Wikidata talk:WikiProject Periodicals/Archive 1
This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Periodical series
Im sorry if this question has been raised elsewhere. I would like to know how to best record that Trends (Q787045) is a series. I have use 'part of' for all the journals in the series (see Special:WhatLinksHere/Q787045). Do we need a new item called 'journal series', which is a subclass of brand (Q431289)? Another example is BMC journals (Q4835951) (See enwiki list), but there are many of these series which have an article on Wikipedia, as English Wikipedia editors have often wanted an article to link to, but not wanted to write a new stub for each journal in the series. John Vandenberg (talk) 10:03, 2 December 2013 (UTC)
- Some more examples are Chemische Berichte (Q902474), Journal of the Chemical Society (Q903605), and Proceedings of the Royal Society (Q1193201). John Vandenberg (talk) 10:23, 3 February 2014 (UTC)
- More examples: Perkin Transactions (Q7169012), Environment and Planning (Q2813257), Comptes rendus de l'Académie des sciences (Q2667573), Acta Crystallographica (Q343155). John Vandenberg (talk) 02:34, 4 February 2014 (UTC)
- More examples: American Journal of Medical Genetics (Q4744254), International Journal of Modern Physics (Q1666666) John Vandenberg (talk) 02:55, 6 February 2014 (UTC)
I am thinking of giving all these an instance of a new item 'academic journal imprint', which would be a subclass of imprint (Q2608849). Other periodical types would need their own subclass. Requesting feedback. John Vandenberg (talk) 02:21, 8 February 2014 (UTC)
- Sorry, I have no idea how to solve this problem, but I think we shouldn't invent new terms. Imho there would be no good translating for "academic journal imprint", since the term "imprint" is used in many languages (probably with different connotation). --Kolja21 (talk) 03:31, 8 February 2014 (UTC)
- Thanks for replying. We need a structural element - I don't mind what we call it, and the name can change later. The above examples are a publishing brand more than anything else, but maybe using 'imprint' is too inventive ;-) e.g. Trends (Q787045) should have a property 'imprint'=>Cell Press (Q5058164), and publisher=>Elsevier BV (Q746413), but we dont have an 'imprint' property.
- "Trend journals" are commonly called the "Trend journal series", I don't like "journal series" as a generic name, the word 'series' is too ambiguous. e.g. Proceedings of the Royal Society A (Q10354104) is often called "Proceedings of the Royal Society of London. Series A", so it would be confusing to call Proceedings of the Royal Society (Q1193201) a series. I'll ask for help from w:WP:AJ. John Vandenberg (talk) 04:28, 8 February 2014 (UTC)
Ah, here is a good quick fix, even if only a interim/temporary solution. We could call them a subclass of (P279) 'academic/scientific journal' rather than instance of (P31) 'academic/scientific journal'. That is what user:FelixReimann did with BMC journals (Q4835951). We would still need to define what they are an instance of, but that isnt an urgent problem. John Vandenberg (talk) 04:48, 8 February 2014 (UTC)
- Perhaps DGG at en:WP has some helpful suggestions here, being a former academic librarian. --Randykitty (talk) 16:23, 10 February 2014 (UTC)
- You do not need necessarily both, subclass of (P279) and instance of (P31). In this case, the class scientific journal (Q5633421) has subclasses BMC journals, Trends, .... Imprint is IMHO not correct, as the publisher remains BioMed Central or Cell, respectively. If we find an item "academic journal series" or whatever, we could at it as target of instance of (P31) but this is not necessary. — Felix Reimann (talk) 10:16, 11 February 2014 (UTC)
- Another option is to use part of the series (P179). John Vandenberg (talk) 12:07, 28 February 2014 (UTC)
- No as the different BMC journals for example do not have an order. P179 is for ordered things (A,B,C,D,...). — Felix Reimann (talk) 16:48, 28 February 2014 (UTC)
- I think it is important to realize that there is a difference between series like Compte rendus, where we have one single journal with either temporally-subsequent series or parallel series with different subjects, and "series" like Trends journals and BMC journals, which are independent journals that are similarly branded by the publisher. --Randykitty (talk) 20:18, 28 February 2014 (UTC)
Part publication of a journal article
Our Fragile Intellect (Q7110639) has a English Wikipedia page, but the publication was split into two articles, with two different DOIs, albeit published back-to-back in the same issue of the journal and on the same date (both electronic and physical). We definitely need an item for 'work' that comprises both articles, as that is the level that 'visible' to the real world, but the academic world will cite one or the other. Do we need each part to have a separate item? John Vandenberg (talk) 10:03, 2 December 2013 (UTC)
Abbreviations
http://jabref.sourceforge.net/ includes some journal abbreviation lists, such as 12. John Vandenberg (talk) 13:15, 5 December 2013 (UTC)
- ISO abbreviations can also be found here. Randykitty (talk) 04:57, 6 December 2013 (UTC)
- We might need an "abbreviation" or "short title" (baroque literature) property: Wikidata:Requests for comment/How to deal with given names and surnames. --Kolja21 (talk) 18:53, 10 February 2014 (UTC)
Template:Cite doi subpages
I've raised a question about 'Cite doi' templates at Wikidata:Project_chat/Archive/2014/01#Template:Cite_doi_subpages. --John Vandenberg (talk) 17:17, 30 December 2013 (UTC)
Original title and language for a newspaper, magazine еtс
Are there properties for a original title and language? -- Sergey kudryavtsev (talk) 09:15, 17 January 2014 (UTC)
PS: w:ru:Шаблон:Журнал uses original language of film or TV show (P364) or language of work or name (P407) in sence of a original language. -- Sergey kudryavtsev (talk) 10:54, 17 January 2014 (UTC)
- @Sergey kudryavtsev: Please use P357 (P357) and original language of film or TV show (P364). --Kolja21 (talk) 07:07, 4 February 2014 (UTC)
Its are not listed in Wikidata:Periodicals task force#Venue item properties, only for a article. I add P357 (P357) and original language of film or TV show (P364) for sake of clarity. -- Sergey kudryavtsev (talk) 12:47, 4 February 2014 (UTC)
Conference proceedings and preprint repositories
Are conference proceedings and preprint repositories within the scope of this proposal? --DarTar (talk) 06:25, 18 January 2014 (UTC)
- Definitely yes for conf-pubs. Pre-prints repos which perform like serial publications would also be within scope, as would pre-print repos which contain a high proportion of outputs which were properly published in periodical. Do you have any specific repo you are interested in? John Vandenberg (talk) 07:12, 3 February 2014 (UTC)
- ACM Multimedia (Q288546) is an example of one that is both a conference and a journal, and a SIG. Do we split it into three items? John Vandenberg (talk) 02:37, 4 February 2014 (UTC)
JVbot uploading sources
JVbot 1 is nearing completion. JVbot 2 needs feedback (please be critical). John Vandenberg (talk) 07:16, 4 February 2014 (UTC)
Classes for periodicity, periodicity property ?
