Old ER offerings showing as currently available on work pages

I was looking over some of my old Early Reviewer wins and noticed that there is a box in the upper right saying they are available in the current batch of books.

Some examples
https://www.librarything.com/work/19651916
https://www.librarything.com/work/20854227
https://www.librarything.com/work/19221834
https://www.librarything.com/work/20306764
https://www.librarything.com/work/19290318

It's also happening on author pages.

https://www.librarything.com/author/stonesgreg
https://www.librarything.com/author/moorekate-2
https://www.librarything.com/author/middletonnick-1

The Skeleton Tree has the same note in the upper right corner, as well as a second note saying that it "was" available. Is one meant to replace the other at the appropriate time?

I see the problem. I am asking Loranne about it, because it seems these batches were not officially "closed" and she may know why.

Bump

https://www.librarything.com/work/19652040 from last summer.

It looks like it started happening with the March 2017 batch.

That fits with >3 WeeTurtle: as that has two copies in the ER account. One is tagged January 16, the other April 2017.

Okay, after poking around, I've found a fix for individual batches, but it doesn't really address why it's happening in the first place. Some details:

>7 amanda4242: is right, the March 2017 batch was the first affected, and all batches after that point were showing the same behavior, for every title I checked. This goes all the way up to the most recent closed batch, September 2018.

I can't figure out what changed that made this start happening—the last major change I made to the Early Reviewers process was well before March 2017. It's possible this change caused it—due to the way the system is set up, I have the final step in ER pretend to automatically send winner lists to publishers. Because I do a fair amount of hand processing winner lists, I need it to skip this step. However, closing an Early Reviewers batch properly should, as far as I know, happen before that point (before I even pick winners).

The fix I've found lies in archiving the batches. Typically, I leave them unarchived for a full calendar year, in case I need to go back and get winner list details again for publishers who've missed their deadline. I hadn't been archiving them lately because I wanted to get this bug nailed down. I'll go ahead and archive through July 2017 now, leaving us a couple extra test months to play with if we need them.

I haven't requested ER books for a while now, but I must say I'm quite discouraged by the fact that it's not, in fact, the objective, algorithmic process we've been promised, but is something involving "hand-processing" and a human "picking winners". There's nothing wrong with having a person, rather than an algorithm, do the selection, but it's quite different from how the process once worked (or how we were told it once worked, I suppose), and a sufficient change that I think it should have been communicated.

>10 lorax: I think you're waaay misreading what lorannen wrote there.

"Hand-processing" is referring to the list of winners (after they've been picked) and there's nothing about the phrase "picking winners" that has to imply a manual (as opposed to automated) process.

>11 saltmanz: is correct. Essentially, I click a button (for each script) to run the scripts that pick winners. There are a handful of them. I don't do anything with the inner workings there, and I have exactly zero say over who gets what book—that's all down to the algorithm, as has been the case since well before my time here. The hand-processing is, in fact, simply formatting the lists of (already picked!) winners prior to sending off to publishers.

>12 lorannen:

Believe me, if Loranne were hand-picking winners, she'd do little else!

Read too much into "I pick {the} winners", I suppose!

bump

bump

bump

https://www.librarything.com/work/19651916 -- alum
https://www.librarything.com/work/20854227 -- june 2019 batch
https://www.librarything.com/work/19221834 -- alum
https://www.librarything.com/work/20306764 -- alum
https://www.librarything.com/work/19290318 -- alum
https://www.librarything.com/author/stonesgreg -- alum
https://www.librarything.com/author/moorekate-2 -- alum
https://www.librarything.com/author/middletonnick-1 -- alum
https://www.librarything.com/work/16290595 -- alum
https://www.librarything.com/work/19652040 -- alum

Are there new examples?

Recent examples

https://www.librarything.com/work/22543392
https://www.librarything.com/work/20523715
https://www.librarything.com/work/22268193
https://www.librarything.com/work/22045033
https://www.librarything.com/work/21960754
https://www.librarything.com/author/rundellkatherine
https://www.librarything.com/author/lucasmatt

Thanks folks, but I don't think I need any further examples. I believe I know why it's happening, but it's beyond my skills to fix. Will take developer time to do. Here's the deal: I typically do what's called "archiving" a batch about a year after it ran. Once I archive, this marker indicates that the book is an alum, rather than current. For some reason, a batch simply being closed isn't enough to do it. And I can't archive batches immediately after they close for other reasons (it makes it harder to track down delinquent publishers, for one).

lorannen (#20):

Good hypothesis. What's the most recent batch you archived? Spot-checking a couple titles, I see titles from the March 2018 batch showing up as "is available" while February 2018 shows up as "was available". Think it's safe to archive March 2018 and see if that flips the switch? It would be a good test.

>21 lorax: And archived. Check or point me at one of the examples?

This was my pick from March batch, 2018.

https://www.librarything.com/work/20315527

It does say that the book WAS available, so at least that one is correct.

lorannen (#22):

The same ones I checked before have now shifted to "was available", so you seem to have pinpointed the explanation.

WeeTurtle (#23):

See above. You checked after she made the fix.

>24 lorax: Thanks! Sounds like we've got it narrowed down, finally. I'll see if I can get Tim to look at this with me and hopefully fix once he's got time. Between redesign and ALA this week, it might be a little while.

In summary, for ER batches that have already closed, but are not yet archived, titles still show the "currently available" message in the upper-right corner on the work page. Archiving the batch changes that to "was available." I archive batches after 1 year—I need to keep them around for a little while to make tracking down late titles, reconciling reviews, etc. easier.

This may be how it was intended to work originally (that's what I'm betting), with archiving happening more frequently, but need a better solution.

>24 lorax: Yup, I saw. I thought part of it was to check that archiving the month did the trick which it seemed to.

bump

Aha...I was just wondering about this today. Thanks to amanda4242 for steering me to this thread and the explanation of what's going on. Didn't really soothe my disappointment at finding that Tea Obrecht's new novel is not, in fact, available on the current list, though.

bump

bump

This bug is on our "to-do" list as soon as we can, thanks for your patience in the meantime!

bump

bump

bump

bump

Okay, thanks for the reminder. I think this is solved. The choice of current batch is a little tricky, so I'm going to ask Abigail to watch if there are any problems as one batch is closed and another opens.

Marking as fixed.

This is happening again.

https://www.librarything.com/work/26327436
https://www.librarything.com/work/11069562

>38 amanda4242: When I take a look at those works, I do not see them as currently available through ER. It says the title in question "was available." Are you seeing something different?

>39 AbigailAdams26: Huh. They're saying "was available" now, but when I looked at them this morning they were still saying "is."

And it's happened again. Almost certainly related to https://www.librarything.com/topic/333457#

>41 amanda4242: Thank you for reporting. As mentioned in the other thread, we're in another ER limbo period, and will investigate.

bump

bump

Closing this. Our ER redesign in February 2022 should have solved this problem. Please reopen, if it recurs.