Source to upload from

https://www2.landesarchiv-bw.de/ofs21/home.php is the starting point of the archive. However, to ease navigation for this purpose, I made a list of the (in my opinion) contents most intresting for commons, mostly photos and maps:

TOTAL 649,253

License

All content is either public domain (age/created by government), and if not, everything is licenced under CC BY 3.0. (see terms of use: Some of the digitized archive records in the finding aid system of the Landesarchiv Baden-Württemberg are still subject to copyright protection, some are exempt from copyright protection as official works, and some are in the public domain. Where they are protected by copyright, the State Archives hold the corresponding exploitation rights and grant a Creative Commons CC-BY license.)
The source "Landesarchiv Baden-Württemberg" and the archive reference ID of the image needs to be mentioned when using the image - this information is also always present on the image when downloading fron the website.

Description

• Do the media URLs follow a pattern?

Yes, from what I can see: Each image can be viewed in an image viewer. The image viewer can be accessed from the archive series page (example)
Those image viewer links look like this: https://www2.landesarchiv-bw.de/ofs21/bild_zoom/zoom.php?bestand=SERIES_ID&id=IMAGE_ID&gewaehlteSeite=FILE_NAME (for example https://www2.landesarchiv-bw.de/ofs21/bild_zoom/zoom.php?bestand=21715&id=2836819&gewaehlteSeite=02_0001128649_0001_2-1128649-1.png)
The image file can be downloaded by using https://www2.landesarchiv-bw.de/ofs21/bild_zoom/download.php?&id=IMAGE_ID&bilddatei=FILE_NAME (example https://www2.landesarchiv-bw.de/ofs21/bild_zoom/download.php?&id=2836819&bilddatei=02_0001128649_0001_2-1128649-1.png)

• Does the site have an API?

No.

• Is there a template that could be used on the file description pages, or should one be created?

Since there is a huge amount of files, creating one might by a good idea.

TheImaCow (talk) 09:28, 19 May 2024 (UTC)

I started scraping and it is going well. Do you know of files that fall under CC BY 3.0 ? -- DaxServer (talk) 07:55, 13 August 2024 (UTC)

All 25,700,000 digitized objects do, per the terms of use.

There is already {{Landesarchiv-bw-image}}, which references

- the archive location (collapsed sections above, see the "Table of identification codes" on the template documentation)

- the archive signature of the object (such as "W 145/4 Nr. 0091")

- the permanent link ID to the digitized object such as "5-790216-1" (http://www.landesarchiv-bw.de/plink/?f=5-790216-1)

This can be used in an information template "accession_number" field.

As for the |source= field, there is clear guidance: We cite the archive signature, if known the author and the permalink, such as

"Landesarchiv Baden-Württemberg, Staatsarchiv Freiburg W 145/4 Nr. 0091 / Fotograf: Leopold Adler"

Example without known/specified author:

"Landesarchiv Baden-Württemberg, Staatsarchiv Ludwigsburg K 414 I Nr 273"

As for the licencing template, I think we should use a custom version of the {{Cc-by-3.0}} template, I created {{LABW}} (for parameter 1, use the same value as for "source" described above)

What data exactly is there to scrape? Is there a way to tell if an object is a photograph or something else (map, drawing, text)? (to decide weather to use {{Photograph}}) ~TheImaCow (talk) 11:03, 13 August 2024 (UTC)

I'm scraping the metadata and the images associated with an object. An object can contain one or more [digitalized] images (of course, only those Findbücher that are digitalized).