Hi, I created monthly magazine (Q15709838) for monthly magazines to be an instance of. Have we a relevant corresponding periodicity property to note the publication frequence for periodical publications and to define better this class ? TomT0m (talk) 22:24, 4 February 2014 (UTC)
- We do not yet have a
frequency
/periodicity
property. It is a generic property, as it applies to at least Infobox journal/magazine/newspaper and also sporting events, and awards. John Vandenberg (talk) 10:38, 5 February 2014 (UTC)- still no property for frequency/periodicity ? this is so essential for periodicals :/ --Hsarrazin (talk) 09:33, 1 June 2016 (UTC)
Invalid ISSNs
The Independent (Q1712632) has an invalid ISSN, because the paper was printed with an invalid ISSN *headdesk* (if the source provided is to be believed). There will be many of these; how do we want to represent these? If they are placed in the ISSN field, they will fill up the ISSN validation report: Wikidata:Database_reports/Constraint_violations/P236. John Vandenberg (talk) 11:00, 15 February 2014 (UTC)
- Same problem with ISBN, see en:Template:Listed Invalid ISBN. We should add a qualifier. --Kolja21 (talk) 23:14, 15 February 2014 (UTC)
ERA journal lists imported
The ERA 2010 journal list (Q15735759) and ERA 2012 journal list (Q15794938) have been imported, and in the process many magazines & newspapers from the Wikipedias have been imported (but not all of them). Wikidata:Database reports/Constraint violations/P236 hasnt been updated for a few days, so I cant see how many erroneous values have been imported in the process. Over at Property talk:P236#Unique constraint, I have recommended that we remove the Unique constraint.
We now have 40539 issns in Wikidata, and 22498 eraids, which means our academic journal database is probably around 25000-30000 items.
In the process I imported some LCCNs, which has resulted in quite a few constraint violations on Wikidata:Database reports/Constraint violations/P1144. Some are badly formatted due to insufficient importing parsing logic, but that property also has a constraint requiring LCCNs are only added to books. Guidance and/or assistance much appreciated on this issue.
Scopus source ID (P1156), CODEN (P1159), ISO 4 abbreviation (P1160), Z39.5 abbreviation (P1161) & Bluebook abbreviation (P1162) have been approved, so I will be attaching Scopus IDs & abbreviations to our journal collection next. John Vandenberg (talk) 01:47, 25 February 2014 (UTC)
April import update
I have imported all of the Danish Bibliometric Research Indicator (BFI) SNO/CNO (P1250) data for journals which have the same ISSNs are journals on the ERA list. i.e. existing items only. I havent added the Danish ID to records which do not also have a ERA Journal ID (P1058) (I will look at the other half of the Danish list next month). See previous discussion at User talk:Fnielsen#Journal aliases and Scandinavian authorities and @Fnielsen:. This has allowed me to compare journal names common to both lists. The majority of the journal titles were identical, or slightly formatted differently, and those were imported first. There were roughly 1000 which had significantly different journal names, or were otherwise skipped by my title name comparison rules. Some were title corrections, others are common aliases, but there are also many interesting ones like journal renames and names in other languages - usually the original language. The English labels were copied to languages Finnish, Danish, Swedish, Norwegian (both) and French, if they didnt already have a label in those languages.
Here are the edits to the last 1000 items.
For non-English journals, it would be great if people could add original language of film or TV show (P364) to the journal, and language of work or name (P407) to the P357 (P357). I can use that to move the appropriate title into the labels of languages other than English.
We need to start establishing label preferences for each language. i.e. Wikidata house rule (Q455282)s. Should English labels use transliterated titles in preference to original language title? (Both Australian and Danish journal lists use ASCII titles). Do we remove Die, Der, El, La, L' prefixes etc? Do we use '&' instead of et, und, etc? Do we want to shorten labels by using '[Organisation]. Journal' instead of 'Journal of the [Organisation]'. When do we use an abbreviation vs including the subtitle, e.g. T C (Q15762478) can be 'TC', 'T C' or 'TC: a journal of biblical textual criticism'.
Then we need title language fallback strategies. e.g. for German journals, e.g. should Danish labels use the German or English title if both are available? For German titles, does Danish prefer the original (i.e. 'für'), or a transliteration (e.g. 'fur' or 'fuer')? The Danish journal list prefers 'fuer' whereas the Australian list usually uses 'fur'!
There are a few Danish Bibliometric Research Indicator (BFI) SNO/CNO (P1250) constraint violations listed at Wikidata:Database reports/Constraint violations/P1250. Many will be the same violations as documented as exceptions at Property talk:P1058.
There are additional ISSN (P236) violations at Wikidata:Database reports/Constraint violations/P236.
The Danish academic book publisher property was rejected at Wikidata:Property proposal/Authority control because only one person supported it. There are a few journal authority control proposals on that page which I fear will soon be rejected unless people vote for them. There are a number of journal property proposals on Wikidata:Property proposal/Creative work which also only have 'one support'. :/ The 'indexed in' property for journals at Wikidata:Property proposal/Creative work#indexed in is especially important in my opinion, and it would allow journal database ID properties (like the Scopus source ID (P1156)) to be moved into qualifiers instead of being a claim on the data item. John Vandenberg (talk) 08:49, 21 April 2014 (UTC)
Proposed property: copyright license
See Wikidata:Property_proposal/References#copyright_license. --Daniel Mietchen (talk) 14:55, 9 May 2014 (UTC)
- For the record, the discussion determined that copyright license (P275) should be used. It would be great if a bot could import this data.
- @Daniel Mietchen: I think open access policy is a property which does not yet exist. It would be a good idea to model it here at the task force first, as there are a lot of strange cases to be considered. John Vandenberg (talk) 01:02, 24 May 2014 (UTC)
Multiple language journals
Global Media Journal (Q11699954) is a fun one. It has ten different language/region editions which are all interwoven in content. Wikipedia will probably only ever have one article for them all. The US/English edition was recently merged by user:Alan ffm into the item for the Polish Wikipedia article, which of course has the Polish edition ISSN and other infobox parameters.
Other simpler cases are Angewandte Chemie (Q538683) and Chinese Physics B (Q3128474) (English language)/ Acta Physica Sinica (Q2200957) (Chinese language). John Vandenberg (talk) 00:56, 24 May 2014 (UTC)
Property with type item where item has no entry
What should we do in the situation where, for instance, an article has an author, but that author has no wikidata entry?
Options that could be done now:
- Create a wikidata entry for the entity
- Either wouldn't link to any wiki page (bad), or
- Could make a wikisource entry; wikisource has a special namespace for authors (i.e. s:Author:Russell_Thorndike)- can orphan pages be created on wikisource?)
- Use a separate author property that takes strings (potentially the least disruptive to wikidata). There is some precedent for this as there are already specialized "author citation" properties that take strings (see Q788451 ). However, this will result in messy citations and possible duplicates if authors become entities at some point.
Options that would require not-yet existing functionality:
- Modify wikidata to allow "red-linking" within i.e. you would type in a name, which wouldn't actually point to a wikidata entry (probably involves major changes to wikidata internals and could be very problematic).
Mvolz (talk) 2014-07-20T15:20:21
- A generic response to a generic question..
