User talk:Stinglehammer

From Wikidata
Jump to navigation Jump to search

This is my talk page.

Wikidata:Property_proposal/broader_concept[edit]

Pinging you (and others) since I suspect through Edinburgh University (and probably data from other orgs like RCAHMS [1]) there are various controlled-vocabulary thesauruses that you are probably familiar with, and could therefore give useful input as to whether or not in your view it would be useful to be able to record the "broader" field from thesauruses like this, on wikidata items matched with entries in such thesauruses, allowing one to reference the thesaurus structure in WDQS queries. Property proposal at Wikidata:Property_proposal/broader_concept. Jheald (talk) 19:44, 10 February 2018 (UTC)[reply]

helpful note[edit]

Hi, I noticed you might have to create multiple accounts for the event. I suggest asking at WD:BN for the account creator right. This allows you to create more than 6 per day. Artix Kreiger (talk) 15:30, 19 February 2018 (UTC)[reply]

Share your experience and feedback as a Wikimedian in this global survey[edit]

Hello! The Wikimedia Foundation is asking for your feedback in a survey. We want to know how well we are supporting your work on and off wiki, and how we can change or improve things in the future. The opinions you share will directly affect the current and future work of the Wikimedia Foundation. You have been randomly selected to take this survey as we would like to hear from your Wikimedia community. The survey is available in various languages and will take between 20 and 40 minutes.

Take the survey now!

You can find more information about this survey on the project page and see how your feedback helps the Wikimedia Foundation support editors like you. This survey is hosted by a third-party service and governed by this privacy statement (in English). Please visit our frequently asked questions page to find more information about this survey. If you need additional help, or if you wish to opt-out of future communications about this survey, send an email through the EmailUser feature to WMF Surveys to remove you from the list.

Thank you!

WMF Surveys, 18:57, 29 March 2018 (UTC)[reply]

Reminder: Share your feedback in this Wikimedia survey[edit]

Every response for this survey can help the Wikimedia Foundation improve your experience on the Wikimedia projects. So far, we have heard from just 29% of Wikimedia contributors. The survey is available in various languages and will take between 20 and 40 minutes to be completed. Take the survey now.

If you have already taken the survey, we are sorry you've received this reminder. We have design the survey to make it impossible to identify which users have taken the survey, so we have to send reminders to everyone. If you wish to opt-out of the next reminder or any other survey, send an email through EmailUser feature to WMF Surveys. You can also send any questions you have to this user email. Learn more about this survey on the project page. This survey is hosted by a third-party service and governed by this Wikimedia Foundation privacy statement. Thanks!

WMF Surveys, 01:40, 13 April 2018 (UTC)[reply]

185 University of Edinburgh employees, with no UID[edit]

# No UID
select ?item ?itemLabel { 
  ?item wdt:P108 ?employer .
  filter not exists {
    ?item ?wdt [] .
    ?wdt ^wikibase:directClaim/wdt:P31/wdt:P279* wd:Q18614948
  } 
  service wikibase:label {bd:serviceParam wikibase:language "en"}
} values (?employer) {(wd:Q160302)}
Try it!

Hope that's useful, Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:04, 14 March 2019 (UTC)[reply]

Witches and calendars[edit]

I saw your edit to Agnes Sampsoune (Q4693114) in which you added a redundant date of death, 28 January 1591 Gregorian. This contradicts the date already in the item, 28 January 1591 Julian.

I have read on Project chat about a big effort to clean up items about Scottish witches, and I have seen your user name, but I can't remember if you are part of the cleanup project. If you are, and you are going to be making extensive use of the source Survey of Scottish Witchcraft Database] it is essential that you learn their policies about stating calendar dates, that is, do they

  • always use Gregorian dates
  • use the calendar in effect in Scotland at the time of the event
  • something else?

When Scotland (along with the rest of the UK, including what would become the United States) switched calendars, Wednesday 2 September 1752 was followed by Thursday 14 September 1752. Along with skipping these days, there was a change of the date when the new year began, at least for legal purposes. In England the last time 25 March was treated as the beginning of the new year was in 1751; 1752 began on 1 January, and that custom has been followed ever since. Scotland had already switched its new year's day to 1 January in 1600. Historians typically use whichever calendar was in force, Julian or Gregorian, for the date they're writing about, but ignore the 25 March new year's day; they always treat 1 January as new year's day.

