| Time |
Nick |
Message |
| 03:48 |
|
eby__ joined #evergreen |
| 04:57 |
pinesol_green |
Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html> |
| 06:04 |
|
serflog joined #evergreen |
| 06:04 |
|
Topic for #evergreen is now Welcome to the #evergreen library system channel! | We are publicly logged. | Large pastes at http://paste.evergreen-ils.org |
| 06:28 |
|
dac joined #evergreen |
| 07:00 |
|
timlaptop joined #evergreen |
| 07:00 |
|
timhome joined #evergreen |
| 07:50 |
|
dac joined #evergreen |
| 08:07 |
|
akilsdonk joined #evergreen |
| 08:32 |
|
ericar joined #evergreen |
| 08:39 |
|
Dyrcona joined #evergreen |
| 08:45 |
|
Shae joined #evergreen |
| 08:46 |
|
mmorgan joined #evergreen |
| 08:49 |
|
kmlussier joined #evergreen |
| 08:54 |
|
tsbere joined #evergreen |
| 09:07 |
Dyrcona |
Kernel updates, oh my! |
| 09:07 |
|
montgoc1 joined #evergreen |
| 09:11 |
csharp |
@dessert add Banana Puddin' |
| 09:11 |
pinesol_green |
csharp: The operation succeeded. Dessert #15 added. |
| 09:12 |
|
Dyrcona joined #evergreen |
| 09:14 |
Dyrcona |
So, you upgrade your mail server from Ubuntu 12.04 to Ubuntu 14.04 to get a newer version of mailman to better handle the DMARC stuff that the big guys are now doing... |
| 09:14 |
Dyrcona |
and, your calendars and other in Horde break because Horde goes from version 3 to version 5. |
| 09:15 |
Dyrcona |
The default storage format for calendars, etc., in version 3 was deprecated in version 4 and removed in 5. |
| 09:15 |
Dyrcona |
Way to go Canonical! |
| 09:17 |
bshum |
Oops. |
| 09:17 |
jeff |
hah. one peril of a 5 year LTS cycle. |
| 09:18 |
csharp |
upgrades+- |
| 09:19 |
csharp |
upgrades-- # most of the time the benefits don't warrant a + ;-) |
| 09:25 |
Dyrcona |
Well, I can't really blame Canonical, but you'd think the package maintainers for Ubuntu would be aware of this and provide some kind of suitable upgrade path. |
| 09:25 |
Dyrcona |
But, then, you'd be wrong. |
| 09:26 |
Dyrcona |
csharp: That article you shared with me from slashdot last week was fairly accurate according to my experience. |
| 09:27 |
|
RoganH joined #evergreen |
| 09:27 |
csharp |
Dyrcona: I thought you'd agree with it |
| 09:28 |
|
jwoodard joined #evergreen |
| 09:28 |
* csharp |
wonders if it's worth trying to backport the new marc_export script to 2.5.1 |
| 09:29 |
csharp |
I'm doing a full DB + holdings export, and that takes us about 48 hours with the "old" script |
| 09:31 |
bshum |
csharp: It's not scientific, but doing a parallel export test when that new marc_export was merged got me my whole DB in less than two hours what used to take 26 hours consecutive extracting. |
| 09:32 |
bshum |
(i.e. it might be worth it, if you're not planning to upgrade again for awhile) |
| 09:34 |
|
yboston joined #evergreen |
| 09:39 |
* Dyrcona |
thinks the new marc_export will just work on 2.5.1. |
| 09:39 |
jeff |
i don't think there's much to backport, yeah. |
| 09:40 |
Dyrcona |
It was basically developed on a 2.5 system. |
| 09:40 |
* jeff |
kicks off export-marc-from-reporter.pl against our reporting db to see what quirky bibs it finds |
| 09:41 |
jeff |
you've all probably seen some interesting data exports when migrating... on Friday, I'm pretty sure I received a CSV file containing MARC-8 encoded characters. That was a new one for me. |
| 09:42 |
bshum |
csharp: For kicks: http://git.evergreen-ils.org/?p=Evergreen.git&a=search&h=HEAD&st=commit&s=LP1223903 |
| 09:43 |
bshum |
That'll get you all the commits if you wanted to poke at pulling new marc_export back |
| 09:53 |
csharp |
bshum++ Dyrcona++ jeff++ # thanks |
| 09:55 |
jeff |
hrm. we still appear to have records that lack an 001. |
| 09:55 |
* jeff |
looks sideways at 245 14 $a Willie Colon the player $h [sound recording] / $c Willie Colon |
| 10:00 |
jeff |
10k sample of bibs, most prolific 100 tags are: |
| 10:00 |
jeff |
43 100 1 $a Various artists. |
| 10:00 |
jeff |
25 100 1 $a Various Artists. |
| 10:10 |
|
gsams joined #evergreen |
| 10:26 |
|
RoganH left #evergreen |
| 10:32 |
|
kmlussier1 joined #evergreen |
| 10:34 |
|
gsams joined #evergreen |
| 10:42 |
|
RoganH joined #evergreen |
| 10:44 |
|
gsams joined #evergreen |
| 10:44 |
RoganH |
Out of curiosity, has anyone else already built a 2.5 mac staff client? |
| 10:45 |
bshum |
RoganH: Not one made out of the official tarballs. We use customized stamping. |