- Creating an author page on Wikisource is a very good solution. It may be an 'orphan', but it should list at least one work which is 'free' (public domain, etc). Notability isnt a serious issue at Wikisource. See s:Help:Author pages for precise eligibility rules.
- If the author is modern and all works are covered by copyright, and they are notable, creating a stub on a Wikipedia is also good.
- String author properties are the anti-thesis of Wikidata, as we want the data rather than one representation of the data. There are many citation formats, and they specify how the name should be formatted.
- 'Redlink' data item is a very interesting idea, but wayyyy beyond the scope of this WikiProject ;-).
- If the author isnt notable, and has no 'free' works, create a Wikidata item, and add a note on the item talk page to say that you believe the author is not (yet) notable enough for a page on any of the Wikimedia projects. Wikidata inclusion policy explicitly allows non-notable subjects if the non-notable item is not an orphan. i.e. On Wikidata, notability is inherited! John Vandenberg (talk) 00:25, 21 July 2014 (UTC)
- "Either wouldn't link to any wiki page" I suppose that if the article has a Wikidata item that Wikidata item would link to the author. I was not aware of the inheritance of notability on Wikidata. I have extended Equation of State Calculations by Fast Computing Machines (Q5384234) and was not sure what to do with the two Wikipedia-non-notable authors on that paper, but as I understand on John Vandenberg it is not a problem to have items for the two missing authors on Wikidata. — Finn Årup Nielsen (fnielsen) (talk) 16:41, 21 July 2014 (UTC)
For journal articles we have instance of (P31) scientific publication (Q591041) and genre (P136) scientific publication (Q591041). I wonder if that is ok or if something else should be done. — Finn Årup Nielsen (fnielsen) (talk) 16:54, 21 July 2014 (UTC)
- @Fnielsen: Journal articles should be instance of (P31) article (Q191067), see Help:Sources. --Kolja21 (talk) 14:29, 20 August 2014 (UTC)
Launch of WikiProject Wikidata for research
Hi, this is to let you know that we've launched WikiProject Wikidata for research in order to stimulate a closer interaction between Wikidata and research, both on a technical and a community level. As a first activity, we are drafting a research proposal on the matter (cf. blog post). It would be great if you would see room for interaction! Thanks, --Daniel Mietchen (talk) 01:48, 9 December 2014 (UTC)
New proterty proposal
See the proposal for "RSL periodical identifier" at Wikidata:Property proposal/Authority control. -- Sergey kudryavtsev (talk) 13:58, 8 May 2015 (UTC)
Newspapers, number of copies, and number of editions
Défense de la France (Q21528945) has been printed 47 times. How do we express that ? author TomT0m / talk page 15:04, 22 November 2015 (UTC)
Wikimania 2016
Only this week left for comments: Wikidata:Wikimania 2016 (Thank you for translating this message). --Tobias1984 (talk) 11:49, 25 November 2015 (UTC)
Scholia
I have created a small website with aggregation of scholarly publication metadata from Wikidata. I can give an overview of "venues" (journals, collections), publishers and authors. For instance, with https://tools.wmflabs.org/scholia/publisher/Q180419 you are able to see journals published by Nature Portfolio (Q180419) and that a lot of main subject (P921) could be set. — Finn Årup Nielsen (fnielsen) (talk) 09:28, 28 October 2016 (UTC)
WikiCite 2017 applications open through February 27
WikiProject Periodicals has more than 50 participants and couldn't be pinged. Please post on the WikiProject's talk page instead.
(apologies for cross-posting) Hey all, we just announced that applications for WikiCite 2017 (Vienna 23-25 May, 2017) are open. WikiCite 2017 is a 3-day conference, summit and hack day to be hosted in Vienna, Austria, on May 23-25, 2017. It expands efforts started last year with WikiCite 2016 to design a central bibliographic repository , as well as tools and strategies to improve information quality and verifiability in Wikimedia projects. Our goal is to bring together Wikimedia contributors, data modelers, information and library science experts, software engineers, designers and academic researchers who have experience working with citations and bibliographic data in Wikipedia, Wikidata and other Wikimedia projects. Members of WikiProject Periodicals are among the most relevant and knowledgeable communities in Wikidata on the topic of structured data for sources and it would be fantastic to see you in Vienna. We have (limited) travel funding available, if you're interested in participating please consider submitting an application. This year's event will be held at the same venue as the Wikimedia Hackathon and we'll be able to accommodate up to 100 participants. Any questions? Get in touch with the organizers at: [email protected] --DarTar (talk) 18:22, 11 February 2017 (UTC)
How do we indicate that a periodical is a member of a list?
How do we indicate that a periodical is a member of a list (listed on a list)? For instance, "Nature Index", seen here https://www.natureindex.com/faq#journals, is a small list of scientific journal. How can we indicate that, e.g., "American Journal of Human Genetics" is listed there? — Finn Årup Nielsen (fnielsen) (talk) 15:01, 17 August 2017 (UTC)
How to deal with journal issue
Hi WikiProject Periodicals has more than 50 participants and couldn't be pinged. Please post on the WikiProject's talk page instead.,
I tried to put data on Charlie Hebdo issue No. 1178 (Q18763302) but I'm unsure how to do it.
What should we use for instance of (P31) ? I used version, edition or translation (Q3331189) (like on the FRBR pattern of Wikidata:WikiProject Books) but maybe there is something better.
How can I indicate the number of issue printed (7.95 million copy, highest ever for the French press), with quantity (P1114) ?
Do you see other data to add and how ?
Cdlt, VIGNERON (talk) 11:17, 10 January 2016 (UTC)
- @VIGNERON: Very good question. I have wondered the same thing. I don't think version, edition or translation (Q3331189) suits well since periodicals have completely different content in each edition in contrast to a book that evolves more slowly. Perhaps we need to create a new specific item for these kind of objects? Ainali (talk) 17:44, 9 August 2017 (UTC)
- I believe issue (Q28869365) is better. Ainali (talk) 20:07, 9 August 2017 (UTC)
- @VIGNERON: part of the series (P179)? Example on Proceedings of the Biological Society of Washington, volume 18 (Q30541626). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:50, 29 September 2017 (UTC)
Last issue
Which property to use to specify the last issue of a magazine/periodical? John Samuel 15:48, 2 September 2017 (UTC)
- @Jsamwrites: Do you mean "final issue", or just "most recent issue"? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:52, 29 September 2017 (UTC)
- @Pigsonthewing: The "final issue". John Samuel 17:03, 29 September 2017 (UTC)
Journal/Periodical data from "KBPlus"
Over the last few years I have worked with a UK based service called KBPlus/KB+ run by Jisc (Q6269240). This collects information about journals/periodicals that are subscribed to by UK Universities. It publishes lists of journals under a CC0 licence. The data made available typically includes ISSNs (and sometimes other identifiers), a textual note on relationships between journal titles (e.g. "preceded by: old title, continued by: new title"), plus information on where you can subscribed to it (or if it is open access). My involvement with the project has been on the technical side rather than on the data side, but I know they have a small team of staff who are responsible for ensuring the accuracy of the data in the system, so I have confidence that the data in the system is well maintained and generally accurate.
