Re: [Evolution] What happened to Evolution?

[Pete Biggs <pete biggs org uk>]

> > That would make no sense for a local desktop application - to have to
> > have a database instance to store [cache] mail?
> There are numerous embeddable SQL transactional databases that provide
> fast, reliable
> and ACID properties that would greatly improve the speed and
> durability of something
> important like email. 

What's the point?  My email server already provides a perfectly reliable
repository for my email, why on earth would I want to have a database
taking up resources on my desktop?

Apart from this there are so many other issues with using databases to
store emails - is there a standard for it? If not, Evo will have to
invent some format to use.  Also backing up databases is an horrendous
minefield - it's hardly reliable and "durable" if the database
containing your emails can't be backed up.

> Since its now clear that Evolution has scaling issues, moving in this
> design direction should be a good thing.

I don't think it's at all clear.  My various copies of Evo (including
2.32.1 on F14) work perfectly fine with 10 or so email accounts, each
containing tens of thousands of messages.  I have no stability issues,
no latency issues and the thing that takes the longest on all my
machines is the initial scan of the IMAP servers.

There were some issues when upgrading from a 32-bit system to a 64-bit
one, there are also some stability issues when upgrading the indexes.
Both these things are covered in the FAQ and there have been plenty of
messages on this list about them.  They are easily solvable.

Finally, if, as I suspect, you are using POP and the messages are being
downloaded into an MBOX format file, then you need to look at the size
of that file.  If that file is big, then the speed of the underlying
filesystem will have a noticeable affect on everything.  Try reducing
the size of your INBOX, or if that's not possible then try using a
Maildir format local repository.

Ultimately, you would probably benefit most from moving to an IMAP
server rather than POP.

P.