| 10:46 |
RoganH |
bshum: I could just do a search and replace for a standard stamp then, or do you do something more than that? |
| 10:46 |
bshum |
I put my Mac back in the closet along with the rest of the random things I own. |
| 10:46 |
RoganH |
lol |
| 10:47 |
RoganH |
No problem, I'll build one, I just didn't want to duplicate effort. |
| 10:47 |
bshum |
RoganH: Our builds of Evergreen are customized, we modify our offline registration forms, and a couple other things. |
| 10:47 |
bshum |
Not everyone's from GA after all. |
| 10:47 |
bshum |
:D |
| 10:47 |
jeff |
we have a 2.5.1 client, but I'm not the one who built it. |
| 10:47 |
jeff |
We were happy to find that with 2.5, the "native" mac app was not outperformed by the winebottled version. |
| 10:48 |
jeff |
We had been running a winebottled version for 2.2 due to performance. |
| 10:48 |
bshum |
jeff: I would think that's been the case since newer xul 14 was introduced. |
| 10:48 |
RoganH |
jeff: I hadn't really bothered since this new laptop of mine runs virtual box at insane speeds but we have a few on older macs that could use the native app. |
| 10:48 |
jeff |
bshum: likely |
| 10:49 |
RoganH |
jeff: could you send it to me, if it'll run with just a stamp replacement it'll work for my users |
| 10:49 |
jeff |
RoganH: similar. i bounce between native mac app and citrix windows app. |
| 10:58 |
|
gsams joined #evergreen |
| 11:19 |
csharp |
RoganH: check with mrpeters |
| 11:20 |
csharp |
he built our 2.5 clients |
| 11:20 |
csharp |
(including Mac) |
| 11:20 |
csharp |
oh, and, incidentally |
| 11:20 |
csharp |
mac-- |
| 11:20 |
jeff |
why the hate? |
| 11:21 |
RoganH |
csharp: I'll drop him a note. I can build it but it's a boring process I'd rather not do if I don't have to. |
| 11:21 |
RoganH |
bah |
| 11:21 |
RoganH |
mac++ |
| 11:22 |
RoganH |
jeff: csharp is just channeling Groucho Marx, whatever it is, he's against it |
| 11:23 |
csharp |
:-D |
| 11:23 |
csharp |
"what are you rebelling against?" "whaddya got?" |
| 11:27 |
berick |
staff client thought.. if all data is delivered via secure websockets and we don't store the authtoken in a cookie (use sessionStorage instead), we could conceivably serve the main page (templates, js, css) as non-SSL (for speed). am i crazy? |
| 11:27 |
RoganH |
that was from the man that said acting is hustling and that no movie is a work of art |
| 11:27 |
jeff |
berick: don't. |
| 11:27 |
berick |
jeff: heh |
| 11:28 |
RoganH |
berick: seems like a bad precedent to set even if valid in that circumstance |
| 11:28 |
berick |
apart from gut reaction / concerns about "why is it not secure?!?", anything specific? |
| 11:28 |
tsbere |
berick: Mixed content warnings alone would be a headache |
| 11:28 |
jeff |
berick: think MITM of the templates/js/css which would then allow you to do pretty much anything. |
| 11:29 |
berick |
tsbere: no mixed content warnings when the only ssl content is websockets |
| 11:29 |
berick |
jeff: hmm... |
| 11:29 |
berick |
yeah |
| 11:29 |
Dyrcona |
RoganH: I made a Mac staff client by installing xulrunner and then doing --install_app with the xpi. Other than that, no. |
| 11:30 |
jeff |
berick: "non-SSL for speed" is fixing the wrong thing -- we should be able to get pretty decent speeds out of https these days, and if speed is a concern we should focus on making it faster, not less secure. |
| 11:30 |
Dyrcona |
That was a couple of years ago, though. |
| 11:31 |
berick |
jeff: MITM.. are you basically saying every web page everywhere should always be SSL? |
| 11:31 |
dbs |
Can't we take advantage of SPDY now? |
| 11:31 |
Dyrcona |
"The quest for speed should not sacrifice the reliability and the stability of the system." -- OpenBSD FAQ 14.19 |
| 11:31 |
jeff |
berick: "every web page everywhere" is a bit of an extreme position. every page in an evergreen system? maybe. |
| 11:31 |
dbs |
"SSL Everywhere" is a thing |
| 11:32 |
berick |
btw, I don't have any specific speed concerns, i'm just thinking out loud |
| 11:32 |
jeff |
berick: for another response to "the secure stuff goes over X, can i serve the rest over unsecure http?" see stripe's docs here: https://support.stripe.com/questions/should-i-serve-my-payment-pages-over-ssl |
| 11:33 |
berick |
.. and i've given up on the idea completely.. now just more talking for my own curiosity |
