ROOT 7, road map, API freeze, ROOT 6->7 guide lines

ROOT
1

Dear Rooters,

as far as I understand the ROOT6->7 is a multi year transition process. Somewhere I read that ROOT7 should be released in 2027, right?

Here I have a couple of questions:

  1. is there anywhere a reliable road map?
  2. when can an API freeze be expected for the already existing ROOT7 classes? I am asking, because we have rather limited man power, and want to minimize the risk of broken code, due to constantly changing API’s.
  3. are you planning to but up a migration guide line for existing ROOT6 code?

Thanks for all you work we profit from,
Andreas

2

Hi Andreas,

Thanks for the post. I perfectly understand the limitations you are coping with and the need to plan ahead to mitigate them.
We plan to release ROOT 7 on time for the start of Run 4, currently scheduled for 2029. We decided to release ROOT 7.00.00 two years before, i.e. during Q1/Q2 2027.
Currently, we do not have a publicly available roadmap but we are working on it. The approach we foresee for ROOT 7 is evolutionary, for example:

  • we will not deliberately break existing code unless forced to, for example by highly favourable cost-benefit ratios.
  • entities will be deprecated, but not in bulk and starting already with a clearly communicated strategy in ROOT 6.
  • A large collection of the current classes, routines and functions will continue to exist in “legacy mode”, i.e. available in ROOT but not further developed and bugs will be addressed on a case-by-case basis. For example, I can anticipate that classes such as TH1* and TTree, will be supported in legacy mode and simply not discontinued.
  • No user will be left alone: the ROOT project commit to make this transition as smooth as possible, not only through the aforementioned strategies, but also supporting our users in the potential migrations.

Moreover, I can anticipate that moving existing worflows to RNTuple and RDataFrame (ROOT 7 components already available in ROOT 6) will help making the 6->7 transition even easier. LHC experiments are following this path.

As a complement to this post, which cannot by far replace the written plan you asked, I invite you to contact me privately, via the Forum or by email, if further discussions at this point is needed. Let me add that your input, coming from a European lab, would be very useful for the the ROOT Project.

Best,
Danilo