Dan Scott @dbs@code4lib.social

Ask for tab-delimited data without specifying an encoding (because who generates anything other than UTF-8 these days?), get pipe-delimited data (!?!) encoded in latin1.

Okay, okay. I'll be liberal in what I accept.

Just this morning I changed all user-facing instances of "catalog" to "catalogue" in our new discovery layer. I know it's annoying for devs but it is genuinely important for cultures.

https://toot.cafe/@nolan/105283827217059237

Highly impressed with @swib 2020 conference (Semantic Web in Libraries - https://swib.org/swib20/) which made attendance free this unprecedented year and are using free software tools (Mattermost for discussion, BigBlueButton for streaming video) for infrastructure.

Kudos @acka47 and Joachim Neubert!

Sometimes I wonder if my employers have deliberately taken advantage of my hyperfocus. Like okaying my buyout of part of my workload to work on research, then assigning me an entire library system migration (of not just one but many constituent libraries) to complete by myself on a condensed timeline.

Which is how I ended up paying my employer to do overtime because I'm not going to let this project fail, as angry as I might be about it, and why my research has been suffering.

Very grumpy that Primo VE is written in AngularJS as that a) hijacks routing and onload so much that vanilla JS just won't work and b) the framework is EOL at the end of next year and c) the performance is awful.

So I have to learn enough of a soon-to-be-obsolete framework to customize things for launch, and will inevitably have to learn something else in a year or so.

I normally enjoy learning but not this kind of situation.

Got our Evergreen instance up to a B+ with about a half an hour of work. Should be pretty generalizable, too.

But the reliance on the ancient Dojo framework is going to kill any chance of applying a Content Security Policy without 'unsafe-inline' and 'unsafe-eval'.

That's the title of the support case I just opened.

I wonder how long it will be before Ex Libris is paying us *not* to be a customer.

Point https://observatory.mozilla.org/ at an Alma or Primo VE instance and see the carnage for yourself.

(Note: Evergreen hasn't implemented much, but out of the box it still gets a D+ compared to Alma's stone-cold F)

Currently editing videos from the command line using ffmpeg because OpenShot is ludicrously slow (exporting a 14 minute video is going to take it an hour?!?)

"in IT, we do not like "scripts" created externally; we rather systemize the data exchange and hence we can support it and be held responsible for its integrity."

Hello, I am a member of the same institution as you, IT, and wrote the script that solved a problem that I've been waiting on you for THREE FUCKING MONTHS without any movement, and posted it so any of your people can review it.

Don't get all pissy over your turf. Perform instead of blocking progress.

So Primo VE is a hosted application over which you have very little control.

It's very frustrating when they push out an update that includes an SVG file with two occurrences of multiple id attributes in a single XML element, which makes Firefox (at least) unhappy.

Another support case opened...

I can't add arbitrary Z39.50 servers as copy cataloguing sources.

Need to submit a support case with the details and confirmation from the source that they would like to be added to Ex Libris's master list of sources.

Oh and actually SRU only, no support for Z39.50 sources.

Wow this is such an improvement over our current library system :(

If you don't configure a campus, then patrons see "NO CAMPUS" as an overall heading when they choose a pickup library. No way to change that (wording or hiding it).*

If you do configure a campus to group libraries, then you can't provide a translation of the campus name. "This is a tentatively planned enhancement for next year."

ARGH

* OK I could hack it via custom CSS but that is garbage and fragile.

Next release will disable a user account that:
* downloads more than 500 megabytes of data in 60 minutes
* downloads more than 100 PDFs in a 60-minute period
(among other triggers)

We proxy streaming video and music.

And Zotero & friends can grab 100 PDFs in a few minutes.

Luckily we will be able to "augment" these rules based on our library's "unique" needs.

Doesn't seem well thought-out. A lot of libraries are going to be hitting this.

https://help-de.oclc.org/Library_Management/EZproxy/EZproxy_release_notes_and_known_issues/2020_EZproxy_release_notes/Information_about_upcoming_version_7.1.0_release?sl=de

Add to that the horrific point-and-click layers of configuration required to set up a new loan policy, with no ability to export the current set of policies in bulk (screenshots are the recommended way to capture settings at any given point in time), with no API for reading/setting/updating loan policies, and this SQL-loving systems librarian is ANGRY

The loan and request policies that Ex Libris migrated for us didn't create item policies that mapped to what we had requested. They just left their default item policies in place.

So there are rules for our item policy RESERVE 2 DAY, but that item policy isn't recognized as a valid item policy. If we mapped two item policies to a single loan policy, then editing them to add a third (valid) item policy erases the two invalid item policies.

Not in my library. Ever.

"Scientific publishers propose installing spyware in university libraries to protect copyrights"

https://www.codastory.com/authoritarian-tech/spyware-in-libraries/