Borrowing From The Lending Library on archive.org
What is this all about?
"If we only have 1 copy of a book, it is only available for 1 hour loan."
"The Internet Archive and participating libraries have selected digitized books from their collections that are available to be borrowed by one patron at a time, from anywhere in the world for free."
https://help.archive.org/hc/en-us/articles/360016554912-Borrowing-From-The-Lending-Library
Also: "Books in this collection may be borrowed by logged in patrons. You may read the books online in your browser or, in some cases, download them into Adobe Digital Editions, a free piece of software used for managing loans." Free as in free beer, but certainly not as in freedom.
I have very mixed feelings about the direction archive.org is taking.
The case in point is the digitized version of a 1942 book, of which I wanted to provide an English version to someone. I cannot simply download the text in a convenient format and forward it to the said person any more, the said person has to create an account, connect and most probably use that Adobe piece of sh*t in order to be able to read that 1942 text for more than one hour at a time. I mean, we are in 2021 and that content has been public domain for some time now in many countries.
I am still struggling to understand the meaning of "if we only have 1 copy of a book, it is only available for 1 hour loan", and of "digitized books from their collections that are available to be borrowed by one patron at a time." This is totally defeating the purpose of making electronic files freely available, and clearly a complete surrender to the DRM idea.
There are fortunately other sources for digitized, public domain books, but archive.org was my main reference for English texts. How to bypass them? Wikisource also has a nice amount of material, but translations are seldom available if the original book is not written in English. That was one of the pros of archive.org for text sourcing.
Someone should write a script that rapidly checks out and downloads the archive's books, screenshots the entirety of each one in seconds, and on the backend turns them into searchable PDFs. Post it all online and call it an Archive of the Archive.
From what I understand, the content itself is accessible to registered users only, and only while logged-in. So it seems that only a script that could create a massive leak would efficiently reverse what amounts to a massive access restriction.
Despair not, though: you can login to archive.org with your Google account. What a relief.
It appears that shared archiving system some have been talking about for a while now is dearly needed. Public domain works could be a good place to start.
Need to decentralize it. Internet Archive (which I use often and love dearly) is a centralized paradigm from the past.
Did you have something like this in mind as output? This could indeed easily be scraped by a bot within the one-hour limit and text-processed offline.
I cannot find the 14-day access anyway. That might be more convenient to process the content, since there would be no need to be logged in while doing the scraping.
Those are some high quality screenshots. I'm sure they could be turned into searchable PDFs.
> Need to decentralize it.
No doubt about that. I would go as far as calling it distributed, because there would be a huge amount of content and referencing workload to distribute evenly (or rather, fairly) among users. I am also reminded of that long discussion about fact checking, which was initiated after a move from archive.org to start flagging content which had been deemed suspicious by random fact checkers from the wild.
Both problems can be solved at once by a decentralized or, possibly, federated model, because they stem from the same centralized paradigm you mention. The more I think about it, the closer I come to believe that federation might in fact have more virtues than some fully decentralized tools for some use cases. If it is fine-grained enough to avoid giving control to a single actor or a handful of actors (see Matrix/Element), then it might benefit from the increased stability and fluidity of an archipelago of user run servers. In theory.
The reason why I usually favor a fully decentralized structure is because federation works perfectly in theory mostly. I can think of only two categories of federation based experiments: those which failed to gain traction (XMPP is an example, as much as I like it) and those which are in fact controlled either by their creator (most trendy "secure" chat apps fall in this category: Wire, Telegram, Signal, which in theory could all be federated) or by a few instances which set the weather for the other ones (Matrix. See also Scuttlebutt: "because the people who hang out there are not jerks," which obviously rules me out anyway). The JitsiMeet and the Lufi instances are good example of successful implementations of the federated paradigm, although I am not sure how much traction they have actually gained.
In fact, I am currently in the critical process of picking a project to focus on. I just cannot even think of initiating the programming of such an endeavor myself, so I'll need to find out how these ideas could be efficiently turned into electronic flesh - by others. I feel a bit like the dwarfs making plans about reclaiming their mountain: "you forget, the main entrance is sealed and the secret door is invisible. Also, there might be a dragon asleep inside". We could indeed use a burglar or ten to reclaim those public domain books.
Then it will only be the federated search engine, the p2p chat app, and very fast we can build that distributed place for public discussion and move most essential material to Gemini. Quick and easy. "Certainty of death...small chance of success...what are we waiting for?!"
Where is that filthy thief Bilbo Baggins when you need him? Always disappearing, so strange...
Indeed, I reckon we shall need all the Hobbiton dwellers to the rescue.
I also hear he has a nephew, who is not shy of walking into dark and inhospitable places. Unlocking the books at archive.org should be a walk in the Shire for him.
