User experience for physics data analysis tools
oshadura opened this issue · comments
Various topics about how to make the life of physicists easier, tools interoperability and etc.
Question about overlap: is this included within #3, "expressing data analysis steps intuitively, to reduce cognitive burden"?
Group #3 is also about performance, which is a distinct topic from user experience, easy to read/write/understand analysis code, etc. At first, I was thinking of making them separate groups, but then decided to put them together because I wanted the people who are interested in these two things to be in the same conversation. If the "easy to use" crowd isn't talking to the "runs fast" crowd, then they'll develop two different ways of doing analysis, which is a disservice for users. (They'd get started on "the easy way" and then have to replace it all with "the fast way.")
There are connections with a few other groups perhaps. My take on what is what:
- #3: sounds like the space of libraries like
awkward: expressing individual steps of an analysis - #4: workflow management and the high level equivalent of #3
- #7: documentation / training
In addition to #7, there is another important point that is not captured yet I think, which is support mechanisms: how do users get help for things they cannot find in documentation? This is also an important UX consideration.
Also there is a UX side to #2 and #5.
Thinking more about this, #9 is perhaps more cross-cutting. An analyzer might be using lots of great individual tools but the UX when putting everything together may still not be great (I think this goes beyond the scope of #4 in principle). This interoperability point requires a broader forum to discuss I think.
Definitely interested in this group!
+1
+1
+1
+1
+1 My main interests here are:
- What is the general user experience for
Pythia8(both C++ and Python), and how can we make it better? - Our documentation is extensive, but uses a rather outdated methodology. What kind of best practice exists, keeping in mind that we want stability on the order of decades?
+1
+1