So about 10 years ago, I worked at The New Yorker and responsible for launching the redesign, paywall, and move to WordPress - we had almost the full archive. The data was mostly there. The real issue was actually permissions and rights a lot of the stories. The contracts from 100+ years ago made no mention of the fact that there was going to be an Internet and that we could publish on a domain in a digital format. I can only imagine their gargantuan task it was to track all of that down and get it covered from a legal perspective. I know that was the issue holding us back then. So glad that’s over.
I’ve long thought about trying to map of how the locations of music and maybe theater events listed in the magazine have changed over time.
There are performances of some kind in pretty much every corner of NYC but it’s interesting to see which neighborhoods have had events deemed relevant to The New Yorker readership in different eras.
It also speaks to what we lose when we lose magazine listings of events (New Yorker effectively gutted this section within the past decade), movie showtime listings via newspaper, etc
We have a very strong archive going back a century until about 2015, but now wading through linkrot circa 2017 is miserable
That's a very neat idea! If you ever have the time to do it you should try it out, in fact you've gave me an idea of trying to do the same for my city, Bucharest, just need to find some relevant data-sources.
I hope this gets incorporated into the existing website. I'm not an active subscriber but I used to be and I always thought there was a very fertile "other articles you might like" grounf that the New Yorker never took advantage of, given it's reputation and legacy.
I’ve happily lost hours to following links at the bottom of one story to the next. The new archive still feels a little clunky (search needs a fair bit of work and the OCR clearly struggled in places), but it’s fun to chase down old classics and they’ve done a great job of highlighting greatest hits from the past 100 years.
Plus the (really high-quality) crossword puzzles often have an Easter egg where the big revealer is linked to an essay from the past.
Slightly different question, but does anyone have any info about Google’s digitisation of Mainichi Shimbun’s pre-war articles? The work was announced 3 years ago, but it’s been radio silence since: https://mainichi.jp/english/articles/20221110/p2a/00m/0bu/00...
I think that a better link (even though it lacks the context) is this new archive (which is mostly good as it lets you quickly see all cover pages) - https://www.newyorker.com/archive
But yeah, without a subscription, this still mostly just leads to walled off pages.
Accessing the actual archived version of every issue at https://archives.newyorker.com/ is truly wonderful as they are fully digitized back to back.
Honestly this got me to subscribe. The back catalog is pretty stellar with pretty much every major writer of the twentieth century making a contribution. Zooming in on PDFs just wasn't how you wanted to read them.
Normally when laymen say "digitized" they mean one of two things: scanned images in a PDF or fully transcribed (and possible formatted) text extracted from the scan. The Complete New Yorker you're thinking of was mostly the former, with a bit of indexing (table of contents pointing to the PDFs if I remember correctly).
This latest digitization project does the latter, transcribing the text into their existing content management system and as far as I can tell, preserving much of the formatting. This comes with full text search, allows cross linking between articles, and all that good stuff.
I suspect that since they include an LLM summary and started this digitization project in early 2024, this was enabled by LLMs.
If I’m reading this correctly, they now have all their historic articles loaded into their CMS. I think they previously just had a system where you could page (and maybe search?) through scans of old issues, which is also cool but not as versatile.
When a lot of content was being put out on CD/DVD, a number of publications did but they are not straightforwardly accessible these days because they're usually on an old version of Windows. (Yes, if you want to make a project of it, you can probably get into them but has never been worth it for me.)
Surprisingly, this has been a project I’ve been tinkering with for years. There is an easy way to get the raw png/jpeg files out, but it does require a windows box. Im planning on working on it more over the long holiday.
The CDs I have seem to be proprietary for Windows from the late 90s. But I also have PDFs through 2005 on my computer which I must have "acquired" at some point.
The browser app might be some outdated Windows application, that's the case with the MAD DVD too, but you can find the actual issue files in some folders
There are performances of some kind in pretty much every corner of NYC but it’s interesting to see which neighborhoods have had events deemed relevant to The New Yorker readership in different eras.
We have a very strong archive going back a century until about 2015, but now wading through linkrot circa 2017 is miserable
Plus the (really high-quality) crossword puzzles often have an Easter egg where the big revealer is linked to an essay from the past.
But yeah, without a subscription, this still mostly just leads to walled off pages.
Accessing the actual archived version of every issue at https://archives.newyorker.com/ is truly wonderful as they are fully digitized back to back.
https://www.reddit.com/r/longform/s/zRJgAEdagi
https://old.reddit.com/r/longform/comments/1e8m5s1/the_250_b...
(old.reddit.com takes you to the old UI)
This latest digitization project does the latter, transcribing the text into their existing content management system and as far as I can tell, preserving much of the formatting. This comes with full text search, allows cross linking between articles, and all that good stuff.
I suspect that since they include an LLM summary and started this digitization project in early 2024, this was enabled by LLMs.
https://old.reddit.com/r/thenewyorker/comments/1jlhrve/instr...
Breaking the DJVU DRM would be the perfect solution though
Here's a link to the guy that broke it:
https://github.com/reconSuave/PlayboyPDF/
https://files.catbox.moe/x4np6u.png
A lot of the gen 1 or so CD content isn't easily accessible although a more industrious person could probably get to it in some manner.