Open Source Integrated Library System

Evergreen on IRC

#evergreen Logs for Wednesday, April 21st, 2010

< Tuesday, April 20th, 2010Raw Log FileThursday, April 22nd, 2010 >
#TimeNickMessage
#01:06:36pmplett has quit IRC
#07:13:34ebyis hashtag o pref evg10
#07:21:38bott-otreby: yes ...eg10 was already taken for twitter :(
#07:31:29ebyjust set the http://www.twapperkeeper.com/hashtag/evg10
#07:32:33bott-otr has quit IRC
#07:38:40bott-drd has joined #evergreen
#07:46:27jeff yawns
#07:47:08AmberJ has joined #evergreen
#07:48:17bott-drdJeff: get it out of your system. Long day ahead
#07:48:59jeffyep. heading down to lobby/reg now.
#07:49:57AmberJ has quit IRC
#07:50:02jeffbott-drd: any idea if it's wired or wireless in the Pearl room?
#07:51:42bott-drdLess
#07:56:47bott-drdUgh. Traffic.
#08:02:19bott-drd has quit IRC
#08:22:34atheos_ has joined #evergreen
#08:26:27atheos_ has quit IRC
#08:33:09atheos_ has joined #evergreen
#08:36:46mrpeters-isl has joined #evergreen
#08:39:01pseubodo1 has quit IRC
#08:53:50wjrwow.. wireless is horrid
#09:00:10wjrok, this AP is much better
#09:00:16wjrs/AP/SSID/
#09:02:05atheos_AV15 is working great for me
#09:02:40mrpeters-islhey all, anyone know of a way i can queue up some "fake" courtesy email prenotices for myself to do some testing?
#09:03:17wjryup AV15 is the bees knees
#09:09:46bott-otr has joined #evergreen
#09:13:46brendan_bywater has joined #evergreen
#09:16:50ebythis place = maze
#09:18:12wjryeah it's pretty crazy
#09:18:30wjrit took us a solid half hour last night just to find registration ;)
#09:18:43ebyi found the aging conference 3 times
#09:18:49wjrhaha
#09:18:51ebyneed more eg signs i think
#09:19:27atheos_here's what I used to figure this place out: http://www.amwaygrand.com/pdf/AGP_FloorPlans_2008.pdf
#09:21:10wjrhandy!
#09:22:46ebyi just followed the nerd looking people
#09:32:36Dmagick_ has quit IRC
#09:45:46atzof interest in hotel door lock security... http://www.youtube.com/watch?v=7INIRLe7x0Y
#09:51:55bshum has joined #evergreen
#10:00:40jeffnerds-looking-people++
#10:01:15jefftypos--
#10:02:17ebythe_coffee_here+-
#10:02:31wjrthe_starbucks_at_the_other_end_of_the_building++
#10:46:35ebythings_blowing_up_at_work--
#10:47:13wjr nods
#10:48:11wjrwith jeff and I here, we have only 1 tech left back home, and he only works afternoon/evening, so being a non-coder here in the hackfest, I get to take care of the things that can't be postponed until afternoon ;)
#10:54:39jamesrf has joined #evergreen
#10:55:26ebymost likely my relatives harassing staff up there about tech
#10:55:31jamesrf has quit IRC
#10:55:36eby is happy that tadl supports them instead of him
#10:56:06wjrhehe we are happy to do it! ;-)
#11:13:01atheos_anyone seen this error before? OILS: Item config script [ARRAY(0x24c58b0) : circ/circ_item_config.js] failed to run: Error: ReferenceError: js2JSON is not defined at line 0: (null)
#11:19:48Dyrcona has joined #evergreen
#11:23:32Dmagick has joined #evergreen
#11:23:48dkyle-mac has joined #evergreen
#11:24:21lisppaste3Dyrcona pasted "Ingest Error in Trunk" at http://paste.lisp.org/display/98088
#11:24:47DyrconaSo, I'm sitting at the hack fest working on my bib ingest problems with trunk.
