Dear Debian Community, This is a personal email, and as such I speak only for myself; I would like to state following regarding a potential inclusion of MATE in Debian:
#1 freeze - I totally agree that the freeze should be respected, and I personally support Debian in enforcing such rule; #2 deprecated packages - MATE 1.4 still uses a lot of the 'dark arts' behind GNOME (ex: libmate, mateconf-2.0, bonobo, ORBit, etc). MATE 1.6 introduces a great effort to move MATE for the GNOME3 infra-structure, thus dropping all this dark packages that very few people really understand (as raised before). So my opinion is that MATE 1.6 is the right release to introduce in Debian (that's what we are doing at openSUSE and Fedora). This to say, I also support your claims in not bringing deprecated stuff back that most of us were happy to see being dropped. #3 use our manpower to contribute in GNOME3 packaging - this makes sense in MATE 1.6, specially because we are moving the underneath technologies to GNOME3 infra-structure, so up to a point working together in this field becomes mandatory for success. Of course we have no interest in maintaining specific packages of GNOME3 as we dont need them, though this should be a decision from the packager in my opinion. If the package wants to help, I'm ok with that. I do this on openSUSE, I was in the past a member of openSUSE GNOME Team, and I still fix stuff on GNOME, despite my efforts are in MATE. I'm sure that any of the other 7 active packagers will fix packages in MATE if they see something needing a fix, it's a part of our workflow. I don't see why Debian couldn't work that way too, but it ends up being people's personal decisions. In this points I do endorse Debian concerns, and currently on a MATE level we're working hard to comply. This to say, we have listened to your (and others) calls, and we're trying to improve. Now, I also have some concerns in the case MATE doesn't roll in official Debian, here's a few: 1) How can a Debian user benefict from introducing foreign (third party, example MATE provided) packages in Debian? Wouldn't this end up having the user running unsupported software in his system? Wouldn't this raise a lot of potential maintenance / support issues for users? Both are bad for MATE and Debian. 2) Our packages are not perfect (I personally don't know nothing of Debian packaging, I do and integrate software in RPM form); We are trying our best, but the improvements required are only possible with Debian's help. Debian has the know-how and expertise to help us improve, we have the manpower and the capability to help Debian users at upstream level, also to help with integration. Wouldn't it be far more plausible if we worked together, instead of throwing everyone into the falacy described in 1) ? 3) At MATE level, we never engaged GNOME3 as a competitor, GNOME3 in fact enables a lot of technology introduced in MATE 1.6. We have had help from several GNOME maintainers, and offers of help to improve, just two examples are Vincent Untz which helped us a lot with the GSetting ports and mate-panel, and recently Federico Quintero Mena. We even started maintaining abandoned packages from GNOME (in GNOME) to use with MATE. I'm sure many of you are already aware Federico sponsored our way in. This to say, we are trying to improve with the help and support from people who __really__ want to help us. 4) MATE isn't a competitor of GNOME3 (unless you are pure marketing fellow and use the marketing defenition: a competitor is a product or service that replaces totally or partially another). We're working together... the relation between MATE and GNOME is quite simple, GNOME is our technology enabler. Lets fall into another falacy and see things that don't exist. Lets not provide free ammunition to the link hunter sensasionlist bloggers... 5) Integration - my experience with integration of MATE in openSUSE (and up to a point in Fedora) has showed me that often a few other packages get a few tweaks. Most of the stuff needed at 3rd parties (ex: XDG) has already been done. Still, a proper integration in Debian will require your know-how. We are aware of a few integration bugs on the Debian packages (upstream), we need your help to fix them while Debian decides what to do. I would personally not like to see Debian users being treated as second class citizens compared to distributions which embraced MATE. This would also help that we get more consistent bug reports and improve deployment capabilities. This is something that can only be achieved by working together regardless of inclusion is accepted or denied. If it gets denied, we're still open for Debian to help us out improving the deployments of MATE managed by upstream (something we can't do alone at the current stage). This to say that in my opinion the only way out I can forsee is one of a symbiotic relation between MATE and Debian. I know that at MATE project everyone looks to the day we can work together, I would like to see it happen. Only Debian can help us make this dream of ours true. Please don't turn your backs to us, lets us work this out together. By bringing the previous to your consideration, I have the honor of expressing also my distinct consideration for the Debian Community, Nelson Marques
