Evergreen ILS Website

IRC log for #evergreen, 2015-10-28

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat

All times shown according to the server's local time.

Time Nick Message
00:47 bmills joined #evergreen
01:49 bmills joined #evergreen
01:51 bmills joined #evergreen
02:22 bmills joined #evergreen
03:22 bmills joined #evergreen
04:23 bmills joined #evergreen
05:25 bmills joined #evergreen
06:26 bmills joined #evergreen
07:33 bmills joined #evergreen
07:44 jboyer-isl joined #evergreen
07:44 collum joined #evergreen
07:45 collum_ joined #evergreen
07:53 rjackson_isl joined #evergreen
08:00 mrpeters joined #evergreen
08:20 ericar joined #evergreen
08:26 rlefaive joined #evergreen
08:31 RoganH joined #evergreen
08:34 bmills joined #evergreen
08:38 Shae joined #evergreen
08:40 mmorgan joined #evergreen
09:00 Dyrcona joined #evergreen
09:28 maryj joined #evergreen
09:33 yboston joined #evergreen
09:35 bmills joined #evergreen
09:41 Dyrcona kmlussier: I'm going to reload my dev database to test some stuff for Comcat. I need fresh data from production.
09:41 Dyrcona <zombie>Brains!!!!!</zombie>
09:41 Dyrcona This means my dev system will be off line for a couple of hours.
09:47 kmlussier Dyrcona: Thanks for the heads up!
09:47 Dyrcona I'll let you know when it is done. The same code will be there.
09:48 Dyrcona I just wants some holds and transits to play with, since I've already canceled them all in my dev system.
09:49 dbs They're not allowed to be called Comcat. Too confusing with Comcast and concat.ca!
09:52 Dyrcona heh
09:52 Dyrcona Commonwealth Catalog: Comcat, CommCat, or in the case of our aou.shortname for them COMMCAT
09:54 * Dyrcona is fixing a hold where the holds get in. ;)
09:55 Dyrcona heah... Should be "fixing a hole...."
09:55 Dyrcona <zombie>Brains!!!!!</zombie>
09:56 * jeff modifies a url so that it ends in /iam/home and is reminded of the book Sphere
10:10 jeff as in, I AM HERE.
10:18 mmorgan left #evergreen
10:20 jwoodard joined #evergreen
10:25 Dyrcona Oy! Aborting a transit resets a hold and resetting a hold aborts the transit....
10:25 mmorgan joined #evergreen
10:25 Dyrcona Of course the latter won't happen if it is already aborted, but still.
10:29 Dyrcona And, resetting a canceled hold seems pointless, but whatevs.
10:34 mmorgan Dyrcona: pointless, true, but can have annoying side effects: lp 1469287
10:34 pinesol_green Launchpad bug 1469287 in Evergreen "Retargeting a cancelled hold can cause problems with uncancelled captured holds" [Medium,Confirmed] https://launchpad.net/bugs/1469287
10:36 bmills joined #evergreen
10:39 Dyrcona mmorgan: I wonder if that affects what I am doing.
10:40 Dyrcona Yep. It probably will.
10:41 Dyrcona Perhaps, _reset_hold needs a line added at the top to bail if ($hold->cancel_time)?
10:42 Dyrcona On of the last things, _reset_hold does is retarget.
10:43 Dyrcona So, perhaps I should abort the transit before canceling the hold.
10:43 berick agreed we don't want to re-target a canceled hold
10:44 berick but we don't want to automatically abort a transit just because a hold is canceled
10:44 berick because the item is still in transit
10:44 berick even if the hold is canceled
10:45 * mmorgan agrees with berick
10:45 berick we used to abort the transit, but that caused confusion
10:45 mmorgan Those items then end up in Reshelving and nobody knows where they are.
10:45 berick exactly
10:46 Dyrcona berick: In my case, I know it should not actually be in transit, and if it is, too bad.
10:46 Dyrcona We get a lot of tickets about "stuck transits" and I don't need more.
10:46 Dyrcona Besides, this is for NCIPServer, not in Evergreen itself.
10:47 jeff anyone know if roomba vacuums are rated for yak hair?
10:48 Dyrcona jeff: Doubt it. You might want their anti-mine model.
10:48 * mmorgan doesn't know, but you can borrow one from one of our libraries ;-)
10:50 Dyrcona Hmm. Guess I could add a configuration option for abort_on_cancel to oils_ncip.xml, but that will make it harder on NOBLE and CW/MARS to update, 'cause they'll need to edit the config.
