Thoughts on expiry

Thoughts on expiry

[Piers Cawley]

There's an off line reader I use on another service which has a very
useful expiry metric, based on threads.

The idea is that, if the most recent article in a thread with no
unread articles is older than the specified age, that entire thread
gets expired/moved to an archive/whatever. I find this particular
behaviour startlingly useful. 

Does anyone know if there's been work done in this area with Gnus? I
can't find anything referring to it in the docs, but I don't want to
go reinventing the wheel if I can possibly help it (although wheel
reinvention can be a fun learning experience...)

-- 
Piers

Re: Thoughts on expiry

[Lars Magne Ingebrigtsen]

Piers Cawley <pdcawley-ding@bofh.org.uk> writes:

> The idea is that, if the most recent article in a thread with no
> unread articles is older than the specified age, that entire thread
> gets expired/moved to an archive/whatever. I find this particular
> behaviour startlingly useful. 

So entire threads get handled as a single entity, instead of handling
each message separately?  That would be useful; yes.  You would always
be able to refer back to the older messages in a thread as long as
there were new messages arriving in that thread.

However, that goes against the grain of Gnus expiration.  The
expiration process handles one message at a time, and has no concept
of threading at that stage.  So I think this would be quite difficult
to implement.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Thoughts on expiry

[Piers Cawley]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Piers Cawley <pdcawley-ding@bofh.org.uk> writes:
>
>> The idea is that, if the most recent article in a thread with no
>> unread articles is older than the specified age, that entire thread
>> gets expired/moved to an archive/whatever. I find this particular
>> behaviour startlingly useful. 
>
> So entire threads get handled as a single entity, instead of handling
> each message separately?  That would be useful; yes.  You would always
> be able to refer back to the older messages in a thread as long as
> there were new messages arriving in that thread.

Indeed. It's how an offline reader for an online service that I use
works and it's great.

> However, that goes against the grain of Gnus expiration.  The
> expiration process handles one message at a time, and has no concept
> of threading at that stage.  So I think this would be quite difficult
> to implement.

I know. I've had a look at implementing it and it makes my head hurt.

-- 
Piers

   "It is a truth universally acknowledged that a language in
    possession of a rich syntax must be in need of a rewrite."
         -- Jane Austen?