I'm not here anymore: https://xitter.cf/t.co/YkJEQfj7PJ https://xitter.cf/t.co/jKrkFqR4iD
Lately
Joined April 2012
- Tweets1.6K
- Following377
- Followers1.1K
- Likes3.5K
Matt Miller retweeted
Putting this out there for anyone who might find it useful: an extension of the marc-json-schema project by @thisismmiller (now with Authority and Holdings too!) github.com/timathom/marc-sch… #MARC21 #XQuery
Matt Miller retweeted
We just covered Matt Miller's GPT experiments with the @librarycongress Susan B. Anthony Papers in our latest blog post and featured it in this month's AI newsletter.
Read here: buff.ly/3ZGDhwv
Sign up for our newsletter: bit.ly/aiforlibrariesnewslet…
Matt Miller retweeted
A government shutdown would pause the paychecks of thousands of federal employees. I compiled FAQs on how this might impact pay, unemployment benefits, health insurance premiums, and more. This link is FREE! Share it with friends who may be impacted.
wapo.st/46iyVh8
Matt Miller retweeted
My first blog post and may I say, it’s a good one
This month's Brooklynology post looks at a battlefield promise made between Union soldiers Josiah Grumman and Alexander Barnie. Click the link to read "I Married the Widow of the Man Who Shot Your Horse." bklynlib.org/3sVOmxo
.
.
[Grumman diary clippings, poem, and POW roster]
Matt Miller retweeted
DJ Phonetic is an open source online tool for making beats with the kicks, snares, and hi-hats that are hidden within historical speech djphonetic.brianfoo.com/
babe wake up, the new hand drawn crowd sourced Bee Movie just dropped thefreemovie.buzz/
Check out this talk tomorrow! Especially if you are interested in using wikibase as a research/DH project platform.
I'm excited to present with @mannatea3: "The @semlabteam's Wikibase: One stop shop for a multi-project knowledge graph" at 2023 LD4 Conference on #LinkedData sched.co/1OPUm @sched July 11 10:50 AM ET @WikibaseHQ @prattischool #LD4_2023
Matt Miller retweeted
Lots of exciting new digital collections up online from @librarycongress! Baseball Publications, Prayer Materials from Africa, the Middle East and Central Asia, historical newspapers, nearly 1,000 OA eBooks and more blogs.loc.gov/thesignal/2023…
Matt Miller retweeted
By transforming E.A.T. @RRauschenberg archival material into #LinkedOpenData, we can ask our #wikibase: What artworks use radios? Query here: tinyurl.com/2oh4n437
#SPARQL #LODLAM #GLAM #ExperimentsinArtandTechnology
Img src: tinyurl.com/bdh38uyy
ALT Yellow lined paper with a block of handwritten text from Robert Rauschenberg. text inscription: "(IDEA OF RADIOS) I WAS ENVIOUS OF CURRENT + ENDLESS CHANGES OF INFORMATION. ORACLE: ALL OF MATERIAL WITH THE EXCEPTION OF THE TECHNOLOGY WERE GIFTS FROM THE STREETS. THE ATTEMPT WAS TO MAKE A MUSICAL INSTRUMENT THAT COULD BE PERFORMED ON WITH OR WITHOUT SOPHISTICATION. SOUNDINGS IS VISUAL AUDIO SPECTATOR CONTROLLED WORK. THE IMAGES ARE CONTROLLED BY THE PITCH OF THE SOUND AND ARE ILLUMINATED FOR THE DURATION OF THE NOISE. THE WORK CAN BE ACTIVATED"
Matt Miller retweeted
Good news for maps! Yesterday @iiif_io & the IIIF Maps Technical Specification Group (TSG) published the Georeference Extension specification. As part of the allmaps.org project, @Sammeltassen and I worked on this spec for more than a year iiif.io/news/2023/05/15/geor…
I bought a .zip TLD: thisismattmiller.zip/
It's like malware, if malware was a zip file of my homepage. Trying to get a copy of this baby on every hard drive out there.
of course, also a new addition to the thisismattmiller family of #brands, joining:
thisismattmiller.com/ (original flavor)
thisismattmiller.club/ ("very cool, very fun")
Matt Miller retweeted
Can you spot any notable names? In the early stages to form #LinkedOpenData from a series of @RRauschenberg E.A.T. edge-notched cards, we look to find what info is expressed about the engineers (front side) and the artists who may have wanted to collab with them (back side)
ALT A scan showing the front and back side of an edge-notched card. It is an Archival document from the Robert Rauschenberg Foundation Archives. The front is typewritten and lists the name, address, professional affiliation and specialty topics; the back is filled with different signatures and dates.
A look at using GPT3/3.5/4 on library and archives collections. Using the crowdsourced transcribed Susan B. Anthony Papers from @Crowd_LOC as a use case.
Using LLM to manipulate and extract metadata and the type of interfaces/data that makes possible: thisismattmiller.com/post/us…
I guess this means my bots will stop working next week? I'm not going to pay muck man to be creative on this platform. Maybe a little thread to send them off, starting with @PDCutup 🫡
And finally @EveryArbys an even dumber bot that required an absurd amount of work. I found a yellow pages dataset and I enjoyed the Arby's Discourse™ on here so...But Google threatened to ban my API account for scraping so many Street View images, so it kind fell out of posting.