[Spambayes] RE: 'SpamBayes fails to filter most-recent message at
Outlook startup' - responses to requests for clarification
tameyer at ihug.co.nz
Fri May 28 01:55:36 EDT 2004
> The Subject-ed above problem does indeed appear (2 cycles)
> to have been corrected in 1.0rc1, though as noted below it
> was not part of FAQs.
The FAQ only covers questions that appear here on this list all the time.
The release notes for each version describe the (major) problems that have
been fixed. These notes are accessible via the download page (click on the
title, e.g. "1.0rc1", although that's hard to see unless you read carefully,
and also mentioned in the announcement that gets made (to, e.g.,
spambayes-announce at python.org). In any case, asking here doesn't hurt, and
if enough people do, it'll just be made a new FAQ.
[posting to spambayes-dev]
> Noted. I used the only contact point we had been provided
> here at Unisys.
Can I ask who provided the contact point? (Just so that I can let them know
what the appropriate methods are). You can tell me this offlist, if you'd
> 1.0rc1; Unisys says that the latest it has tested and offers
> for download is 1.0a9 (? whatever the numbering styles mean).
The development goes through various stages - alpha, then beta, then final
(each stage is more thoroughly polished and tested - adding features and the
like tends to be pre-alpha, or alpha, rather than later on). The "a" or "b"
in the versions stands for "alpha" or "beta", and if there's more than one
(eg) alpha, then there'll be a1, a2, and so on. Each time one of these is
put out, there are also "release candidates" - basically these will be
exactly like the final version, and are just to shake out any problems in
packaging. Since the 1.0 (final) release is a big deal, the "release
candidate", or 1.0rc1, version is more widely available than normal.
After 1.0 comes out, you should upgrade to that, and then you can probably
ignore any version that has a letter after it, unless you need/want to be on
the bleeding edge (or you encounter a serious bug that doesn't effect many
other people, and so need a version that fixes it). SpamBayes has gone
through more development (and so alpha and beta versions) than a lot of
software does before getting to the initial 1.0 version, which is why so
many people are using these "pre release" versions.
I gather that someone at unisys is taking care of testing each version and
making it available for people there (which is absolutely fine by us).
Presumably they'll get the 1.0 version when it comes out (and it doesn't
appear that the plug-in will be changing between 1.0rc1 and 1.0), test it,
and let you know that they're happy for you to use that.
[my signature about reply-all'ing]
> Sorry - I'm not clear how to avoid this yet respond to your
> requests for clarification. Is it spambayes at python.org alone
> - or one of websites you note above - you wanted me to reply-to?
Sorry - that just means that any replies should be sent to me
(tameyer at ihug.co.nz) *and* to the list (spambayes at python.org), which is
exactly what you did. Typically this is done with the "reply all" action.
Often people just reply to me, so it never goes to the list, which isn't so
good (takes longer to get a response, isn't archived, doesn't get help from
anyone else, and so on).
More information about the Spambayes