Open Source Integrated Library System

Evergreen on IRC

#evergreen Logs for Tuesday, May 1st, 2012

< Monday, April 30th, 2012Raw Log FileWednesday, May 2nd, 2012 >
#TimeNickMessage
#01:22:47moodaepo_indy has joined #evergreen
#01:39:57hopkinsju has joined #evergreen
#02:39:53denials pushes kmlussier's release note updates; then syncs up the dev docs & DIG docs for release notes & install, and cleans up a bunch o' stuff at the same time
#02:39:58denialsand then goes to sleep
#02:56:52kivilahtio1 has joined #evergreen
#02:56:52kivilahtio has quit IRC
#03:01:49denialsberick: a bit late for this year, but asciidoc's slidy backend does support notes on slides / handouts: http://groups.google.com/group/asciidoc/browse_thread/thread/3e7f1e4180e56694
#03:17:36Callender has quit IRC
#04:50:53hopkinsju has quit IRC
#04:51:33moodaepo_indy has quit IRC
#07:07:59timlaptop has joined #evergreen
#08:13:55akilsdonk has joined #evergreen
#08:17:19fortin has joined #evergreen
#08:19:27kmlussier_ has joined #evergreen
#08:34:43gdunbar has joined #evergreen
#08:37:02Dyrcona has joined #evergreen
#08:37:33AaronZ-PLS has joined #evergreen
#08:39:45berickthanks denials
#08:43:08hopkinsju has joined #evergreen
#08:48:16Callender has joined #evergreen
#08:50:49tsbere hates the first of the month
#08:51:59bradl remembers the old PINES unicorn system would crash every 1st of the month from monthly reporting
#08:54:38atheos loves the first of the month. I get to submit billing for my moonlighting projects
#08:55:15tsbereI get all the bounce messages from bad mailing list subscriptions on the first of the month.
#08:55:40tsbereBecause actually checking the email addresses they are subscribing to things might actually take effort
#08:55:42tsbere>_>
#08:58:53hopkinsju has quit IRC
#09:04:57denials@herald add dbwells Why so serials?
#09:04:57pinesol_greendenials: Error: I can't find dbwells in my user database. If you didn't give a user name, then I might not know what your user is, and you'll need to identify before this command might work.
#09:05:18denialsmeh
#09:06:59Meliss has joined #evergreen
#09:09:09jeffdenials++
#09:10:58dbwells:)
#09:13:43wlayton has joined #evergreen
#09:21:57gmcharltdenials: time to update the release notes HTML on the website?
#09:22:54denialsgmcharlt: we should automate the update of that, eh?
#09:23:10gmcharltdenials: indeed
#09:23:26gmcharlthmm, do I have access to the webserver...
#09:24:14denialsupdated the release notes HTML, will now update the README too.
#09:24:52gmcharltdenials++
#09:24:53denialsWhich we might want to rename the title of to "Building and installing Evergreen" so that it fits a little nicer into the Table of Contents...
#09:25:23denialsgmcharlt++ # memory & keeping us on track
#09:27:27denialswould be slightly better SEO too
#09:27:40denialsokay, README is updated on the website too
#09:28:54denials sighs at the number of silo-y lists already on http://open-ils.org/listserv.php
#09:29:43denialsI guess we can stop calling the evergreen & opensrf commit lists the "SVN commit list"s
#09:29:46denials:)
#09:30:16gmcharltAND DESTORY ANY SENSE OF HISTORY?!?!
#09:30:17gmcharlt;)
#09:30:29denials registers destorify.com
#09:30:36gmcharltheh
#09:30:49collum has joined #evergreen
#09:31:05gmcharltactually, that's a surprisingly nice typo, come to think of it
#09:31:19gmcharltde-story _should_ be a word
#09:33:00denials"The reason that it took so long for us to figure out why acquisitions matching during bib loading was slow was because the pieces had been destoried across the acquisitions list, sysadmin list, cataloging list, general list, and technical discussion list, and nobody could piece it together"
#09:33:49jeffyou left out the reports list.
