Anouk Lang

@ffrwelin Followed by me on white nationalism in Narnia and the implications for computational text analysis, and finishing up with @qad on adaptations of Shakespeare for young readers using the YRDL database #DH2023

@ffrwelin Now Nichole Nomura on “readability formulas” and the YRDL database (Young Readers Database of Literature) #DH2023

It’s our #datasittersclub guest Data-Sitter ⁦⁦@jbyszuk⁩! using stylometric methods to investigate ghostwriters #DH2023

@semames1 @RuthAhnert There was a 'tick if you need childcare' box on the registration form! But I don't know what was provided

@RuthAhnert I was actually thinking/hoping that future conferences might provide a space for mini DH-ers! It’s a big enough conference, so there would surely be demand…

Having major DH Graz envy and reminding myself that one day the baby will not need me to snuggle him back to sleep at night and I will miss those days he did.

#DH2023 Of interest to those wanting to use NLP on TEI documents: standoffconverter github.com/standoff-nlp/s… Changes TEI XML to plain text and identifies sentences with <s> which go back into the TEI

⁦@thatbudakguy⁩: Questions around labor turned out to be central to the Humanistic NLP project #DH2023

@ttasovac on humanists learning to tolerate a certain amount of error and imperfection in their datasets: "it's a litmus test of adulthood" <nervous laugh from audience> #DH2023

Q: "So do we now know the features by which fiction can be classified?" Ryan Dubnicek: "I hid that slide" #DH2023

Other takeaways: Sampling logic is more significant than statistical models; publication date is perhaps the most important factor when classifying; human-produced metadata is not infallible #DH2023

Room bursting at the seams for Ryan Dubnicek & @Ted_Underwood's paper on identifying English-Language fiction in HathiTrust. Takeaway 1: ML classification can work, but edge cases are challenging even for humans #DH2023

the wild outer limits of Victorian fantasy fiction

Tell your friends / colleagues / room-mates / enemies / passing motorists / local town crier / etc. This'll be a difficult but important role to fill, requiring a technical generalist's breadth and a humanities heart. Help me help the humanities by finding the perfect person.

@KatherineBode thinking about intersectionality in DH not only as a matter of attending to different groups of people, but also of different modes of collaboration, different disciplinary norms, different data protocols #DH2023

Thomas O’Connor presents the data model for the CLERICUS project, which is looking for collaborators #DH2023

'Not Just Books! Digital Collections and Resources' - Doctoral Open Day at The British Library (25 August) Web archives, metadata, born-digital collections and more - come and speak to our digital research experts! Programme & booking link: bl.uk/research-colla… Please RT!

Here's the link to an article about the Rigasche Zeitung Newspaper project by @krkryger and Kaarel Vanamölder, for those who missed the paper: kirj.ee/acta-publicati… #DH2023

Great paper from ⁦@krkryger⁩ on visualizing data from a corpus of C19th newspapers. This map shows the time taken for news from different places to reach Riga. (He also recommends Top2vec for topic modelling a newspaper corpus) #DH2023

@agiati A new bar has been set for keeping a DH audience’s eyes off their phones and off Twitter #DH2023

Silence in #DH2023 twitterverse is evident of Sarah Kenderdine’s captivating, fabulous keynote.

#DH2023 Sarah Kenderdine describes the astonishing Songlines exhibition which began when Australian Indigenous community members sought help to preserve vital songs & stories. An important step in Indigenous/cultural heritage relations. Go see it on tour! nma.gov.au/exhibitions/so…