41 * dbs can't think of a future history... 2011-03-10T17:58:06 phasefx and I have an extra 3 years or so on you 2011-03-10T17:58:28 and I've been working on the Other Big Free ILS, of cousre 2011-03-10T17:58:46 yeah, I was going to mention that :) 2011-03-10T17:59:26 hmm, perhaps point to the -dev list instead of -general, if the discerning judges want to see development discussions 2011-03-10T17:59:38 http://paste.lisp.org/+2KWA/6 # revised version 2011-03-10T18:00:03 http://markmail.org/search/?q=list%3Aorg.georgialibraries.list.open-ils-dev is a cleaner link 2011-03-10T18:00:21 awesome 2011-03-10T18:02:33 gmcharlt: my link ID is the clever "danscott" 2011-03-10T18:04:28 http://paste.lisp.org/+2KWA/7 # answer to the anything else qusetion 2011-03-10T18:13:33 Typo in #1 s/maling/mailing/ 2011-03-10T18:14:26 http://paste.lisp.org/+2KWA/8 # why are we applying; this response is the one I'm least happy with so far 2011-03-10T18:15:05 fixed (and ouch!) 2011-03-10T18:16:37 gmcharlt: oh - should we point to the FLOSS Weekly interview somewhere? "Randal vouched for the project by putting us on TWiT, we've gotta be worthy!" 2011-03-10T18:17:36 Hah, for #8 you forgot to mention "THE FILTHY LUCRE" 2011-03-10T18:17:42 sssssh! 2011-03-10T18:18:00 we could put in links to publicity in the final question 2011-03-10T18:18:07 what's the interview link? 2011-03-10T18:18:52 I think for #8 we could probably work in something about the academic connection with Evergreen - 2011-03-10T18:18:54 http://twit.tv/floss132 2011-03-10T18:19:34 dbs: I was just going to suggest that angle for #8 2011-03-10T18:21:03 http://paste.lisp.org/+2KWA/9 2011-03-10T18:21:54 "One of the mentors works in an academic library and is eager to help students gain practical software development experience." ish? 2011-03-10T18:23:02 gmcharlt: If you both don't mind me butting in "to initiate work on some long term enhancements to Evergreen for example extending it from its current role of managing physical items for libraries to also handling electronic resources." 2011-03-10T18:23:39 moodaepo: go ahead 2011-03-10T18:24:13 Do we also want to mention that when the Evergreen project unsuccessfully applied for the GSoC in 2008(?), the project picked itself up by its bootstraps and funded two students to deliver internationalization and install & configuration enhancements to Evergreen."? 2011-03-10T18:24:40 yes 2011-03-10T18:24:42 Maybe in the "anything else" column, if we opt to use that 2011-03-10T18:25:37 (natural follow-on is "why the heck didn't you apply for the last couple of years then, dummies!") 2011-03-10T18:26:47 http://paste.lisp.org/+2KWA/A # current state of anything else 2011-03-10T18:29:35 dbs: The answer to that follow-on is "because we were too busy coding" ;) 2011-03-10T18:30:45 dbs: I can see some of what you both are posting being used on the website (also: it's just nine sorry bastards working on the code, dummies and none of them is Dirk Gently) 2011-03-10T18:30:53 tsbere++ 2011-03-10T18:31:37 *** yboston has quit IRC 2011-03-10T18:32:16 http://paste.lisp.org/+2KWA/B # proposed application template for the students 2011-03-10T18:33:18 Should we play up the angle of wanting to improve the options available to libraries and their patrons as part of the "why applying"? 2011-03-10T18:33:35 tsbere: certainly we should mention the patrons 2011-03-10T18:33:45 (will nobody think of the patrons?) ;) 2011-03-10T18:34:07 moodaepo: btw, what were you driving at with your 18:23? 