Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Can you read this cursive handwriting? The National Archives wants your help

smithsonianmag.com

Post reply on HN
Print view View original

221–230 of 267 posts

Re: Can you read this cursive handwriting? The National Archives wants your help

#221

Post by tptacek »

Avshalom wrote: post #119
tptacek wrote: post #39

Earlier quoted context omitted.

OK, fair enough, but can you find one in this article that's hard for an LLM? The gnarliest one I saw, 4o handled instantly, and I went back and looked carefully at the image and the text and I'm sold. Like if this is a crowdsourcing project, why not do a first pass with an LLM and present users with both the image and the best-effort LLM pass? Later I signed up, went to the current missions, and they all seem to pos…

Real quick, how long do you think chatgpto4 has existed? How long do you think the National Archive has been archiving?

It's 4o. The crowdsourced transcription project dates back to 2012. My comment is mostly on this article.

Re: Can you read this cursive handwriting? The National Archives wants your help

#222

Post by FireBeyond »

wkjagt wrote: post #150
ChrisMarshallNY wrote: post #120

They should ask a medical school for help ;) My family is Ivy-League, all the way, and has the worst goddamn cursive writing I've ever seen. It can take me an hour to read a Christmas card from my sister.

I've always wondered how pharmacists can read those prescriptions. There must be some kind of course in university that they followed.

A lot of it is understanding the abbreviations.

"2T BD IAF UF", 2 tablets, twice a day, immediately after food until, finished"

Re: Can you read this cursive handwriting? The National Archives wants your help

#223

Post by miltonlost »

Dylan16807 wrote: post #192
patrick451 wrote: post #152

Earlier quoted context omitted.

These kinds of ideas just sound to me like "Suppose you had to use broken technology X. How do you make work?"

I don't think you're wrong, but that's because there are no alternative technologies. The only alternative is leaving much more of the archive inaccessible for a much longer period, possibly forever.

> The only alternative is leaving much more of the archive inaccessible for a much longer period, possibly forever.

No, the alternative is volunteers transcribing. Like this project.

Not every problem needs a computer.

Re: Can you read this cursive handwriting? The National Archives wants your help

#224

Post by flir »

sandworm101 wrote: post #179
prng2021 wrote: post #178

Earlier quoted context omitted.

It was never “solved” unless you can point me to OCR software that is 100% accurate. You can take 5 seconds to google “ocr with llm” and find tons of articles explaining how LLMs can enhance OCR. Here’s an example: https://trustdecision.com/resources/blog/revolutionizing-ocr...

By that standard, no problem has ever been solved by anyone. I prefer to believe that a great many everyday tech issues were in fact tackled and solved in the past by people who had never even heard of LLMs. So too many things were done in finance long before blockchains solved everything for us.

In my experience the chatbots have bumped transcription accuracy quite a bit. (Of course, it's possible I just don't have access to the best-in-class OCR software I should be comparing against).

(I always go over the transcript by hand, but I'd have to do that with OCR anyway).

Re: Can you read this cursive handwriting? The National Archives wants your help

#225

Post by asveikau »

sandworm101 wrote: post #179
prng2021 wrote: post #178

Earlier quoted context omitted.

It was never “solved” unless you can point me to OCR software that is 100% accurate. You can take 5 seconds to google “ocr with llm” and find tons of articles explaining how LLMs can enhance OCR. Here’s an example: https://trustdecision.com/resources/blog/revolutionizing-ocr...

By that standard, no problem has ever been solved by anyone. I prefer to believe that a great many everyday tech issues were in fact tackled and solved in the past by people who had never even heard of LLMs. So too many things were done in finance long before blockchains solved everything for us.

OCR is very bad.

As an example look at subtitle rips for DVD and Blu-ray. The discs store them as images of rendered computer text. A popular format for rippers is SRT, where it will be stored as utf-8 and rendered by the player. So when you rip subtitles, there's an OCR step.

These are computer rendered text in a small handful of fonts. And decent OCR still chokes on it often.

Re: Can you read this cursive handwriting? The National Archives wants your help

#226

Post by astura »

kopirgan wrote: post #195
astura wrote: post #142

Earlier quoted context omitted.

