This is a discussion better held on spambayes-dev, so I'm cc'ing that
list and removing the spambayes list. Since most of the people here
won't have seen the original note, I've included it below in its
entirety.
On Mon, Jan 14, 2013 at 5:15 AM, Thomas Hruska <thruska(a)cubiclesoft.com> wrote:
> On 1/13/2013 9:13 PM, Skip Montanaro wrote:
>>
>> Perhaps you have forgotten that SpamBayes is open source software.
>> You (or anyone else) are more than welcome to grab the source and do
>> what needs to be done to get SpamBayes working again with more recent
>> versions of Windows. I have put out the plea for developers various
>> places more than once and have always been met with silence.
>
>
> This is where you officially move the project(s) to GitHub and pay attention
> to pull requests. You're more likely to get people contributing on GitHub
> than SourceForge. Forking, modifying, and submitting pull requests is just
> as easy as merging and accepting the pull request into the main branch. If
> you don't want to do any development (the hard part anyway), the key is to
> stay on top of pull requests and don't let them sit around in the queue for
> more than a couple weeks. The work on your end becomes rather minimal -
> taking more of a hands-off managerial role.
>
> Automating builds is also important so that someone can grab the latest
> binaries and run them. Once the automated system works, it is also fairly
> hands-off.
>
>> As one of the lesser SpamBayes developers, I actually find I no longer
>> need SpamBayes. Over time, the various organizations which provide me
>> with my daily diet of email (pobox.com and Gmail) have gotten better
>> and better at filtering out spam. Before I actually stopped using it,
>> those providers did such a good job that it was pretty rare that the
>> software was even exercised. I suspect some of the other SpamBayes
>> developers may also be former SpamBayes users. Time marches on.
>
>
> I suspected this was the case. A move to GitHub would at least afford new
> life into the product. There's always room for improvement. Spambayes is
> still useful as a tool in corporate environments where most companies have
> their own mail servers and half of the anti-spam appliances are terrible at
> dealing with spam (legit mail gets hung up more often than not and spam
> leaks like a sieve). For example, Postini is owned by Google, but it does a
> rather terrible job at filtering spam. Google has plans to shut it down
> though. At that point, companies might come running to solutions like this
> if they don't like the results. Positioning this project for that possible
> eventuality would be a smart move.
>
> The key to developing software is to use it daily. You might want to
> consider dropping pobox.com and Gmail if their anti-spam measures are so
> good that you no longer feel the need to develop this product. Enterprises
> still need good server and client-side anti-spam tools and that likely won't
> change any time soon. To that end, this product should shift focus from
> meeting individual needs to the needs of the enterprise.
>
> --
> Thomas Hruska
> CubicleSoft President
>
> I've got great, time saving software that you might find useful.
>
> http://cubiclesoft.com/
I know about the technical advantages of moving from centralized to
distributed version control. That, by itself, won't change anything.
Besides, Sourceforge supports Git, so we could, in theory, convert
from svn to git and leave the code at SF. Does GitHub offer some sort
of "help wanted" or publicity features that make it likely that those
pull requests will just start happening?
I don't understand either you advocate dropping my pobox.com and Gmail
email services. My original reason for needing/using SpamBayes was
that my inbox was littered with spam and none of the other options I
had tried up to that point were effective. Now that they are, I see
no reason I should throw them out just so I can keep using SpamBayes.
I have plenty of other things to do with my spare time.
Skip
