Re: [inn] How to have more than one server

[Harry Putnam <reader@newsguy.com>]

Russ Allbery <rra@stanford.edu> writes:

> Harry Putnam <reader@newsguy.com> writes:
>
>> Thanks for clearing that up.  The documentation with inn is massive, but
>> doesn't really contain an overview that explains explicitly what it does
>> or can do.  At least not in language I understood.  The INSTALL file
>> alone is some 1200 lines, and there are 82 manpages.
>
> INN really isn't targetted for the small user or for someone not already
> very familiar with news.  I'd like to increase its friendliness, but it's
> difficult to do and requires a bunch of infrastructure improvements.
>
> Many individual users are probably better off with something considerably
> smaller, like leafnode.

Hello Russ, I saw your name figured prominently in the documentation.
Just the INSTALL file alone looks like it was a major piece of work.
I fully understand now, how something with this much too it would be
quite difficult to present in a `friendly' way.  And don't envy you
the task one little bit.

It might forstall other dummies like me, if the FAQ (in the overview
section) were to say point blank.  That inn isn't for the faint of
heart and not for light news downloading.  I didn't really get that
message from reading the overview.  Maybe just unfamiliar with the
terms used there.

Jorge Godoy <godoy@conectiva.com> writes:

[...]

> If you need that, I have made RPM packages. Drop me a message if you
> want them. 

Thanks for the offer.  Only problem I have with leafnode.  (And I
don't have it running) is that it wants to second guess what gets
downloaded based on usage.  Is it still the case that you have to
trick it into downloading stuff you don't actually read as news?

I want to just say download these groups, put them in a newsrc or
something and thats that, no tricky heuristics or the like.

Suck can do that, but lacks much in the way of interface to newsrc.
It wants an active file from HOST, and actually expects inn
configuration files.  Although it doesn't depend on them.

Last time I got involved with suck, I wrote shell scripts to convert
gnus .newsrc so it would update the one suck uses.  But lost track of
all that and didn't want to redo it.  It would take rewriting anyway
since it was doing quite a lot more.