iiinova — where your digital self keeps evolving

Josh Hadro @hadro.bsky.social · 07/10/2026

Today in OCR/HTR fun: I'm working with entity extraction over an HTR corpus, and the entity extraction differentiated between versions of the term with and without a superscript "t" -- so "War Dept" and "War Dep<sup>t</sup>" are different entities. Attached image is the latter, in different hands.

A collage of extracted handwritten snippets that all feature "War Dep<sup>t</sup>"

052

Josh Hadro @hadro.bsky.social · 03/10/2026

Elder millennials love waxing nostalgic about printing out MapQuest directions, but 1917 has us all beat: people bought books of fun drives that helped them navigate an era before named highways, let alone interstate highways. Here’s the 1917 Automobile Blue Book 1917 vol 3 trip from NYC to Philly

Screenshot of the drive from NYC to Philly form automobile blue book of 1917 vol 3

172

Josh Hadro @hadro.bsky.social · 01/10/2026

I have yet to see a VLM that can do the following: 1.) handle the entire page at this level of complexity, on what are often degrades page scans, not just clean clips: tile.loc.gov/image-servic... 2.) Preserve the reading order, or at least not jumble it beyond recognition

The full page 4 of The New Hampshire gazette, September 23, 1800, with some complex tables with text going in many directions, and brackets, and numbers

140

Josh Hadro @hadro.bsky.social · 30/09/2026

You: "OCR is a solved problem" Me:

A screenshot of a page from The New Hampshire gazette, September 23, 1800, with text going in every possible direction, and brackets, and numbers

35711

Josh Hadro @hadro.bsky.social · 30/09/2026

How do you show a whole digitized book in one thumbnail? Modern books have illustrated covers, but in research collections the cover is often a plain library binding. So I made Flipbook: hover over a book from any #IIIF source and flip through its pages. hadro.github.io/flipbook/ #GLAMsky 📚📜

410030

Josh Hadro @hadro.bsky.social · 23/09/2026

A true galaxy brain marketing maneuver: Brooklyn Public Library adding a free ebook PSA onto the end of the Jury Duty orientation video Well played @bklynlibrary.bsky.social! Love the suggestion to visit the nearby Center for Brooklyn History as well.

Promo on a video monitor suggesting jurors should visit the Brooklyn Public Library on their lunch break

010

Josh Hadro @hadro.bsky.social · 22/09/2026

If I had a time machine, after I finished all the moral imperative work etc etc., I would go back to 1930s-1950s and figure out how to be a fly on the wall for all the conversations about microfilming library collections to try and grok how it all went down

An essentially impossible to read microfilm image of a 1913 Manhattan and Bronx city directory title page, under exposed to the point of being illegible

1110

Josh Hadro @hadro.bsky.social · 21/09/2026

And once you get into tight multi column pages of city directories, phone books, and more? With indents and ditto marks and ads that cut across columns etc? Digitized from sketchy microfilm? Yowza. Still, fun problems to work on!

A page from a 1906 Upington Brooklyn city directory with two full columns of text and ads that cut across the columns

070

Josh Hadro @hadro.bsky.social · 21/09/2026

You could OCR this page in 2013 just fine — but would the word order be useable? Even now, it’s not simple! That one ad box that cuts across two columns is a beast.

Image of a page from the 1948 national directory of morticians that presents ocr and reading order challenges

2120

Josh Hadro @hadro.bsky.social · 21/09/2026

“For anyone who needs reminding: The ability to do this is, and has always been, the promise of open data.” This could apply to basically every useful and interesting thing I’ve done in my career!

Screenshot of the post from which the quote above was sourced

162

Josh Hadro @hadro.bsky.social · 18/09/2026

Spent the last few days at #ff2026 talking about neat advancements in OCR technology, among other things. Here’s a pic of 2016 me showing the time I used GNU parallels to chain together 10 computers in NYPL Labs over LAN to do mass distributed tesseract jobs. I basically created an OCR botnet.

Josh standing in front of a display screen with command line output on it showing the results of a large scale Tesseract job

160

Josh Hadro @hadro.bsky.social · 12/09/2026

So here's where we've gotten so far, getting pretty clean in terms of excluding ads (both vertical and horizontal), identifying surname start rows, and identifying indented rows that should adopt the surname of the line above. This is complex stuff, especially if you're trying to generalize!

