Thanks for all the links! Yes I have been using IA-upload quite happily, though it took me a while to figure out the time lag. I assumed there would be other copies, but 17 is a lot. I wish there was a way to find out which is the best copy to upload to Commons - we should have at least one for everything that exists on archive.org in, say, over 4 copies. Another generic question I had about Commons .djvu files was why these can't by default show up in Wikisource on a simple index page if the .djvu files are linked from Wikidata items as being instance of a book, with a statement link to Commons file and a statement link to the author's Wikidata item, which in turn is linked to Wikisource author id.

On Thu, Mar 7, 2019 at 1:27 PM Federico Leva (Nemo) <nemowiki@gmail.com> wrote:

Jane Darnell, 07/03/19 13:09:
> Thanks for this thread, guys, super interesting! Where can I subscribe
> to these notices ?

The notices go to the email address of the account which uploaded to
archive.org, in this case BUB's email.

> I have never received an email from archive.org
> <http://archive.org>.

I've not had one in many years either. Publishers are getting more
aggressive these days:
<https://www.theguardian.com/books/2019/jan/22/internet-archives-ebook-loans-face-uk-copyright-challenge>

> Also, how do you download from Google books (I
> have downloaded google .djvu files from archive.org <http://archive.org>
> before, but the OCR tends to be pretty lousy).

This specific book was uploaded with <https://tools.wmflabs.org/bub/>
when it still worked.

> Next, how do you use ABBY
> to convert a .PDF to .djvu?

I'm not sure I'd recommend ABBYY for the DjVu creation. I've recently
updated the instructions
<http://en.wikisource.org/wiki/Help:DjVu_files>, switching them to a
focus on image quality rather than compression. Some simple tweak in the
command line has a huge impact.

> Lastly, if the counter notice doesn't work,

The counter-notice probably works but I don't want to flood the IA with
extra work in case these takedowns become more frequent. I've sent this
example to the list to see if it makes sense or we should just drop it
in this case. The takedowns are handled by IA staffers in a rather
manual way, I think, so I'm not sure how many they can sustain.

(There's also the possibility that the upload was wrong and the book at
that URL is not actually the PD book found on Google Books but something
else entirely. It has happened a time or two in the past that I know of.)

> can't you just re-upload another version?

Sure, in fact there are already 17 more. :-D
<https://archive.org/search.php?query=%22a%20manual%20of%20ancient%20history%22%20rawlinson>

But better play nice, no need to dodge the counter-ticket process.

> I agree though that we should
> be uploading these files to Commons, and I do this for the ones I care
> about, just to be on the safe side.

Definitely, and this should be easy enough with IA-upload after the
recent fixes: <https://tools.wmflabs.org/ia-upload/>.

There is no need to panic and transfer millions of books now!

Federico