| 11:34 |
berick |
hm, so, i guess one concern w/ MITM is that even though we don't deliver the authtoken, MITM js, could access the authtoken in sesionstorage |
| 11:35 |
jeff |
that and likely other things, yes. |
| 11:36 |
dbs |
https://istlsfastyet.com :) |
| 11:41 |
|
gsams joined #evergreen |
| 11:47 |
|
RoganH left #evergreen |
| 11:59 |
jeff |
berick: not that i'm giving stripe any specific authority here, just that it was one example that came to mind that closely mirrored what you had been suggesting. :-) |
| 12:01 |
berick |
jeff: of course |
| 12:01 |
berick |
and thanks for the feedback, everyone. a little reaffirmation now and then is good. |
| 12:02 |
jeff |
sometimes the answer to "why do we do this?" isn't "oh, right. of course." |
| 12:02 |
jeff |
sometimes it's "hey, that's a really good question -- let's change that!" |
| 12:19 |
|
rfrasur joined #evergreen |
| 12:25 |
|
ktomita joined #evergreen |
| 12:28 |
|
krvmga joined #evergreen |
| 12:29 |
krvmga |
as a rule, at cwmars central site, we're not dealing with patrons. except i get to deal with them fairly regularly because i answer the mail that comes into the cwmars general account. |
| 12:29 |
krvmga |
today a patron accused us of providing false information |
| 12:30 |
krvmga |
in an earlier email, he had complained at length that he was not getting the books he had on hold |
| 12:30 |
krvmga |
easy problem: every book he had on hold was just ON ORDER. nobody owned them yet. |
| 12:31 |
krvmga |
when i gently explained this to him, he came back with "but when i log into my account it shows 9 circulating copies for this one and 13 for that one" follwed by ranting about we don't know what we're doing and giving him false information |
| 12:32 |
krvmga |
i responded saying that we couldnt find the pages he said he found, gave him the urls so that he could see in the catalog the books were on order, and asked him for a screen shot. i also gave him a little cheat sheet on how to do a screen shot and save it. |
| 12:32 |
krvmga |
i think he was actually getting exercised over notes like "0 of 9 available" |
| 12:33 |
jeff |
the feedback from that interaction could probably be "make it more clear that copies are on-order", possibly just "in more places" |
| 12:33 |
krvmga |
sort of going to the it-says-available space |
| 12:33 |
jeff |
oh, but you're still mid-story -- sorry! |
| 12:34 |
krvmga |
jeff: no that's ok. but i have to say it seems harder to make it clearer for instance |
| 12:34 |
krvmga |
https://bark.cwmars.org/eg/opac/record/3292661?query=mr%20mercedes;qtype=title;locg=1 |
| 12:34 |
|
jihpringle joined #evergreen |
| 12:34 |
krvmga |
ON ORDER is in capital letters after every copy |
| 12:34 |
bshum |
Fwiw, we ripped those details out of our system to avoid questions like those. :D |
| 12:35 |
bshum |
Though I think we accidentally left it in the My Account area, so I have a sticky note on my desk to rip it out of the columns there too. |
| 12:35 |
csharp |
bshum: what did you rip out? |
| 12:35 |
krvmga |
bshum: so you don't display any "x of X available" messages? |
| 12:35 |
bshum |
csharp: The current holds count |
| 12:35 |
csharp |
ah |
| 12:35 |
kmlussier |
bshum: On all copies or just on orders? |
| 12:35 |
bshum |
I don't like listing out how many holds there are for a given bib because it gives patrons false hope. |
| 12:35 |
bshum |
And staff |
| 12:35 |
bshum |
:) |
| 12:36 |
bshum |
But that's a whole other matter |
| 12:36 |
kmlussier |
As a patron, I like to see the holds count. Even though it's not entirely accurate, it gives you a general sense of whether it's going to be a long wait or not. |
| 12:36 |
bshum |
We keep the available counts as a general detail. |
| 12:36 |
bshum |
Yeah we've been discussing bringing it back for certain libraries as a toggle on/off. |
| 12:36 |
csharp |
interesting - we've not had any issues with that |
| 12:36 |
bshum |
Like skip it for the schools (who don't let kids place holds anyways) |
| 12:36 |
* csharp |
knocks on wood |
| 12:36 |
krvmga |
bshum: the problem here, as i see it, is that this patron is getting confused over "available" |
| 12:37 |
kmlussier |
You could say 0 copies available and take out that second number? |
| 12:37 |
csharp |
krvmga: yeah, I can see that being too fine a point for a patron like that |
| 12:37 |
krvmga |