• The metadata can be accessed by clicking the Details in the Strukturansicht view. URL format - https://www2.landesarchiv-bw.de/ofs21/olf/druckansicht.php?id_titlaufn=IMAGE_ID&bestand=SERIES_ID In the example above from permalink http://www.landesarchiv-bw.de/plink/?f=5-790216-1 - click on the Details, it will open a small popup with info. The direct URL would be https://www2.landesarchiv-bw.de/ofs21/olf/druckansicht.php?id_titlaufn=3092053&bestand=23318
  • From this view, this particular object has good metadata to determine what it is, for example:
    • Art der Infomation = Bild
    • Art der Vorlage = Glasplatte
    • Format = 16 x 11 cm
  • Not all objects have such good info.
• The info on the images can be obtained from the thumbnail view - that's the magnifying glass icon in the Strukturansicht view. URL format - https://www2.landesarchiv-bw.de/ofs21/bild_zoom/thumbnails.php?bestand=SERIES_ID&id=IMAGE_ID An example would be
  • https://www2.landesarchiv-bw.de/ofs21/bild_zoom/thumbnails.php?bestand=7564&id=7278223 for multiple images all of which have the same Bestellsignatur
  • https://www2.landesarchiv-bw.de/ofs21/bild_zoom/thumbnails.php?bestand=23318&id=3092053 for single image

I haven't yet explored the data that I've gathered so far, so I've limited info atm. -- DaxServer (talk) 12:13, 13 August 2024 (UTC)

Things I noted regarding file names - some images have no title (eg "Keine Angabe"), then we should only use the archive ID (PL 723 DK 94-2), and if the images have title, we should probably start the file name with the archive ID, because they are sometimes sorted by date or relation etc, and this preserves the original file order in the categories, e.g. "PL 723 DK 55-20 - Bf. Kupfer", where the "20" is a running number.

I removed some series from the above table...

- T 1 (Zugang 2008/0013)/SERIES ID 22257 only text documents

- T 1 (Zugang 2008/0032)/22664 only low quality scans of personal photos, probably out of scope

- T 1 (Zugang 1983/0018-01)/10461 only text documents

- J 153/political party advertising often recently collected by the archive, unlikely if actually free

- Q 2/50 - 17,000 negative photo bags containing ~520,000 individual photographs, unfortunaly low scan quality, so extracted images are of very poor quality

All objects seem to have their "category tree" as part of the description, such as

Staatsarchiv Ludwigsburg

\/

Deposita, nichtstaatliche Archive und Nachlässe / 1335-1997

\/

Nachlässe (ohne Deposita)

\/

PL 723 Nachlass Hans Noller: Sammlung zum Eisenbahnwesen in Württemberg / Ca. 1844-2011

\/

2. KB-Dias

\/

Stuttgart - Horb ("Gäubahn")

\/

Diakasten 18: Stuttgart - Horb I

at this object: http://www.landesarchiv-bw.de/plink/?f=2-5340602-1

I think it would be a good idea to automatically copy these category trees as commons categories, as this would make the manual categorisation needed for the images significantly easier. Such a category tree has already been created for the US National Archives upload, see Category:US National Archives Record Groups. ~TheImaCow (talk) 12:42, 22 August 2024 (UTC)

TheImaCow I was able to collect 375,357 records (many with multiple files). Here is a statistic of the metadata:

• Bemerkung shall be merged with Bemerkungen
• I'm not sure if we want to merge "Autor" and "Autor/Fotograf/Künstler"
• Do you want them to be uploaded with translation into EN? Or simply use the {{de template and be done with that?

Let me know how you want to proceed now. -- DaxServer (talk) 22:00, 20 December 2024 (UTC)

I added some {{Information}}/{{Photograph}} template values for the original keys to the table. I think we should only use the original "{{de" values, english translations should be done manually. To proceed, probably a test upload? (there is also the {{Landesarchiv-bw-image}} template for |accession_number=, see above)~TheImaCow (talk) 22:49, 21 December 2024 (UTC)

@TheImaCow Here're the stats into Archivalienart:

• Amtsbücher - 3
• Fotos - 228950
• Karten und Pläne - 47749
• Nachlässe - 7
• Personenakten - 7062
• (blank) 91585

-- DaxServer (talk) 09:30, 24 December 2024 (UTC)

I wanted to do some test, but the website has some issues https://www2.landesarchiv-bw.de/ofs21/olf/bildexplorer.php?bestand=21569 (on archive). Tweeted to them, but maybe they're all in Holidays. Will wait and see. -- DaxServer (talk) 09:44, 24 December 2024 (UTC)

Possibly that collection got removed intentionally from the website for whatever reason, it must have been between "FAS K" and "FAS Pa ST" here, at the bottom. Now this collection is missing from the list. (only 90 items affected) ~TheImaCow (talk) 22:00, 24 December 2024 (UTC)