I'm wondering if there would be any interest in using the data from KB+ to either create new records in Wikidata (where the journal doesn't already have a wikidata entry), or adding data to existing wikidata entries. I've done some initial work and KB+ holds information on approximately 268.5k journal titles. I've matched about 48k of these to existing wikidata entities based on the ISSN.
Is there any interest, and if so any advice on the best approach? Owenpatel (talk) 11:43, 5 October 2017 (UTC)
Is the state of the description for issues outdated
In the isse section it says to use P364 but the examples use P2439 instead (which makes more sense to me).
It also states to use properties P433 and P478 but in the examples P478 is use as a qualifier for P433. The latter is not useful for issues where the it is unknown when the first one got out. I also found examples where P478 is a qualifier for P361 which somehow makes more sense to me.
I'd be really nice if someone with a bit more experience could explain that for me or give me some pointers. Ogmios (Tratsch) 17:12, 3 November 2017 (UTC)
Importing all articles from the EFSA journal
The EFSA journal is an open-access publication by the European Food Safety Agency. It has periodic evaluations of food additives, and it would thus be very useful to link those articles to the food additives that exist in Wikidata, paving the way for sourced recommandations in outside projects like Open Food Facts.
http://onlinelibrary.wiley.com/doi/10.2903/j.efsa.2017.4787/full
I have no idea how the import of periodicals is done in Wikidata. Is that a request for titles ? Does it require tedious work ?
Teolemon (talk) 19:28, 7 November 2017 (UTC)
Topic: main subject (P921)?
Why do we not have a indication of how the topic should be added? I suppose that main subject (P921) would be the obvious, but I note here and there are used field of work (P101), see, e.g., [1]? — Finn Årup Nielsen (fnielsen) (talk) 11:32, 30 May 2018 (UTC)
After some thinking out loud at Project Chat as to the correct understanding of annual collection (Q19816504), and its relationship to volume (Q1238720), I have made the following changes:
- Changed the English label of Q19816504 from "volume" to "annual volume"
- Made annual collection (Q19816504) a subclass of volume (Q1238720)
- Added annual collection (Q19816504) to the Volume item properties table as an explicit alternative choice of instance of (P31) for a periodicals volume -- typically, for a volume that doesn't have a volume number, but is referred to by year.
I hope these changes make sense, and are acceptable. Jheald (talk) 13:06, 1 June 2018 (UTC)
Biodiversity Heritage Library
A few weeks ago Magnus's Reinheitsgebot created Wikidata items for about 60,000 titles from the Biodiversity Heritage Library. These items include all sorts of things -- books, periodicals, catalogues, individually bound article reprints, technical reports, etc -- but also quite a few periodicals.
About 2000 of the items, that can be retrieved with this query: tinyurl.com/yby29epl
were tagged with keyword "Periodicals" at BHL, and I have made instance of (P31) periodical (Q1002697).
There are also some items which did not have the 'periodicals' tag, but nevertheless do appear to be periodicals, based on their having a large number of volumes, eg this list of about 180 with more than 10 volumes: tinyurl.com/y9nlvkrg
(I didn't catch this in time, before creating huge numbers of links to the Internet Archive for them, one for each volume. Per the design recommendations on the front page here, these probably ought to be split out into separate items for each volume).
There's also a further set of items that had multiple volumes, numbered by year. I'll try to tag these and add a query for them soon.
There are a few issues with these items:
- Many of them are probably duplicates of items we already have for periodicals.
- I see that User:PieterJanR has caught and merged quite a few, but I suspect there may be quite a lot more still to do. I am not sure what the best work-flow for identifying these would be.
- UPDATE: Thanks to User:Rdmpage showing me where to get an (incomplete) list of ISSNs out of BHL, here's a query for some of the merge candidates, that match a publication with an existing ISSN:
tinyurl.com/yatr5upl
Jheald (talk) 17:46, 8 June 2018 (UTC)
- UPDATE: Thanks to User:Rdmpage showing me where to get an (incomplete) list of ISSNs out of BHL, here's a query for some of the merge candidates, that match a publication with an existing ISSN:
- A lot of the items could use a better instance of (P31), but it's not clear to me what it should be.
- For example, what should The Garden: an illustrated weekly journal of gardening in all its branches (Q51481557) be? Just a magazine (Q41298), or something more specific? While Proceedings of the Bournemouth Natural Science Society (Q51522324), a proceedings series of a local amateur society, would seem not to be a scientific journal (Q5633421) or a proceedings (Q1143604); but what should it be identified as?
- Some sort of clean-up project is going to be needed; but before that, a good guide to help classification might be useful.
- Inevitably there are a lot of further statements that probably could/should be added.
- A first iteration of a dashboard, with counts of the statements currently on the items, can now be found at Wikidata:WikiProject BHL/Progress:Titles, which may inspire people as to what can be added.
- Please do add yourself to the #Participants section at Wikidata:WikiProject_BHL#Participants if you are interested! Jheald (talk) 15:10, 8 June 2018 (UTC)
Values for distribution format (P437)
It would be useful to agree some values for distribution format (P437), in particular how to represent eg "print" and "online" as a qualifier on ISSN (P236).
IMO, existing values aren't particularly useful for this:
- direct values
tinyurl.com/y8ayv9ea
: paperback (Q193934) (2), softcover (Q990683) (5), Portable Document Format (Q42332) (1), retail (Q126793) (1), subscription business model (Q322948) (1), printing (Q11034) (1) - qualifier values
tinyurl.com/y7cynrtg
: CD-ROM (Q7982), web page (Q36774) (2), print (Q11060274) (work of art) (2).
I am tempted to go with printed matter (Q1261026) and digital distribution (Q269415) as possibly the best options at the moment, but am happy to change if we were to find (or create) better choices. Jheald (talk) 09:36, 15 June 2018 (UTC)
- online publication (Q1714118) may be better for online. I'll change qualifiers using digital distribution (Q269415). Jheald (talk) 17:41, 19 June 2018 (UTC)
Start time / End time as qualifiers
I find that it is quite often that I want to specify a range of years or volumes as a qualifier -- for example: under an external ID, to indicate that a particular date-range of the journal is available with that ID, or that the institution uses that ID for a particular date-range, perhaps with another ID for another date-range.
Is it appropriate to use start time (P580)/end time (P582) for this? Or would that be interpreted that this a former ID for the journal, that was used between those years, but which has now been replaced (for all copies) by the new one? Jheald (talk) 11:42, 15 June 2018 (UTC)
- In my view the use should be okay, because if an identifier has actually been withdrawn, it should be marked reason for deprecated rank (P2241) = withdrawn identifier value (Q21441764), and deprecated. Jheald (talk) 11:51, 15 June 2018 (UTC)
Importing articles on cosmetic science
We're trying as part of Open Beauty Facts to provide more insights on cosmetic ingredients, and a big part of that would be to be able to indicate research on various cosmetic ingredients. Would it be possible to import article metadata from those journals into Wikidata, so that we can link the Article items them to Ingredient items? The 3 major journals seem to be:
The International Journal of Cosmetic Science https://onlinelibrary.wiley.com/journal/14682494
The Journal of Cosmetic Dermatology https://onlinelibrary.wiley.com/journal/14732165
Cosmetics (Open Access) http://www.mdpi.com/journal/cosmetics
On the Open Beauty Facts side, our goal is eventually to let you scan your beauty cream, and provide you contextualized open data about it, including the titles of those articles (which often highlight risks and benefits)
Teolemon (talk) 09:01, 1 July 2018 (UTC)
Journal splits
How is it recommended to deal with journals that become split into multiple parallel series?