since the system considers ON ORDER as Available (because you can place holds on things that are just ON ORDER) |
| 12:38 |
krvmga |
like tossing a pebble in a pond, though, this kind of patron complaint has a ripple effect |
| 12:38 |
kmlussier |
Well, just because it's holdable doesn't mean the system considers it to be available. Checked-out copies are holdable and not available. |
| 12:38 |
krvmga |
the next thing i hear is from a librarian that patrons hate the catalog because it's so inaccurate |
| 12:38 |
* csharp |
goes back to trying to create a template for an item list that displays multiple stat cats |
| 12:38 |
bshum |
I think your folks are just too fickle, krvmga. |
| 12:38 |
bshum |
:) |
| 12:38 |
bshum |
(not a real answer, I know) |
| 12:39 |
csharp |
"be less fickle!" close ticket |
| 12:39 |
krvmga |
lol |
| 12:39 |
bshum |
I think we see less of these because of how our catalogs get their own vhosts per library and most libraries point patrons to that. The savvy ones start learning to search consortially, so they just have to learn how to read the screen more precisely. |
| 12:39 |
bshum |
I'm not sure I view "on order" as not available, per say. Maybe less available :) |
| 12:39 |
bshum |
Or soon to be available. |
| 12:39 |
csharp |
mostly that's just 3-year-old behavior: "I'm not able to get what I want, so I'm going to yell at you" |
| 12:39 |
bshum |
It's a weird middle ground that I hate a lot. |
| 12:40 |
bshum |
I imagine it's possible to alter the way that copies are counted for availability. |
| 12:41 |
krvmga |
we recently did a survey on network speeds that was open to the public. several thousand patrons wrote in comments. |
| 12:41 |
csharp |
our vocabulary around items and holds is pretty confusing, even from the technical end ("available" vs. "eligible" vs. "holdable", etc) |
| 12:41 |
krvmga |
the overwhelming majority or respondents were very happy with the catalog |
| 12:41 |
krvmga |
the word "love", or some variation of it, was used repeatedly |
| 12:42 |
kmlussier |
I'm confused. The system isn't counting those on order copies as available. The catalog says 0 are available. |
| 12:42 |
csharp |
@love CWMARS |
| 12:42 |
pinesol_green |
csharp: The operation succeeded. csharp loves CWMARS. |
| 12:42 |
csharp |
kmlussier: maybe the item status is set differently between your two catalogs? |
| 12:43 |
csharp |
item status settings, that is |
| 12:43 |
krvmga |
kmlussier: yes, you're right but the patron is seeing 0 of 9 available as being "9 available that you're not giving me" and not "0 available for me to have" |
| 12:43 |
bshum |
Yeah, well. |
| 12:43 |
bshum |
Maybe we really don't want to give them any of them. :D |
| 12:43 |
bshum |
"working as intended" |
| 12:44 |
krvmga |
is that the Ron Swanson approach to the catalog? |
| 12:44 |
csharp |
"they're available, just not to YOU" |
| 12:44 |
krvmga |
<reference>Parks & Recreation, U.S. comedy television show</reference> |
| 12:44 |
* csharp |
is being flip - will shut up now ;-) |
| 12:45 |
csharp |
krvmga: love that show! |
| 12:45 |
bshum |
I should watch that someday. |
| 12:45 |
* Dyrcona |
reads some scrollback... |
| 12:45 |
krvmga |
bshum: i find it very funny. |
| 12:45 |
Dyrcona |
Being too accurate leads to inaccuracy. |
| 12:45 |
csharp |
there's even a crazy librarian character ;-) |
| 12:46 |
krvmga |
csharp: yes. and my wife (who has been a director and is currently an assistant director) and I get big laughs out of it. |
| 12:46 |
csharp |
krvmga++ |
| 12:48 |
csharp |
krvmga: http://mattlcostello.tumblr.com/post/7792435851/all-of-the-anti-library-quotes-from-season-2 |
| 12:48 |
krvmga |
Dyrcona: next thing you're going to tell me that Wikipedia isn't a good source |
| 12:48 |
krvmga |
csharp++ |
| 12:50 |
krvmga |
on another topic...i've been wondering about doing some formal release management at cwmars |
| 12:51 |
krvmga |
when we upgrade, we're using evergreen as the rest of the world does but then we have so many customizations to our installation, all over the place |
| 12:51 |
Dyrcona |
We use git to manage customizations and build our upgrades as branches. |
| 12:52 |
krvmga |
the past couple of upgrades we have naturally run into bits and pieces of customization that we missed upgrading |
| 12:52 |
krvmga |
Drycona: we're having a meeting about that in the next couple of weeks. |