It seems some information has been changed from the last time I gathered the data. I'll have to do a refresh on the 375k items collected. This will take a few weeks it seems 😵‍💫 -- DaxServer (talk) 18:29, 1 January 2025 (UTC)

@TheImaCow Can you give me an example for merging the values for medium= and collection=, for example this one: http://www.landesarchiv-bw.de/plink/?f=3-558838

Also, how should the <br/> be translated into wikitext? - line break, wikitext break, special character, or something else -- DaxServer (talk) 10:15, 14 March 2025 (UTC)

Medium: medium=Bild - 9,9 x 6,0 cm //// (medium=Art der Information - Format)

Collection: collection=Kommunalarchive im Hohenlohekreis<br>Kreisarchiv Hohenlohekreis<br>Sammlungsbestände<br>Foto-, Dia- und Postkartensammlung<br><br>KrA SF 2<br>Fotosammlung: Ehemals eigenständige Gemeinden Adolzfurt - Lassbach / um 1860 - 2011<br>10. Bitzfeld<br>10.2 Gebäude, Bauten :::

/// collection=Einordnung des Bestands<br><br>Aus Bestand

Hope this is understandable/answers the question, here is an example on how it could be rendered:

I had the idea with the little arrows ↓↓↓ on the collection tag spontanous, they could be added too. ~TheImaCow (talk) 21:52, 15 March 2025 (UTC)

@TheImaCow I wanted to normalize the Laufzeit field to fit into {{complex date}} but it seems to take too much of my time. Would you be able to convert them? If so, I can email you an export with that data. If not, I'd add the original data into some other_fields parameter. Let me know -- DaxServer (talk) 12:10, 18 March 2025 (UTC)

Unfortunaly I'm quite clueless with such data processing, I don't think I could make that work. Lets do the solution with other fields. ~TheImaCow (talk) 22:45, 18 March 2025 (UTC)

@TheImaCow It seems all pre-work is done. Maybe I'll add couple of SDCs. Also, request for whitelisting domain: MediaWiki talk:Copyupload-allowed-domains#Allowlist request - Landesarchiv Baden-Württemberg Online-Findmittelsystem -- DaxServer (talk) 12:53, 30 March 2025 (UTC)

BTW, can you design some categorization schema? -- DaxServer (talk) 13:25, 30 March 2025 (UTC)

I suggest creating a category for each series (the ones listed in the collapse boxes above, first value in the "Aus Bestand" field e.g. Category:KrA SF 2 Fotosammlung: Ehemals eigenständige Gemeinden Adolzfurt - Lassbach / um 1860 - 2011

These categories should be categorized into already existing Category:Collections of Staatsarchiv Ludwigsburg, Category:Collections of Staatsarchiv Freiburg, Category:Collections of Staatsarchiv Sigmaringen etc. ~TheImaCow (talk) 11:28, 5 April 2025 (UTC)

@TheImaCow Extracting the Bestand info seemed a bit more complicated with the data format I have right now. So I fellback to using Category:Generallandesarchiv Karlsruhe Findbuch O format.

I uploaded 10 test files: Special:Search/hastemplate:LABW file:

Let me know how they look. -- DaxServer (talk) 13:43, 10 April 2025 (UTC)

Looks good overall! Two things:

