-------- Original-Nachricht -------- > Datum: Wed, 28 Nov 2007 11:01:33 -0500 > Von: Dereck Martin <[EMAIL PROTECTED]> > An: > CC: [email protected] > Betreff: Re: [dspam-users] SPAM: Re: [dspam-dev] Dspam
> Well, I know i use dspam 3.8.0-r4 and gentoo portage system has an update > for > 3.8.0-r7. So, someone is doing code updates. > It is not that simple. Some of the revisions are fixes to the purge job (aka the dspam.cron file) and other stuff around DSPAM. The newer editions (currently we are at 3.8.0-r8) are mainly code fixes (we need to act on Bugzilla reports). > I am not sure if it is > gentoo > devs making modifications or if it is someone else. I know they have a > tendency > to provide their own fixes to packages in portage. > Well... this used to be true in the old days. Now we don't do that heavy patching anymore. We try to be as conservative as possible and only include patches we report (and get approved) upstream. With DSPAM it is another issue since DSPAM was +/- frozen development for about a year and then 3.8.0 came out and now again things seams to be stopped. And our Bugzilla gets flooded by bug reports and we need to act or remove DSPAM from Portage (and I would not like to see DSPAM to be removed from Portage). > This package really needs community support. If it is closed from out > side > development, it will not get the proper attention it needs and the people > will > stop using the product. I have outlined some important things I have > realized > when installing and using the software below. I feel that only community > based > development will get this done in an effective way. > > The main reason is this product can't just be installed (like a lot of > closed > development applications tend to be / some binary to install with config > limits) > it has to be specially configured per system based on the MTA being used > and any > other content filtering application like virus scanners and malware > scanners. > Sometimes the mail server resides on the same box as dspam (me), or others > use > it and then forward mail onto a windows exchange server or something. All > these > variety of setups require a lot of knowledge of various particulars. I > feel a > closed developed product just doesn't have the diversity to satisfy this. > A > community based effort can usually reach all these areas. > > The other main reason is documentation. Current Documentation is good... > If you > are in a lab with a perfect distribution setup. Other than that you might > as > well throw it out the window. I have read various setups docs all over > the web, > and I have come to the conclusion that no one has it setup the same. The > docs > don't take in for variety of systems and applications. A community based > wiki or > doc system with real world examples would help with this i believe. It > would > continue to grow as the product would grow as well. > > So two major things, that ideally would require full community > participation... > > minor thing is to fix some issues with the web interface. I am still > having to > fix permissions on folders for every new user. > > ~D > Steve > Yan Seiner wrote: > > David Rees wrote: > >> On Nov 28, 2007 12:47 AM, Andreas Schneider <[EMAIL PROTECTED]> wrote: > >> > >>> Mick Johnson wrote: > >>> > >>>> We'll be putting out the next patch release early next year, and > >>>> would like > >>>> anyone interested in being a tester to please drop us a line! You > >>>> can always > >>>> > >>> Sorry, I don't understand why this is required. Create a source > package > >>> release it to the public. People will try and test it and report back > >>> to the > >>> dev mailing list. There are a lot of people doing patches atm. > >>> > >> > >> Again, I concur. At the very least, there are enough patches floating > >> around to warrant a 3.8.1 release. > >> > >> I'm afraid that if development doesn't happen out in the open, it will > >> be more difficult to attract outside developers and testers. If that > >> is not a priority and development remains behind closed doors, then I > >> would suspect that would motivate people to fork the project. Just > >> look at all the discussion that has take place about forking dspam > >> recently because of the lack of response from Sensory Networks. > >> > > Yes. I participated in one of those "open but closed" tester programs > > for a different product; after a year of testing, submitting bug reports > > and developing what we thought was an excellent rapport with the > > company, we got a terse 'Thank you and the beta mailing list is now > > closed' note, and no more support from the company unless we paid. > > > > It really left a bad taste in my mouth; either treat me as a co-owner / > > co-developer or as a customer, but don't do this halfway open/closed > > thing. I won't do it again. > > > > > >> I am excited to see some response from Sensory Networks. I hope that > >> development can continue as dspam is an excellent spam filter. > >> > > Yes. I'd like to see better installation / setup automation. > > > > --Yan > > > > !DSPAM:474d8842131711213287347! > > > > -- GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS. Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail
