Re: Gnus: Permanently killing a thread?

["Björn Bidar" <bjorn.bidar@thaodan.de>]

Joel Reicher <joel.reicher@gmail.com> writes:

> Christopher Howard <christopher@librehacker.com> writes:
>
>> Robert Pluim <rpluim@gmail.com> writes:
>>
>>> ..kill files are allegedly slow. So what you do instead is 'L',
>>> which allows you to lower the score of a thread, and then you can
>>> use 'V x' to set the score below which messages are automatically
>>> hidden (I donʼt remember offhand if the effect of 'V x' is
>>> permanent or not, as I donʼt really use scoring).
>>
>> The score approaches sounds find, except I was just wondering: Is
>> the score somehow portable across multiple computers? I use multiple
>> backends, including nnimap, nnatom, nnrss, and nntp. However,
>> focusing on the IMAP messages, it seems that if I mark a message as
>> read (which I think is basically the same thing as "killing" it)
>> then the message is still marked as read when I login via Gnus on a
>> different computer. Would hiding still work if I was just marking
>> messages with a low score?
>>
>> I only use nnatom, nnrss, and nntp on one computer, but the nnimap,
>> including the mailing list groups, are used from both computers.
>
> IMAP is a bit of a special case; where possible (always, perhaps?)
> Gnus will use IMAP flags stored on the server for representing things
> like readedness marks, ticks, etc.
>

It could do that but doesn't at the moment. Nnimap Gnus IMAP backend
can already query capabilities. If it implements rfc5464 IMAP METADATA it could be
possible to store these inside the metadata of severs which support the
extension.