Discontinue MAGIC support in FACT/Mars

[dneise]

In a private discussion with Thomas Bretz he mentioned, that it might be a good idea to stop supporting MAGIC in our version of Mars. I welcome this decision very much, however I this needs of course to be agreed on by the majority of users/developers of Mars in FACT.

So, I put up this ticket mainly to have a place to discuss this (if a discussion is needed at all). As soon as the majority agrees, we will officially drop MAGIC support. That's it for the moment. No code will be deleted or so.

We just create a snapshot of the trunk, that is known to be the last version of FACT/Mars, where MAGIC support has not been officially dropped. (Of course we know, that it is not clear whether MAGIC data analysis was fully supported before). After this point in time, developers are free to throw away code, that is not needed by FACT anymore.

An important question of course is, are there any MAGIC members in FACT, that do use the FACT/Mars to analyze some MAGIC data? If this is the case please inform us now.

At the same time I would like to drop support for ROOT versions smaller than 5.0. So if anybody thinks we should still support ROOT 3 or 4, please speak up.

I propose to leave the discussion open for a couple of weeks (say 2 weeks) and if nobody is against this, I will advance with this task.

[Daniela Dorner]

I'm not sure whether this is the right place to discuss this, as there might be people who are not in FACT, but use the software to analyse MAGIC data. I know that for example Stefan R. was still using it, but as I haven't seen him for a while, I'm not sure what his plans are.

Personally, I definitely interested in still analysing MAGIC data, but I doubt that I will find time for it :(

Do we still have support for old ROOT versions? With my 5.18 it doesn't compile anymore since a loooong time.

[dneise]

Do we still have support for old ROOT versions? With my 5.18 it doesn't compile anymore since a loooong time.

When I said, "dropping support for ROOT versions smaller than 5" I was referring to parts of the FACT/Mars code base, that explicitly contains work arounds for very old ROOT versions. But of course, just because there are dedicated work arounds, that treat some problems in ROOT 3.14 in one file of FACT/Mars this does not have any implications for any other combination of FACT/Mars and ROOT versions.

Since there probably aren't many of these places nor is it likely that despite those dedicated work arounds any of these old versions does still work with FACT/Mars, it might not be worth discussing this matter to a large extend.

Anybody can of course still try to use FACT/Mars to analyse any data from any experiment. All I'm saying is, that we as the FACT collaboration should, for the sake of code cleanliness, focus on FACT only. And therefor explicitly allow fellow developers to drop code, that is not FACT related, if they feel the need to.

No that I think about it, I believe it might make FACT/Mars development much much easier, when we would decide to only support one single version of ROOT per FACT/Mars release. And ask developers to explicitly mark work arounds that are needed to make Mars work with ROOT in the source, so when ever we create a new release and happen to decide to support a new ROOT version, we grep for those work arounds and see if they are still needed....

[Daniela Dorner]

with my comment, I wanted just to mention that with root 5.18 Mars does not compile on my laptop anymore since at least two years