For example, Proceedings of the Royal Society of London (Q21385304) has, since 1905, been published as Proceedings of the Royal Society of London. Series A, Mathematical and physical sciences (Q27721410) and Proceedings of the Royal Society of London. Series B, Biological sciences (Q27712767) (give or take some naming tweaks).
There seem to be two different models currently in use:
- Consider that Proceedings of the Royal Society of London (Q21385304)dissolved, abolished or demolished date (P576)"1905"
- Proceedings of the Royal Society of London (Q21385304)replaced by (P1366)Proceedings of the Royal Society of London. Series A, Mathematical and physical sciences (Q27721410)
- Proceedings of the Royal Society of London (Q21385304)replaced by (P1366)Proceedings of the Royal Society of London. Series B, Biological sciences (Q27712767)
- Consider that Proceedings of the Royal Society of London (Q21385304) is still active, and should have no end date; and write
- Proceedings of the Royal Society of London. Series A, Mathematical and physical sciences (Q27721410)part of (P361)Proceedings of the Royal Society of London (Q21385304)
- Proceedings of the Royal Society of London. Series B, Biological sciences (Q27712767)part of (P361)Proceedings of the Royal Society of London (Q21385304)
- (and vice-versa with has part(s) (P527))
I quite like the second model, which emphasises that the two journals are still connected and part of the same stable.
However, many bibliographic sources (including eg BHL, JSTOR, Library of Congress, British Library) treat the title as having concluded in 1905.
The first model has the advantage that we can flag an error if a paper after 1905 is recorded as published in Q21385304, rather than one of the two split titles. This is straightforward if Q21385304 does has a P576 date = 1905; otherwise another mechanism will be needed to make this possible.
A possible third option would be to have an item for the 1856-1904 run of the journal, in the same way that we sometimes do for a particular 'series' of a journal; but to additionally have another item, for the whole (ongoing) run of the journal, that the first item and Q27721410 and Q27712767 would all be part of (P361).
This seems a possible way forward, but raises some questions, eg: (i) what should be the instance of (P31) for this item, preferably distinguishable from the instance of (P31) for the 1856-1904 run, and also flag that no article item should be published in (P1433) it; (ii) the articles in many wikis might correspond to the "long" journal (cf eg en:Journal of Microscopy). These articles would miss out on current publication details, if the item they were linked to represented the "long" journal, rather than the item for its current incarnation. A bespoke template for journals might be able to get round this, by following links to such further items, but how widely would such a template be available?
Would appreciate opinions on the best way to deal with situations such as this. Jheald (talk) 23:04, 18 July 2018 (UTC)
Please see Property_talk:P98#For_newspapers_?.
--- Jura 05:08, 31 July 2018 (UTC)
English Wikipedia project worth knowing about
I encourage members of this group to take a look at w:en:Wikipedia:WikiProject Newspapers. (Note, the focus on the United States is not axiomatic; if and when members want to incorporate other countries, I think all members of the project will support the notion enthusiastically.)
The problem we perceive is, while Wikipedia often serves as a valuable resource in learning the background behind news sources, several thousand local U.S. newspapers have no Wikipedia entry. We aim to create 1,000 high quality stubs about U.S. newspapers by December. We're working with university instructors, to get their students working on things like drafts and infoboxes. You might also be interested to read w:en:user:Michaelacaulfield's blogging on the topic at http://hapgood.us
Thank you @Bluerasberry: for pointing this Wikidata WikiProject out to us! -Pete F (talk) 21:38, 30 July 2018 (UTC)
- Items used: newspaper (Q11032) , United States of America (Q30) , Oregon (Q824)
SELECT DISTINCT ?item ?itemLabel ?sl ?st ?placeofpublicationLabel
{
?item wdt:P279*/wdt:P31 wd:Q11032 .
{ ?item wdt:P17 wd:Q30 } UNION { ?item wdt:P495 wd:Q30 }.
?item wikibase:sitelinks ?sl ; wikibase:statements ?st .
# ?item wdt:P291 ?placeofpublication .
# ?placeofpublication wdt:P131* wd:Q824 .
SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
}
- It might be easier to start this by creating items on Wikidata for every newspaper. Once done, one would already have full infobox and could easily generate a stub from that.
--- Jura 21:57, 30 July 2018 (UTC) - @Peteforsyth: above a query with what already available (at least if the country=US is defined). Looks like many were created recently.
--- Jura 22:03, 30 July 2018 (UTC)
- @Jura1: This makes sense logically, but in practice I have no idea how to go about creating a Wikidata item for each in anything remotely like an efficient way. I'm not enough of a data guy, I suppose (and I don't know that any of our core project members are, either). Can you sketch it out with more specifics, or point me toward some kind of resource on how to do something like that?
- I can see the potential value of the kind of query you provide, but I don't know if there's a way to filter it by U.S. state, which would probably be the most useful to the project itself. I tried adding to the query "located in the administrative territorial entity" and narrowing it down to Oregon, but got results from all over the USA. -Pete F (talk) 22:44, 30 July 2018 (UTC)
- I do think these lists would form a good basis for automatic creation of Wikidata items, if I knew how to do it. -Pete F (talk) 22:46, 30 July 2018 (UTC)
- I added two lines in the query above that would allow filtering (just remove the #). You might need to change P291 to P131. @99of9: seems to have created quite a lot of items recently. Maybe all are already available. Personally, I use QuickStatements to mass create items. I takes a couple of trial-and-error attempts. There is also a newer version available since.
--- Jura 22:54, 30 July 2018 (UTC)- I will have to figure out the syntax of these queries, I get a "query is malformed" error. But I will see what I can learn on my own, thank you for the idea. QuickStatements looks like a useful suggestion too. Thank you! -Pete F (talk) 23:06, 30 July 2018 (UTC)
- I have created nearly all of the items these USA lists on wikidata already, and the last lot require careful matching to ensure I'm not creating too many duplicates. See my post at the en-wiki-project. Here are some direct links to queries of interest: USA newspapers missing USNPL ID map. --99of9 (talk) 02:45, 31 July 2018 (UTC)
- I fixed my query above and also updated Wikidata:Lists/newspapers_by_format.
--- Jura 23:01, 31 July 2018 (UTC)
- I fixed my query above and also updated Wikidata:Lists/newspapers_by_format.
- @99of9: At [2], I merged "The J. Journal" with "J. Journal". It might be worth standardizing them one way or the other. Label and alias can easily have both. I'm not entirely sure about the use of P131 on these items. Place of publication or headquarters location might be the better fit.
