Dear all,
Without in any way or another trying to start the old discussion, I would just like to announce that I have created a free external service that monitors SME servers. For now it does DNS blacklist monitoring and SMTP test to see whether server is responsive. So far 68 servers from 17 users have been registered so there is some life out there J If the userbase grows and admins find this service useful, I will start to spend time on creating some more features as per my previous email. The site for those who cares can be found at: http://smeoptimizer.swerts-knudsen.dk Best regards, Jesper From: [email protected] [mailto:[email protected]] On Behalf Of Jesper Knudsen Sent: 6. marts 2013 22:19 To: 'Shad L. Lords' Cc: [email protected] Subject: Re: [discussion] Future of SME - CentOS 6 why? Hi Shad, I understood from Chris Burnat that Smolt was dead and that your therefore did not know much about the installed base - I was wrong in that assumption. Having seen the numbers I still think that something is not right - how come only so few have upgraded to SME8 roughly 1 year after the final release came out (or are these new installs)? The CT6 move might be a good enough idea - to keep with the flow, stay updated and support future HW (and f.ex Joomla 3x) but will that make SME more popular and provide more and new users? I understand making an appstore is a major task and very likely bigger than moving to CT6 but clearly some way seem to be needed to make some money for contribs.org and potentially also the contributors. As simple way to get started on this would be: - I would make a section on contribs.org (in a new skin and away from wiki style....) where contribs.org "approved" contribs were located. All contributors placing contribs here have to sign a "partner agreement" where they agree to donate 30% (Apple helped us all to set this standard) of their income to contribs.org. This way we do not need to make the transaction ourselves and avoid hassle with taxes, etc. - Transactions would for now have to handled by the contributor (via PayPal account or something). - contribs.org are obviously not liable for anything (payment or quality) - We can chose to police the payments in the future if we desire but initially we just need to know whether it works. This could be policed through a special SME Smolt version 2 that knew something about the running software on top of HW specific items. Personally I think that the mail system does not, out of the box, meet the needs of 2013. It ought to have DKIM and SPF "helpers" build in to help users in mail delivery. I know that some info on this is documented here and there but I think it should be standard. On the inbound part I can also see improvements that could help reducing spam levels - on top of that a quarantine system would be very helpful (next-gen unjunkmgr). Especially the inbound stuff would require a major rewrite of the qpsmtpd plugins to function well. I know I could just raise a bug and I could also just implement it as I have all the pieces. I would be happy to do so if I got rewarded somehow and I would be happy to pay 30% cut to contribs.org. Additionally I would want to provide subscriptions to all the special rules and signatures which is a part of the ScanMailX filter system - again for mammon. Maybe I am all wrong and I am an outlier - maybe I am not. Best, Jesper From: Shad L. Lords [mailto:[email protected]] Sent: 6. marts 2013 00:44 To: Jesper Knudsen Cc: [email protected] Subject: Re: [discussion] Future of SME - CentOS 6 why? On Tue, Mar 5, 2013 at 2:33 PM, Jesper Knudsen <[email protected]> wrote: Again, if someone with DNS access would redirect smolt.contribs.org to smolt.swerts-knudsen.dk (using a CNAME), then I would start gathering data and give us all an overview of the amount of installations out there. Why not just help get the smolt server on contribs.org working? Smolt should also give us an overview of the various version users are running. We can already get this. Here are the number of systems running the various versions: 215 8.0beta5 261 7.1 263 7.0 440 7.2 480 7.5 1123 7.6 1174 7.3 1202 8.0beta6 2156 8.0beta7 3948 7.4 7695 8.0 14372 7.5.1 Does that not make sense? I don't see what stats have to do with getting a new version out there. Maybe it's just me, but why are we trying to take another step upwards and spending all the energy in migrating to CentOS 6 with very little benefits to the users? What is it in the CentOS 6 base that is critically lacking in the CentOS5 (which is not planned for discontinuation)? I my humble opinion SME 8x has an OK MySQL and PHP level that allows for many 3rd party packages and the added HW that CentOS 6 supports is, in my opinion, not critical for the SME users generally running on rather old HW. There is a lot of new hardware coming out (ivy bridge) that isn't supported on el5 series. 5.x does have okay apps but there are some packages out there that will only run on latest 6.x release. Why are we not spending the resources on integrating the most commonly used contribs and improving the current platform - giving real benefit to the user? Where do you see improvement needed? Contribs are easy to add and if there is any functionality that needs changing then you should raise a bug. Firstly I think that we need to understand whether the distribution is actually dead already (I think not but..) and I therefore suggest we revive smolt for that purpose alone. The pure number of active server out there might motive all to move forward. If you have any information on getting smolt running please pass it along. There is no need to move smolt to a new outside location. We just need help getting it to run reliably. In theory I would enhance the smolt report with more relevant data to figure out more about the use of SME out there. Actually I would suggest we make a special SME Smolt that also collects information (100% anonymous) about: Smolt is in CVS. If you have enhancements or fixes for it please post a bug and attach patches. - Amount of users - Amount of Domains - installed contribs - properly more... I don't really see what use these numbers would be. They would be interesting but I don't see us making any changes to the system based on them. We might need to build a web service for this, but that is quite easy. I did that a while back (SME 6x) collected data on that spam statistics packages. I would be happy to build this if its gets released generally. Once the data from smolt is in the database then it would be easy to write any number of reports against it. With this at hand we can motivate developers to continue their efforts and target our development towards the right needs. "right needs" are subjective. Right now the effort that the community/developers want is to get things based on el6. It is moving along very nicely and everyone is working together very well. Additionally I would suggest that we think of means for the serious contributors to monetize on their SME work (and motivate them to keep the base bugfree)- maybe some kind of SME AppStore where a cut of each transaction goes back to contribs.org? That is no small undertaking. In order to do something like this you would need to develop quite a bit more infrastructure to host and manage the apps, keep track of who has paid for what, accounting, taxes, etc. Not saying it can't be done. But, I think that the effort to do something like this would be much more that the effort required to move to an el6 base. -Shad
_______________________________________________ Discussion about project organisation and overall direction To unsubscribe, e-mail [email protected] Searchable archive at http://lists.contribs.org/mailman/public/discussion/