• File name: A better format would probably "<Titel> - LABW - <Bestellsignatur>"
  We need to use some amount of description per Commons:File naming. It looks to me like the "Titel" field is a good fit for that. "Landesarchiv Baden-Württemberg" can be abbreviated to LABW (to avoid too long file names). The "Bestand"/"Permalink" IDs can be left out, as they are only the ID numbers automatically assigned by the digital finding aid, without a direct weblink, they are pretty useless. The "Bestellsignatur" archive signature should be sufficient.
  Example:
  Photograph from Landesarchiv Baden-Württemberg Hauptstaatsarchiv Stuttgart M 700-4 Nr. 100 Bestand-6587 permalink 1-852418-1.jpg
  ↓to↓
  Soldaten im Schützengraben (Gruppenbild) - LABW - Hauptstaatsarchiv Stuttgart M 700/4 Nr. 100.jpg
  This will result in some duplicate file names where multiple images are assigned to the same object, so it looks we can keep the digit(s) after the last "-" in the "Permalink" ID to make the filename unique e.g.
  Photograph from Landesarchiv Baden-Württemberg Hauptstaatsarchiv Stuttgart J 317 Nr 10 Bestand-23838 permalink 1-1335908-2.jpeg
  to
  Herbst 1902 - LABW - Hauptstaatsarchiv Stuttgart J 317 Nr 10 -2.jpg
  Exception i've seen: On media where "Archivalienart" is "Personenakten", we must use "Name" field instead of "Titel", as a title is often not present there.
  Hope I explained this understandable
• Collection field/Arrows ↓: These arrows are at every line break, not just at the "collection level", which can lead to situations like here Easiest to probably just leave them away then.

~TheImaCow (talk) 11:54, 11 April 2025 (UTC)

@TheImaCow There are some entries where they put description as Titel like this one: https://www.landesarchiv-bw.de/plink/?f=5-348952 I don't think I can extract a meaningful filename for such. Any ideas? -- DaxServer (talk) 14:49, 14 April 2025 (UTC)

@DaxServer If it's possible we could limit the title to maybe the first 200 characters or so. However, it is probably better to just leave out that particular collection. It seems to be consisting entirely of text documents which are difficult to properly curate on Commons anyway. Are there more collections with such long titles? (I've seen collection A 25/1 Landgericht Freiburg with the same issue, lets leave that out too)~TheImaCow (talk) 19:47, 14 April 2025 (UTC)

@TheImaCow Here's the updated test:

• https://editgroups-commons.toolforge.org/b/OR/3249b63c509/
• https://editgroups-commons.toolforge.org/b/OR/b2bb6b3e574/

-- DaxServer (talk) 11:27, 15 April 2025 (UTC)

Looks good! ~TheImaCow (talk) 15:57, 15 April 2025 (UTC)

Applied Commons:Bots/Requests/CuratorBot (4) -- DaxServer (talk) 17:40, 15 April 2025 (UTC)

Assigned to	Progress	Bot name	Category
DaxServer (talk · contribs)	Done	CuratorBot (talk · contribs)	Category:Files from Landesarchiv Baden-Württemberg uploaded by CuratorBot

There're about 35k images for which the copyright is not mentioned which is assumed to be the default CC-BY 4.0 as mentioned in their Nutzungsbedingungen. I think I'll skip them for now. From what I understand, LABW is working on making more archives available digitally. I'd rather come back to this project, perhaps at the end of next year to do another batch of updates. Until then, I'd call it a day for the uploads. It'd be a good idea to categorize them -- DaxServer (talk) 14:01, 19 September 2025 (UTC)

Opinions

@DaxServer: As of now, categorization isn't done as suggested above by User:TheImaCow. E. g. This file was added to Category:Collections of Hauptstaatsarchiv Stuttgart AND to the non-existent Category:Hauptstaatsarchiv Stuttgart Findbuch J 319. The latter should be created as a subcategory of the former, which means that, in order to avoid overcategorization, Category:Collections of Hauptstaatsarchiv Stuttgart shouldn't be added to the individual files, as it is redundant. Also, the main category overcrowded with tens of thousands of files is quite useless. Do you plan to fix the overcategorization? --Sitacuisses (talk) 00:13, 6 June 2025 (UTC)

These "Category:Collections of Archive" categories are added automatically because they are part of the {{Landesarchiv-bw-image}} template, which is used in the "|source=" parameter. I suggest we remove that function from the template. ~TheImaCow (talk) 00:21, 6 June 2025 (UTC)

@Sitacuisses I'll create the categories once I'm finished with the uploads. -- DaxServer (talk) 13:31, 7 June 2025 (UTC)

TheImaCow, The files already uploaded that are under CC-BY have SDC license stating CC-BY-3.0 (example file) since I retrieved the info before the license change between 22 March 2025 and 30 March 2025. Do you think the SDC 3.0 can be removed and just let 4.0 stay? Also, can you update the {{LABW}} with the license change info? -- DaxServer (talk) 10:30, 8 July 2025 (UTC)

