There is also http://tenox.pdp-11.ru which archives and hoards tons of very old software.
From the readme.html file:
# Tenox Operating Systems Archive
This is an archive of operating systems, firmware, drivers and documentation.
We mostly focus on RISC systems of the 90s and early 2000s.
## Directory structure
The main part of the archive is the /os directory that is sorted by operating system name. Each os subdirectory will have install media, apps and docs for that specific OS. For systems that have multiple names eg. `OSF/1`, `Dec Unix`, `Tru64`, we would name it from first to last, eg. `osf1-tru64`.
The other top level directories like /apps /hw and /docs are for categories that do not fit with an OS. For example apps that run on multiple operating systems or docs for hardware that runs many OSes.
Yes, Tenox's archive is awesome and you can find some very rare and interesting bits on there.
But it is also terribly un-curated. Some directories look like someone just slapped a bunch of ZIP archives in there without checking. Some archives don't have version numbers at all, sometimes the version numbers are wrong, sometimes you end up with three archives containing the same disk images just in different formats.
Still, it'S the best place to find rare and unexpected things, so I can definitely recommend it, if you have enough time to do the sorting/curating yourself
In general, bitsavers is a treasure trove for computing archaeology.
There is so much to learn about alternative computing realities, outside what was happening at Bell Labs.
The two decades that predated UNIX, what was happening at Xerox, DEC, IBM, GE,..., while UNIX was being developed, and the early 8 and 16 bit home computing.
So many alternative realities and what ifs to discover.
I looked at the xerox/sdd folder constantly for my first book. If you want to know why the hardware was bit-sliced processors instead of a 68K or 8086, it's all right there.
I think a user friendly modern desktop OS built upon a foundation of OpenVMS would be amazing.
Ultimately of what you listed I think the stuff developed at Xerox shaped modern computing the most. The rest of them focused too much on mainframes, minis, and backend enterprise computing instead of personal computing.
Can files included in a torrent be updated after the fact? I guess the issue is the archive is constantly growing and rsync deals with that gracefully.
Don't quote me on it, but I think this would be possible with v2 of the protocol. Changes were made so that file hashes are also part of the protocol (in addition to piece hashes), so versions of the torrent with added files would not change the hashes of the existing files, enabling users with older versions of the torrent metadata to still join the current swarm and get information on the new files.
This is the case, but with fairly severe caveats. There is no way for a client who wants to download a dead/unhealthy torrent to find other healthier torrents containing the same file(s).
You'd need some kind of central torrent directory to be able to do this, because the DHT has no way to search for torrents containing a file hash.
Torrent swarms still aren't a good way to do distributed archiving.
No, but if you point the new torrent at an old data directory it should skip downloading the existing data.
Although you will only download from people on the same version as you.
IPFS us probably a better fit here as content will be downloaded from anyone who has it even cross-version. IPNS could also be used for distributing updates.
The site archives a lot of old content, from software to print that helped define many of our youths. As an example, I can go to Bitsavers and find copies of all the magazines that I used to read in my high school library in the 1990s.
If you weren’t alive (or into tech) back in the day, you might get a laugh out of some of the predictions. If you were alive/into tech, it’s a place for nostalgia.
From the readme.html file:
# Tenox Operating Systems Archive
This is an archive of operating systems, firmware, drivers and documentation. We mostly focus on RISC systems of the 90s and early 2000s.
## Directory structure
The main part of the archive is the /os directory that is sorted by operating system name. Each os subdirectory will have install media, apps and docs for that specific OS. For systems that have multiple names eg. `OSF/1`, `Dec Unix`, `Tru64`, we would name it from first to last, eg. `osf1-tru64`.
The other top level directories like /apps /hw and /docs are for categories that do not fit with an OS. For example apps that run on multiple operating systems or docs for hardware that runs many OSes.
## Access information
- http://osarchive.org
- ftp://osarchive.org
- rsync://osarchive.org/osarchive
## Public Mirrors
- http://tenox.pdp-11.ru/
## Legal Status
This site operates under a DMCA exemption for obsolete computer media and archives. For more info refer to:
- https://www.federalregister.gov/documents/2021/10/28/2021-23... - https://www.copyright.gov/1201/docs/1201_recommendation.pdf - https://www.copyright.gov/1201/docs/librarian_statement_01.h...