10:51 Dyrcona And would require some explanation in the non-existent documentation. ;)
10:53 Dyrcona It's all in a development branch. I haven't pushed anything to working/user/dyrcona/better_abstraction, yet. Not for the abort... The cancel fixes are there.
11:03 gmcharlt joined #evergreen
11:04 gmcharlt joined #evergreen
11:04 ericar joined #evergreen
11:06 Dyrcona I'll make it optional. That isn't very difficult.
11:08 mdriscoll joined #evergreen
11:09 JimT_ joined #evergreen
11:09 bshum gmcharlt++ # website stuff
11:11 JimT_ Does anyone have the browse index working for 710 tags which contain a subfield t?  I think I have fixed all the author browse indexes that weren't working except for this one.  Nothing I do seems to make a difference.
11:20 Dyrcona And, aborting transits when holds are canceled is now optional in NCIPServer.
11:20 Dyrcona Of course, I haven't tested it, yet. ;)
11:20 * Dyrcona waits on the database reload to finish.
11:21 Dyrcona jboyer-isl: So, I had a net removal of 156 lines of code the other day, and now a net addition of 121 lines since then.
11:21 Dyrcona jboyer-isl: In case you're keeping score. :)
11:22 bshum @marc 710
11:22 pinesol_green bshum: An added entry in which the entry element is a corporate name. (Repeatable) [a,b,c,d,e,f,g,h,k,l,m,n,o,p,r,s,t,u,x,3,4,5,6,8]
11:22 * bshum doesn't think he has anything like that in his DB
11:22 Dyrcona Well, I can't answer that, 'cause we're not really using the browse indexes.
11:23 bshum And 710t is title, not author?  According to https://www.loc.gov/marc/bibliographic/bd710.html anyways
11:23 Dyrcona bshum: You must have some records with 710s. Doesn't mean you've indexed them.
11:23 bshum Dyrcona: True... that would be the more accurate statement to make.
11:23 bshum Yes, I meant that I do not believe we have any browse index like that.
11:24 Dyrcona I might have added something like that to a keyword index if it wasn't there already, but we're mostly using stock indexes.
11:25 * bshum glances at kmlussier and her probable mountain of knowledge with non-stock indexes.
11:29 JimT_ I'm no expert and going with what the cataloger has told me should be there...710 t is a title index but supposedly the a,b,c,t,d,g,n will show in the author index as 'conference' listing.
11:31 JimT_ ...or maybe corporate?...gets mixed up in my head after a while.
11:34 JimT_ as far as indexing...I tweak the stylesheet...and then resave my test record.
11:35 dbwells joined #evergreen
11:38 JimT_ The 710 t index is part of the stock indexes...it just doesn't seem to work.   I'm not trying to do anything fancy...just, mostly, trying to make things work that are already defined but do not.
11:40 JimT_ Was hoping to find someone who had this working and who would share what they did to make it work.
11:45 kmlussier JimT_: I think there have been previous discussoins about the 710t in the irc logs. Bmagic and miker have had this discussion once or twice. I can't recall the gist of it, and, unfortunately, I'm tied up with other things right now.
11:46 JimT_ okay...I'll try to track it down in the logs.   Thanks.
11:46 kmlussier My recollection is that sometimes it is indexed and sometimes it isn't based on some other thing that's in the 700 field. Sorry, not very helpful at the moment. :)
11:46 kmlussier At some point, we should mine that info from the logs and put it in documentation somehwere. :)
11:46 JimT_ Having a direction to look is always helpful.
11:47 kmlussier Oh, actually, the previous discussions may have been in relation to keyword indexes. But I suspect the issue is the same whether you're looking at browse or keyword indexes.
11:48 ericar_ joined #evergreen
12:14 Christineb joined #evergreen
12:25 JimT_ left #evergreen
12:38 bmills joined #evergreen
12:48 Dyrcona kmlussier: My dev system is running again in case you want to look at anything.
12:49 mmorgan joined #evergreen
13:25 jihpringle joined #evergreen
13:42 jlitrell joined #evergreen
14:25 graced joined #evergreen
14:29 * bshum hates billings
14:30 bshum I found 479 cases where the entry in money.materialized_billable_xact_summary where total_paid < total_owed AND xact_finish IS NOT NULL.