2011-03-10T18:34:08 I figure pointing out that improvements will directly and indirectly help people not even using the software themselves is a good thing ;) 2011-03-10T18:34:57 moodaepo: other than the sentence being a awkward 2011-03-10T18:35:42 gmcharlt: Yea there was another awkward line one sec 2011-03-10T18:36:06 and yes just that it was awkward great stuff otherwise 2011-03-10T18:36:58 moodaepo: yep, thought so. breaking it up helps: "To initiate work on some long term enhancements to Evergreen. In particular, we want to expand it from its current role managing physical items for libraries to also handling electronic resources" 2011-03-10T18:39:19 gmcharlt++ # the other line: "The Evergreen project previously applied for the GSoC in 2008, although the project was not accepted, the community funded two students to deliver internationalization and install & configuration enhancements to Evergreen." 2011-03-10T18:39:49 s/, a/. A/ on that one 2011-03-10T18:40:23 Well you can loose the second coma unless you do want to break it into two lines. 2011-03-10T18:40:36 "The Evergreen project unsuccessfully applied for the GSoC in 2008. Although the project was not accepted, we went ahead and funded two students to deliver internationalization and installation and configuration enhancements to Evergreen." 2011-03-10T18:41:17 I was trying to loose the word "unsuccessful" : ) 2011-03-10T18:42:01 applied but was turned down as alternate wording? 2011-03-10T18:42:16 gotcha 2011-03-10T18:42:40 "The Evergreen project applied for the GSoC in 2008. Although the project was turned down, we went ahead and funded two students to deliver internationalization and installation and configuration enhancements to Evergreen. 2011-03-10T18:43:18 *** _bott_ has left #evergreen 2011-03-10T18:44:35 http://paste.lisp.org/+2KWA/C # current state of anyting; note bit about dealing with the money 2011-03-10T18:45:34 s/we went ahead and/the community/ (who is we?) 2011-03-10T18:45:54 s/either for the fund/either into the fund/ 2011-03-10T18:46:52 gmcharlt: Even though I think ESI funded the students (not sure though) 2011-03-10T18:47:03 moodaepo: thanks for corrections 2011-03-10T18:47:20 moodaepo: I suspect we did, but I'm not sure; before my time 2011-03-10T18:52:22 I've submitted the draft (don't worry, it can be edited right up to the deadline) 2011-03-10T18:52:28 gmcharlt: Also should we change "internationalization and installation and configuration enhancements to Evergreen" to "internationalization and installation/configuration enhancements to Evergreen" that two ands are making it look weird. 2011-03-10T18:52:31 gmcharlt++ 2011-03-10T18:52:44 stepping out for a bit, feel free to keep up the suggestions 2011-03-10T18:53:13 moodaepo: done 2011-03-10T18:53:41 yay 2011-03-10T18:55:58 *** _bott_ has joined #evergreen 2011-03-10T19:06:03 For those that care, threw up a new grace periods patch/branch 2011-03-10T19:12:25 @later tell gmcharlt ESI funded one student, Laurentian funded the other. Also, in the "anything" section (or perhaps the student socialization section?) we might want to add the 3rd annual international Evergreen conference - as a "Hey, we're not rinky-dink" marker 2011-03-10T19:12:25 dbs: The operation succeeded. 2011-03-10T19:12:47 * dbs goes back to getting kids to sleep 2011-03-10T19:24:44 dbs++ # Clarifying the funding sources. 2011-03-10T19:24:45 back 2011-03-10T19:26:23 gmcharlt: What about noting KCLS adoption of Evergreen? 2011-03-10T19:28:29 me <-- heads home 2011-03-10T19:28:41 Oops missed a / 2011-03-10T19:28:59 moodaepo++ 2011-03-10T19:29:44 size, numbers and location "Evergreen in the Evergreen state!" Or "In Microsoft's backyard Evergreen open source get's adopted." (but I bet Koha was there long before : ) 2011-03-10T19:30:16 *** rkobus has joined #evergreen 2011-03-10T19:30:32 *** rsinger has quit IRC 2011-03-10T19:30:44 *** rkobus has quit IRC 2011-03-10T19:33:26 *** phasefx has quit IRC 2011-03-10T19:33:27 dbs: thanks; added both mentions 2011-03-10T19:33:46 *** phasefx has joined #evergreen 2011-03-10T19:38:01 moodaepo: Bill Gate's library uses OSS software! 