--- Jura 10:55, 1 August 2018 (UTC)- @Jura1: Maybe just deprecate any title from USNPL that you don't think is right? I don't have the local expertise to know which truly use the "The", and I wouldn't want to make a general call. I think you're right about using place of publication (P291) (assuming we consider the items to be about the newspaper not the newspaper company), but if they were about the company, then maybe HQ... I was previously used to working on geographic items, so gravitated toward a redundant located in the administrative territorial entity (P131) essentially as a way of ensuring it is caught by general domain queries, but I can desist if preferred. You'll be pleased to note that in the map query above I used place of publication (P291). --99of9 (talk) 11:39, 1 August 2018 (UTC)
- I have created nearly all of the items these USA lists on wikidata already, and the last lot require careful matching to ensure I'm not creating too many duplicates. See my post at the en-wiki-project. Here are some direct links to queries of interest: USA newspapers missing USNPL ID map. --99of9 (talk) 02:45, 31 July 2018 (UTC)
- I will have to figure out the syntax of these queries, I get a "query is malformed" error. But I will see what I can learn on my own, thank you for the idea. QuickStatements looks like a useful suggestion too. Thank you! -Pete F (talk) 23:06, 30 July 2018 (UTC)
- I added two lines in the query above that would allow filtering (just remove the #). You might need to change P291 to P131. @99of9: seems to have created quite a lot of items recently. Maybe all are already available. Personally, I use QuickStatements to mass create items. I takes a couple of trial-and-error attempts. There is also a newer version available since.
Journals that change their name
Related but distinct from the question above: How to deal with journals that vary their name?
For example The Gardeners' Chronicle (Q905864) -- Wikipedias in English, German, French and Dutch regard this as a single title, that was published for nearly 150 years, from 1841.
Spanish wiki has duplicate articles
- es:The Gardeners' Chronicle and es:The Gardeners' Chronicle & Agricultural Gazette for 1841-1873 (cf Tropicos: 1224) (duplicated article?)
- es:The Gardeners' Chronicle, new series (1874-1886) - (Tropicos: 18135)
- es:The Gardeners' Chronicle, ser. 3 (1887-today) - (Tropicos: 18136)
BHL distinguishes:
- The Gardeners' chronicle (1841-1843)
- The Gardeners' chronicle and agricultural gazette (1844-1873)
- The Gardeners' chronicle [n.s.] (1874-1922)
- Gardeners chronicle & gardening illustrated (no entry)
Library of Congress seems to have one entry, from 1855 sn89049009 -- no indication up until what date.
British Library appears to have
- The Gardeners' Chronicle : (Horticultural Trade Journal.) (1841-1975)
- GC & HTJ : The horticulture & amenity weekly (8 Oct. 1976-3 Jan. 1986)
- Horticulture week : gardener's chronicle & horticultural trade journal. (1986-ongoing) [as "Vol. 199, no. 2" onwards]
though there are also separate catalogue entries for
- The Gardeners' chronicle : a weekly illustrated journal of horticulture and allied subjects. (1874-1956);
- "New ser., v. 1 (1874)-v. 26 (1886); ser. 3, v. 1 (1887)-v. 138 (1955); v. 139, no. 1 (Jan. 7, 1956)-v. 139 no. 9 (Mar. 3, 1956)."
- "Numbering peculiarities: Series numbering dropped with v. 139, no. 1."
- but with holdings notes running from 1841 to 1976
- "Earlier Title:Gardeners' chronicle and agricultural gazette 0261-3441"
- "Later:Gardeners chronicle & gardening illustrated"
- (Gardening Illustrated appears to ceased as an independent title also in 1956, so presumably in that year the two titles were merged).
- Gardeners chronicle, gardening illustrated & the greenhouse (Jan-Feb 1964)
- Gardeners chronicle (1964-68)
- Gardeners chronicle & new horticulturist (1969)
- Gardeners chronicle, the horticultural trade journal (1969-1976)
Conclusions?
What am I supposed to do with this?
- If one goes for the most atomic approach, (and ignoring everything after 1956 until there are scans to link for it), does that mean separate items for
- The Gardeners' chronicle (Q51481592) (1841-1843), The Gardeners' Chronicle & Agricultural Gazette (Q6144503) (1844-73), The Gardeners' Chronicle, new series (Q6144501) (1874-86), The Gardeners' Chronicle, ser. 3 (Q5484179) (1887-????) ?
- Should there be another different item, representing the whole history of the title (linking to the various wikis) ?
- Does it make sense to keep 1841-1843 separate, when there is so little of it ?
- How to represent BHL bibliography ID (P4327) on The Gardeners' Chronicle, new series (Q6144501) and The Gardeners' Chronicle, ser. 3 (Q5484179) when the same BHL title entity (cf Q51481585) covers both? What to state for start time (P580) and end time (P582) qualifiers, the full range at the BHL page (1874-1922) for both, or something different? What qualifier to add, to note the link is shared, to prevent a 'distinct value' constraint error ?
- How to represent "new series", "series 3", etc on items, in a way that is friendly for citation, thinking of both on-wiki citation templates, and off-wiki citation? Jheald (talk) 12:43, 19 July 2018 (UTC)
Other cases
- Just Bulletin des Sciences, par la Société Philomatique (Q5735509) at Tropicos; but Bulletin des sciences, par la Société philomathique (Q51451284) / Nouveau Bulletin des Sciences, publié par la Société Philomatique de Paris (Q6045821) / Bulletin des Sciences, par la Société Philomatique (Q51451281) at BHL Jheald (talk) 15:40, 19 July 2018 (UTC)
- Proceedings of the Royal Irish Academy (Q6087105) combines all three parts, split since 1902 & subsequently renamed several times: cf en:Proceedings of the Royal Irish Academy Jheald (talk) 17:42, 19 July 2018 (UTC)
Discussion
@Jheald: I don't have a good answer for you, but I agree that this is a very important question. Also in discussion here. Perhaps we can have a framework which tolerates both levels of atomicity? Yes, another different item, representing the whole history of the title. If that is created first, that's fine too, and editors can work out if they want to break out to the subitems or not. I think it's worth keeping in mind that other types of entity (e.g. businesses, buildings, humans) do not usually get a new item when they change name. --99of9 (talk) 07:14, 16 August 2018 (UTC)
Best way to add WorldCat items?
The Portland Bee (Q55833876) existed from 1876 to ~1880. There were both daily and weekly editions; and there were two owners. (Founder sold to second owner; second owner sold it back to the founder.) These facts, I think, account for the fact that OCLC/WorldCat has three entries:
- https://www.worldcat.org/title/portland-daily-bee/oclc/31861637&referer=brief_results
- https://www.worldcat.org/title/daily-bee/oclc/10527729&referer=brief_results
- https://www.worldcat.org/title/weekly-bee/oclc/10527776&referer=brief_results
Should I add all three to Q55833876? Or is there a better way to handle this? -Pete F (talk) 22:33, 30 July 2018 (UTC)
- Seems like a common issue with newspapers. The Sellwood Bee has the same problem -- it's been a continuous paper since 1906, but has used the names "Sellwood Bee," "Sellwood-Moreland Bee," "The Bee," and briefly "The Milwaukee Bee". OCLC seems to have a different identifier for each (or at least for some of them. Wikidata wants to only have one OCLC identifier per item; to me, it seems silly to create a separate data item here for each, simply because WorldCat/OCLC chose to do so. But maybe I'm wrong. What's the best practice? If we did make several different data items, how do we go about linking them to Wikipedia? -Pete F (talk) 23:47, 30 July 2018 (UTC)
- One can easily add several identifiers per item. The "named as"-qualifier can be used to indicate how they differ.