#09:33:54jeff;-)
#09:34:25kmlussier_Correction: no acquisitions list. denials was there to talk us out of it. :)
#09:34:33denialsjeff: hah - right, so the underlying problem was a report running against master that was causing locking escalation? :)
#09:35:20gmcharltit is a tradeoff -- keeping everybody in one place versus having separate "places" for focused discussion
#09:35:32jeffthere are good/bad bits
#09:35:40denialskmlussier_: and thanks for listening to my pleas! It was an exaggeration for dramatic effect
#09:36:11gmcharltand regarding the sysadmin list in particular -- heard a few people say last night that they're scared of the developers^W^W^W^W not comfortable overwhelming a technical list with a bunch of operational concerns
#09:37:01denialsI thought that was why we had called it the "technical discussion list" on the listserv page, but the description is currently very code-oriented, I would concede
#09:37:04jeffi believe there may be benefit to having less separation between 'dev' and 'sysadmin'. that said, i'm making no arguments for/against new lists.
#09:37:57gmcharlt is in favor of anything that gets discussion happening among catalogers, sysadmins, etc. -- particularly if some feel constrained at present -- and between all the various subgruops, but doesn't claim to know any One True Ways to encourage that
#09:38:05jeffs/'dev' and 'sysadmin'/the 'dev' and 'sysadmin' communities/
#09:38:07denialsAFAICT, the dev list is currently mostly about install & sysadmin and less about dev stuff
#09:38:38jeffi found it interesting that "irc == devs only, and no sysadmins"
#09:39:09fortin has quit IRC
#09:40:38denials tries pointing at FUDforum's http://cvs.prohost.org/index.php/Mailing_List_Manager again for the forum-y web front end for mailing lists, should a sysadmin want to try setting it up as a test
#09:41:02jeffif more lists help people communicate with each other, i'm all for it. i would encourage any that can to participate in multiple lists and act as bridges where appropriate / useful.
#09:42:19jeffand if at some point some of the many lists merge with each other, great! ;-)
#09:43:28swills_ has joined #evergreen
#09:43:29denialsI did wonder whether it might make sense for people to generate mailing lists via data points. E.g. prefix subjects with "Acquisitions: blah blah" or "Sys Admin: blah blah". but maybe that's just the filtering/data-oriented geek in me
#09:43:45jeffmailman has support for topic-prefixes
#09:43:52jeffand users can say "i don't want to see cataloging stuff"
#09:44:07jeffbut that adds complexity to the act of posting / replying
#09:45:12denials will step away from the discussion and let the communication/web people deal with it as they best see fit
#09:45:13swills_ is now known as StephenGWills
#09:46:18collumdenials: Does the mailing list manager consolidate all the lists into a nice little format. Possibly just sorting by date?
#09:46:47denialscollum: http://fudforum.org is self-hosted, so you can play with it there
#09:46:56collumwill do
#09:47:26denialsbut quick answer is "yes, I think it does what people want from a forum"
#09:47:29tsbereI already have to scroll my folder list due to list overload. I don't know if I want to add a sysadmin list to that. :P
#09:48:17denials reserves the right to snap quietly if the bulk of the mail on the -general list becomes "You should ask that question on the <name> list!"
#09:48:45denialsOkay, well and truly done now!
#09:49:59denialsbtw, kmlussier++ # great addition to the release notes!
#09:50:10denialsand sfortin++ # likewise
#09:51:58kmlussier_Thanks! It was eye-opening to see how much new stuff is coming with this release.
#09:58:52denialstsbere: I still think gitstats is broken at http://git.evergreen-ils.org/gitstats/Evergreen/index.html; it looks like it's running the stats against an old checkout dir. either there's a missing "git fetch / git pull", or it's pointing at the wrong directory, or it's failing (maybe due to user perms)?
#09:58:59jeffkmlussier++
#09:59:47denialskmlussier_: btw, I made you the "author" of that particular commit, so you'll take your rightful place in the git stats once the stats get updated :)
#10:00:09tsberedenials: I made note to look at it after my vacation is over.
#10:00:52denialstsbere++ # go vacation! get out of channel! :)
#10:01:51tsberedenials: My host is still sleeping, so I am emptying my inbox of the first of the month insanity. >_>
#10:01:58tsbereand the channel is a welcome distraction from that insanity.
