On 21/02/13 07:19, [[w:en:User:Madman]] wrote: > Does anyone know what the status is of the OTRS project on Labs? Given > a contact, I'd be happy to do what I can to help; I have some limited > experience configuring/deploying OTRS (up until the end of the 3.0 > branch last year, nothing with 3.1 or 3.2 unfortunately). > > I think opportunities for *volunteer* help have to consciously be > maximized, especially for volunteers who are or are willing to be > agents and/or identified to the Foundation. It's not going to get done > otherwise. > > -Madman/ea
I don't see much future in that, sadly. Yes, a puppetization from a volunteer could help the WMF, however they won't give you access to the current setup that you would be replicating. And that's a point that has been barring any volunteer help for years on this topic. Only ops can work on it, but nobody is assigned to otrs, and they have other tasks. There's a mixture of technical needs, legal issues and too-risky-to-touch it. Then Martin Edenhofer appeared offering to help with it, but there was delay after dealy: a NDA is needed, then separate machines, later he needs to provide the ssh key... And no work is done. On 21/02/13 07:32, James Alexander wrote: > Yeah, I have to agree sadly that we need more tech support and this has > been a thing that has been ongoing for a while. I personally think it > should remain in the foundation for many reasons (the least of which is > relatively large legal reasons) but we REALLY need to focus on it, or a > replacement, more. > > OTRS is the public face of not only the projects but the foundation in > general and answers an absolutely insane amount of email every year and > that has been the case for a while. When I first started applying to work > at the foundation my big interview ended up being about 8 hours (with a > liquor break in the middle) explaining to Philippe how I thought OTRS > needed to be replaced. I thought, and continue to think, that the system > underserves the job and we would be better served with something else that > could take much better advantage of modern advancements and clarity in > purpose. > > Sadly at the time they didn't have the money for me to work on OTRS (and so > I came to do the fundraiser) and since then I have heard rumors of it's > upgrade or replacement every single year (multiple times) only to be told > later that the resources aren't available. I've seen us look at the upgrade > multiple times, I've heard it be called both new "ceiling wax and cake > frosting" but not necessarily called a good option. It may be, I don't know > and we (as usual with outside products) overwork it beyond measure. Even > the professional OTRS folks when we were talking to them about helping > upgrade basically said "errr, you have HOW much in the database?" and told > us to just abandon it and start fresh with their new version. That said > even their internal OTRS version wasn't upgraded yet last year .... > > We need to do something though, it is disappointing to me that it hasn't > been a bigger priority because I think it should have been and I think it > should be now. I'm not sure if an OTRS upgrade is the best option... but it > is probably better then what we have. For a long while I thought we should > wait and not upgrade so that we can just replace it... but clearly it's > been too long for that now. > > James Thanks for your insight, James. It's very interesting. As you have dealt with it, can you clarify why is the upgrade such a big problem? Yes, we have tons of emails. So what? Does the upgrade use O(2^N) operations?? Even if not-too-efficient, I would expect the upgrade to have finished in three years :) I don't even know about a test upgrade being performed ever. I agree that OTRS is kind-of inefficient. We could easily build a replacement in 1-2 months *keeping the old data*. If OTRS works quite well on a single server, just imagine what we could do in a multiple server setup. I find hard that such version would perform worse. Not to mention the “handy” improvements we could add based on our usage. But just a newer OTRS version would be an improvement. _______________________________________________ Wikimedia-l mailing list Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
