Leszek Wroński of the Jagiellonian University has kindly sent me a list of typos in IGT2. Quite cheering as these things go, as the needed corrections are tiny, and certainly no readers are going to be led astray by them.
Joseph Jedwab of Kutztown University has even more kindly sent me a second list of typos for the supposedly corrected version of IFL. This isn’t so cheering, as he has noted some silly foul-ups which could confuse students as well as a whole rash of smaller mistakes. The composite list of needed corrections is now quite embarrassingly long. The good news is that CUP will print a corrected version at the next reprinting (maybe later this year).
To make the corrections for IFL, I’ll have to edit the PDF which is a real pain compared with editing the source document, were that possible. But the original was typeset using a proprietary program that no longer runs on macs under OSX, and exporting to another program wouldn’t preserve fine details of layout. Huh. Moral: don’t, really don’t, use proprietary programs (whether document processors, layout programs, bibliography managers, or whatever) for stuff that needs to last: you don’t know what the future will bring. LaTeX and its companions rule, OK?
I wonder whether, in this high-tech age, there is a market for print-on-demand versions of books, alongside the versions that are printed in large runs and sent to bookshops. You could opt to ask the publisher for a print of this month’s version, with all typos known up to last month corrected. The principle could extend to e-book versions.
I suppose that the main obstacle would be the need to go through the formal process of control over revisions each time.A press probably does not want to let the author, acting alone, tweak the files that feed straight into the printing press. And some corrections would add or remove lines, potentially messing up the layout. But perhaps we are not too far from the continually revised book. The HoTT book illustrates how existing technology may be put to good use:
http://homotopytypetheory.org/book/
I endorse your call to avoid proprietory software. To those who are unconvinced, I regularly point out that open source software is not just cheapar and likely to be longer-lasting, but (nearly always) better.
Here is a more useful link for anyone who is interested in the HoTT production process:
http://math.andrej.com/2013/06/20/the-hott-book/
I’ll just treat the errors in IFL as extra-credit exercises for the students (plus a little bit of “I just wanted to see if you were paying attention…”.)