#10:03:36kmlussier_denials++ I feel like I should break out a bottle of champagne. :)
#10:05:36moodaepo_indy has joined #evergreen
#10:08:01wlaytonkmlussier_++
#10:08:14moodaepo_indy_ has joined #evergreen
#10:11:48moodaepo_indy has quit IRC
#10:15:23fortin has joined #evergreen
#10:23:00moodaepo_indy_ has quit IRC
#10:45:29StephenGWillsKathy, I knew that managing MassLNC would eventually get you committed, congrats ;-P
#10:45:54phasefxhehe
#10:46:47akilsdonk has quit IRC
#10:48:50bkuhn has quit IRC
#10:49:54kmlussier_StephenGWills: heh...you'll know where to find me in another few weeks.
#11:05:55AaronZ-PLS has quit IRC
#11:19:55moodaepo_indy has joined #evergreen
#11:24:14moodaepo_indy looks around
#11:32:07bshumI'm not fully awake yet, but I'm writing up my objections to the sys admin-only list as a -1 reply in a bit.
#11:33:24bshumI had thought of denials' points, but will read the scrollback to see if I missed anything else.
#11:33:48moodaepo_indy has quit IRC
#11:36:31moodaepo_indy has joined #evergreen
#11:41:23moodaepo_indy has quit IRC
#11:42:18gmcharltkmlussier_++
#11:45:30AaronZ-PLS has joined #evergreen
#11:49:48akilsdonk has joined #evergreen
#11:57:00bwicksall has joined #evergreen
#11:58:01denialssenator++ # nice response
#12:03:37senatorthanks denials
#12:03:44senatordbwells: i'm turning to rel_2_2 pullrequests, but let me know if you could use my attention on any of the 2.2-targeted serials bugfix work we were dealing with at the conference
#12:04:01senators/but/so/ # you know what i mean
#12:06:15denials wonders if there's any chance of a maven approach towards Java dependencies coming to an OpenSRF & Evergreen git repo - getting the deps in place for the OpenSRF 2.1.0 RC2 was annoying on Sunday, never quite finished that off to be able to test :(
#12:06:26jlamos has joined #evergreen
#12:06:39wlayton has quit IRC
#12:07:25denialskmlussier_: we need to get either your name, or MassLNC's name, or both, onto http://docs.evergreen-ils.org/2.2/_attributions.html
#12:10:39senator+1
#12:40:58akilsdonk has quit IRC
#12:43:24collum has quit IRC
#12:46:38_bott_ has quit IRC
#12:47:22_bott_ has joined #evergreen
#12:48:18csharp comes in late to the sysadmin list discussion
#12:49:31csharpI agree with the hesitation about starting another list for the same reasons mentioned here
#12:50:00csharpI *do* think there are local system admins that don't have many community-wide resources at their disposal
#12:50:29denials hopes it's possible to turn evergreen-governance-l into an auto-responder telling people that evergreen-governance-l was supposed to be dead around a year ago and pointing them at eg-oversight-board :)
#12:50:56csharpthe admins like me and bshum and mrpeters-isl in this channel and comfortable on the mailing lists weren't really who I had in mind when I mentioned the sys admin "silo" on Saturday
#12:51:13csharpdenials: I'll look into that
#12:51:18denialscsharp++
#12:51:43csharpI'd actually like to deactivate it altogether and just keep the archives available
#12:51:50akilsdonk has joined #evergreen
#12:52:50bshumcsharp: To be fair, two years ago, I was strongly advocating for a sys-admin mailing list, but was pursuaded to try using existing methods of communication in the interest of open community sharing. I stand by that now, hence my reply on the topic.
#12:53:10bshum(and/or I got lazy and things are working so don't fix what isn't really broken for me)
#12:53:20bshumBut I'm just one vote :)
#12:55:08csharpmy thought was maybe we set up an IRC meeting to guage interest in a regular group and try to keep traffic on the existing lists
#12:55:32csharpif there is interest, we could have regular IRC meetings ala devs or reports
#12:56:29wlayton has joined #evergreen
#12:56:39denialscsharp: good idea!
#12:57:18phasefxcsharp++ call yourselves wardens
#12:57:22csharpheh
#12:57:29phasefxrangers
#12:57:31csharp has to run to a meeting
#12:57:33denialsberick: I note that http://evergreen-ils.org/~denials/eg2012_tpac.html#(28) doesn't identify "Granular search indexes" as a TPAC gap :)
#12:58:47berickdenials: are you referring to the all-index advanced-level-2 search builder?