| 12:52 |
krvmga |
we've wanted to be using git for a long time but somehow it just hasn't happened |
| 12:54 |
krvmga |
i use git, personally |
| 12:54 |
Dyrcona |
You could start with a local repository, then. |
| 12:54 |
Dyrcona |
I would suggest you figure out how you would like to separate your customizations into branches. |
| 12:55 |
Dyrcona |
Like, OPAC customizations, SIP configuration, Z39.50, etc... |
| 12:55 |
krvmga |
Dyrcona: difficulty on a scale of 1-10? |
| 12:56 |
krvmga |
5. say 5. it's a safe number. 5 |
| 12:56 |
Dyrcona |
The technical side is simple if you use git. The hard part may be deciding how to split your customizations. |
| 12:56 |
krvmga |
Dyrcona: yeah, i see that. |
| 12:56 |
krvmga |
it's gonna take some mulling over. |
| 12:56 |
Dyrcona |
Yep. |
| 12:56 |
csharp |
krvmga: fwiw, we manage ours in one big pile of commits in a pines_rel_2_X_X branch, though we have a separate kpac repo - Dyrcona's suggestion for granularity is better overall though |
| 12:57 |
Dyrcona |
You could also throw everything into 1 branch, but that can lead to more conflicts come upgrade time. |
| 12:57 |
Dyrcona |
csharp++ |
| 12:57 |
krvmga |
csharp: i had originally been leaning in that direction. |
| 12:57 |
krvmga |
csharp++ |
| 12:57 |
krvmga |
Dyrcona++ |
| 12:58 |
Dyrcona |
To continue, for the sake of the logs... |
| 12:58 |
csharp |
I think it's a natural progression to start with one big branch, then split out as you need to - upgrades are a little tedious having to sort out the pile from version to version |
| 12:58 |
Dyrcona |
Once decided how to parcel out your customizations, I would checkout a new branch for each customization branch you need. |
| 12:59 |
krvmga |
Dyrcona: this sounds messy merging branches...? |
| 12:59 |
Dyrcona |
If you're currently using 2.5, I'd base the branches on rel_2_5, or even 2.5.1 or whatever you're running now. |
| 12:59 |
Dyrcona |
Nah, it isn't. Git is designed for it. |
| 12:59 |
bshum |
I learned with specific branches for each area of customization. Then I got lazy and just started rebasing a big branch filled with changes :) |
| 12:59 |
* bshum |
does it the "bad" way. |
| 12:59 |
Dyrcona |
bshum lives dangeroulsy. |
| 13:00 |
bshum |
"Yeah, well, I think it'll be okay." |
| 13:00 |
krvmga |
we'll be going to 2.5 next month |
| 13:01 |
Dyrcona |
When it comes time to upgrade, make a branch for production or whatever based on the new branch you are going to. |
| 13:01 |
csharp |
eeevil: around? |
| 13:01 |
Dyrcona |
Then you can git merge the custom branches into it. |
| 13:02 |
Dyrcona |
We base our custom branches on the master branch, because when we upgrade, we checkout from the master branch. |
| 13:02 |
Dyrcona |
For instance, we're upgrading this coming weekend. |
| 13:02 |
Dyrcona |
We also loaded everything on our training server in advance to give our members a chance to look at things and find any bugs that we missed during preliminary testing. |
| 13:03 |
eeevil |
csharp: sorta ... sick day :| |
| 13:04 |
Dyrcona |
We made the branch on the 16th of April and named it mvlc_20140416. |
| 13:04 |
csharp |
I'll just ask my question to the full channel (though I'd be happy if eeevil sees it) - I am making some edits to the reporter.classic_item_list definition in fm_IDL.xml, and wanted to know if it is correct to denote the "rel" attribute of the link between reporter.classic_item_list and our "legacy" stat cats as "might_have" rather than "has_a" |
| 13:04 |
csharp |
eeevil: cool - sorry you're feeling bad :-/ |
| 13:05 |
Dyrcona |
It has had two or three bug fixes applied as well as our custom branches. |
| 13:05 |
|
tjspindler joined #evergreen |
| 13:05 |
csharp |
for reference http://git.evergreen-ils.org/?p=working/Evergreen.git;a=commit;h=0e114d1d33e4f23e28c43c3a5ef04bc8718348e3 |
| 13:06 |
Dyrcona |
It is a good idea to rebase your custom branches periodically to keep them up to date with changes or when you want to test a new branch, like going from 2.5 to 2.6 . |
| 13:06 |
eeevil |
csharp: checking something before answering ... working from memory is not a good idea today ;) |
| 13:06 |
csharp |
eeevil: understood |
| 13:08 |
|
mrpeters joined #evergreen |
| 13:08 |
krvmga |
Dyrcona: it sounds like we have much to learn in this regard. |
| 13:09 |
Dyrcona |
krvmga: How do you manage your customizations now? Do you keep patches around? |