If the files are all tagged with 4.0, I don't think it is nessescary to remove the 3.0 - it is still valid as licences are not revokable and I don't think there is a practical difference here anyway. Regarding the template, do you mean that it should also state that the licence was previously 3.0? It states correctly that the licence is 4.0. ~TheImaCow (talk) 20:03, 15 July 2025 (UTC)

Thanks for clarification. I'll resume the uploads. -- DaxServer (talk) 13:25, 17 July 2025 (UTC)

Hi! As mentioned on DaxServer's talk page, I think it would be great to add {{Check categories}} to the image pages. This makes it significantly easier to find images that have not been properly categorized. In particular in connection with the relatively easy categorization by location (see e.g. Category:Photographs of Baden-Württemberg by Willy Pragher), local users could then find all images from their city/region that still need proper categorization. Some further thoughts:

• In particular, I care for the >100,000 images by Willy Pragher. They are high-quality and in many cases with encyclopedic value (e.g. many portraits of people with Wikipedia articles that have no images yet), but are not usable at the moment as they cannot be found without categories.
• I am not sure about other images. In theory, the template could be added for the same reasons on all LABW files, but I am not sure whether there will ever be someone who wants to categorize every single file in categories such as this one. Maybe that would clutter Category:Media needing category review too much. Does anyone know of other collections with images of particular interest?
• As >100,000 of these images have already been uploaded, I think it would be important to add the template to those as well. The "problem" here are the files which have been categorized by users in the meantime. It probably is easy to add the template to all files which have not been edited since their upload? But this would omit all files that I have already roughly categorized by location, e.g. the 5500 files in Category:Photographs of Bucharest by Willy Pragher, none of which have been properly categorized by what they each depict. Is it possible to automatically find out if any "meaningful" categories have been added to a page? If not, I think it might still be worth it to add the template simply to all your uploads. I would guess that only 1 % have been properly categorized in the meantime and it is easier to remove the template manually from the respective pages after a short check than to manually go through 10,000s of images and checking whether they need categorization.
• Somewhat unrelated: If the bot touches all images by Willy Pragher again anyway, it could also add the appropriate subcategories of Category:Photographs by Willy Pragher by year to all files with a date.

--Entbert (talk) 13:32, 13 August 2025 (UTC)

Thanks for the detailed suggestions, @Entbert!

• a tracking category for review, similar to Category:Images from USACE to check. Perhaps one category per each collection, to begin with?
• the {{tl2|Check categories}} on all images. I've already changed the workflow to add it to the new uploads. For existing ones, I can look into how to add the template using OpenRefine or other tools.
• I have all the metadata in OpenRefine and extracting meaningful categories - like Willy Pragher by year or the others - takes a little bit of work. I can't promise to do that right away, but once we refine what we want, I can invest the time.
• There are ~116k more images left to upload of which ~1k are from Willy Pragher. I updated the data for the new uploads, but for the already uploaded ones, I'll have to do them in another iteration.
• There is an active German community who have good interest in the images. I think we/you might want to reach out to them: Category talk:Staatsarchiv Wertheim#Tausende neue Bilder - wertvolle Schätze

-- DaxServer (talk) 14:47, 14 August 2025 (UTC)

• Pinging @TheImaCow, @OhneEisen for inputs on ideas on how to further categorize the images, now that the uploads are done in this batch. -- DaxServer (talk) 14:03, 19 September 2025 (UTC)

  Big thanks for your work! I don't think it is really possible to "automatically" categorize most of the images - but looking at Special:RecentChangesLinked/Category:Files_from_Landesarchiv_Baden-Württemberg_uploaded_by_CuratorBot (warning, loading this may take a while), various people are already busy manually categorizing images every day. ~TheImaCow (talk) 09:27, 28 September 2025 (UTC)

User:Goldmull has now removed a large number of files from Category:Hohenlohe-Zentralarchiv Neuenstein Findbuch KrA SF 2/1. Shouldn't those files stay in that category? --Rosenzweig τ 11:57, 3 November 2025 (UTC)