A full page of a 1906 Brooklyn city directory, with boxes around just the entries we want to extract

120

Josh Hadro @hadro.bsky.social · 12/09/2026

The whole page, in case you want to see the glorious context and the other ads

The whole page surrounding the Mahoney's National Detective Agency ad, from 1906 Upington Brooklyn City Directory

140

Josh Hadro @hadro.bsky.social · 12/09/2026

Stop what you are doing and look at this amazing ad from the 1906 Upington Brooklyn City Directory! Nothing screams "all kinds of legitimate detective work undertaken" more than a floating, radiant, disembodied eye. 📜📚

Black and white 1906 advertisement for Mahoney's National Detective Agency, with a floating eye that appears to be radiating light. The banner behind the eye reads, "All kinds of legitimate detective work undertaken; corporations, bankers, & merchants work a specialty."

1166

Josh Hadro @hadro.bsky.social · 12/09/2026

This is my copy of the Iliad I was reading 25 years ago today. I’d woken up early to finish reading Book 16 before class. My dad called my dorm phone, and I made a heavy mark in the margin on the left side to keep my place. He told me “something is happening” and don’t leave your dorm.

Close up shot of a page of the Iliad with a heavy pen mark in the left margin

1112

Josh Hadro @hadro.bsky.social · 06/09/2026

My spouse’s commute for five years was a ferry ride, Greenpoint to DUMBO. I’ve always been jealous of that commute, NYC was meant to be navigated by waterway.

View of the Williamsburg bridge from Domino Sugar Park with ferries navigating the river.

040

Josh Hadro @hadro.bsky.social · 31/08/2026

I want you to think in your mind what a "magnetic physician" could possibly be, and be prepared for it to be so much sillier than whatever you're imagining (From an 1884 NYC city directory I'm transcribing)

Screenshot of a city directory transcription tool I've built, with an entry highlighted, and a hand-drawn circle around "magnetic physician" listed as the occupation for Phebe F. Dusenbury in an 1884 NYC city directory

010

Josh Hadro @hadro.bsky.social · 27/08/2026

This is a staggering read, and this image from the video is sadly not satire

A screenshot from a video in the quoted tweet that says “forward Brookyln” [sic]

012

Josh Hadro @hadro.bsky.social · 26/08/2026

Different kid, same playlist — that year we were top 0.1% global listeners to Jolene

A screenshot of Spotify wrapped showing that we were among the top 0.1% pf global listeners to the song Jolene

031

Josh Hadro @hadro.bsky.social · 26/08/2026

The year my kid was born Dolly was on in our house on constant rotation because we, like countless other parents, found that her angel voice calmed our crying kid, and we were apparently among the top 2% of Dolly listeners that year.

A screenshot of Spotify wrapped showing that Dolly Parton was our top artist, and we were in the top 2% of Dolly listeners

010

Josh Hadro @hadro.bsky.social · 18/08/2026

I've been trying to do a better job of documenting my personal projects and evening explorations, so in July I wrote up a little bit about the background on the Brewery Guides data explorer project I worked on earlier this year. hadro.github.io/blog/brewery...

Screenshot of the Brewery Guides data explorer showing the entry for the Brooklyn Budweiser Brewery on Dean St

131

Josh Hadro @hadro.bsky.social · 11/08/2026

Ok! I can post more thorough thoughts tonight along with all the files, but for now the headline is: this is indeed a *beast* of an OCR/NER challenge, but it's doable. My pipeline out-of-the-box did so/so, but with just a bit of tweaking it can be done relatively cleanly. Teaser screenshot attached

Screenshot of a data explorer that is the output of a Directory Pipeline tool, run on an 1841 business directory section of booksellers

110

Josh Hadro @hadro.bsky.social · 05/08/2026

Film you've watched more than six times with a gif. Hard mode: no Star Trek, Star Wars, LOTR, Marvel, Disney, or Ghibli.