| 13:11 |
krvmga |
Dyrcona: it's all done the hard way now. i save directories of the customizations then, after the upgrade, go in and add the files where needed. |
| 13:11 |
krvmga |
we also rely on Equinox to take care of patching |
| 13:11 |
eeevil |
csharp: short answer is yes, but the next two columns (stat_cat_X_value) give you the value of the thing pointed at by the links |
| 13:11 |
bshum |
I remember those days. Back in our 1.6/2.0 era when we'd manually write down which lines changed before/after for every file we touched. |
| 13:12 |
bshum |
Before git. |
| 13:12 |
eeevil |
csharp: what's the use case of a left join from statcat to entry via the link? |
| 13:12 |
krvmga |
http://upload.wikimedia.org/wikipedia/en/archive/f/f4/20100830193250!The_Scream.jpg |
| 13:13 |
eeevil |
(in the report, I mean. the view already does that) |
| 13:13 |
krvmga |
that about covers it for me, atm |
| 13:13 |
|
dMiller_ joined #evergreen |
| 13:13 |
krvmga |
things will be better when we start using git for everything |
| 13:16 |
csharp |
eeevil: there's not one I can think of |
| 13:17 |
eeevil |
csharp: in any case, might-have with a link is more correct, yes |
| 13:17 |
csharp |
eeevil: thanks for checking! |
| 13:18 |
csharp |
@dessert eeevil |
| 13:18 |
* pinesol_green |
grabs some Coconut Cream Cake for eeevil |
| 13:18 |
eeevil |
I'll save the for when I can taste again :) |
| 13:18 |
csharp |
eeevil++ |
| 13:22 |
jeff |
Is anyone else in the process of doing or starting to do regular reports to find likely data quality issues requiring attention? Anything from bib QA reports to "these shelving locations should have a circ modifier in this list" or "these are acceptable call number patterns for call numbers with copies in this location"? |
| 13:23 |
jeff |
Or even "these copies are deleted but still checked out", "these copies are 'Available' but have an open transit or circ", "these patrons have an address alert which probably means that they should be in a specific profile group (but they are not)", etc? |
| 13:25 |
tjspindler |
I have thought about this but we have not done anything |
| 13:25 |
Dyrcona |
jeff: Not regular but we've done some reports when issues come up. |
| 13:26 |
Dyrcona |
jeff: And not "reports" in the reports editor sense, but usually SQL or Perl to list the problems, and sometimes fix them along the way. |
| 13:26 |
jeff |
We're currently doing this but not across every shelving location / department / branch. I'm starting to get more interest from staff, so was 1) looking for ideas of other things that we're not yet doing yet others have found useful and 2) gauging interest so that i can wrap some of these up to share |
| 13:27 |
tjspindler |
Dyrcona: that is what we have done but i have been thinking we should do more |
| 13:27 |
jeff |
Dyrcona: yeah, these are SQL reports in Jasper |
| 13:27 |
csharp |
jeff: we are not in the process, but I'm interested |
| 13:27 |
jeff |
Dyrcona: any kinds of things you've done that I didn't list above that you'd care to share, in terms of "we went looking for X with Y and not Z"? |
| 13:28 |
jeff |
csharp: figured you might be. :-) |
| 13:28 |
csharp |
our cataloging staff here does some QA spot checking based on reports, but the reports themselves aren't set up to find errors |
| 13:28 |
jeff |
csharp: similar question as for Dyrcona -- have you any examples of things you'd be likely to want to report on? |
| 13:28 |
csharp |
jeff: I don't have specifics right now - I'll ask around |
| 13:29 |
jeff |
cool -- appreciated, but no rush! |
| 13:29 |
Dyrcona |
jeff: Something we did recently was to change a bunch of "laserdiscs" to DVD because until there was a DVD vr format and even after, many DVDs are cataloged as laserdisc. |
| 13:29 |
jeff |
Dyrcona: nice. the change was made at a biblio.record_entry.marc level, or somewhere else? |
| 13:30 |
Dyrcona |
jeff: Yep, in bre.marc. |
| 13:30 |
jeff |
Dyrcona: and did you go through the trouble of using MARC::Record or did you use regexp_replace? :-) |
| 13:30 |
Dyrcona |
MARC::Record |
| 13:31 |
|
jeff__ joined #evergreen |
| 13:31 |
|
jeff__ joined #evergreen |
| 13:31 |
jeff |
Dyrcona++ |
| 13:32 |
jeff |
Dyrcona: assuming that was using perl, and not a custom DB function? |
| 13:32 |
Dyrcona |
jeff: Yes, a Perl script. I should share it somewhere. |