14:30 bshum Those seem to indicate that the xact_finish was set too early on certain billings and there's still meant to be money owed.
14:30 jeff time for forgiveness.
14:31 bshum Well, the way it seems to manifest in the patron record is that the patrons show up as owing nothing, cause it doesn't add to the total amount owed, due to the xact_finish being set
14:31 bshum But yeah, weird stuff transpired.
14:31 bshum SELECT COUNT(*) FROM money.materialized_billable_xact_summary WHERE total_paid < total_owed AND xact_finish IS NOT NULL;
14:32 bshum I'm sure there's other logical reasons for this to happen
14:32 bshum Or maybe it was a phase of production where a bug happened
14:33 berick what's the time frame?
14:33 bshum berick: Hmm, good question
14:33 tsbere bshum: My lookup is slightly different. SELECT COUNT(*) FROM money.materialized_billable_xact_summary WHERE balance_owed > 0.0 AND xact_finish IS NOT NULL;
14:34 * tsbere isn't sure he wants to investigate the != version of his query with over 2800 returns, the 1 from > though may be worth checking
14:34 bshum tsbere: Good call, I get the same result.
14:34 bshum berick: The one I was looking at happened during the summer.
14:34 bshum THey kind of look like theyr'e from all different dates
14:35 jeff the fact that ansible uses cowsay by default if present is amusing to me.
14:35 bshum xact_start for some range from 2013 through present
14:35 bshum So, meh
14:35 bshum These are weird cases
14:35 mmorgan bshum: lp 1484989 ?
14:35 pinesol_green Launchpad bug 1484989 in Evergreen 2.8 "Fines are not calculating until after circulation is closed" [Medium,Fix released] https://launchpad.net/bugs/1484989
14:35 mmorgan We were affected by this after going to 2.8
14:36 berick bshum: hm, ok, so it's a persistent issue.  not something that happaned way-back-when
14:36 tsbere bshum: So, I have a whopping one with a positive balance_owed, for a 0.50 balance. But negative balance_owed.....
14:36 jeffdavis bshum: don't you have a nonstandard circ.lost.xact_open_on_zero setting?
14:37 jeffdavis Yours was the opposite to what is in master or something.
14:37 bshum jeffdavis: I think we do, but because the setting is named wrong, the effect is that having ours set is effectively like not having it set at all.
14:37 bshum Which is good, cause I don't want it to leave xacts open on zero anyways.
14:38 bshum mmorgan: Ah yeah, I remember that bug.  It's possible that some of these more recent ones are a result of that, but those older ones shouldn't be related.
14:39 jeff keep in mind that there are a few dates to look at, and which one you care about may depend on which bug you're looking at.
14:39 bshum Right.
14:39 jeff xact_finish vs the most recent billing_ts, etc.
14:40 * jeff watches ansible go
14:41 csharp any reason we'd need a 5 minute delay for emailing a receipt for an online payment?
14:41 csharp (that's the default in the a/t setup for money.payment_receipt.email)
14:41 tsbere csharp: To hopefully catch multiple payments into a single email?
14:41 bshum csharp: I imagine any delay is just to group ... yeah, what tsbere says
14:41 csharp ah
14:41 berick i imagine 30 seconds would cover it
14:41 csharp that makes sense
14:42 csharp berick: yeah, I agree
14:42 berick i mean, unless you expect them to make a series of separate CC payments
14:42 berick but that would be unusual
14:42 berick or would it?
14:43 jeff it's grouping on usr, not something like email, right? the most common "several payments in a row" we see is families paying each account.
14:43 jeff but then again, that's skewed by the fact that we don't permit partial payment.
14:44 jeff but we also don't permit payment on open circs, so if you paid everything, then renewed some open circs, you could pay again.
14:44 csharp jeff: we're about to have the "do we allow partial payments" discussion in PINES soon too, methinks
14:44 csharp right now we allow them, and it's leading to mucho confusion
14:44 jeff in terms of user expectations, i think they'll expect to get one receipt per credit card transaction, regardless of how many "payments" evergreen turns that into.
14:44 * bshum hates partial payments
14:44 * berick also sees no problem w/ someone receiving multiple emails of this type
14:45 berick pretty sure most sites would send one email per CC transaction, which a short timeout would give you
15:07 tsbere A quick question: Is anyone currently around that would be interested in a captive portal system to protect wifi?
