Javier Cha @javiercha.bsky.social · 05/10/2026I've been brainstorming ways to change the interface so that LLMs work like a very advanced thesaurus. On the question of what percentage of the writing is to be done by humans, my personal take is that the final touches still need to be done by a human. The final task shouldn't be copy and paste. 140
Reposted by Javier ChaLauren Klein @laurenfklein.bsky.social · 30/09/2026Opinion-havers! Historicists! Nerd friends! Leftists! All/none of the above! Submit a proposal to GENERATIVE DIGITAL HUMANITIES, a new volume on AI and DH to be edited by @mkirschenbaum.bsky.social @lnakamura.bsky.social @zentralwerkstatt.org and me! Deadline 11/1. Please share! 11411
Javier Cha @javiercha.bsky.social · 30/09/2026This is going to be an important volume, but you're going to face the same challenge the critical infrastructure studies volume faced: it'll be at least two years between the completion of the first draft and the actual publication of the chapters. 110
Javier Cha @javiercha.bsky.social · 29/09/2026You can do this in Python. codecarbon.iocodecarbon.ioCode CarbonOpen tools to measure the environmental footprint of computing and AI — CodeCarbon for local compute, EcoLogits for GenAI APIs. 130
Reposted by Javier ChaRebecca Solnit @rebeccasolnit.bsky.social · 28/09/2026Campus rape is really widespread; statistics suggest a significant percent of women who go to college will be raped there; feminism opened up a good conversation about it around 2012-13; it's weird when people treated Cornell like an isolated rather than dramatic example of a pervasive problem. 412411525
Reposted by Javier ChaMaria Antoniak @mariaa.bsky.social · 26/09/2026I get this instinct, but in the world of AI, philosophy has been privileged for a long time. These aren’t questions with final answers, and we’ve had rigorous conversations about them for thousands of years. I find going to other parts of the humanities and social sciences more practical. 2809
Javier Cha @javiercha.bsky.social · 26/09/2026100%. Give me more empirical research that helps us make sense of the messy world. 050
Javier Cha @javiercha.bsky.social · 26/09/2026Not the first time we've seen academics engage in police roleplaying--and upon facing a serious academic misconduct case, a lot of them will be inclined to protect their own tribe. (I hope I wrote that in a clunky enough manner.) 010
Reposted by Javier ChaArtjoms Šeļa @artjomshl.bsky.social · 24/09/2026it might be my bad mood speaking, but most universities on the planet are more involved in producing civil servants and managers than producing knowledge. It's kind of a wonder we still call all this "higher education", hire PhD students, deal grant money, run academic presses. for how long? 073
Reposted by Javier ChaArtjoms Šeļa @artjomshl.bsky.social · 24/09/2026nothing is more devastating than hearing your colleague, for whom English might be, idk, fourth language, read out all those llm-speak americana cliches in a conference talk. It doesn't lead to more egalitarian spaces, or more just treatment of talent. It inevitably signals status. 1121
Reposted by Javier ChaArtjoms Šeļa @artjomshl.bsky.social · 24/09/2026use of LLMs is pervasive in academia, everyone is depressed about it. but who can afford ~not~ to rely on AI? those who already hold high social capital: good English, good schools, good networks. For everyone else the 'baseline' is already far out of reach without LLMs 3235
Javier Cha @javiercha.bsky.social · 25/09/2026Revisit materials from the 1960s and 1970s. A few weeks ago, I revisited Isaiah Berlin's "scientific history" essay, for example. STEM students also seem to like John Gaddis. 040
Javier Cha @javiercha.bsky.social · 23/09/2026In 4 years, we've come from whatever this was to... Claude Code. 021
Reposted by Javier ChaTransactions on Machine Learning Research @tmlrorg.bsky.social · 16/09/2026Spoilers: all 10 ended up being desk rejected. - 1 paper withdrawn - 1 author was too busy for a call - 1 author no-showed - Authors of 3 papers couldn't answer basic qs - Authors of 3 papers could, but not technical qs - 1 author answered all qs, but paper had a major flaw 26114
