A new universal database (K10plus)

[Philologist]

Since March 2019, the name K10plus stands for a large universal data pool. It consists of data from universities, public libraries, research institutions such as Max Planck Institutes, museums, archives, public institutions and what not. The number of the participating institutions is more than 2000.

This database can be searched directly in Bookends with File > Import Filter Manager > Germany - GBV (GVK).fltr , and the hits can be imported. It just requires that you update the server address, as explained here:
viewtopic.php?t=6045

Search (in English) is also possible directly in your browser:
<https://opac.k10plus.de/DB=2.299/SET=2/ ... 209.203,FY>

The number of records in this joint database is currently more than 235 million. The collections of the Library of Congress contain approximately 173 million items (Wikipedia).

[Dellu]

Dear Philologist; thank you for bringing this powerful resource to our attention. This is very interesting system.
I am relieved to find a new working search engine after my failure to deal with Worldcat.

Having said that, I have a little observation I want you know (in case you are somehow involved with the project, as you said you want to hear a comment in the other thread).

I have been looking to fill up incomplete metadata of my books from Amharic language. Turns out, the results from the Browser are better than the one that comes using the Server. I don't know why that is: the Server connection is showing only the Transliteration; not the original language; while web-search shows the important fields (the author and title) in the original language.

I know the issue is not from BE because the search using Google Books show in the original language.

https://take.ms/mQq4b

Monosnap Image 2023-09-22 17-24-32.jpg (229.85 KiB) Viewed 341199 times

I personally don't like to store references using Transliteration because they are inaccurate; and can never be accurate as there are different ways of doing it (no universal convention). I want to import them in the native language; and then insert the transliteration for my personal use in the comment or note field).

I am glad BE has a browser. I can use that one to get the references. But, I am still puzzled why the server gives a different result from the webpage.

[Jon]

You should be looking at what they return from the server using Bookends direct search, not Autocomplete Paper.

Jon
Sonny Software

[Philologist]

in case you are somehow involved with the project

No, I'm not involved in any way. I'm just a happy user. :–)

the Server connection is showing only the Transliteration; not the original language

As Jon said, use Bookends direct online search. I searched for the item you pointed to with the red arrow, and this is what I get:

1.png (242.17 KiB) Viewed 341172 times

As you can see, the original title is displayed in the Geʽez script (Amharic). When I press the Import button this information is not imported. Perhaps we need to adjust the MARC filtering in the Import Filter Manager, but I'm not sure how to do that. Or there is something else we need to do to get the original title into the right field. When I say "the right field" I mean the field we have determined for this purpose in Bookends.

Perhaps Jon can help…

Until we have fixed this, you can easily copy the original title and author directly from the online search (see the screenshot above.)

For all other languages I have tested, the information is imported in the original language, as you can see on this screenshot.

2.png (35.85 KiB) Viewed 341172 times

The screenshot also shows an anomaly in the Editors field (which I mentioned earlier.) Henrik Ibsen is the author, not the editor, and Georg Schulte Frohlinde is the translator; not sure whether he should also be classified here as "editor".

[Philologist]

OK, I got it.

0.png (36.65 KiB) Viewed 341169 times

00.png (134.37 KiB) Viewed 341169 times

[Jon]

Thank you, Philologist, that's exactly the right solution. I didn't know how to enter those characters from the keyboard so I could see the MARC output for myself.

@Dellu. Please enter unusual Unicode characters in the actual post, not just pictures of them. Then one can copy/paste them to reproduce what you are seeing. The entire MARC record returned by a direct library search is shown in the lower pane, so you can see what information is available. Which fields are actually imported is determined by the import filter. For those interested, the description of MARC records and how to parse them is covered in detail in the User Guide. Modifying them is actually quite simple, as this example illustrates.

Jon
Sonny Software

[Jon]

@Philologist

I think I should rename this filter to reflect its new capabilities. Should I call it Common Union Catalogue (K10plus GVB/SWB).fltr?

Or simply Common Union Catalogue?

Or something else?