#11:25:23DyrconaI was getting errors about bad 785 tags, so I split all of my bibs with 785 tags into their file and loaded them 1 at a time into the trunk database.
#11:26:03DyrconaNow, instead of getting a message about 785 tags missing a subfield, which none of my 785 tags actually do. I have verified that all of the 785 tags have at least 1 subfield.
#11:26:18DyrconaI get the message that I posted at lisp paste.
#11:26:48DyrconaThe question is, has anyone seen this when importing records into trunk, and what did you do to resolve it?
#11:27:58DyrconaMaybe, I should just shout, "Hey! Anybody seen this?" :)
#11:28:05atheos_disregard my question. I don't need any help, I just need to be in proximity of Evergreen developers in order to fix my issues.
#11:28:35bshumDyrcona: That sounds like it would be fun ;)
#11:29:44Dyrconaatheos_: missed your question.
#11:30:08atheos_well, I didn't fix it after all - crap - I'll ask it again on your behalf :)
#11:30:16atheos_anyone seen this error before? OILS: Item config script [ARRAY(0x24c58b0) : circ/circ_item_config.js] failed to run: Error: ReferenceError: js2JSON is not defined at line 0: (null)
#11:32:46DyrconaHmm. No, I haven't seen that. You're getting it when you run the circulation scripts.
#11:32:50Dyrcona?
#11:33:31bott-otr has quit IRC
#11:51:40DyrconaEureka!!!!!
#11:51:42atheos_Dyrcona, getting that error when checking out via SIP. media type, magnetic media flag is not being passed because it's choking on circ_item_config.js
#11:52:17bshum has quit IRC
#11:52:21brendan_bywater has quit IRC
#11:53:27Dyrcona has left #evergreen
#11:57:07tsbere has joined #evergreen
#12:04:42BobW has joined #evergreen
#12:05:02atheos_ has quit IRC
#12:47:50bott-drd has joined #evergreen
#12:49:18natschil has joined #evergreen
#12:52:45bott-drd has quit IRC
#12:53:08bott-drd has joined #evergreen
#13:07:39atheos_ has joined #evergreen
#13:13:10bott-otr has joined #evergreen
#13:21:02Dyrcona has joined #evergreen
#13:25:06Dyrcona_ has joined #evergreen
#13:28:16Dyrcona has quit IRC
#13:28:16Dyrcona_ is now known as Dyrcona
#13:30:21brendan_bywater has joined #evergreen
#13:30:21moodaepo_amway has joined #evergreen
#13:33:09mck9 has joined #evergreen
#13:40:32bshum has joined #evergreen
#13:42:59natschil has quit IRC
#13:47:42BobW has quit IRC
#13:49:51jamesrf has joined #evergreen
#13:58:04ejk has joined #evergreen
#14:05:13jamesrf has quit IRC
#14:22:52pmplett has joined #evergreen
#14:25:31jamesrf has joined #evergreen
#14:27:36bshum has quit IRC
#14:28:37brendan_bywater has quit IRC
#14:40:16jamesrf has quit IRC
#15:01:34brendan_bywater has joined #evergreen
#15:09:42csharp_GPLS has joined #evergreen
#15:17:55pmplett is now known as pmpafk
#15:23:32atzbummer... "use of inherited AUTOLOAD is deprecated"
#15:30:21wjrI hear a lot of people saying "coffee" in my vicinity.