I have emailed J. Goodare of that project about these questions, I'll let you know if I find anything. If you can't find any calendar policy stated on the database website, you might try comparing some dates that are unequivocal from other sources and comparing them to the database's values. Jc3s5h (talk) 11:07, 2 August 2019 (UTC)[reply]

I have researched this further and put my results at Talk:Q4693114#Deprecation of date of death. Jc3s5h (talk) 13:15, 2 August 2019 (UTC)[reply]

Reply received from Prof. Julian Goodare. Prof. Goodare, one of the database editors, promptly replied to my email.

The calendar is that used in Scotland at the time, i.e. the Julian calendar. I doubt whether there are any post-1752 dates in the Survey, but if there are then they will be in the Gregorian calendar. As for the dating of the new year, this is taken to begin on 1 January throughout (even before 1600).

Jc3s5h (talk) 13:25, 2 August 2019 (UTC)[reply]

@Jc3s5h: Many thanks for looking into this and bringing this to my attention. Yes, myself and emmacarroll3 are working on a project to import and tidyup the data from the Survey of Scottish witchcraft database. Looks like OpenRefine has automatically set the dates to the Gregorian calendar on import but Emma has now taken note of this and will rectify accordingly. many thanks again, Stinglehammer (talk) 14:03, 2 August 2019 (UTC)[reply]
I'm glad I was given some hints about where to look for better sources. I still have doubts; I recall reading that the execution was on a Saturday but 28 January 1591 Julian proleptic calendar was a Thursday. Jc3s5h (talk) 14:23, 2 August 2019 (UTC)[reply]

OpenRefine tutorial: restricted access[edit]

Hi!

I just wanted to let you know that access to https://media.ed.ac.uk/media/OpenRefine+Beginners+Tutorial/0_y5bxsswq seems to be restricted to members of the University of Edinburgh. Thankfully the video is still available on YouTube though.

− Pintoch (talk) 08:04, 21 October 2019 (UTC)[reply]

@Pintoch: thanks for highlighting this.I've had a look at the permissions and it seems to be published openly. Can you check the link again to see if it is still hitting the Edinburgh Uni paywall please? many thanks Stinglehammer (talk) 12:57, 22 October 2019 (UTC)[reply]
Yes it works! Thanks a lot for fixing this, it's a great resource and I always try to avoid linking to Google services when I can. − Pintoch (talk) 14:28, 22 October 2019 (UTC)[reply]

Ritual objects and witchcraft trials[edit]

I noticed that you added a bunch of ritual object (P8706) to Agnes Sampsoune (Q4693114). I have not read the source you cited. Perhaps, in the source, it was clear in context these were alleged objects. But in the context of Wikidata, they are presented as objects that Sampson really used as ritual objects. I object to these additions. Jc3s5h (talk) 14:08, 25 November 2020 (UTC)[reply]

Argylls[edit]

(since this has dropped off PC bringing it here - @Jheald: as well)

So I've had a quick go at mocking up a couple of battalions 1st Battalion, Argyll & Sutherland Highlanders (Q111907264), 2nd Battalion, Argyll & Sutherland Highlanders (Q111907276). I think this model of independent items for battalions seems to work OK; we can definitely describe most of what we need to using part of (P361). I've gone with James's recommendation of keeping both "operational" and "regimental" links in the same property, but haven't added qualifiers for the "regimental" ones just yet.

The merging of the two battalions is shown by a merged into (P7888) link on the junior one, since it was merged "upwards" into the other, rather than replaces (P1365).

Some quick notes on what we can't currently represent:

  • There's no obvious way to indicate the "class" (for want of a better word) of the unit (these are both Regular Army) though I suppose we could do something with instance of (P31) to distinguish regular, territorial, reserve, etc.
  • Similarly there's no obvious way to indicate "role" where it served as a specialist unit, though I've experimented with it on the 1st A&S item.
  • The 2nd A&S ceased to exist in Feb 1942, and then in May the 15th A&S was redesignated the 2nd. Modelling both of these as the "real" 2nd makes sense (with a new item to cover the 15th up to redesignation) but there's no way to say "the battalion did not exist for a few months". This is a general problem with modelling concepts where an item has two distinct periods of existence, but it's not one we have a great solution for.