#12:59:09denialsberick: I am
#12:59:25denialsor at least drop-downy access to a granular index
#12:59:35berickright
#12:59:44berick has some LP tickets to create
#13:02:21denials was thinking back to phasefx's question about splitting up large batches of reingests and such
#13:02:55phasefx doesn't even remember that question. Need more vitamin b12
#13:02:56denialsThere might be some value to wrapping up tsbere's autosuggest ingest approach into a function, and then call it on progressively larger subsets
#13:03:40denialse.g. 1% and commit, then 5% and commit
#13:04:18senatorthat might have been my question
#13:04:20senatorand i agree
#13:04:21denialsthinking that getting that 1% in first would at least provide a quick payoff in terms of some functionality for autosuggest. heck, maybe 100 batches of 1% each
#13:04:51denials says this as the reingest on his test server is 10% of the way through after ~24 hours
#13:05:00denialslast time it died on a malformed MARC record :)
#13:05:09senatorergh
#13:05:33berick+1 to batches
#13:05:51bshumReminds me of the 1.6.1-2.0 upgrade when we ran the reingest SQL on the side in batches of 1000 bibs each.
#13:06:19senatorright yeah. not making admins have to start all over for things like that at upgrade time was my main thought
#13:06:33senatorhaving a subest of records workin early is a side benefit
#13:06:43senatorblah. lazy typing.
#13:08:05jeff"our artisan bibliographic records are re-ingested in small batches by functions that source their arguments locally from sustainable sources"
#13:08:17senator*mppphbbbt!*
#13:08:19senatorjeff++
#13:09:03phasefxwhere's the surgeon general's warning for MARC?
#13:34:55jeffI'd like to expose acp.copy in the item attribute editor. I believe up to this point it has only been used for Acq. Are any more familiar with Acq of the opinion that exposing that column would be Bad?
#13:35:35jeffI can see a permission check when updating it from the item attribute editor, but of course that might limit acq's ability to update the value...
#13:35:54jeffMusing in irc before opening a launchpad wishlist before starting work on it.
#13:35:58phasefxacp.copy?
#13:36:04jeffer.
#13:36:09jeffacp.cost :P
#13:36:12jeffsorry.
#13:43:24senatoronly thing i'd suggest is that good labels in the UI would be desirable to explain the difference between price and cost
#13:43:43senatorand since i imagine those labels are IDL based, I guess we're really talking about reviewing the labels /there/
#13:57:03jeffsure thing
#13:57:33jeffand with the new/upcoming "hide fields in item attribute editor", they could even be hidden.
#13:58:01jefffwiw, "price" == list price billed to patron when lost/etc, "cost" == amount of money library paid to acquire item
#14:05:06senatorthat's good for the logs. i usually forget which is which
#14:06:28bericksketching out a potential 2.3 release schedule. thinking about using months as natural boundaries. http://snipurl.com/23bhiwp
#14:06:37fortin has quit IRC
#14:10:06berickthat's assuming we want to get back into the march/sept schedule, of course, which we probably still need to vote on
#14:13:20phasefxseems like a sane sketch to me
#14:31:10senator is glad that was encrypted with 256-bit AES
#14:31:22bericksenator++
#14:31:43senatoractually though that is a neat service to be aware of
#14:35:07jeffselect null from auditor.zer0knowledge;
#14:43:22jeffpoor little sub-utility db server.
