Open cranmer opened 5 years ago
It is a good idea to have a few user-stories to help us focus on a well-defined set of things. After all, we can add more cases as we progress.
Lets make different .md files for different user stories and have the initial README point to them.
Thanks for getting started on this.
The discussion about the front-end, back-end, and API comes before the context of what "the system" is trying to do. Some example user-stories are probably needed first.
My impression is that you have in mind mainly the user-facing part of the query system, where the user is specifying cuts and the like and then running over lots of data.
We should remember the broad scope of the Analysis Systems group and probably imagine multiple analysis systems aimed at different use cases. For instance, one might be specifying cuts, running over data, and producing histograms. Another might go further to include the statistical analysis (eg. HistFitter, TRexFitter, etc.). Another might be reinterpreting an analysis (eg. RECAST). Similarly, we might want to think about where analysis optimization of the analysis or more combined-performance tasks fit into this.
It's not clear yet if this repo should be trying to collect all of these or focus on a restricted set use cases and their corresponding "systems".
Maybe we should focus on the user-stories first.