Let me know what you think. It's a little fiddly to make the battalions (especially if you want to tidy up brigades as you go...) but it's relatively straightforward to sort them out. Andrew Gray (talk) 19:38, 8 May 2022 (UTC)[reply]

Great thanks @Andrew Gray:, I think I need to bring Clare from the Museum in on this chat too. Just to see what the rest of their data could work with the data model. I'll email her and see if we can't all converge here. Stinglehammer (talk) 16:22, 10 May 2022 (UTC)[reply]
Sure - or happy to shift to email if that would be easier (will need to write all this up onwiki somewhere anyway). I guess the desired data in part depends exactly what they're wanting to use the items for! Andrew Gray (talk) 18:23, 10 May 2022 (UTC)[reply]
Great @Andrew Gray, Jheald: just chatted to Clare on MS Teams and she'd be very appreciative if you could help suggest ways of modelling the data on regiments and battalions but feels email might be better in first instance. She's going to email me with (1) a query issue and (2) how she'd like to model the data and if possible I'd like to email you both to get your thoughts on how she could progress. Stinglehammer (talk) 10:49, 18 May 2022 (UTC)[reply]
Sounds good - email away. Andrew Gray (talk) 20:09, 18 May 2022 (UTC)[reply]

Did a few more of these tonight.

As before some apparent gaps in the model - needs a bit more of a complicated vocabulary to use in end cause for what happened to them, and there is not an obvious way to tie 2/ and 3/5th back to their parent units. But on the plus side, official name (P1448) to handle the nuances of naming seems to work pretty well. Getting somewhere. Andrew Gray (talk) 22:26, 22 May 2022 (UTC)[reply]

Deletion of superfluous witch trial items[edit]

Hi, I started to delete a couple of these items you requested at RfD. I noticed that some are linked from items on the accused witches, for example Margrat NcIllduy (Q43391790). Are those items to be kept? Lymantria (talk) 07:10, 29 July 2022 (UTC)[reply]

Hi @Lymantria: yes very much so to be kept. The accused witch items should all be retained. The deletion request relates to only a very specific set of 803 items in total which were created 3 years ago as if they about witch trials but it turns out, on further investigation, that they are merely "mentioned in trial" items i.e. just the recording of the act of being named or mentioned in a witch trial. This is of course useful information but it can be modelled very differently through the links between accused witch items and actual witch trial items. Hope that helps. Stinglehammer (talk) 09:50, 1 August 2022 (UTC)[reply]
Yes, Thank you. Lymantria (talk) 19:43, 1 August 2022 (UTC)[reply]

Could you try to delete the links (batch mode)? It’s tedious to delete hundreds of items if you have to fix every incoming link first … --Emu (talk) 10:50, 16 August 2022 (UTC)[reply]

Hi @Emu:, yeah what's the best way to get these items deleted in your view? You meaning it would be better if they had all the statements removed via QuickStatements first?? Stinglehammer (talk) 15:07, 16 August 2022 (UTC)[reply]
@Stinglehammer Take for example Q43403198. There are two incoming links from Bessie Friece (Q43391033) and witchcraft investigation of Bessie Friece (Q112914798). To do the job properly, an admin first has to delete the incoming links (that is, edit Bessie Friece (Q43391033) and witchcraft investigation of Bessie Friece (Q112914798)) and then delete Q43403198 – no need to change or delete statements in Q43403198). It’s a lot easier if the admin just has to delete the item itself, so I wonder if it might not be wise to come up with a way to identify incoming links and delete the statements from those items. I hope I managed to explain it well enough :-) --Emu (talk) 17:27, 16 August 2022 (UTC)[reply]

Query on Scottish Witchcraft Trial data and consistency[edit]

Recently I was trying out some SPARQL queries which utilise the Scottish Witchcraft Trial data here. My fear is that I may be merely reinventing the wheel, using my limited SPARQL to create diagrams which may already have been covered in the original 2018 exercise, though lacking an Edinburgh Uni login I can't see the visualisation pack from that time. Anyway, I have a broader question: what is the ongoing situation nowadays regarding the data in both the original MS-Access database and in Wikidata? Are there processes for ongoing refresh, for resolving emerging contradictions in data adjusted in each place, etc.? AllyD (talk) 20:06, 1 October 2022 (UTC)[reply]