140

Josh Hadro @hadro.bsky.social · 30/07/2026

You can see amidst the rubble the remains of the Statue of Liberty illustration that was painted over the old Sherita sign at Atlantic Av and Classon Av in Brooklyn

Closeup of some rubble with black on white illustration lines at the top of the pile

Medium distance shot of the same rubble, clearly the remains of a building just demolished

A shot of a building with scaffolding around it, about to be torn down

000

Josh Hadro @hadro.bsky.social · 28/07/2026

I don't generally trust automated historical geocoding because streets change often, but for a work-in-progress the NYC addresses are basically stable and useful as a broad indicator of where listings clustered.

A screenshot of a map data viz, with bubbles emanating from the center of NYC neighborhoods, sized by the number of entries contained in that neighborhood. Central Harlem (772), Midtown (255), and Bed-Stuy (196) stand out as the largest bubbles.

120

Josh Hadro @hadro.bsky.social · 25/07/2026

The first picture I took of my kid outside the NICU had the Sherita sign in the background

Picture of a very tiny baby in a car seat, with the Sherita sign on a Brooklyn street corner in the background

010

Josh Hadro @hadro.bsky.social · 25/07/2026

Fare thee well, Sherita sign and the building upon which you stood Your history was sketchier than almost anyone knew (www.thecityreporter.nyc/2025/05/19/s...), but you were a Brooklyn landmark nevertheless

The blank billboard where the Sherita fuel oil alien thing sign was once painted, on a building about to be demolished, at the northeast corner of Atlantic Ave and Classon Ave

120

Josh Hadro @hadro.bsky.social · 15/07/2026

And a companion view that collapses all those individual listings into a listing of "establishments" -- or at least does the best that can possibly be done programmatically for ~100,000 individual listings written by humans and typeset into travel guides in the middle of the 20th century.

An "establishment" view of a search for "Dew Drop" which shows 17 establishments across 95 individual listings, in cities across the country.

030

Josh Hadro @hadro.bsky.social · 15/07/2026

Sneak peak at the desktop view

A data explorer view of an entry for "Liberty Hotel" in Atlantic City in 1959, with a data viz showing the entry for the same business in 7 different travel guides across a 32 year span

140

Josh Hadro @hadro.bsky.social · 15/07/2026

Getting pretty close to finishing a first pass on a fun personal project that brings together travel listing across a bunch of different publications from ~1930-1966

A data visualization of an entry for “Liberty Apartment Hotel” in Atlantic City listed in seven different travel guides

180

Josh Hadro @hadro.bsky.social · 22/06/2026

I cannot urge you strongly enough to spend a random evening every now and then hand-transcribing 150 year old city directories from where you live, because you will find out that an inventor from 1876 lived around the corner from you!

A screenshot of a custom-built transcription tool for Brooklyn city directories, showing an entry for a Lucius K Gibbs who was in inventor in 1876

130

Josh Hadro @hadro.bsky.social · 22/05/2026

Maybe if we went back to the era where license plates were printed in annual indexes along with your name and home address, there’d be more accountability for drivers running over pedestrians E.g., digitalcollections.nypl.org/items/0b4ed4...

Screenshot of the first page of the 1915 “Automobile Digest and Register” for the greater NYC area

030

Josh Hadro @hadro.bsky.social · 16/05/2026

21 entries in the explorer … you may be able to guess what source material I’m working with

Screenshot of a search for “dew drop” in a historical travel guide explorer m

010

Josh Hadro @hadro.bsky.social · 16/05/2026

I’m working on a data explorer for some mid-19th century travel guides, based on the output of my Directory Pipeline tool, and I’m really jazzed about how it’s coming along!

A screenshot of the mobile view of a page showing a snippet of an entry from a travel guide, along with a “see all likely match listings” feature and an Open Streetmap “approximate location” feature

261

Josh Hadro @hadro.bsky.social · 20/04/2026

Everyone wants to live in Roku City, but I woke up last night and couldn't get "Roku Suburb" out of my brain so now you have to witness an approximation of it

An AI generated image of "Koru Suburb," a riff on the "Roku City" screensaver on roku devices. it depicts a bleak, desolate American suburban wasteland. Foreground: a shuttered strip mall with flickering half-dead neon signs, a vast empty parking lot with faded yellow lines and a lone shopping cart. Midground: a dark closed big-box store, a fast food pylon sign with several letters missing, utility poles with tangled wires. Background skyline: a water tower, a cell tower, and the underside of a highway overpass.