#15:40:11bott-drd has quit IRC
#15:43:39brendan_bywater has quit IRC
#15:44:06brendan_bywater has joined #evergreen
#15:58:03csharp_GPLS has quit IRC
#15:59:32moodaepo_amway has quit IRC
#15:59:34brendan_bywater has joined #evergreen
#15:59:55mck9 has left #evergreen
#16:00:51bott-otr has quit IRC
#16:07:34dkyle-mac has quit IRC
#16:09:43jamesrf has joined #evergreen
#16:14:28Dyrcona has left #evergreen
#16:17:00jamesrf has quit IRC
#16:18:11jamesrf has joined #evergreen
#16:18:19atheos_ has quit IRC
#16:35:55jamesrf has quit IRC
#16:39:54pmpafk is now known as pmplett
#17:20:16atheos_ has joined #evergreen
#17:25:47tsbere has quit IRC
#17:41:17brendan_bywater has quit IRC
#18:06:29atheos_ has quit IRC
#18:27:03jamesrf has joined #evergreen
#19:00:14jamesrf has quit IRC
#19:55:37eby drinks some bourbon and hopes others are too
#19:56:00jenny has joined #evergreen
#21:09:22bott-drd has joined #evergreen
#21:12:31bott-drd has quit IRC
#21:45:40jenny has left #evergreen
#22:19:27bott-otr has joined #evergreen
#23:05:37jeff7~/me yawns
#23:05:55jeffoh dear.
#23:06:00jeff yawns
#23:06:01jeffthat's better.
#23:06:25gmcharltjeff: where did you find the cloning machine?
#23:06:26ebyall about the typos
#23:06:52ebyjeff: remind me to talk to you about holds more tomorrow
#23:06:59ebyafter i depressed atz
#23:07:58jeffgmcharlt: cloning machine? uhoh. do i have clones running around?
#23:08:14gmcharltmaybe, I just heard 7 of you yawn
#23:08:19jeffeby: willdo. i am testing my FIFO holds now.
#23:08:50bott-otr is sending a scathing email to staff regarding holds
#23:08:56jeffeby: then to test some "send age hold protected copies home" and a "send copies home to fill holds if they haven't been home in a while"
#23:09:27jeffturns out jamesrf implemented fifo holds almost exactly the way i did this morning. he tested it recently, was to put in production soon.
#23:09:39jeffso...
#23:09:46jeffindependent_confirmation++ i guess!
#23:10:05jeff is amused at the concept of "homesick" copies
#23:10:12ebyin my spare time i think i realized how to make this a clusterfuck
#23:10:25jeffand is that how you depressed atz?
#23:10:33ebyyes
#23:10:38jeffis atz here, by the way? i'd think so, but i don't know if i've encountered him yet.
#23:10:45ebyno he isn't
#23:10:49ebyhe did c4l
#23:10:53ebyalready grilled him
#23:10:53jeffah, got it.
#23:11:13ebymore on the combining diff holds models
#23:11:25ebythat i poorly discussed with you
#23:11:31jeffi'm intrigued and terrified at your suggestion earlier of doing some kind of external opensrf callback for local hold mangling.
#23:11:38bott-otrhow do you justify sending two copies from A to B and B to A to fill holds, when they could have filled holds without transit?
#23:11:41ebywell i explained that poorly
#23:11:48ebywasn't about external
#23:11:51jeffbut i really think the "hold priority" queue option has potential.
#23:12:12ebywas more about having more of a encased "hold" logic functions
#23:12:22ebythat you can mix and match
#23:12:30jeffessentially "you have X priority to burn on holds. hold goes to the patron who 'wants it more'"
#23:12:37bott-otr now runs off to bed
#23:13:00ebyso that a checkin event can cause a cascade in logic
#23:13:15ebyrather than expanding one function
#23:13:22jeffbott-otr: if that was a question for me, i'm not sure i can justify it, but then B checks in at A, we don't yet know that A will check in at B
#23:13:23eby realizes still not doing well
#23:14:05jeff'strict FIFO' holds are an experiment, IMO. one that we'll try, and hopefully pay close attention to in terms of trying to see if we're doing a lot of unnecessary transits.
#23:14:15ebyjeff: more of how to handle situation where you have multiple queue stiles of fifo, priority, etc in one system
#23:14:46ebywhere hold logic is similar to circ modifyer
#23:14:55jeffyeah, that gets tricky. is the user who didn't opt into the 'wants it more' style queue disadvantaged because all their holds are at priority zero?