| 13:34 |
jeff |
Dyrcona: I'm debating the usefulness of a db function that uses MARC::Record to enable batch updates where regexp_replace is sometimes used. Doing it with perl and DBI instead of a database function might make more sense. Not sure yet. |
| 13:36 |
Dyrcona |
jeff: I believe I used a regexp match in the initial query to find the bib records to fix. |
| 13:37 |
Dyrcona |
jeff: I'll share the script somewhere public after I finish an email about what is still not working on our mail server. ;) |
| 13:40 |
|
jeff_ joined #evergreen |
| 13:41 |
jeff |
sorry for the noise on that client. :P |
| 13:42 |
|
jeff_ joined #evergreen |
| 14:01 |
|
vlewis joined #evergreen |
| 14:02 |
Dyrcona |
jeff: I should actually make some changes to the script before I share it. |
| 14:02 |
Dyrcona |
We found that adding the rejected or unfixed bib records to a bucket didn't work so well because we had over 800 of them. |
| 14:02 |
Dyrcona |
Our catalogers ended up dumping the bucket as a CSV. |
| 14:03 |
Dyrcona |
I should probably change that bit of the script to dump to a CSV instead of putting the records into a bucket. |
| 14:06 |
|
vlewis joined #evergreen |
| 14:17 |
|
akilsdonk joined #evergreen |
| 14:36 |
dkyle |
jeff: I'm interested. I've written a lot of queries similar to what you're talking about, although most have been probably been particular to grpl in some way. |
| 14:38 |
jeff |
dkyle: cool -- like i said, I'm interested in other things that people have had success reporting on, or wished that they could report on. I'll be sure to keep you posted. I'm hoping to write some things up and share some more Jasper Reports examples focused on bib/catalog/patron QA. |
| 14:38 |
dkyle |
we do a lot of things based on various call number parsings, so one thing I do is create views to encapsulate those parsings and then write functions or queries against those views |
| 14:38 |
jeff |
Oh, I forgot the patron QA bits... things like reporting on patrons with syntactically invalid email addresses (before we enabled input validation on that, but still useful for future if/when some sneak in via imports/updates/etc) |
| 14:40 |
jeff |
other patron bits: (i guess i mentioned one already -- the address alert + profile group disagreeing) we have some stat cats that are currently required but not controlled, and we have a report that exposes the "freeform" entries -- again here, the idea is to prevent those freeform entries in the near future |
| 14:41 |
jeff |
as well as "patrons about to go into collections", "patrons over collections warning limit but are otherwise exempt", etc. |
| 14:42 |
dkyle |
jeff: I like the idea, I'd be glad to contribute SQL for some of the "hit list" you come up with if you'd like |
| 14:45 |
kmlussier |
gsams: Are you around? |
| 14:47 |
dkyle |
speaking of reporting, is anyone else using pl/r to generate report graphics form the database? |
| 14:47 |
jeff |
no, but I heard from either you or _bott_ that you had delved into that. How do you like R? |
| 14:49 |
|
collum joined #evergreen |
| 14:50 |
dkyle |
Jeff: R is awesome, strange syntax and data types if you are not a stats person, but very powerful and lots of packages. and i think its cool to call a Pg function and get a nicely formated pdf graph as the output. |
| 14:52 |
jeff |
From some of the R output I've seen, I'd wager that it's tons prettier than the "gnuplot ascii output from the psql command shell" tricks I've seen. :-) |
| 14:57 |
dkyle |
Jeff: I've not done a lot of such work, but the ggplot2 R package makes generating nicely formated graphics easier than anything else I've seen. It's often more work getting the data munged properly than getting the graph |
| 14:57 |
|
hbrennan joined #evergreen |
| 15:09 |
dkyle |
so like Jeffs idea for a collection of QA reports, I can envision a collection of pl/r functions for typical stats and collection development stuff. maybe even an CHQ type EG module someday |
| 15:16 |
|
RoganH joined #evergreen |
| 15:26 |
|
jwoodard joined #evergreen |
| 15:47 |
gsams |
kmlussier: I'm here now |
| 15:48 |
kmlussier |
gsams: The SMS issue you brought up over the weekend? I was able to replicate it in our catalog. If you file a bug in LP, I can confirm it. |
| 15:48 |
kmlussier |
I just didn't want the issue to get lost in the IRC logs. |
| 15:48 |
* kmlussier |