Hi@AllyD:, apols for the delay in responding. Been off work sick with a lurgy. Just to check, what aspects of the data are you interested in? Not sure what you mean by 'visualisation pack'. There is the original MS Access database and our website witches.is.ed.ac.uk and our Wikidata guide at thinking.is.ed.ac.uk/wikidata-workshop which has a page about the witchcraft work. The MS Access database website is in need of migration, probably to a new MySQL database but we'll maintain all the links to/from wikidata. and we have a MS Form for anyone who wishes to contact us esp. if noticing anomalies/contradictions in the data. That's a big job so doing some QA ourselves but it really needs more eyes on it to help us notice the issues. Hope that helps.Stinglehammer (talk) 08:37, 11 October 2022 (UTC)[reply]
The previous data visualisation I'm wondering about is this one, if it (or its underlying queries) are available anywhere outside the Uni wall.
I have a few emerging data additions such as Marjorie / Marion Thomsone and Margaret Kentestone in the 1643 Culross Kirk records. It is possible they appear elsewhere under variant names as they had (not unreasonably) fled Culross. Once I have checked a bit more, I will probably feed these through on the online form. AllyD (talk) 08:19, 14 October 2022 (UTC)[reply]
I have avoided adjusting these Wikidata items, apart from fixing one obvious typo in a description, but what I am wondering is whether Wikidata has any cautionary processes in place for dealing with situations on imported data where the original source is still actively updated? Unless there is a Wikidata->Access feed in place as well as Access->Wikidata updating, should there be some provision for flashing up a warning when editing Wikidata if updates are better applied to the other datastore, to avoid data inconsistencies? AllyD (talk) 08:51, 14 October 2022 (UTC)[reply]
Hmm, that's a big question. I've worked with two interns importing the data from the Survey of Scottish Witchcraft from 2019-2022. Certain elements of the data are easy to check on (number of the accused, number of trials etc.) but as open data I can't prevent others editing. Maybe monitoring all ~10,000 items in my watchlist. Doesn't seem easy task. Recruiting new intern to help me quality assure the data before it goes live on the updated website so we can have faith that the data viz are accurate. Original Survey site is being mothballed so we need to migrate its data to a new MySQL site by March 2023 and preserve links from Wikidata to it which I'm assured will happen. But Survey data has some inaccuracies/eccentricities so there is some increased interest in students improving it - which then has implications for ensuring Wikidata keeps pace with these changes too. I'll discuss with my manager anyway but it's maybe also one for @LydiaPintscher: and my other Scottish historical data colleague @James nayler is real: to think about whether some data quality options or security/warnings/flags are appropriate. Thanks for highlighting Stinglehammer (talk) 12:55, 20 October 2022 (UTC)[reply]
Thanks for copying me in @Stinglehammer. I don't have a straightforward answer to this issue. The situation with our work is a bit different in that we're actively building the dataset on WD, rather than importing an existing database. The problem with our approach is we don't have any authority IDs and are relying solely on detailed references to manuscripts & other documents to verify the datapoints. Users have to trust us that the info taken from the manuscripts is accurate, especially as most of our users aren't able to check and re-check the manuscripts themselves.
It does make me wonder about being able to develop batches of items in a sandbox of sorts -- thereby allowing the whole lot of data to be flagged en masse as 'in progress'.
There are so few of us (relatively speaking) working on this stuff that we're in desperate need of guidance for best practice. We try to blog about our processes, but not sure how much traction it gets (especially when we're having to explain WD for the first time to many in our audience). James nayler is real (talk) 08:58, 27 October 2022 (UTC)[reply]

Use of 'part of'[edit]

Please stop misusing part of (P361) to point items to your wikiproject. All sorts of geolocatable items in Scotland are not part of your project. The two properties established to link items to wikiprojects are maintained by WikiProject (P6104) and on focus list of Wikimedia project (P5008). I beg to suggest that the second of these may be the more appropriate for you.

I have moved all such P361 statements to P5008. --Tagishsimon (talk) 08:46, 26 April 2024 (UTC)[reply]

The other thing, whilst I'm here. If you want to add a statement for the location of an item, such as the Dornoch Witch Stone (Q123249315), please use the location (P276) property. The located in the administrative territorial entity (P131) property is for 'administrative territoral entities', per its label, and should be used for the council area in which the item is found. It is extremely unhelpful and extremely disruptive to use the wrong properties for statements. --Tagishsimon (talk) 08:55, 26 April 2024 (UTC)[reply]
Retrieved from "https://www.wikidata.org/w/index.php?title=User_talk:Stinglehammer&oldid=2137583768"