I leave it to the more active project participants to decide what you consider the best way to handle this. Personally, I wouldn't create a new item on each change of ownership (sample Q7743126#P127 just gives a start date). I think it's debatable if each new name should get a separate item.
--- Jura 10:35, 1 August 2018 (UTC)- Thanks @Jura1:, it sounds like you and I have the same instincts, and I like the idea of pairing it with "named as." I agree it would be good to hear from more people. In early discussions about Wikidata (~2013) I recall hearing from a couple people that there was no such thing as a fact too insignificant for Wikidata. So, for instance, every grain of sand might have its own Wikidata entry. That seems several steps beyond what's practical or useful to me, but I don't know how much it is part of the defining spirit of Wikidata. Under that theory, of course we would have separate entries for each name a publication had...but then, it becomes challenging to link the overall publication (the thing that carries the significance and the identity in "human-scale" terms) to other Wikimedia projects etc. When I add a second OCLC number, I get this warning: "single value constraint - This property should only have a single value with the same distribution qualifier." I wonder if that is merely a cautionary note, or if it means that some day somebody will come along and change it to comply with a policy. Difficult to tell, and I'd like to understand better. Perhaps Property talk:P243 would be a better place for me to ask. I see a similar question there from June 2018, but no answer yet. -Pete F (talk) 17:04, 1 August 2018 (UTC)
- One can easily add several identifiers per item. The "named as"-qualifier can be used to indicate how they differ.
Title language of journals
Minor thing: I just noticed that these are English by default (sample fix).
--- Jura 20:46, 26 August 2018 (UTC)
Mismatch in taxonomy of Wikidata and English Wikipedia
This is a vexing issue in my work on newspaper articles, though it would apply to other subject areas as well:
There is not (and should not be) a perfect 1:1 relationship between Wikidata items and Wikipedia articles. Case in point:
w:en:German language newspapers of Oregon discusses several newspapers.
Wikidata should have items for each of those newspapers. (Currently, here are two: Q55627861 and Q57207037.)
It seems to me that these items and the Wikipedia article should be linked, but it's not possible to do so. I tried to link each of the Wikidata items to the Wikipedia redirect page for each newspaper, but Wikidata "resolves" the redirect to the article it points to, so it's not possible to add a Wikipedia link to the second (or third, etc.) Wikidata item related to that article.
What's the best approach here? -Pete F (talk) 19:45, 10 October 2018 (UTC)
- A possible, if incomplete, approach: I could create a Wikidata item for "German language newspapers of Oregon," and make each newspaper item an "instance of" it. I suppose that's probably the only viable way to do it that fits into the way Wikidata works...yes? -Pete F (talk) 19:49, 10 October 2018 (UTC)
- The article you mention seems more like a list. I added a sparql equivalent to the item that could be used to generate the list (provided that the items have place of publication and language). You could try to add an automated w:Template:Wikidata list at Wikipedia. --- Jura 21:24, 10 October 2018 (UTC)
- I don't understand in what way it seems like a list; it's entirely prose. (I do see that there are two separate EN Wikipedia articles for w:en:Newspapers in the United States and w:en:List of newspapers in the United States, for whatever that's worth.) -Pete F (talk) 22:22, 10 October 2018 (UTC)
- @Peteforsyth: You could tag the article as <instance of> Wikipedia article covering multiple topics (Q21484471) or <instance of> "group" <of> "newspaper"; either one could have "has part" for each of the associated newspapers, and the newspapers could have "part of" the enwiki article item. - PKM (talk) 23:40, 27 November 2018 (UTC)
P123: Should this be used for people, organizations, or both?
I've opened a discussion about whether Property:P123 ("publisher") should be used to indicate the person who serves as a newspaper's publisher, the [person or organization] who owns the paper (for which we already have Property:P127 ("owner")), or both (as it seems to be now). Please join the discussion over there if you have thoughts! -Pete F (talk) 22:23, 27 November 2018 (UTC)
Section or rubric
Hey, I rose up this question at WikiProject Books, but they point me here it fits here more. I am looking for property which would describe section within the Magazine. Lets say I have an article called XY. I can point out it was published in e.g. Forbes magazine, using published in (P1433), but can I somehow also point out that the article was the part of section called Technology? Is there such way? What about to use section, verse, paragraph, or clause (P958)? --Juandev (talk) 20:37, 27 April 2019 (UTC)
- I think on an item about the article you'd have something like
published in |
| ||||||||||||
add value |
- With magazine (Q41298) and column (Q267628) with the appropriate items if they exist (many newspaper columns, in particular, often have no title). Circeus (talk) 04:41, 29 April 2019 (UTC)
- That looks good.--Juandev (talk) 08:40, 29 April 2019 (UTC)
Daily journal
Hi y'all,
WikiProject Periodicals has more than 50 participants and couldn't be pinged. Please post on the WikiProject's talk page instead.
I finally put data in Q19180761 and Q19180763 (which were blank for 4 years!) and I'd like some advice. I particular, I see that the documentation say to use part of (P361) between but here L'Ouest-Écair has more than 17000 daily issues so it doesn't seem reasonable to use it (it would make L'Ouest-Éclair (Q3204738) - by far - the largest item of Wikidata). So, what do you think?
Cheers, VIGNERON (talk) 18:23, 5 June 2019 (UTC)
- @VIGNERON: Maybe use part of (P361), but make an exception to using the reciprocal property has part(s) (P527), and instead catch them all with a single has part(s) of the class (P2670). But as far as I can see, this is currently hypothetical, because we only have a handful of the 17000 potential daily issue items? --99of9 (talk) 22:22, 5 June 2019 (UTC)
- How about published in (P1433)? We have other "articles" floating that are really entire supplementary issues treated as a single article by the publisher's metadata (e.g. Q56836084) Circeus (talk) 06:06, 6 June 2019 (UTC)
- If we expect many issues finally becoming part of Wikidata/Commons/Wikisource/etc., another solution could be to model the volumes (indicated by "Première année", "QUINZIEME ANNEE") explicitely. This would be in line with most scientific citation rules, and establish a three level hierarchy (periodical - volume - issue), and perhaps article as a fourth level. Downside is, you may not have values for volume in the sources (which is often the case for articles in 20th Century Press Archives (Q36948990)). So I'm not sure if we can come to one general solution. Jneubert (talk) 08:10, 6 June 2019 (UTC)
- @99of9: well, I do have the idea (not fully planned yet) to import all the issues (after checking there is 16216 for the Rennes edition but there is also editions for other cities, for a sum total of more the 50k+ issues), that's why I'd like to have a proper structure first
- @Circeus: hmm, interresting... not entirely convinced but why not.