I learned cursive in 2nd grade and was very strictly REQUIRED to use it up until high school, where they stopped requiring cursive. 1) My cursive was always slower than print. I was happy to go back to print so I could write fast. I went to school in the "analog" era, so 100% of all assignments were hand written and not typed. 2) I noticed that literally only 1 person in my school stayed with cursive when printing wa…

Very interesting.. Frankly did not know most of what's said in replies.. That it's not compulsorily taught and more surprisingly it's slower to write! I thought having to lift pen repeatedly would be slower? Anyway I need to try to really know I guess! Versus the time taken to add those extra links. Like most others I've not written much in years perhaps decades, that has screwed up my handwriting as even minor notes…

>I thought having to lift pen repeatedly would be slower?

The extra strokes required for all those fucking loops more than make up for having to pick up the pen.

Cursive probably made a lot of sense when people were writing with quill pens, but in modern times each individual has their own comfort level and preferences.

>Cleared a few misconceptions... One of them being writing in blocks is somewhat 'childish' and cursive is more literate.

I was taught exactly that when I was growing up, which is why cursive was required for all school assignments pre high school. I always thought it was bullshit though because books aren't written in cursive and I only knew a single adult that used cursive in their every day lives. It seemed like a weird academic script.

I think a big reason I was so frustrated with being forced to use cursive in school was because after I learned to write in print and before I learned cursive I wrote a LOT. Like I'd write stories almost every day. I loved writing so much and then they gave me this new script that I needed to use for writing that slowed me down. It's like... Stop changing things on me.

I'm really glad cursive is no longer required in a lot of places. My school years would have been so much better without being forced to use cursive.

Re: Can you read this cursive handwriting? The National Archives wants your help

#227

Post by Aloisius »

astura wrote: post #176
awithrow wrote: post #154

Earlier quoted context omitted.

I'm in the US and learned it in school. I just never really needed to use it consistently. Assignments and papers that were still handwritten could be done either way. Cursive never felt noticeably faster for me to write. I'm sure it would have had I been forced to do it. By the time I was in high school (1999), i remember typing most long form assignments. Now the only time I ever read cursive is on letters from my…

>Cursive never felt noticeably faster for me to write. I'm sure it would have had I been forced to do it. I was forced to use cursive and it was still slower than print.

I’m not sure why cursive would be faster given the letterforms require a lot more travel.

Maybe it would be when writing with a quill where splatter and breakage were a concern, but surely not with a ballpoint pen.

Re: Can you read this cursive handwriting? The National Archives wants your help

#229

Post by dpb001 »

kmoser wrote: post #193
dpb001 wrote: post #134

Earlier quoted context omitted.

Same here. Old enough to remember when your signature on a credit card receipt would be given a quick look to compare it to the scrawl on the back of the card. If this was still being done I’d probably fail 50% of the transactions I attempt.

Nobody has checked the back of my credit card for the presence of a signature in decades, let alone whether the signature matches. (I also haven't bothered to sign my credit card for this reason, but also because why would I want somebody to have my actual signature if my card is stolen?) These days my "signature" on a credit card purchase is usually a smiley face. Nobody has ever complained.

Yup, it’s been decades - I remember it happening with the carbon copy imprinting devices and it may have been more common in the US rural South where I was working at the time. The squiggles I fingerpaint on checkout screens now are my version of your smiley face.

Re: Can you read this cursive handwriting? The National Archives wants your help

#230

Post by Modified3019 »

iterance wrote: post #177
lukeschlather wrote: post #159

Earlier quoted context omitted.

LLMs improve significantly on state of the art OCR. LLMs can do contextual analysis. If I were transcribing these by hand, I would probably feed them through OCR + an LLM, then ask an LLM to compare my transcription to its transcription and comment on any discrepancies. I wouldn't be surprised if I offered minimal improvement over just having the LLM do it though.

Are you guessing, or are there results somewhere that demonstrate how LLMs improve OCR in practical applications?

Someone linked this above

https://trustdecision.com/resources/blog/revolutionizing-ocr...

> Our internal tests reveal a leap in accuracy from 98.97% to 99.56%, while customer test sets have shown an increase from 95.61% to 98.02%. In some cases where the document photos are unclear or poorly formatted, the accuracy could be improved by over 20% to 30%.

While a small percentage increase, when applied to massive amounts of text it’s a big deal.

Post reply on HN