Log file Viewer

[02:26:34] <sumanah> pradyunsg: the Chan Zuckerberg Initiative grant proposal asks: "Number of software projects that depend on the project (if applicable)"

[02:26:53] <sumanah> you & Paul noted:

[02:26:54] <sumanah> - All projects containing a file named “requirements.txt” on GitHub will depend on pip.

[02:26:56] <sumanah> - Any project using tox for testing will depend on pip

[02:27:39] <sumanah> EWDurbin and I will be submitting this in less than a day and I could use help -- could you maybe spend 20 min trying to get a very rough lower estimate?

[02:32:32] <pradyunsg> sumanah: yep yep - I'll take a look in my break hour after this class (after I'm done with breakfast 🙃).

[02:34:14] <sumanah> Thanks pradyunsg - I rephrased our answer and left a big _______ where you can fill stuff in

[02:35:12] <pradyunsg> Sure thanks!

[02:42:41] <toad_polo> sumanah: pradyunsg: At this point pyproject.toml is also a good indicator that you rely on pip.

[02:42:54] <sumanah> toad_polo: makes sense!

[02:43:16] <toad_polo> Actually wait, does poetry use pip under the hood?

[02:44:09] <pradyunsg> toad_polo: I'm thinking of requirements.txt, tox.ini, pyproject.toml and Pipfile.toml.

[02:44:16] <pradyunsg> No sure about poetry tbh.

[02:44:21] <pradyunsg> *not

[02:44:26] <toad_polo> There are a few other things that use that file, now that I think about it. Black, town crier, etc.

[02:44:44] <sumanah> So we don't need to be INCREDIBLY comprehensive here IMO

[02:44:56] <toad_polo> Yeah.

[02:45:05] <sumanah> let's not let precision lead us down this particular path

[02:45:26] <sumanah> my answer starts:

[02:45:29] <sumanah> "It’s very hard to gauge this because we don’t have telemetry to observe closed-source development organizations that use pip as part of their development and deployment processes. However, to a first approximation, the entire ecosystem of Python software projects — at least hundreds of thousands of projects, if not millions — depends on pip. Any software project that uses..."

[02:45:31] <toad_polo> Even those files are a pretty conservative estimate, honestly.

[02:45:32] <sumanah> and goes on from there

[02:45:34] <sumanah> right

[02:46:04] <toad_polo> Yeah, that sounds good.

[02:46:12] <pradyunsg> ^ +1

[02:47:10] <sumanah> pradyunsg: and please don't spend more than 20 min trying to get that estimate of GitHub projects that have those kinds of files, like requirements.txt and so on

[02:47:26] <sumanah> you have other stuff to do, like work on your monthly pip report and work on your PyCon Korea keynote and work on the resolver

[02:47:42] <sumanah> I really appreciate all the stuff you've already done

[02:47:45] <pradyunsg> Yea - I won't.

[02:47:50] <pradyunsg> I'm actually wondering if we should say GitHub ecosystem or something to that effect, since we're only collecting these metrics from GitHub.

[02:48:14] <sumanah> yeah sure

[02:48:19] <sumanah> don't worry too much about the wording, I'll edit it

[02:49:14] <sumanah> toad_polo: I think you know, but this is toward https://chanzuckerberg.com/rfa/essential-open-source-software-for-science/

[02:49:59] <sumanah> toad_polo: earlier this week I met with the head of Simply Secure who said they could be our partner on the UX side of a pip project

[02:51:27] <sumanah> gonna try to sleep now and then be heads-down on finishing it tomorrow morning ET

[02:51:36] <sumanah> bye for now

[06:17:46] <pradyunsg> in case someone is following, I've updated the relevant section in the Dropbox Paper document.

[20:03:02] <sumanah> Grantwriting is project planning (which I can do) plus making things sound persuasive (which I can do) with very specific word limit constraints (which I can do) on a deadline (which I can do) but all of it together can be hard

[20:03:46] <sumanah> this message brought to you by the Chan Zuckerberg Initiative deadline in ~4 hours

[20:06:36] <sumanah> we won't know till late October whether we get the funding.

[20:47:45] <sumanah> is there a pip roadmap?

[20:48:23] <sumanah> I know the resolver is key but is there anything else other than a few things on https://www.pypa.io/en/latest/roadmap/ ?

[20:49:54] <sumanah> EWDurbin: ^ I see in the Quality selection criteria for CZI that "Existence, clarity, and recency of project roadmap" is a factor. So since there's no place to put that in the application, and they are gonna take some time to read all these applications....

[20:50:08] <sumanah> we have time to work on the pip roadmap and make sure they see it once they start looking at pip :)

[20:51:08] <EWDurbin> heh, clever hack.

[20:53:15] <EWDurbin> sumanah: do you need anything from me on the above?

[20:53:43] <sumanah> EWDurbin: wanted to give you a heads-up, but also, you agree that this is the way to go? you agree that there's really no place in the CZI application itself to put the general project roadmap?

[20:53:55] <EWDurbin> lemme double check on that

[20:56:49] <EWDurbin> sumanah: all i see is references to "how the work fits into the existing project roadmap" for the Work Plan section of the Proposal

[20:57:11] <sumanah> OK yeah

[23:31:46] <sumanah> yay it is submitted! Thank you pradyunsg for helping research the pip metrics and metadata!

[23:33:54] <sumanah> looking forward to writing/helping improve the pip roadmap soon