Wikisource talk:Monthly Challenge/Archives/2021-06

Long-term series

We probably need to thing about long-term series, in that they're so enormous that they'll block out chunks of MC effort for years. For example, even if we can do a complete Carlyle every month, it'll be nearly 3 years.

I think we should make it a "thing" that if work fizzles on a series (for some informal and flexible definition of fizzle), it gets rotated at the next month and can be reinstated if people re-nominate it. Incomplete series can go on a "we tried, you wanna go?" pile for picking over and re-nomination.

The means we get rapid turn over (every month) unless people really like a series and show it by proofreading it, then it will stay as long as that continues. TL;DR play with it or put it back in the toybox, don't leave it on the floor. Inductiveload—talk/contribs 18:01, 7 June 2021 (UTC)

@Inductiveload: Hmm, you're probably right. Maybe we could put them in a separate section. One for "To proofread (Long-Term Series)" and one for "To Validate (Long-Term Series)".

My thought behind these series is that they works are important enough that they should be done no matter how long it takes. I know that interest might fizzle out, but it's really hard to know when it will return. It's a sort of always there if someone wants to pick it up. What I don't want to happen is to have a huge pile of abandoned series that only have a few volumes done. Languageseeker (talk) 03:45, 8 June 2021 (UTC)

I'm trying to limit the number of spots that these series take by only allowing one in To Proofread and one in To Validate. I don't think that one will be added each month unless there is significant community interest. Languageseeker (talk) 03:45, 8 June 2021 (UTC)

@Languageseeker: I think leaving "fizzled" series on a priority pile for easy pick-up would be OK. So resumption of a previous series is preferred to a whole new series (assuming the same level of interest). I'm just wary of blocking out slots for months and months until we can refresh. I'd rather start a new series than keep a stalled one blocking the slots (but I'd rather finish an active series than start a new one).

Validation is easy to deal with: the volumes get three months (from the start) and then they drop off naturally as normal. If people want to validate any MC past proofread work, they can come back at any nomination. Without auto expiry, we'll end up buried in yellow works and spread validation so thinner and thinner and hardly anything will ever get done. We'll have to accept that many works will expire either un-proofread or un-validated, but make it easy to bring them back if interest sparks. Inductiveload—talk/contribs 08:22, 8 June 2021 (UTC)

@Inductiveload: I see your point. Instead of taking a slot forever, maybe we can make a list of unfinished series. Let's treat long term series as normal works and give them three months. However, I still don't think that we should have more than one volume from a series in proofread or validation. Would it be ok if the previous volume isn't validated by the time the next volume is finished, it gets removed from the list? Languageseeker (talk) 04:22, 9 June 2021 (UTC)

Yeah, we should have a list, and reinstatement from that list should be preferred over new series if there is appetite (but better to have an active work than not).

I don;t think we should remove from the "to be validated" section early, even if new volumes get proofread. Validation is usually slower than proofreading, so that would pretty much ensure it was impossible to validate anything. After 3 months, we'll likely reach a steady state of the inflow of proofread volumes and the outflow of validated or expired volumes (I reckon we'll probably have ~2-4 volumes in the validation queue, mostly depending on the rate of proofreading). Or the series stalls.

Also, the validation queue is segregated by age, so probably the series will be split across the three sections (and the section is at the end). Inductiveload—talk/contribs 14:48, 10 June 2021 (UTC)

Redundant works contained in series

Also, we should probably think about checking that volumes in the MC aren't covering material which we already have. For example, HG Wells v2 is Moreau and Sleeper, but we have complete, scan-backed versions of both of those. Same for v3. Whereas v4 is Anticipations, which we don't have a copy of at all. I'm not against filling in the series, but I'd rather focus efforts on things we do not have scan-backed versions of. After all, one of the stated goals of the MC is core text building. Inductiveload—talk/contribs 03:21, 11 June 2021 (UTC)

I think that it might feel as if we're replicating works, but these are not exactly the same. For me the long-term series have to be more than a series of mere reprints. Instead, they either have to contain a contribution from the author, an illustrator, or an editor that make them a critical edition. For example, for the Atlantic Edition, H.G. Wells revised all of the material. Most of them have never been reprinted because they were in copyright. They are unknown versions of very well know stories. The rich paid to have slightly better versions of these texts and Wikisource is making them available to everyone.

I'm also wary of breaking up a series. At which point, do we stop? Should we skip only volumes or also remove parts of volumes? Realistically, nearly every long-term series will contain some duplicate title. I'd rather proofread a few extra volumes than have a bunch unfinished series. Languageseeker (talk) 12:38, 11 June 2021 (UTC)