[page request] Applications Gallery

[matthewfeickert]

@kratsg recently broached the idea of having a way to demo some interesting public applications of the Scikit-HEP stack to potential users. For example, he recently made a Binderized GitHub Gist of Reproducible ATLAS SUSY Summary Plots which is a great example.

I had suggested that one approach might be an application gallery similar to matplotlib's. This could be a way to both demonstrate actual public uses of the Scikit-HEP stack out "in the wild" and be a handy thing to have on hand to point to in talks and demos.

I'm not sure if it would live directly on the website or somewhere adjacent.

Thoughts from @henryiii and @eduardo-rodrigues?

[eduardo-rodrigues]

Thanks for bringing this idea up, @matthewfeickert and @kratsg! I like the idea, though I have mixed feelings; let me try to spell them out:

• The matplotlib gallery is something that obviously suits perfectly a plotting/visualisation library but many of Scikit-HEP's packages are not so easy to showcase in a gallery.
• On the other hand, some of those libraries can often be showcased in examples of code that build on several packages, and that's the main idea behind https://github.com/scikit-hep/scikit-hep-tutorials/, which in fact is being rendered as a Jupyter Book on our website - scikit-hep.org/scikit-hep-tutorials - though not explicitly mentioned in scikit-hep.org/, something that we overlooked.
• The Gist you mention above - very nice one, BTW! - would seem to fit rather well in the tutorials IMO.

Let's see what you think of these comments ...

[matthewfeickert]

The matplotlib gallery is something that obviously suits perfectly a plotting/visualisation library but many of Scikit-HEP's packages are not so easy to showcase in a gallery.

I more meant that the gallery view (before visiting them) allows for a quick visualization of available use case studies to click on — visualizations not necessarily needed, depending on the case study / use example.

The following is just what I am thinking in my mind at the moment and not trying to dogmatically advocate for 🙂:

In my mind the case study / use examples and the tutorials would serve two different but complimentary purposes:

• Tutorials are just that: tutorials. They should be in depth explorations and demonstrations of the basic abilities of the different tools.
  • Examples include all of the nice tutorials that are already up on the Scikit-HEP Tutorials Jupyter Book and things like @HDembinski's very nice PyHEP 2019 Introduction to iminuit
• Case studies / use examples would be examples of what people have gone and done with the tools in real world scenarios.
  • A real world use example would be something like @kratsg's Gist or maybe some of the examples of things we've seen people do at the PyHEP conference series.

[eduardo-rodrigues]

OK, I see your point about case studies. Then I assume you would even make a call for, or at least encourage people to submit, proposals of usage examples for us to include in that gallery. I guess we can give this a go! Having one or two first examples to show seems like the best way to give the idea a go.

I would still say that we should not forget about tutorials and there some material of pyhf + other packages would be welcome too :-).

[eduardo-rodrigues]

So @kratsg and @matthewfeickert are you going to give this a go? Happy to look at a temporary build as an example :-).

[matthewfeickert]

I would still say that we should not forget about tutorials and there some material of pyhf + other packages would be welcome too :-).

Very much agree here. 👍

are you going to give this a go? Happy to look at a temporary build as an example :-).

I plan to eventually if no one beats me to it, but I think it will have to wait until I have at least something done on our CHEP 2021 proposal/paper. 😬

[eduardo-rodrigues]

No worries. And vCHEP 2021 is certainly a priority! Great to hear that Scikit-HEP will be indirectly represented 👍 .

[eduardo-rodrigues]

Hello @matthewfeickert and @kratsg. Am looking at long-standing issues related to the site, having in mind the "review and status report" as a paper. Have you got any clock cycles to put in place some proof-of-concept for such a gallery? Once that's in place others can then more easily contribute :-).

[klieret]

Late to the party, but I wanted to quickly advertise (and link for posterity) that we'll have a dedicated meeting today at 5pm CERN time to discuss how to combine the "analysis gallery" ideas together with a revamp of our training center.
This follows up with the discussion at pyHEP.dev.

[eduardo-rodrigues]

Hi @klieret, many thanks for the info 👍! I'm also "late to the party" since back today at work after some needed hols. I will try and follow at the link above.