Look what I found, while strolling in the archives:
"A new build from pfeco from the Hungarian Puppy community, Etionapup compatible with trisquel packages."
Oooohhh, who's a cute little puppy? Who's a cute little boy? We'll just have to download and try this puppy out I guess...
Ahhh, yes, just like Trisquel - with Waterfox browser! And Google as the home page! Maybe not so much like Trisquel.
And no apt or apt-get! This is very interesting. It actually points to Trisquel sources for its packages, but doesn't use apt, so I'm not real sure exactly how this package manager is working. But I was able to install a package, so there's that.
And a lot of the text in the distro is written in Hungarian language, so it's taking me a bit of time to try to figure out what I'm reading and where I'm supposed to click.
Here's a screenshot. Not the prettiest color combination I've ever seen, but that should be easy to fix.
> Google as the home page!
That's when one suddenly realizes the cozy computing environment that Trisquel ships with. I also sometimes forget how unfree and privacy-less the average computing life can be out there.
> Not the prettiest color combination I've ever seen
Indeed. Hungarian Puppies might have a different color perception than the rest of us.
Anyway, I always had a soft spot for Puppy Linux, and I think it is nice that these iso are being hosted on archive.org. Talking about which, the Trisquel 6 Toutatis screenshots are missing from the list: https://trisquel.info/en/screenshots. I find all these disappearances very disturbing indeed.
On the plus side though, it's only using 50mb of ram!!! Fantastic.
On the not so good side, I gave it 1500mb of memory in a vm, and it's running out when I open 3 browser tabs and two small programs. So I think Waterfox could be a very bad choice for memory management. I tried to check into memory use per tab open, but the second time I started Waterfox it applied updates and then refused to start up again, telling me "Error - Invalid desktop entry file: '/usr/share/applications/waterfox.desktop'. And I can't use the Puppy package manager to install it or uninstall it for some reason.
But, oddly enough when I go to the Puppy package manager, abrowser is available for installation. Guess I'll have to give it a try.
Edit: No, false alarm, abrowser would not install. Puppy package manager said various files needed for installation were not available.
I just noticed that Puppy cleaned its kennel since last time I checked: https://puppylinux.com. Good puppy.
Since the Etionapup hybrid is not exactly bringing the house down, the recommended path would then be to get the last Ubuntu 20 derived version "FossaPup64 9.5" and do your usual libretization tricks (get Linux-libre, remove non-free repo, install Abrowser/TorBrowser if possible, etc.) and stay on the bright side of the RAM.
Some more archive.org curiosities:
https://ia600108.us.archive.org/34/items/mags2017-08_the-last-supper-a-whodunnit/TheLastSupper-AWhoDunnit_2.png
https://ia802905.us.archive.org/35/items/adventure-island-v3/Adventure_Island_v3.jpg
> Since the Etionapup hybrid is not exactly bringing the house down, the recommended path would then be to get the last Ubuntu 20 derived version "FossaPup64 9.5" and do your usual libretization tricks (get Linux-libre, remove non-free repo, install Abrowser/TorBrowser if possible, etc.) and stay on the bright side of the RAM.
That might be hard to do since I can't even begin to understand package management on puppy. I've never spent more than a few minutes looking at these tiny distros out of curiosity. And running a distro with 50mb of ram (before you open a web browser and increase that by 6-8 times of course) may be fun, but it's certainly not necessary on any machine with more than about 1-2gb of ram. Seems like the puppies might be best suited for hardware that's so ancient that even I don't have any of it anymore.
I'm nearly done with my latest respin - antiX on a Debian Sid base with runit as the init manager and DWM and Icewm as the window managers. It's got some really cool technology and advanced package versions, and it can fit on a single 700mb cd. And after that I really want to do the Devuan libretization I've been talking about. That's the one that I could really see myself using for a long period of time.
Sure. I was not suggesting that you do that thing yourself, but that if someone was interested, they'd better try to take these steps than go for the Hungarian hybrid (IMHO). I have not tried any Puppy distro for ages, but the strange RAM-eating-browsser-monster behavior might be caused by some respin choices. I could not start Fossapup from USB, but did not have much time for it when I tried, so I might give it another try. That would allow me to check the current default browser in Puppy. I would bet it is not Wetcat.
I too do not see any necessity to run with less than 500MB of RAM for my current daily computing, I rather see Puppy as an interesting curiosity worth keeping an eye on. I still have a spare eeepc which might soon run better on Fossapup than on any recent version of Debian/Devuan or Trisquel, though. That day, Puppy might help keep it alive.
Other than that, I'd rather go for libretized Devuan or your AntiX respin.