Kathy Harris

Fantastic article! It reinvorces exactly why I'm taking so infernally long to create metadata for my @archive.org collection of literary annuals -- bc once s/o quotes it, that knowledge gets re-quoted & calcified!

Building the metadata for a crosswalk to Internet Archive, Curran Index +

I'm enjoying that he basically is re-publishing all of those original periodical articles and adding an intro which is really a mini-monograph. So many reviewers were aghast when I included big excerpts in my book, but it was bc no one had access to the literary annual texts!

What a boon!

The 60pp intro to the collection 🤯
His economic take here is so interesting and the final section of the intro tackling what I think is very similar to your take the power dynamics of India's cultural growth in the early 19c.

Bonus that this intro is such an expansive history lesson.

I had already gotten up and wandered halfway down the hallway before remembering that he's not there anymore

Today was spent wading thru a fab book by Rahul Sagar *To Raise a Fallen People* that resurrects 19c articles from India's periodicals. When my brain is bursting like today, I usually wander down the hallway to visit my wonderful Philosophy colleague, Anand Vaidya. He passed away suddenly Oct 2024 😥

In searching for a family contact for a deceased colleague's family to ensure we can re-purpose his dataset, I bumped into my 2012 post lamenting lack of $$ & time to do what I'm doing now in 2025 😪 triproftri.wordpress.com/2012/03/06/l...
Thnx to our #DH center, the dream lives 13 yrs later 🤩

Japan #DH was incredible! Next up, discussing critical cataloging at CA Library Assoc for feedback on the line b/w agnostic vocab & scholarly apparatus = metadata

So grateful for ongoing guidance from the community incl @gworthey.bsky.social 🙌

cla2025.eventscribe.net/fsPopup.asp...

@rs4vp.org

[I think I need another year of research leave to continue having fun on this #dh project! ]

After borrowing workflows to tag those 1k images

THEN I'll launch into asking really interesting questions about the development of the short story in early 19c, representations of colonial voices, warping of beauty standards, inclusion of non-English texts & more

@rs4vp.org #digitalhumanities #dh

I now have a much better grasp on what's going to be required to clean the OCR from @archive.org HCOR files that feeds into the workflow for converting into TEI so the corpus embraces the material (serialized) form -- all prepped to share w/other #periodicals scholars

#dh / Minimal Computing FTW!

Wrapped the Japan #Digitalhumanities conf & am so incredibly grateful for this community! Such interesting and engagement people with very poignant and important projects.

Thanks for @rs4vp.org for the support from the @patrickleary.bsky.social Field Development Grant award to attend!

#dh

As is the thing in Japan, heading to the local Family Mart (or 7-11) to pick up something delicious for dinner before jet lag makes me completely incapable of moving & sends me to be very hungry...

Between these 3 days of incredible presentation at Japan #digitalhumanities , BlueSky #DH & sketching workflows w/ChatGPT (using it for good), this feels more & more like getting my hands dirty into the depths of bigger #dh project than I've ever been able to perform 😍😍🤩🤩🤩

#grateful #dreamproject

would you recommend exporting to PageXML (would have to convert HCOR bc Internet Archive doesn't have that format) *before* doing OCR post-correction using PageXML format? or after?

oh ty!

ty! I'm in Japan, so open side of the world. Just now night fall and jet lag somnabulance state

I was just talking to someone about this -- Vinayak Das Gupta suggested the same thing. Thnks for the reference!

well, this is much further along in discovering the process than I thought I would be able to learn this year. Thank you!

Also, we have edu chatgpt 5 access....wondering if there mt be some good uses for it in this work?

oh ffs...I'm going to have to dive into TEI, aren't I...bollocks...

also, it looks like it was later developed to use running headers in some titles, but it really depends on the editor and the year bc they improved the form as they figured out how ppl were reading/using the vols each year

I haven't investigated if the files leave the location of the engravings intact within each vol. though. One of the things about working on these serialized vols as material texts is ensuring that we get those markers so we can run some larger analysis on the entire corpus with that intact.

Oh geez, I was looking at the OCR search text file ...and it's a mess. But, the HCOR file retains formatting and doesn't lose the paratextual markers (like page numbers)...at least in this file: dn721803.ca.archive.org/0/items/forg...

no, but I can find out

Also, HELLO @melissaterras.bsky.social

Internet Archive digitized 83 duodecimo volumes (literary annuals 1820-1860) in June 2025! I've got the OCR files -- in fact, once we go public with IA collection, will deposit clean OCR files into GitHub. Hoping to get it right to encourage more people to work on these vols.

Oh yes...if she checks BlueSky?

I'm wondering if I should take samples of errors from 1 or 2 of the first couple of years of each title (since the titles published annuals for 20 years have the same editors, for the most part). But, I would like to preserve the paratextual pieces (eg pg number & fns) too.

and also realizing that we should use a program that we can train from our own dataset (or maybe borrow from Digital Victorians or maybe Women's Writer's Project, if they have one?) that's specific to 19c print volumes & poetry?

cc @tedunderwood.me another question for you?

Recommendations for automating OCR corrections of scans from 19th print volumes?

Transkribus keeps coming up - anyone have experience with this?

#dh #digitalhumanities

Ok, at a DH conf right now and got a quick overview -- and input these into our sjsu edu Chatgpt to see what it came up with for definitions, tools, humanities applications. I think it just did the work, but how can I know that it's right?