A mix of both might not be a bad idea.  Having some sort of automated script for most of it but then hand-editing when something like an author is changed?  I'm no programmer but I'm sure there is a way to automate some of it.  That being said, as a person with a disability, most of my work right now in RL is on a volunteer basis while I get some help looking for a paid job. These files usually seem to be updated once a week as it is, so i'm sure an automated script running 1x a week to automate most of this and then a manual hand-edit when correction posts are made, seems like something you wouldn't have to devote a lot of time to.  I'd volunteer to do that if no one else can, but I'd have to find someone to write a programming script for automating it, first.

Jared

On Sat, Jul 19, 2025 at 1:41 PM Greg Newby <gbnewby@pglaf.org> wrote:
The long-time maintainer, Larry Copenhaver, has had to step back from
this role due to health issues.

The GUTINDEX.ALL files under "offline catalogs" at
https://www.gutenberg.org (and companion files like GUTINDEX.2000) are
Project Gutenberg's accession files. They list all our publications in
numerical order.

Many people use these frequently (including me).

To continue to maintain them, either:
1. A new maintainer needs to step forward
2. We could try to develop automation to maintain the files

The maintainer role is to simply copy Posted announcements to the
right place in the GUTINDEX files.

An automated solution would plug into the suite of "workbench" tools
the production team (WWers) use. If you've seen the Posted email list,
you can see that the top line of a Posted announcement is what goes
into the GUTINDEX files.

Generally the lists simply grow with every new posting. Occasionally
they need some hand-editing, such as when we remove items or fix
the title or author.
  ~ Greg

On Sat, Jul 19, 2025 at 12:31:28PM -0400, Dan Lowe wrote:
>
> On Thu, Jul 17, 2025, at 1:21 PM, Greg Newby wrote:
> > Indeed, we might have lost our maintainer. I've reached out...
> >   ~ Greg
>
> The format of those files is not overly complex. Do we have clean enough metadata to generate them programmatically? I assume the book title and book number are low-hanging fruit, but I'm not sure about other information, e.g. illustrator(s), editor(s), subtitle...
>
> Dan