#23:15:28ebywell priority based on local definition
#23:16:03ebybut if you support series this is going to come up i think
#23:16:12jeffdoes having three holds give you the option to have 1, 0, -1, and then you get a huge boost by having 40 holds and having your 'top' hold be 39 and your bottom be -39, and you just use it to jump to the front of the line on that prio 39 hold?
#23:16:31ebysomeone who is waiting for disc 2 versus fifo disc 2
#23:16:56jeffyeah. different hold priorities don't seem to mix well at first glance.
#23:17:03jeffi had this discussion with miker the other day.
#23:17:44ebywhich puts me towards hold "plugin/modules" with various weights
#23:17:56ebyfor local weighting
#23:18:43ebyjeff: but in general i see large overlap so coexistance is already there
#23:19:11ebyyou have the i want x queue which has paused holds
#23:19:23ebyfifo is just i want everything
#23:19:48ebyand then priorty is a subset of both
#23:20:20jeffif SYS1 wants to do holds as FIFO and SYS2 wants to do them as FIFO+PRIO, and SYS3 wants to do them as PROX (the current behavior)... as long as SYS1, SYS2, and SYS3 all limit holds to copies within their systems, making them all essentially independent, probably fine.
#23:21:00ebyi'm obviously biased towards system
#23:21:03jeffmixing two methods at the same level is what seems tricky to me, and i'm still not sure if i need to tweak any FIFO into the hold targeter itself. I don't *think* so, but I haven't thought it through yet.
#23:21:19jeffSomething like the priority stuff sounds like it would need to be in the targeter also.
#23:22:01jeffokay, i see priority as being a subset of active/suspended holds.
#23:22:26jeffbut the latter is more of an "i don't want this right now" as opposed to "i don't want this as much as i want this other thing"
#23:22:33jeffthen again, maybe it's not that different.
#23:23:19jeffbut an "inactive" or frozen hold is pretty absolute. it will never have anything captured... as opposed to the "i don't want this as much" would still be captured as long as nobody wanted it "more"
#23:23:46jeffit starts to get into enough complexity that i can't find myself explaining it simply to staff/patrons.
#23:24:23ebywell similar here
#23:24:38jeffbut i do want to put more thought into the 'give me this series, in order' and "don't give me more than X at a time of these"
#23:24:46ebythere's no way to give good info on say a hold page
#23:24:58ebybut 1 of 1 also doesn't give much
#23:25:10ebyyou can't say when '1' is coming back
#23:25:51ebyjeff: but what i was trying to ask you was whether there was a good opensrf way to build "module" type things
#23:26:02ebywhere you can have multiple things listen for an action
#23:26:06ebyand cascade
#23:26:36jeffwhich does strike me as more of the drupal idea of a 'hook'
#23:27:13jeffand no, i'm not aware. there was a module/versioning proposal a while back, but i don't remember if it addressed that.
#23:27:16ebyso tadl could say fifo is more important than x of x
#23:27:26ebybut we could say x of x is higher
#23:27:37ebybut still allow us to develop the fifo together
#23:27:43ebyas a unit
#23:27:54ebyand not duplicate full workflows
#23:28:42ebyif that makes sense
#23:28:48jeffcreating different hold capture (or hold opporunistic capture) logic can be done in mainline code after experimentation and an org unit setting can be used to select which to use.
#23:29:50jeffit works best if you say SYS1 holds have a hard boundary such that they only consider SYS1 copies -- and SYS2 has a similar where they only consider SYS2 copies... and then SYS1 and SYS2 can have different capture logic.
#23:30:33ebywith the explanation of opportunistic vs targeted holds it seemed like multiple processes don't work well currently but if you could choose priority then maybe they could
#23:30:48ebyi think the easiest would be to choose one
#23:30:58jeffSYS1 and SYS2 allow holds to target copies at each other, then having two different capture logic results in what miker_ was warning about earlier... spooky action at a distance and 'unfair' capture practices.
