Huntster

References on copied statements

Just curious what you think about this: I noticed your edit which copied a statement on the length of the C-32 to the C-32B. I think this is a good effort to do, but my only question is if it is appropriate to retain the reference on the target item, when the cited source does not specify that it is applicable to that variant. In this case, I am sure the length of all C-32s may well be the same, regardless of variant, but the cited source, so far as I read in it, does not ever indicate that its stats apply to the C-32B in particular (or even mention any A/B variants at all), only ever referring to 'C-32' in general. I wonder if it would be best, when doing these copies, to simply remove the citation unless the source material does indeed specify the exact item. I'd like to get your opinion on this. Josh Baumgartner (talk) 19:20, 10 September 2023 (UTC)

@Joshbaumgartner: If you wish to remove the reference, feel free. There are very few source of data specifically about the 32B. My interpretation is that when a source speaks on the "C-32" it's referring to both models, since there is no plain "C-32", just A and B models. For example, this also seems to state both the A and B are the same dimension; the overall article is about the 32 in general, but specifies the specific differences when applicable. — Huntster (t @ c) 23:16, 10 September 2023 (UTC)

Yeah, I don't think there is any problem with the data itself, just don't want to misrepresent the source material. I guess there is always a certain amount of interpretation when it comes to sources. Like you say, there is no actual sans-suffixe 'C-32', so when a source says the 'C-32' is X m long, how exactly should that be reflected here. Obviously C-32 is a super simple example, and probably most physical stats are the same for both A and B, so this is kind of an academic discussion on that example. For say the '747' though, different models have different lengths so if a source isn't clear on which variant is being detailed, do we really put it at the '747', or do we extrapolate and say they must mean the -400 so put it there, or what? Do we just assume any source that can't be bothered to be specific probably isn't a reliable enough source material to bother with? Most good data sources are kind enough to give an idea which model the data for but unfortunately USAF data sheets get a little lazy about this. I'm in the middle of moving so I'm divorced from my reference library for a while. Josh Baumgartner (talk) 02:19, 11 September 2023 (UTC)

@Joshbaumgartner: Well, when there's evidence that individual models don't all share certain measures, then there's really only two options for the family item: leave them off entirely and only apply to the specifically referenced model, or include them and add "applies to part (P518)" for the representative model. I think often it's honestly best to just leave those measures off the family item, especially those families that include models with significant deviations from norm, but I did just do the second thing the other day when adding masses to Apollo Command and Service Module (Q680027) (every one of those things were highly customized and had significantly different masses, so I picked Apollo 15 as "most representative" of the mature design). The trick I used to not overwhelm the property is to put applies to part (P518) in the reference. — Huntster (t @ c) 02:50, 11 September 2023 (UTC)

@Joshbaumgartner: Also, what do you think about including serial numbers under the total built property when those models have only super limited runs? I just realized after comparing some notes that the C-32A actually has 8 constructions rather than just the 4 that's often quoted, and I kind of want to list those 8 s/ns so others don't fall down the same confusion trap I did. — Huntster (t @ c) 02:52, 11 September 2023 (UTC)

Good points. I agree that it seems best if a family item is only a container for the models, and that the data, wherever possible, be at the model level. I hadn't seen applies to part (P518) used in the reference before, that's an interesting technique.

I was looking around, and found object named as (P1932) and object stated in reference as (P5997). One of these can be placed in the reference to specifically list how the source calls out the object, so maybe this could work:

• Boeing C-32B Gatekeeper (Q113682342)
  • length (P2043) > 47.32 m
    • Reference:
      • stated in (P248) > (source item) or other normal citation properties
      • object named as (P1932) > "C-32", or
      • object stated in reference as (P5997) > "C-32"

It seems like P1932 and P5997 were separately created to do the same thing probably without knowing the other one existed, so I'm not sure if one is better than the other. However, if this is included in the reference, I think there is zero issue, since we know that the data applies to C-32B, but the datasheet just says C-32 because well, they are aiming at an audience that probably rarely cares how a tri-service designation is parsed, and at the same time we aren't giving users the mistaken impression that source is explicitly claiming the C-32B specifically.