#14:59:10wlayton misses his zer0knowledge t-shirt
#15:04:26kmlussier_ has quit IRC
#15:25:36fortin has joined #evergreen
#15:27:47jeffexposing acp.cost would of course eventually help with my current situation, where i have seven copies with multiple values for "cost" :-)
#15:48:04dbwellsdenials: I finally got around to updating Library::CallNumber::LC to remove the "deprecated use of qw(...) as parentheses". Thanks for pointing it out. http://search.cpan.org/~dbwells/Library-CallNumber-LC-0.22/
#15:52:04timlaptop has quit IRC
#15:52:13moodaepo_indy has joined #evergreen
#15:54:49moodaepo_indy_ has joined #evergreen
#15:55:49jeffdbwells++
#15:56:30collum has joined #evergreen
#15:58:00moodaepo_indy has quit IRC
#15:59:13Meliss has quit IRC
#15:59:44dbwells_ has joined #evergreen
#16:03:09dbwells has quit IRC
#16:08:21akilsdonk has quit IRC
#16:12:08wlayton has quit IRC
#16:19:32collum has quit IRC
#17:01:12moodaepo_indy_ is now known as moodaepo_nb
#17:09:25AaronZ-PLS has quit IRC
#17:26:21Dyrcona has quit IRC
#18:32:41hopkinsju has joined #evergreen
#18:50:32dbwells__ has joined #evergreen
#18:56:40artunit has quit IRC
#18:56:40mcarlson has quit IRC
#18:56:40b_bonner_ has quit IRC
#18:56:40tsbere has quit IRC
#18:56:41jeff has quit IRC
#18:56:41dbwells_ has quit IRC
#18:56:41Callender has quit IRC
#18:56:42gmcharlt has quit IRC
#18:56:43kbeswick1 has quit IRC
#18:56:43moodaepo has quit IRC
#18:56:44gdunbar has quit IRC
#18:56:45bshum has quit IRC
#18:56:47kivilahtio1 has quit IRC
#18:56:47ldwhalen has quit IRC
#18:56:49berick has quit IRC
#18:56:49_bott_ has quit IRC
#18:56:51jeffdavis has quit IRC
#18:56:52mtate has quit IRC
#18:56:53dmagick_ has quit IRC
#18:56:54shadowspar has quit IRC
#19:00:3917SAA1UE2 has joined #evergreen
#19:00:39Callender_ has joined #evergreen
#19:00:39mcarlson has joined #evergreen
#19:00:39Callender has joined #evergreen
#19:00:39gdunbar has joined #evergreen
#19:00:39kivilahtio1 has joined #evergreen
#19:00:39ldwhalen has joined #evergreen
#19:00:39gmcharlt has joined #evergreen
#19:00:39tsbere has joined #evergreen
#19:00:39bshum has joined #evergreen
#19:00:39berick has joined #evergreen
#19:00:39moodaepo has joined #evergreen
#19:00:39jeff has joined #evergreen
#19:00:39artunit has joined #evergreen
#19:00:39b_bonner_ has joined #evergreen
#19:01:27_bott_ has joined #evergreen
#19:01:27jeffdavis has joined #evergreen
#19:02:00mtate has joined #evergreen
#19:02:00dmagick_ has joined #evergreen
#19:02:00shadowspar has joined #evergreen
#19:02:31Callender has quit IRC
#19:02:40Callender_ is now known as Callender
#19:08:42fortin has quit IRC
#19:16:49shadowspar has quit IRC
#19:17:16shadowspar has joined #evergreen
#19:25:13plux1 has joined #evergreen
#19:25:15_bott_1 has joined #evergreen
#19:26:17plux1 has quit IRC
#19:26:54jeffdavi1 has joined #evergreen
#19:32:39_bott_ has quit IRC
#19:32:40jeffdavis has quit IRC
#19:34:52dbwells_ has joined #evergreen
#19:42:31atheos_ has joined #evergreen
#19:42:31dbwells__ has quit IRC
#19:45:01senator_ has joined #evergreen
#19:50:57atheos has quit IRC
#19:50:57senator has quit IRC
#19:51:01atheos_ is now known as atheos
#20:20:48moodaepo_nb has quit IRC
#21:51:27hopkinsju has quit IRC
#22:07:38denials is bending OpenSRF Java build to his will. Somewhat.
#22:16:31jeff is preparing ac by rec id for tomorrow
#22:26:56denialsjeff++
#22:44:16denials1% of records browse-ingested makes for reasonable autosuggestion; http://laurentian-test.concat.ca/eg/opac/home?loc=105
#23:06:10jeffdavi1 is now known as jeffdavis
#23:10:58bshumdenials: Wowsers, that autosuggest is super slow :(
#23:22:54denialsbshum: well, to be fair i've moved on to ingesting more records; so it's in the middle of the 4th percent
#23:23:16bshumHeh, it started to speed up a bit after the initial stabs I did.
#23:23:18bshumBut true.
#23:23:25denialsactually doesn't seem to be too slow to me
#23:23:32denialsright
#23:23:39bshumI think the initial loading took a few moments longer than I expected.
#23:23:41bshumBut now it's fine.
#23:23:50bshumAnd pretty cool.
#23:23:53denials blames dojo :)
#23:23:54bshumI can't wait to turn on ours
#23:23:57denialssenator++
#23:24:06bshumSo that our patrons can tell us how crappy our bibs are :)
< Monday, April 30th, 2012Raw Log FileWednesday, May 2nd, 2012 >