- @Jneubert: yes, I though of creating fictious volume for years (which fit this URL https://gallica.bnf.fr/ark:/12148/cb32830550k/date1899 ).
- PS: I've added the Gallica ID (P4258).
- I'm still at the early stages gathering informations and data, so any remarks and comments are welcome!
- Cdlt, VIGNERON (talk) 12:18, 6 June 2019 (UTC)
- Yes, expressing the volume via a year is a much safer bet for other periodicals then via volume count. Wouldn't volume items mitigate the problem very much? You wouldn't get more than a few hundred volumes, and not more than a few hundred issues/volume, so part of (P361) could work. Jneubert (talk) 12:33, 6 June 2019 (UTC)
@Jneubert: So if I understand correctly, the part of (P361) tree would be :
- L'Ouest-Éclair (Q3204738)
L'Ouest-Éclair, Rennes (items to be created)L'Ouest-Éclair, Rennes (Q64441564)L'Ouest-Éclair, Rennes, 1899 (items to be created)Q64441577
That could work. @99of9, Circeus, Hsarrazin: what do you think? Cdlt, VIGNERON (talk) 14:40, 6 June 2019 (UTC)
- So the complete hierarchy for this article would be
- The Shanghai Gazette (Q46995846)
- The Shanghai Gazette, 1921 (item to be created)
- The Shanghai Gazette, 4. April 1921 (item to be created)
- The Shanghai Gazette, 1921 (item to be created)
- The Shanghai Gazette (Q46995846)
- I suppose that the intermediate levels should be considered optional: For single articles, people will normally not want to create the full hiearchy. But I don't suppose that this will cause much harm - if somebody lateron finds a level overcrowed, it should be achievable to insert intermediate levels derived from the publication date and update the items automatically. (If different editions exist, like the Rennes edition, or it turns out that the "Shanghai Gazette" have separate English and Chinese editions, that may be not achieveable.) However, for mass imports with prior deep knowledge of the newspaper, it could be planned well in advance and differentiated properly. Later users will profit from this, as they get aware that differnt editions exist, and may assing the proper part-of item. Jneubert (talk) 15:32, 6 June 2019 (UTC)
- Just remember a list of articles by journalist Kim Wall, which in the article items mostly use published in (P1433). Jneubert (talk) 15:47, 6 June 2019 (UTC)
@VIGNERON: Individual articles are not structured that way by default. They use published in (P1433) to the journal and (normally) no intermediate items are created just to serve as target. However, where the issue does already exist separately because of reasons (e.g. because it constitute an item on another Wikiproject), the individual issue should probably the target of published in. Now, at this point, the question remaining is whether the article should also link to the serial in addition to the issue.
As for the issues, I think this tagging for the 29 September 1913 issue works is pretty good. If there was a volume number (which I assume is lacking here), a volume (P478) should also be present. Circeus (talk) 20:10, 6 June 2019 (UTC)
- In the possibly analogous case of individual maps from a book with multiple volumes, I have used published in (P1433) to link the map to the overall atlas (with volume (P478) and page(s) (P304) as qualifiers), and part of (P361) to link the map to the volume. I only created an item for the volume because it was useful to have a category for it on Commons, and to be able to give that category an infobox. Our templates appear to expect the overall work to be the target of P1433. Jheald (talk) 21:56, 6 June 2019 (UTC)
- @Jheald: that sounds like a very good idea, especially as the same articles can be published in multiple edition (so the published in (P1433) would be unique but the part of (P361) would be multiple to be more precise). @99of9, Circeus, Hsarrazin: what do you think of this idea? I also created L'Ouest-Éclair, Rennes (Q64441564) and Q64441577 to try; it would require discussion and checking (for instance, the instance of (P31) which I'm not really satisfied with). @Circeus: no there is no volume number, and in fact there is no volume at all (plus, I don't like P478 who confuse the notion of tome and volume which is problematic in theory). Cdlt, VIGNERON (talk) 08:44, 7 June 2019 (UTC)
- for Revue des Deux Mondes (Q1569226), of which we already have over 13000 article (Q191067) and collection of articles (Q17518557) from fr-wikisource, the chosen structure is :
- for an article : published in (P1433)Revue des Deux Mondes (Q1569226) + volume (P478) + publication date (P577) + page(s) (P304) (+ part of (P361) if part of a series...)
- for a series : published in (P1433)Revue des Deux Mondes (Q1569226) + number of parts of this work (P2635) + has part(s) (P527) links to articles -> see example Q17357108
- the data are automatically retrieved from wikisource header with a script, and the dates are to be completed manually (reason why a lot of them are only year for now... --Hsarrazin (talk) 09:30, 7 June 2019 (UTC)
- for Revue des Deux Mondes (Q1569226), of which we already have over 13000 article (Q191067) and collection of articles (Q17518557) from fr-wikisource, the chosen structure is :
- @Jheald: that sounds like a very good idea, especially as the same articles can be published in multiple edition (so the published in (P1433) would be unique but the part of (P361) would be multiple to be more precise). @99of9, Circeus, Hsarrazin: what do you think of this idea? I also created L'Ouest-Éclair, Rennes (Q64441564) and Q64441577 to try; it would require discussion and checking (for instance, the instance of (P31) which I'm not really satisfied with). @Circeus: no there is no volume number, and in fact there is no volume at all (plus, I don't like P478 who confuse the notion of tome and volume which is problematic in theory). Cdlt, VIGNERON (talk) 08:44, 7 June 2019 (UTC)
Capsfix
labels of items about journals seem to follow the scheme of labels generated by default for English Wikipedia pages (on any topic): an initial uppercase letter followed by lowercase letter. In the absence of a description, this makes it difficult to distinguish an item about a journal from the item about the topic.
To reduce the problem, I added a description to most items about journals, but I think we should try to find a way to correct capitalization of journal titles (to follow the one used by the journal itself). --- Jura 14:02, 24 June 2019 (UTC)
Editions of volumes
WikiProject Periodicals has more than 50 participants and couldn't be pinged. Please post on the WikiProject's talk page instead.
How should one handle an item for a book such as [3]? This isn't the first time I've encountered this kind of book. It's periodical literature, but I don't believe it has anything resembling an issue (Q28869365) - just volumes (it could be argued that these are editions?).
Furthermore, I'm inclined to think each volume should have respective version, edition or translation (Q3331189) items since the books have a specific publisher and other properties typically associated with an edition item. Does this seem correct? --SilentSpike (talk) 17:09, 12 August 2019 (UTC)
- To clarify my question, I see three possibilities:
- Each "volume" is a different written work (Q47461344) item
- Each "volume" is a different version, edition or translation (Q3331189) of the same written work (Q47461344) item
- Each volume (Q1238720) of the periodical (Q1002697) comes with version, edition or translation (Q3331189) items
- --SilentSpike (talk) 17:22, 12 August 2019 (UTC)
- @SilentSpike: Better ask your question to Wikidata:WikiProject Books: we already had that discussion, see Wikidata_talk:WikiProject_Books#How_does_Has_Part_interact_with_the_work/edition_distinction?, without any clear conclusions. Snipre (talk) 22:11, 15 August 2019 (UTC)