15:14 csharp tsbere: while I'm not personally particularly interested, it is a common request from our libraries, and several (if not many) are purchasing solutions
15:14 jeff nope. we actively killed all of ours in the interest of having a better patron experience.
15:14 csharp jeff++
15:15 tsbere We have at least one library that believes it has local businesses using the library wireless instead of paying for an internet connection. :(
15:15 csharp our libraries appear to be mainly interested in collecting usage stats
15:15 csharp and that allows them to sort cardholders vs. guests
15:15 jeff yeah, we do that without a captive portal.
15:16 jeff cardholders vs guests is something we're not doing right now.
15:16 tsbere We are also interested in the potential for "those actually intentionally doing things vs those that just happened to walk by with a device that knows the network"
15:17 jeff tsbere: does the library in question have concerns about local businesses checking out books for the use of the business?
15:18 jeff tsbere: are you interested enough that you care to make the experience of using the library wireless harder for every user, and add greater technical complexity to the process?
15:18 jeff I guess the answer is "yes" since we're having the conversation, but I like to ask. :-)
15:18 tsbere jeff: I think part of the concern was when they noticed "The library is CLOSED, staff are on a different network, nobody is hanging around outside, and there is no bandwidth available on the patron side"
15:18 miker tsbere: invest in copper mesh...
15:19 jeff I've had this kind of conversation so often with so many people I've lost track of who -- so forgive me if I've talked about this with you before. :-)
15:19 tsbere We also have a couple of libraries that are visited by people from out of state who have no card, no interest in getting a card, and want to do nothing but use the wifi.
15:20 jeff Cool. Does the library kick people out for reading books in the library without being an active cardholder? :-)
15:20 tsbere Depends. Are said people being highly disruptive while doing so?
15:20 jeff If the patrons brought their own network connectivity would the library still feel the same about them?
15:21 jeff If the people are being disruptive, have a policy that lets you deal with them being disruptive, and make use of it.
15:21 miker csharp: re your short timeout on CC receipts, as a practical matter, the maximal minimum delay will be 1 minute -- the resolution of cron's timer. but it could (would often) be shorter. a 30s delay could be up to 1.5mins
15:21 csharp miker: yeah - that's what I'm settling on - 30s delay plus a cron job that runs every minute
15:22 tsbere jeff: Let me put it a different way: For whatever reason they feel they can't ask the people to leave (or perhaps they aren't even physically in the library) but they are preventing others from being able to function on the wifi.
15:22 jeff tsbere: of course, all of this is in theory and in the abstract. sometimes you have to be practical, also. it just annoys me when i see libraries spending time and money making their services more difficult to use.
15:23 tsbere jeff: I was asked to find a way to limit the wifi to card holders only. Captive portal is the best solution I have come up with.
15:24 jeff if you venture into requiring a card to use the wifi, at least consider giving added benefit to the patron at the same time, by adding meaningful security to the network :-)
15:24 jeff (again, for reasons of practicality i know that isn't always possible -- there's only so much you have in terms of resources, etc)
15:25 csharp I would consider using QoS to create a guest network of throttled bandwidth for non-cardholders
15:25 csharp (if given that particular request)
15:25 jeff still not sure i get the cardholders distinction. do you card people before allowing them to plug in a laptop to power?
15:25 csharp jeff: *I* don't care, but our libraries *really* care (in many cases)
15:26 tsbere Some of our libraries don't have places they let people plug in laptops
15:26 csharp that's probably true of our libraries too - most of my knowledge on this point is purely anecdotal
15:26 jeff an oddly discordant philosophy for libraries, but i guess if i'm honest with myself it's actually a common philosophy. :P
15:27 csharp jeff: yeah - I agree with the discordance
15:27 csharp jeff: meaning - I agree with you that it is discordant and they are blind to that perspective ;-)
15:28 jeffdavis jeff: better yet, "do you card people before letting them use books in the library?"
15:28 * csharp wants to live in a utopia of freely available wifi and endless access to information
15:29 csharp for instance, I cringe at this kind of thing: https://www.youtube.com/red
15:29 jlitrell I'd think if power were as limited as bandwidth can be, they'd limit access more.  :)  Got a 150/150 connection here, and it's fairly trivial to max it out.
15:29 tsbere jeff: MVLC allows out-of-state individuals to, for a fee, get a MVLC library card. Many out-of-state individuals are showing up at some libraries and using the wifi to an extreme.
