🚀 go-pugleaf

[Please repost this wherever you reposted the previous message]

The spelling checker patent that I posted contains several claims.
Each of these claims acts like an independent patent, in that prior
art defeats only the claims to which it applies.

Claim 1 is the most general; thus, it does the most harm.  It is also
the easiest to defeat, since anything that fits claim 1 would work.
I think we have prior art now for claim 1 at least.

Claim 2 is more specific, and thus a little harder to defeat.  And so
on.  Note that most of the claims refer to previous claims.  This
means that the earlier claim is, in effect, part of the later claim.

If you think you have found prior art, please try to determine which
claims might be affected, and give us this information in your message.

Some people have mentioned examples such as the Tenex command
interpreter, based (as far as I can tell) only on the user-visible
feature of command completion.

These examples are not necessarily relevant at all.  The patent covers
a particular kind of data structure.  So a program is relevant only if
it uses that kind of data structure.  This is something you could not
determine just by running the program; you have to know how it works
inside.  If I were implementing command completion, I would use a
simple list of command names, which is not what the patent is about.

Aside from that, a program that uses this data structure only counts
if it was *published* that it uses this data structure.  Keep in mind
that only publication *before 1985* counts, since that's when the
patent application was.

If you send just a lead, not a particular publication, then someone
may eventually get back to you to ask you to follow it up.  That's
because the LPF doesn't have a lot of volunteers to do work like this.
Finding someone to look through the messages to find the promising
ones will be difficult enough.  So if that person asks you to do a
piece of the rest of the work, please help out.