161

Josh Hadro @hadro.bsky.social · 19/03/2026

Ha, it is there, but it was called Lytle-Stoppenbach Company at the time! Most volumes have an entry for it, e.g. hadro.github.io/brewery-guid... and hadro.github.io/brewery-guid... What a fun connection!

1896 volume brewery and malster directory entry for Lytle-Stoppenbach Co. (Maltster) in Jefferson Junction, WI

101

Josh Hadro @hadro.bsky.social · 19/03/2026

I'm fascinated by digitized historical directories, and I'm a big #IIIF booster, so naturally I spent some recent evenings building the tool I wanted to see in the world: a "Data Explorer" view of 12 brewery directory guides from 1896 to 1918: hadro.github.io/brewery-guid...

Screenshot of the Data Explorer tool applied to 12 brewery directories, published from 1896-1918

2103

Josh Hadro @hadro.bsky.social · 10/03/2026

If nothing else, it makes it possible to explore the dozens of breweries that a century ago were all within a few miles of where I live!

Screenshot of a brewery guide data explorer I put together recently

010

Josh Hadro @hadro.bsky.social · 03/03/2026

Calling all technologists, artists and other creative visionaries: The Library of Congress has opened the call for the next Innovator in Residence! Apply by 2pm ET April 10 newsroom.loc.gov/news/library...

Library of Congress Innovator in Residence graphic

11626

Josh Hadro @hadro.bsky.social · 28/02/2026

“I wonder if it made the deadline for the print edition…” “Oh.”

The bottom left corner of the Saturday Feb 28 NYTimes front page that reads “Iranians Cite Progress in Talks, p.A9”

000

Josh Hadro @hadro.bsky.social · 28/12/2025

post a perfect album from the 90s that isn't nirvana, pearl jam, soundgarden, or alice in chains

The album cover for Cure for Pain by the band Morphine

191

Josh Hadro @hadro.bsky.social · 06/12/2025

Pretty close, cat and opossum (hell yeah) make an appearance

Screenshot of the comments of the chicken ragu bianco recipe from NYTimes cooking

010

Josh Hadro @hadro.bsky.social · 20/10/2025

On one hand: furloughed 👎 On the other hand: afternoon pinball with a 4 year old, hell yeah (2-player cooperative grand champion to boot 💪)

A Jaws pinball machine with the top of a four year old’s head in front. The pinball display shows a screen for entering the initials because the players achieved the two-player cooperative grand champion high score 💪

181

Josh Hadro @hadro.bsky.social · 02/10/2025

However, unexpected dagger through the heart that this is a museum piece 📰

A old newspaper box where you could buy paper copies of NYTimes for $0.75

0263

Josh Hadro @hadro.bsky.social · 02/10/2025

On one hand: furloughed On the other hand: weekday visit to NY Transit Museum with a 4 year old, hell yeah

3825

Josh Hadro @hadro.bsky.social · 20/09/2025

Tired: getting into 3D printing and generating cheapo schlock Wired: Marrying a sculptor/ceramicist and asking for ridiculous custom hand-crafted artisanal kitchen gadgets based on hare-brained ideas

A custom drip-catcher for pour-over coffee with a teabag holder extension, all made out of wet gray clay

050

Josh Hadro @hadro.bsky.social · 16/09/2025

This is the hat I wear every day. RIP to an all-time great who led the ensemble cast of one of the greatest caper films of all time.

A basic black baseball cap with “setec astronomy” spelled out in Scrabble letter

060

Josh Hadro @hadro.bsky.social · 10/09/2025

Crystal Hudson just sent out an email blast telling ebike riders to "SLOW DOWN!" Where's the email blast to car drivers responsible for 99% of the injuries & deaths? Where's the outrage about ghost plates? Or how about the single 2024 Kia Forte terrorizing her district and racking up speeding tix?

In bubble chart map of NYC infographic showing the top 10 super speeders by color, with substantial bubbles in every borough including an orange color identifying a white 2024 Kia Forte racking up tickets in Crystal Hudson's district 35.

220