#23:30:59ebynot yet convinced it would be the best
#23:31:35ebysince you already have things like the holds that are skipped for vacation , etc
#23:32:01ebyso there is already things skipping over people
#23:32:13ebyand really you are changing who is being skipped
#23:32:29jeffsure, but right now that's just active/inactive holds (frozen/thawed to use the old terms)
#23:32:46ebyyeah
#23:33:00jeffthere's no current 'this patron is on vacation', and i think if there ever grew something like that, it would be implemented as a "suspend all these holds"
#23:33:08ebyyeah
#23:33:11jeff(and would just be a convenience UI)
#23:33:18bott-otr hasn't read the whole scroll, but...
#23:33:33ebyand a this person is not ready could also just be "suspend all these"
#23:33:40bott-otrwhat about the possibility of a wildcard pickup location? ...obviously for small systems
#23:33:42jeffbecause inspecting each hold user for a 'on vacation' bit would probably be too much of a pain.
#23:34:00bott-otrwhere a patron would choose to get the next available, and travel to get it
#23:34:01jeffbott-otr: "your item is ready for pickup, but we're not going to transit it for you"?
#23:34:05ebywas thinking along the lines of logic that says "this person is ready"
#23:34:09bott-otrjeff: right
#23:34:22jeffbott-otr: dunno, but i'll keep it in mind as i poke around.
#23:35:00bott-otrpreserves efficiency, but reduces convenience
#23:35:00ebyso it can be reused
#23:35:19jeffcurrently if a patron tries to place a hold on an uncommon item that is age hold protected to not leave BR2 but the patron attempts a hold for pickup at BR1, it fails with the "one of these five things may have prevented this hold from being placed"
#23:35:28ebysomeone with disc 1 of 5 of series has 3-5 suspended
#23:35:38ebysomeone with a netflix style hold has many suspended
#23:35:52ebyand various other levels
#23:36:02jeffi would like to see it offer a "if you opt to pick this item up at BR2 you'll be able to place the hold" -- or something almost as helpful but not as specific, perhaps.
#23:36:03ebyit would be logic of deciding who was ready
#23:36:43ebywe luckily don't have those
#23:37:03ebywe allow holds as soon as entered in the system even if months from fullfilment
#23:37:22jeffi think giving patrons the option of wildcard pickup location would be nice, but for our district we're a bit too farspread. it's more effecient and more convenient for the courier to transit something from branch1 to member3 to fill a hold.
#23:37:58jeffeven though branch1 and branch2 might be close to each other, and both might be equally "on the way" for a patron on their way to work/school... that's not something to rely on.
#23:38:26jeff"please select the libraries that you would be willing to pick this item up at"... save this as my default...
#23:38:43jeffi want to try 'simple' fifo first and see how it goes.
#23:38:50ebydef
#23:38:56ebythat's what we have now
#23:39:09ebyi could see something like what i'm thinking helping with that
#23:39:22ebyyou could weight the types based on location as well
#23:39:59ebyand then you have the -10 fudgie factor
#23:40:03jeffi don't want to study data and say "hey, these two holds would have been more efficient/convenient if the system did FOO", if things were pretty darn convenient/efficient for the other 3000 holds that month.
#23:40:43ebywasn't thinking that level
#23:41:16eby@karma fudgies
#23:41:16pinesoleby: fudgies has neutral karma.
#23:41:23jeffwell, i'm considering spot checking things if we try this experiment, so i'm making a note for myself to not try and be perfect.
#23:43:15ebywell fifo is def top priority for us
#23:43:20ebyso we don't go backwards
#23:43:33ebythe rest is things we've just discussed going forward
#23:43:42eby@karma permafudge
#23:43:42pinesoleby: permafudge has neutral karma.
#23:45:02phasefxwho coined homesick copies?