Reposted by Javier ChaTransactions on Machine Learning Research @tmlrorg.bsky.social · 16/09/2026TMLR has faced a deluge of submissions, necessitating stricter desk rejection policies due to limited reviewer capacity Co-EiC Nihar Shah reached out to authors of 10 papers slated for desk reject. Could they answer questions about their *own* submission? medium.com/@TmlrOrg/ask...medium.comAsking Authors About Their Own PapersBy Nihar B. Shah 116965
Javier Cha @javiercha.bsky.social · 16/09/2026After four weeks of nonstop coding, it's time to slow down a little and engage in some reflections. As I regularly remind my students and lab members, a digital humanities scholar who only engages in technical work without reading and reflecting isn't really in the humanities. 0101
Reposted by Javier ChaEugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 15/09/2026"LLMs are just n-dimensional classifiers" has a very similar flavor to "everything is just quantum mechanical interactions." www.science.org/doi/10.1126/...science.orgMore Is Different 410512
Reposted by Javier ChaChristof S. 🇪🇺 @christof.fedihum.org.ap.brid.gy · 13/09/2026Time flies! I've arrived here in Hong Kong already a week ago. I'm here for a month as a first part o my two-year role as guest professor at the University of Hong Kong's Faculty or Arts, specifically in the context of their "Arts Tech Lab". See: www.atlab.hku.hk 184
Reposted by Javier ChaACL 2027 @aclmeeting.bsky.social · 11/09/2026ACL Sustainable Reviewing Policy: We are introducing changes in the @ReviewAcl reviewing and submissions. The changes will cap authors and introduce changes in the reviewing to keep our community sustainable. #NLProc #NLP 12712
Javier Cha @javiercha.bsky.social · 04/09/2026I also ask that journal editors and conference organizers please stop sending requests to review a manuscript or an abstract that obviously should've been desk-rejected or sent back to the author(s) for revision. I must be doing close to 40 reviews a year. This is unsustainable. 010
Reposted by Javier ChaScott McGrath @smcgrath.phd · 04/09/2026🧪 We are staring down a crisis with peer-review. Editors needed 4.5 invitations to secure a single peer reviewer in 2025, double the effort required back in 2018. Meanwhile, we have half of surveyed scientists admit to using AI to draft their reviews. This isn't sustainable. #AcademicSkyscience.orgCan AI help solve the peer-review crisis? Here are its promises and pitfallsAmid a shortage of human reviewers, tests indicate AI can assist—but is not a savior 1174
Reposted by Javier ChaSeth Rudy @sethrudy.bsky.social · 03/09/2026Me: to study the humanities is to cultivate a fuller understanding of the mind as we seek to find meaning in the past, present, and future of our individual and collective existence Them: are you happy Me: god no 1399
Reposted by Javier ChaTed Underwood @tedunderwood.com · 03/09/2026Imagine if we decided, since the humanities == the good life, we have to hire whatever theorists empirically make ppl happiest. 3202
Javier Cha @javiercha.bsky.social · 01/09/2026I love how I now have the ability to visually express my research philosophy. Over the years, I've experienced so much difficulty trying to convey this to my designers. 000
Javier Cha @javiercha.bsky.social · 01/09/2026Thanks to AI agents, I've managed to completely redesign two of my existing websites and create a brand new one ... in 6 days. Brave new world. And I'm beyond pleased with the results. Check out: javiercha.com bigdatastudies.net samhan.ai 110
Javier Cha @javiercha.bsky.social · 28/08/2026Wow! Retro to the max. Brings back so many memories. 010
Reposted by Javier ChaRetro Tech Dreams @retrotechdreams.bsky.social · 28/08/2026Download dot com's most popular downloads at the end of 1996 832664
Reposted by Javier ChaHillary Rodham Clinton @hillaryclinton.bsky.social · 27/08/2026It’s Lake Ontario. 1907351445588
Reposted by Javier Chawomensartbluesky.bsky.social @womensartbluesky.bsky.social · 27/08/2026Yayoi Kusama (1929-2026) RIP 3065651430
Reposted by Javier ChaInternet Archive @archive.org · 27/08/2026📚 What looks like the work of a solitary author can be the product of a system. In BIG FICTION, Emory professor Dan Sinykin argues books are shaped by a larger system: agents, editors, marketers & others whose decisions shape the page. 🎧Listen: futureknowledge.transistor.fm/episodes/big... 🧵 29322