2011-03-10T19:38:24 (not sure if that's actually his home library, though, but it would be a nice thought if true) 2011-03-10T19:44:55 http://paste.lisp.org/display/120405 # paste of entire GSoC application as it exists now; note that formatting looks better in reality 2011-03-10T19:57:17 *** frank_ has left #evergreen 2011-03-10T20:53:49 *** moodaepo_ has joined #evergreen 2011-03-10T20:58:23 gmcharlt: I see, a quick Google search didn't find anything about a Gate's library using open source. I did find this [ http://techrights.org/2010/08/23/greek-public-library-captured/ ] which seems disturbing since I do hold him in high regard for his philanthropy. No mention of MSFT in the foundation news release. 2011-03-10T20:59:14 gmcharlt: back. Took a look at the actual submission in GSoC page (good test of my backup admin skills) - the line formatting is bizarre, otherwise looks good 2011-03-10T20:59:47 I wouldn't push the Seattle/Microsoft/Gates thing, although it is musing 2011-03-10T20:59:52 amusing 2011-03-10T21:00:05 dbs: heh I was just joking 2011-03-10T21:02:51 dbs gmcharlt: In the application should there be a mention that developers in/for libraries is a small, close knit group and somewhat hard to break into therefore a good opportunity to introduce possible comp sci students into the subject while providing the library/Evergreen land with new blood? 2011-03-10T21:04:07 moodaepo_: not sure about that; might make us sound too niche-y 2011-03-10T21:04:09 They can use the experience and apply it with Koha, Vufind, Blacklight, or heck any closed source ILS company. 2011-03-10T21:04:37 Will do just thinking out loud 2011-03-10T21:05:23 *** rickd_ has quit IRC 2011-03-10T21:05:59 dbs: I'm fixing up the line wrapping now 2011-03-10T21:07:03 gmcharlt: I tried fixing some line wrapping, where there was just one or two words on a line, and when I saved it it went crazy-short on every line. Sorry man. 2011-03-10T21:10:49 gmcharlt: Did you think it was a good idea to even mention that KCLS a big library went with Evergreen recently or was it not necessary to include? (which is fine) 2011-03-10T21:11:07 moodaepo_: it was a good idea and I added it 2011-03-10T21:11:46 Could we get/did we get any feedback from the students ESI/Laurentian funded? 2011-03-10T21:12:09 moodaepo_: good question. dbs? 2011-03-10T21:12:17 Oh ok I didn't see it in the paste from 19:44:55 2011-03-10T21:13:09 both students are gainfully employed 2011-03-10T21:13:27 you can ask kbeswick what he thought :) 2011-03-10T21:13:59 dbs: I've finished fixing the wrapping; how does it look now? 2011-03-10T21:15:16 gmcharlt: As for vouching for us, did Chris Cormack have a go with GSoF for Koha? I'm sure he'd vouch for you : ) 2011-03-10T21:15:28 GSoC! 2011-03-10T21:15:34 gmcharlt: disappearing mentors is the only section that seems to have weird wrapping now 2011-03-10T21:15:45 Google Summer of Fun! 2011-03-10T21:16:33 moodaepo_: rangi's company did a GSoC this summer (NZ time, that is), but not GSoC itself 2011-03-10T21:16:40 err, that makes no sense 2011-03-10T21:16:49 they did a GSoC-*like* thing, but on their own hook 2011-03-10T21:18:08 Aaaah right that's because he said the time diff was somewhat of a problem 2011-03-10T21:18:45 dbs: ok, got that fixed 2011-03-10T21:18:53 gmcharlt++ 2011-03-10T21:20:57 s/time/season/ 2011-03-10T21:27:59 *** moodaepo_ has quit IRC 2011-03-10T22:25:10 *** rsinger has joined #evergreen