needs to disappear for a little bit. |
| 15:48 |
gsams |
alrighty |
| 15:49 |
gsams |
I was asking in case it got solved between 2.3.5 and 2.6 because I couldn't find a bug report specific to that |
| 15:49 |
gsams |
I will file one though |
| 15:50 |
kmlussier |
gsams: I confirmed it in 2.5, and I don't think it's been fixed since then. |
| 15:50 |
gsams |
kmlussier: thanks! |
| 15:53 |
|
bmills joined #evergreen |
| 16:00 |
bshum |
I'm reading this discussion on #code4lib about the creation of more lists (and the replies) and I can't help but wish we had a few LESS lists in our community :) |
| 16:01 |
gsams |
bshum: We need a place that we can bring all these lists together. I know! Let's make a list! |
| 16:01 |
jeff |
firmly tongue-in-cheek: sure! i'd love to form a committee to discuss a taskforce to utilize the LESS css framework in Evergreen! I'll create the open-ils-LESS-discuss-L mailing list right now! |
| 16:02 |
bshum |
Ahahaha :\ |
| 16:04 |
|
akilsdonk joined #evergreen |
| 16:11 |
gmcharlt |
bshum: jeff: we also need a list to debate "less" vs. "fewer" ;) |
| 16:22 |
|
krvmga joined #evergreen |
| 16:24 |
grammarpolice |
gmcharlt: Who needs a debate? It obviously should be "fewer" in this instance. |
| 16:25 |
gmcharlt |
grammarpolice++ |
| 16:25 |
krvmga |
i found the source of the confusing message one of my patrons brought up. the offending words appear to be hard-coded |
| 16:25 |
gsams |
grammarpolice++ |
| 16:25 |
krvmga |
the case was that the title he was placing a hold on was ON ORDER. every copy. |
| 16:26 |
krvmga |
when he placed a hold and then looked at the hold in his account, he could see the message WAITING FOR COPY 46 HOLD ON 9 CIRCULATING COPIES |
| 16:26 |
krvmga |
except, at this moment, there are not circulating copies |
| 16:26 |
krvmga |
i now understand why he felt irritated |
| 16:27 |
krvmga |
this wording seems to be hard coded in files in locale |
| 16:27 |
krvmga |
is there any reason why i shouldn't fix this? |
| 16:35 |
* bshum |
thinks individual sites should feel fine in changing wording to work best for them as desired... |
| 16:36 |
krvmga |
bshum: i just want to take out the word "circulating" since that's what's causing the confusion |
| 16:36 |
bshum |
That said, if you feel strongly to advocate for changing the wording by default for the next Evergreen release (2.7, etc. since we're string frozen technically for existing releases), then I'd say worth filing a wishlist bug to tack on some code for consideration. |
| 16:37 |
* bshum |
has seen that done before, like for the reset password button : |
| 16:38 |
krvmga |
bshum: thanks. after i've made sure to document where the wording should be changed, i'll do that. thx. |
| 16:38 |
krvmga |
bshum++ |
| 16:38 |
bshum |
The example I mentioned from the past: https://bugs.launchpad.net/evergreen/+bug/1032646 |
| 16:38 |
pinesol_green |
Launchpad bug 1032646 in Evergreen "Password Reset Button Text" (affected: 3, heat: 14) [Wishlist,Fix released] |
| 16:38 |
bshum |
In case you want to see a bug that changed wording successfully for the community at large. |
| 16:40 |
|
mrpeters left #evergreen |
| 17:10 |
kmlussier |
Doodle poll for May developers meeting - https://doodle.com/ncdke3xitdzvwfwa |
| 17:12 |
|
mmorgan left #evergreen |
| 17:13 |
bshum |
kmlussier++ |
| 17:25 |
jeff |
bshum, gmcharlt: I forgot to ask -- should we have a separate open-ils-LESS-discuss-L-jobs-L list for job postings? |
| 17:25 |
* jeff |
ducks |
| 17:27 |
|
vlewis joined #evergreen |
| 17:27 |
|
vlewis joined #evergreen |
| 17:29 |
kmlussier |
All kidding aside, the thing that concerns me about all of our lists is that a new person entering the community might send a request for help to a low traffic list not knowing that he/she may get more responses if the same request were sent to the general list. |
| 17:30 |
kmlussier |
I'm primarily thinking of the sys admin list, where I've seen a couple of questions languish. |
| 17:30 |
pinesol_green |
Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html> |
| 17:34 |
|
krvmga joined #evergreen |
| 18:58 |
|
mjingle joined #evergreen |
| 20:02 |
|
doug joined #evergreen |
| 21:29 |
|
mjingle joined #evergreen |
| 22:57 |
|
dbwells_ joined #evergreen |
| 22:57 |
|
remingtron joined #evergreen |
| 23:21 |
|
dMiller_ joined #evergreen |
| 23:57 |
|
dMiller_ joined #evergreen |