#23:45:09jeffheh, me i think.
#23:45:21ebyanyone with iii floating collections
#23:45:30phasefxis that where the copies must transit back to their circ lib at some point?
#23:46:44jeffidea would be that i would put a time since last home attribute on the copy, and there might be a ou setting for the circ lib of the copy... which was consulted when checking in at a remote library... if now() - time_last_home > homesick_setting, AND if there were any holds for pickup at home, send it home.
#23:47:01jeffbut don't send it home just to send it home... because then it would be captured and transit right back out.
#23:47:12ebywith iii it is more that everyone is at the party and no one can find a home
#23:47:28jeffbut i DO want to verify and implement (because i don't think it's there) a "send age hold protected items home when they check in at a remote lib"
#23:47:41ebycarts_full_of_stuff--
#23:48:50ebyjeff: glen has plenty of campfire horror stories of floating collections if you ever need lessons
#23:48:58phasefxjeff: sounds cool to me. I was thinking a simpler "total transits" threshold that just sent items back home regardless of what was going on
#23:49:44jeffWELCOME TO THE EVERGREEN HOLD LOGIC SETUP WIZARD. ANSWER THE FOLLOWING MULTIPLE CHOICE QUESTIONS:
#23:49:52jeff1. Our courier system is:
#23:49:59jeffa. fast and efficient
#23:50:06jeffb. slow as molasses
#23:50:12jeffc. what courier?
#23:50:29jeff2. We would prefer that items:
#23:50:31phasefxd. expensive
#23:50:36jeffa. be checked out to patrons
#23:50:43ebyjeff: for our floats it depends on whether school is in session and which places have the easiest dropoffs for X mode of transportation
#23:51:02jeffb. sit on the shelf, because they're "ours"
#23:51:02jeff(etc, etc)
#23:51:30jeffeby: oh, we've already had the question of "can we have zero fines for students when they check out during school hours at library X"?
#23:51:45ebyyouth dvd return = where storytime is scheduled this month
#23:52:58ebywith floating dvd collection + iii = work
#23:53:31ebyjeff: you were messing with floating for awhile weren't you
#23:53:40ebyi need to look more into that for eg as well
#23:54:25jeffwe had floating collections in the past, we do not have them any longer.
#23:54:42jeffthat's the official version of reality.
#23:55:14jeffactual reality is that there are still some in the system, but call number is about the only thing that clues us to them being leftover rotating items.
#23:55:21ebyi'll bug you more tommorrow
#23:55:31ebyhopefully over beverages
#23:55:38jeffi'm not sure if it's a case of them not having attributes changed properly when they were made 'non-rotating', or what.
#23:55:41jeffindeed!
#23:55:42phasefxsome old threads re: floating collections, http://markmail.org/message/nsq2tb6m6qhxyk2e http://markmail.org/message/y7wmzrss3zbxgjj5
#23:56:40jeffalmost the opposite of bott's 'wildcard pickup_lib' idea would be 'homeless copies'... where they just go back on the shelf wherever they happen to check in.
#23:57:01ebythat would be iii
#23:57:10jeffalmost need a configurable 'room at the inn' / 'lot full' to keep things balanced, though.
#23:57:18ebyexcept if there is no room it would be endless carts
#23:57:24jeffhah!
#23:57:35ebywhere someone manually sends carts somewhere else
#23:57:44jeffwell, see... i anticipated that problem and fixed it with a cute name and no code!
#23:57:57phasefxwe have homeless style now as an option in trunk I believe
#23:58:21jeff patches 1.6.0.1 image with fifo holds logic
#23:58:41jeffhere's where i get to see if i'm doing my ou setting correctly.
#23:58:51jeffi think i'll need help making an upgrade script if this works.
#23:58:58eby looks forward to checkbox of whether he wants "carts full of shit" on homeless option
< Tuesday, April 20th, 2010Raw Log FileThursday, April 22nd, 2010 >