Reposted by Javier ChaAlliance of Digital Humanities Organizations (ADHO) @adho-org.bsky.social · 25/08/2026📢 AI and History Conference 2026 📍 October 15–16, 2026 | Johns Hopkins University, Baltimore, MD 🔗 The conference program is available at: proflouishyman.github.io/ai_conferenc... ⏰️ Registration closes September 15, 2026. 🔗 More info & registration: proflouishyman.github.io/ai_conferenc...proflouishyman.github.ioAI and History Conference 2026A practitioner-focused conference for historians using AI and computational methods, October 15–16, 2026 at Johns Hopkins University. 056
Reposted by Javier ChaCanadian History Ehx @cdnhistoryehx.bsky.social · 25/08/2026Before Europeans arrived, these were some of the names for The Great Lakes: Superior: gichi-gami (great sea) Erie: erielhonan (long tail) Michigan: michigami (long body of water) Ontario: Ontarí'io\kanadario (great lake\beautiful water) Huron: karegnondi (fresh water sea) 293121264196
Reposted by Javier ChaBrett Bobley @brettbobley.bsky.social · 25/08/2026Highly recommended reading from my wonderful former colleague at NEH, Molly Hardy. (Related aside: I once asked one of the DOGE boys "who would ever want to work here after what you've done?" To which he replied, "oh, NEH will be so much better soon." Yeah, fuck off.) 36116
Reposted by Javier ChaAlliance of Digital Humanities Organizations (ADHO) @adho-org.bsky.social · 25/08/2026📣Call for Papers: Hong Kong Association for Digital Humanities Annual Conference (HKADH) | Model/Making: Toward Generative Humanities ⏰️Submission deadline: September 1, 2026 🔗 More information: 2027.hkadh.org#cfp 🔗 Learn more about HKADH: hkadh.org2027.hkadh.orgHKADH 2027Model/Making: Toward Generative Humanities. 21-24 January 2027, Hong Kong. 077
Javier Cha @javiercha.bsky.social · 25/08/2026I think I know that journal! I had a good laugh when it happened to me. 110
Javier Cha @javiercha.bsky.social · 24/08/2026CFP deadline is approaching! In my original post, I shared the wrong deadline. 🫣 Please submit your abstract by 1 September. #HKADH2027 031
Reposted by Javier ChaJames Cummings @jamescummings.bsky.social · 28/07/2026Very happy that the Hong Kong Association of Digital Humanities (HKADH) has joined @adho-org.bsky.social! Announced at #DH2026. 0154
Reposted by Javier ChaAlliance of Digital Humanities Organizations (ADHO) @adho-org.bsky.social · 29/07/2026🎉 We are delighted to welcome the Hong Kong Association for Digital Humanities (HKADH) as the newest constituent organization of the Association for Digital Humanities Organizations (ADHO)! Please join us in welcoming HKADH to the ADHO community! Learn more about HKADH: hkadh.org #DH #ADHO #HKADHhkadh.orgHong Kong Association for Digital HumanitiesAdvancing New Pathways in Digital Humanities in Asia’s World City 0156
Reposted by Javier ChaRetro Tech Dreams @retrotechdreams.bsky.social · 21/08/2026Sid Meier's Civilization II (1996) 720522
Javier Cha @javiercha.bsky.social · 21/08/2026Eventually, I'll look for a suitable local open-weight model to take GPT's place and release the entire source code online. Then others can fork it and continue customizing the web application according to their needs. A software paper, along with my methodological reflections, will soon follow. 021
Javier Cha @javiercha.bsky.social · 21/08/2026This is how NER should be done. I used Claude Code to add LLM integration. For the time being, I'm using GPT-5.6 Luna, which costs literally just pennies to process these requests. Notice how this approach makes provenance visible and invites fellow historians to inspect the results. 191
Javier Cha @javiercha.bsky.social · 16/08/2026After human verification of all the records (though I don't expect to find many errors), I will release the whole thing so that you can use it for research or in class. 010
Javier Cha @javiercha.bsky.social · 16/08/2026There is an option to export this exactly visualization as PNG or SVG, so that I can include it to my book or article, as is, and I have the option to toggle between colour and black & white. 100
Javier Cha @javiercha.bsky.social · 16/08/2026If you know anything about premodern Korean history, this is truly mind-blowing. I'm still in disbelief that this whole thing is ready to be deployed after spending less than half a day working on it. 100
Javier Cha @javiercha.bsky.social · 16/08/2026After 5 minutes to produce the prototype and 5 hours of debugging and refining the features with my input, this is the result. It's called Sebo: Koryŏ Kinship Workbench. The name itself is Claude's own suggestion. 110