As for serial number use, I think that's fine idea. I think ultimately it is best if each instance (s/n) has its own item which is instance of (P31) model (ex. A-93 (Q119662088) for an old BuNo), but of course that is a long way off.

I note you added the popular name Gatekeeper to the C-32B. Did you just uncover this? I hadn't seen it before, so it's new knowledge for me and just wondering when they came out with that. Josh Baumgartner (talk) 04:11, 11 September 2023 (UTC)

@Joshbaumgartner: object named as (P1932) would be used as a property qualifier, and object stated in reference as (P5997) would be used in references, but yes, the latter is really redundant to the former. Otherwise, I think that's a fine way of going about things.

Unfortunately, while I agree that each airframe should have an entry here (since WikiData's original charge was documenting all the world's things), the deletionist movement has very rapidly grown here, and now those in leadership are pushing the idea that only notable or cross-linked items should exist. I firmly disagree with this notion in principle, but there's a very real issue of the Foundation running out of server resources (memory, not storage) rather quickly. There is a hope that implementing the MUL language type will alleviate some strain, though.

Gatekeeper has been an alias for a long time in the item, I just added it to the label to match general usage amongst other aircraft families and models. — Huntster (t @ c) 04:49, 11 September 2023 (UTC)

Sounds like P5997 is the one to go with. I don't like the deletionists getting ammunition for their cause, but I get that physical limitations do exist. I will say though, even when I add a whole batch of new items, I can tell by the Q numbers assigned that it is like drops in a hurricane, so even though we do have a lot of aerospace items, the volume of new data being added for some fields is orders of magnitude greater than anything we are doing.

Practically speaking though, while it may not be reasonable to add every F-16 or 737 serial to the database in this lifetime, I would think 8 C-32As to illustrate things would be entirely reasonable. Josh Baumgartner (talk) 16:10, 11 September 2023 (UTC)

Call for participation in a task-based online experiment

Dear Huntster,

I hope you are doing well,

I am Kholoud, a researcher at King's College London, and I am working on a project as part of my PhD research, in which I have developed a personalised recommender model that suggests Wikidata items for the editors based on their past edits. I am collaborating on this project with Elena Simperl and Miaojing Shi.

I am inviting you to a task-based study that will ask you to provide your judgments about the relevance of the items suggested by our model based on your previous edits.

Participation is completely voluntary, and your cooperation will enable us to evaluate the accuracy of the recommender system in suggesting relevant items to you. We will analyse the results anonymised, and they will be published in a research venue.

The experiment should take no more than 15 minutes, and it will be held next week.

If you agree to participate in this study, please either contact me at kholoud.alghamdi@kcl.ac.uk or use this form https://docs.google.com/forms/d/e/1FAIpQLSfA1wfdBfCRlcG3WhDyc-V8lzgPNx3fDFCNXkyn4CSwahXZ_A/viewform?usp=sf_link

Then, I will contact you with the link to start the study.

For more information about my project, please read this post: https://www.wikidata.org/wiki/User:Kholoudsaa

In case you have further questions or require more information, don't hesitate to contact me through my mentioned email.

Thank you for considering taking part in this research.

Regards Kholoudsaa (talk) 11:51, 6 October 2023 (UTC)

Atlas problems

Sorry about the Atlas stuff, its because the Atlas family was marked as a weapon due to its short career as an ICBM, and my bot ran with the idea for the variants. I spotted the problem and fixed the parent yesterday, but didn't notice its children had bern changed. Thanks for fixing Vicarage (talk) 05:22, 4 January 2024 (UTC)

@Vicarage: No worries! I figured that was what happened. I think everything's cleared up now. The Atlases were certainly interesting with their very mixed history. — Huntster (t @ c) 05:24, 4 January 2024 (UTC)

starfighter

It got changed in a big job because it was a "fighter", which is an aircraft type. I'll remove that Vicarage (talk) 18:09, 17 January 2024 (UTC)

@Vicarage: Again, no worries, it happens. — Huntster (t @ c) 18:40, 17 January 2024 (UTC)