Source references in individual tindallgrams

[muvster]

It would be a lot easier to fix typos in already edited tindallgrams if a reference to the source document and page number was available. As it is, it's hard to know if a certain misspelling was part of the original text or not. Besides, making it easy to go from the edited version to the scanned original would be a nice feature in itself.

A deluxe version of this might be to make it possible to make individual source pages part of the site somehow, and link directly to those.

Or are there some pesky copyright issues or something that would make this tricky?

[seanredmond]

Good idea. I'm really not sure about the copyright issues (government memos, plus they've been circulating on the internet for a while). Other than that, it's just feasibility and labor.

It might be nice to jump right to the correct page in the entire file in some viewer like Scribd. They're big files, though, and sometimes they run in reverse order or are upside down. The other option would be to split up the big files up and be able to link to the single relevant memo. Someone would have to do all that chopping while also correcting the page order and orientation.

[muvster]

I might have some time to spend on this now. I'm thinking it could be done in steps:

1. Add page information for each memo source reference to https://github.com/seanredmond/Tindallgrams/blob/master/_data/tindallgrams-index.yaml
2. Display page info (if available) on the "by source" pages (e.g. http://tindallgrams.net/source/1). Could be as simple as a tooltip on each source ref.
3. (Optional) Display source and page info on individual memo pages.
4. (Optional) Link directly to source from relevant pages. Requires figuring out a good way to link to specific pages of a source doc.

Steps 1 and 2 could be done without having to modify the source docs, and it seems to me it should be possible to keep the page information accurate even if source docs are later modified to flip upside down pages, for example. Step 3 should be easy enough to add if desired, and step 4 is really just a bonus (my gut feel is that splitting the source docs is probably the way to go unless something clever can be thought of that makes it possible to only download the relevant parts of the pdf).

Thoughts? If you could figure out a way to add page information to the source refs in the yaml index that keeps the structure decent, and then display that info on the rendered pages when available, I would be happy to start going through the sources and updating the index with page references.

[muvster]

Actually, once the page information is in place in the index, it should be a simple matter to script the splitting of the sources into individual memos using something like pdftk cat: https://www.pdflabs.com/docs/pdftk-man-page/. Might be worth looking further into.

[seanredmond]

It's taking me a while to get to this (and to your PR) -- sorry! This deserves a thorough answer and I just need to find the time.