Re: Idea: Parent group

On Wednesday 26 September 2007 14:20:21 Andrew wrote:
> Nothing particularly important, but - just for consistency's sake - it
> occurred to me that ESC-P could be generalized to work in the subscribed
> group list too.

Can this not be done with a macro, using slrn's existing search functionality?

The parent/child analogy isn't a particularly useful one for newsgroup names. 
Every Usenet article that has a References header can *usually* be assumed to 
have had a "parent", but there's no way of knowing whether a newsgroup has 
a "parent" or not. Newsgroup naming conventions aren't consistent between 
hierarchies, or even in the same hierarchy at different dates.

> Let's say I have selected the 'comp.unix.windows.x' newsgroup, pressing
> ESC-P could show and select the 'comp.unix.windows' group.

Neither of those groups is in the canonical Big-8 list, nor are "comp.unix" 
or "comp".

comp.windows.x exists but "comp.windows" doesn't.

comp.sys.laptops and comp.sys.laptops.thinkpad are a rare example of what 
might be called a parent/child relationship, but in fact the groups have no 
necessary connection with each other: comp.sys.laptops could be removed and 
comp.sys.laptops.thinkpad would be unaffected. 

I don't think there are enough examples of such a relationship between group 
names to make what you suggest useful, though as I said it might be possible 
to do it with a macro if it would be useful to you. And you certainly have 
more chance of finding "parent" groups in some otehr hierarchies, such as 
alt.*.

Finding "siblings" might be more useful in general; there are lots of examples 
of them.

> Right now, within the group list, ESC-P calls up a box where you can
> select postponed articles to post, which is kind of inconsistent with
> the pager mode behaviour.

I don't think it's possible to be consistent in the naming of key bindings; 
there are just too many slrn functions and too few letters in the alphabet. 
Besides, as I've said, newgroups don't have "parents" in any generally 
meaningful way, so the "p" association wouldn't be a particularly good one. 
Why not "e" for "[E]xpand hierarchy tree"?

But I'm not a developer, so I'm not the one you need to persuade. :-)

-- 
PJR :-)

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/