15:29 jeffdavis we've had libraries where the staff client becomes unusable at a certain time of day because a young person was coming into the library every day at that time and watches instructional Youtube videos.
15:29 tsbere jeff: Thus, the idea is likely "get them to actually pay for their constant use of library resources or go find somewhere else to get their free internet from"
15:30 jeff avoid the increased support and maintenance burden of a captive portal, and invest some of those resources in having reasonably secure wifi that is architected in such a fashion as to allow full utilization, but throttles down to something not-incredibly-slow when there's contention/competition. :-)
15:31 jeff tsbere: do you think the issue would be different if they were "showing up at some libraries" and using their own internet connection, not making use of the library wifi?
15:32 kmlussier jcamins: Have you decided if you'll be making it to the hack-a-way?
15:32 jeff also, i do wonder what "using the wifi to an extreme" means. are you talking about disproportionately high bandwidth utilization, or something else?
15:33 jcamins kmlussier: I have, and I will not be able to make it. :(
15:33 kmlussier jcamins: Mint chips will be arriving just in time for me to make a bunch of cookies before everyone meets up next wee.
15:33 jcamins Oh, man! That sounds delicious.
15:33 tsbere jeff: I have one claim of someone showing up in a van every day, parking next to the library, and using enough bandwidth to cause issues for other patrons.
15:34 tsbere jeff: I also have several claims of out-of-state individuals basically treating the library as their place to run the "internet needed" tasks of their businesses every day, to the point of claiming they are losing money when the internet is down.
15:34 kmlussier jcamins: That's too bad. Maybe next time.
15:34 berick kmlussier++
15:34 jeff tsbere: How will the captive portal help with that issue?
15:35 bshum kmlussier: Mint chips, eh?  /smiles
15:35 bshum kmlussier++
15:35 tsbere jeff: As far as staff can tell none of these people have in-state cards in the first place.
15:35 kmlussier bshum: I'm thinking choc. mint cookies, the saltine things I made for the doc hackfest, and some thumbprint cookies you haven't seen yet.
15:36 tsbere jeff: I know one of the "running their business" people found the wifi down (the AP had taken a hit during a storm), was told they could use the in-library computers, but didn't have a card to plug into the reservation system
15:36 notkyle joined #evergreen
15:36 kmlussier I'm not quite sure when I'll have time to make them, with pumpkin carving and a fencing tournament on Sunday. But I'll make sure they're done by Wednesday.
15:36 notkyle left #evergreen
15:37 jeff tsbere: What would the solution be in these two examples if the individuals were in-state cardholders?
15:38 jeff tsbere: and would it be different if they were out-of-state "paid the fee" cardholders?
15:39 tsbere jeff: Excessive bandwidth people are a problem in general that they want a solution to as well, but don't want to spend a lot of money on.
15:39 jcamins We're going to travel around Thanksgiving to avoid A) the fact that I've got scads to do at work, and B) losing out on the opportunity for an 8-day vacation that uses only 4 days of vacation.
15:39 tsbere jeff: Otherwise, "pay state taxes" or "paid the fee" make them otherwise fine with the use of the network.
15:40 tsbere (in most cases)
16:33 ghostin joined #evergreen
16:36 ghostin left #evergreen
16:42 csharp tsbere: re low-cost solutions: buy several of http://goo.gl/AMXptq and use DD-WRT (https://www.dd-wrt.com/site/)
16:45 jeff or keep the existing access points and stick pfsense on any supported hardware.
16:45 jeff etc.
16:45 jeff (RIP m0n0wall)
16:50 kmlussier Stompro: Are you around?
16:51 Stompro kmlussier, yes, what is up?
16:52 kmlussier Stompro: I'll shoot you a pm
17:11 mdriscoll left #evergreen
17:31 mmorgan left #evergreen
17:33 jwoodard "instructional Youtube videos"? Such a thing exists?
17:54 jeffdavis jwoodard: Sitka support folks have made a bunch...
18:31 jihpringle jwoodard: https://www.youtube.com/channel/U​CiMYBCQG4QJVT-B3Ruk0Ncg/playlists
19:12 remingtron joined #evergreen
19:54 jihpringle joined #evergreen
21:44 gmcharlt courtesy of the Let's Encrypt beta, https://evergreen-ils.org now has a free SSL cert
22:05 csharp gmcharlt++

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat