General questions about the project

[Michal Wysokinski]

Dear rootpy Developers,

I'm working on an online monitoring application for the LHCb experiment which will use PyROOT for GUI and histogram plotting. I've stumbled upon your project after doing some research for one of my presentations and got quite qurious.

It's really noble that someone is trying to make interactions with ROOT a little bit more convenient, so thank you for your effort :)

If you don't mind I'd like to ask you a few questions concerning the current status of the rootpy project:
1) Is this project completely 3rd party without any "official" involvement from CERN?
2) Would you say that it's already quite stable and will be supported for many years or is it just an experiment in the early phase of development and its future is unknown?
3) Our monitoring application will be working to at least 2018, so is your project already in a state that apart from new functionalities it won't change much in new versions?
4) Do you think that it's mature enough to use it in the critical part of the experiment's software?

Cheers,
Michal 

[Peter Waller]

Hi Michal,

It's great that you found us, thanks for getting in touch.

On 14 March 2013 12:32, Michal Wysokinski <wy...@onet.pl> wrote:

1) Is this project completely 3rd party without any "official" involvement from CERN?

Yes.
 

2) Would you say that it's already quite stable and will be supported for many years or is it just an experiment in the early phase of development and its future is unknown?

I would tend towards the latter, though it's a community driven project and people are welcome to join - so the answer to this is in your hands as much as ours. You also have control.
 

3) Our monitoring application will be working to at least 2018, so is your project already in a state that apart from new functionalities it won't change much in new versions?

We have changes planned, but we'll do our best to maintain backwards compatibility where it is needed.
 

4) Do you think that it's mature enough to use it in the critical part of the experiment's software?

In my view the main problem we have right now in my view is that we have a few intrusive behaviours on ROOT, which means that we modify the behaviour of it. That could cause problems for other libraries which were written against PyROOT and don't expect the new behaviours we add. Ideally, we shouldn't affect the behaviour of any code unless it directly imports rootpy (or gets passed a rootpy-flavoured object). This might be a good target for a 1.0 release.

I would say, give it a go, but exercise due caution and try to make it break before depending on it for massive infrastructure right now.

And if you find problems, by all means please let us know [1]!

All the best,

- Peter

[1] https://github.com/rootpy/rootpy/issues/new