[asterisk-users] Discussion: Are we ready to leave 1.4 behind?

[Olle E. Johansson]

>> 
> 
> I don't think there's anything inherently wrong with the bug tracking system.  It's more of a resource issue with many conflicting priorities.  Officially letting off some of the pressure from older branches does help.  I would like to be making faster progress through bug reports and patches.  I do have an open position for another full time Asterisk developer at Digium in case anyone is interested.  :-)
> 

I agree with Russell here, we have resource issues in the bug tracker but that's nothing that can be solved by another piece of software. If you have issues that is not handled timely, why don't you spend some time with other issues to help out? Surely there are issues where you can give a helping hand.

In answer to an earlier email that I felt was kind of attacking me I would like to point out that I am very happy and grateful about the resources that Digium put in the project, and continue to do. Just to clarify that this discussion was not about trying to paint Digium as a company as evil - which I was accused of. Digium is a very old business partner to my company and we've done great together. That doesn't mean we can't critizise each other or not want to discuss issues in the open.

To answer another attack, I have been contributing code and bug fixes to both 1.8 and trunk. Most of my code exist in versions for trunk and 1.4. Customers pay me for 1.4, I forward port it to trunk when I have time and resources over. It's not a personal choice that most of my development work still is based on 1.4. Of course I would love being doing development freely, creating great new code for the new release. There's a lot of stuff to do in Asterisk trunk, but no one out there that wants to put resources towards it in my direction. Asterisk trunk development is sadly too far away from my customers current business. The 1.6.x release schedule widened that gap and we need to discuss how to close the gap again. We do not need a large number of maintained releases between the long term support releases.

So far I haven't seen more than a few people that chimes in to this discussion saying we need to have 1.4 open, I haven't seen many people running 1.8 in production either. I have seen a lot of important issues being reported with 1.8 which to me confirms that it's still not ready.

I have been working in commercial software companies for a long period in my life. A product manager that called for end-of-life of the 1.4 release at this stage would be out of a job very soon. Migrating a customer base from one version to another is very, very hard. It seems much harder in telecom software than in the rest of the software world. We need to continue to work on 1.8 and do a lot of marketing for upgrading as soon as we're comfortable with it and have resources to manage the bug reports that will come in. We really need to push and shove. What I can't do to my customers is forcing them to upgrade to something that doesn't work. Customer will simply stop paying me if I do.

I will not continue to push this issue, just realize that I will have to manage my own 1.4 branch fixing the issues that affect my customers, which will exclude management of a lot of modules that are not used at all in our installations. As I said before, I have no resources to support all of the code base for everyone. That's just life, painful as it is. In the ideal world, there would be resources to help everyone. Unfortunately, I still have to have money to bring home at the end of the day.

Thanks for a very good discussion. As usual, I learned a lot from it. Keep reporting issues so that all of us can move forward to new releases.

Feel free to contact me off-list if you want to discuss this further.

/O