I think "Common Union Catalogue" is how it is referred to on this page

https://bibliothek.htw-berlin.de/en/lit ... atalogues/

I think Germany should be removed from the name because non-German users will never try it. I'd mention the origin of the consortium in the Notes, though.

Jon
Sonny Software

[Philologist]

@Dellu. It wasn't MARC; a "tag" had to be added to the Import Filter. So how did I find the right tag?

1. Use Bookmark's direct online search.
2. Select the item you want
3. At the beginning of each field you see a tag number (for example 800).
4. This tag number must be followed be the character which you see next after $, which in our case is "a".

1.png (30.15 KiB) Viewed 341165 times

The Import Filter then looks like this:

2.png (62.08 KiB) Viewed 341165 times

And the import result looks like this:

3.png (79.2 KiB) Viewed 341165 times

I notice that the author and the title seem to be identical here. I don't know the language so I don't know where the author name ends and where the title begins, or if there is any author in the fields at all. I would have to make more tests to figure that out.

[Jon]

The output uses the same tag and subfield for original author and title. 880a. So Bookends can't distinguish between them and imports the first one it encounters. That's a problem with the way the information is provided. The field/subfield tags must be different for different pieces of information for Bookends to know how to deal with the data.

Jon
Sonny Software

[Philologist]

CORRECTION
Now it's perfect, I think. The author, written in the Geʽez script (Amharic), requires 880a and the original title needs 880b.

4.png (48.89 KiB) Viewed 341156 times

5.png (14.73 KiB) Viewed 341156 times

[Philologist]

I think I should rename this filter to reflect its new capabilities. Should I call it …

I'll have to think about that. I would then suggest something that indicates the universal character of this consortium, a speaking name. It doesn't have to be a translation of the real name(s), especially because the abbreviations are hard to understand and to remember, even for native speakers. This is a work in progress, and the latest name they came up with is K10plus. LOL.

'Common Union Catalogue' can mean everything and nothing.

How about:
Universal Catalogue
World Catalogue
Worldwide Catalogue

I think Germany should be removed from the name because non-German users will never try it.

I agree.

[Philologist]

I didn't know how to enter those characters from the keyboard

macOS doesn't seem to come with a Geʽez script (Amharic) keyboard, so one has to use the Character Viewer and drag each letter into the application one is working in.

I stumbled on a website which allows the user to use his or her normal keyboard, and each time a character is typed it's immediately transformed into Amharic. Quite nice.
https://www.lexilogos.com/keyboard/amharic.htm


@Dellu. You probably don't need it in your work, but I thought I should mention it anyway. A free keyboard and an Amharic font can be downloaded from SIL International (formerly Summer Institute of Linguistics.)

Abyssinical SIL
https://software.sil.org/abyssinica/
https://software.sil.org/abyssinica/download/

[Jon]

The problem I have with Universal or Worldwide is that, if I understand correctly, the consortium is broad but only includes German institutions. Of course literature knows no borders (except in Russia), but the implication is that the search would include international institutions as well. They refer to GBV as the Common Library Network. Maybe that's a provocative enough name that will draw eyes and still be accurate?

Common Library Network

Jon
Sonny Software

[DrJJWMac]

> I think Germany should be removed from the name because non-German users will never try it. I'd mention the origin of the consortium in the Notes, though.

By analogy then, users outside the US will never try the Library of Congress site so we should rename that filter????

FWIW, I'd call the filter the K10plus filter unless you can suggest that the GBV/SWB storage formats are used at other websites as well.

I just an observer on this as my field of study does not send me to such sites in any case.

[Jon]

I don't agree with that analogy. I think the Library of Congress is universally recognized (and certainly by scholars) as an internationally valuable resource. Few outside of Germany have heard of GBV or GVK. Almost none of the filters' names start with the country of origin -- that was actually implemented first by the German users who contributed them. By calling attention to the country, though, a natural inference is that the source is for German literature. Therefore, unless that's what they're after, most users won't even try it.

I agree with Philologist that K10Plus means nothing, so unless you know that "brand" you'd be unlikely to try it.

I'll probably leave a filter with the name Germany - GBV (GVK), so current users will still have access. And add another with a more inclusive name, such as the ones we've discussed.

Jon
Sonny Software