Participate in Hacktoberfest? #71

Open JeremyRand opened 2 years ago

Should we participate in Hacktoberfest in 2022?

Possible benefits:

Attracts code contributions that we might not get otherwise.
Some new contributors might stick around afterward; increasing Namecoin's contributor count is desirable.
Attracts other publicity.

Possible drawbacks:

Might attract spam PR's.
We would need to be attentive during October and respond promptly to PR's.
Depends on GitHub or GitLab.com, which is against our principle of decentralization.

Thoughts?

I suspect the spam PR problems will be minor. We already participate in Tip4Commit, and while we do occasionally get spam PR's due to Tip4Commit, they aren't a significant time sink for us.

If we decide to participate, we should figure out which repositories and issues should be tagged as Hacktoberfest participants.

Also note that a nontrivial number of package repos participate in Hacktoberfest, so this could be a way to get into more package repos.

I don't think the GitHub/GitLab dependency should be a blocker; realistically speaking, even once we transition to Gitea or Sourcehut or something, we will probably keep our GitHub account around for discoverability purposes, so it's not a major conflict.

@domob1812 @brandonrobertz @yanmaani Thoughts?

I don't think the Github/Gitlab dependency is an issue for this, either. I personally would prefer not to apply with Namecoin Core (nor Electrum), as that is mostly dependent on upstream anyway, a complex codebase to start with, and also (honestly) I'm not sure how much time I will have for Namecoin Core mentoring.

But I can imagine that some of the more application-focused projects could be well suited (but that's up to you to decide of course).

I personally would prefer not to apply with Namecoin Core (nor Electrum), as that is mostly dependent on upstream anyway, a complex codebase to start with, and also (honestly) I'm not sure how much time I will have for Namecoin Core mentoring.

@domob1812 FWIW my guess is that there are a decent number of simple Namecoin Core issues that we could use for Hacktoberfest without overwhelming participants. Based on a quick glance, the following look like decent candidates:

Would reviewing PR's for those issues during October be too much of a time sink for you?

I am fine with reviewing Hacktoberfest PR's for Electrum-NMC; I suspect there is a decent amount of low-hanging fruit there. I'm also fine with the following applications repos:

certinject (e.g. support more properties)
encaya (e.g. more tests, especially on platforms like macOS)
ncdns (e.g. improved integration tests)
ncdns-nsis (e.g. improved integration tests)
ncdns-repro (e.g. improved signing methods like Signify / Split-GPG / CodeCrypt / OpenTimestamps)
ncp11 (after pkcs11mod is more mature, not sure if that'll happen by October but it seems likely)
ncprop279 (e.g. more tests)
pkcs11mod (e.g. improved tests and more complete spec coverage)
StemNS (e.g. fingerprinting resistance and STS)

Thoughts?

I think it could potentially be cool and I can commit to being on spam patrol during the event. I, like Daniel, have heard about people trying to get a bunch of spam PRs in to artificially boost their contributions.

That list you came up looks pretty good. Some of them can be quite simple. I'm not sure what my bandwidth will be like, but I can probably help review some stuff if we get PRs in earnest.

I can surely review simple PRs for issues like the ones you listed, if that is all that is required. I'm not sure I can provide full mentoring for participants while creating PRs, though, in case that is required as well.

@domob1812 Full-fledged GSoC/Outreachy-style mentoring isn't a thing in Hacktoberfest AFAIK, but it is possible that participants will have questions for us in order to make sure their approach is likely to pass review.

I've applied Hacktoberfest topics/labels everywhere that looked appropriate. Here's the list of repos that I've marked as participating in Hacktoberfest, and here's the list of issues that I've marked as favorable for Hacktoberfest participants (61 issues in 10 repos). I haven't yet created issues for packaging in distros; will do that in the next few days. Feel free to glance through the lists and let me know if you think I've made any unwise decisions.

I think the tests related ones are particularly good and some of the maintenance stuff like updating stuff (and the testing that would be required). A lot of the windows and specific packaging stuff is over my head, but I'll be willing to help where I can. Thanks for putting this together.

Is https://github.com/namecoin/namecoin-core/issues/382 a good candidate? I'm not qualified to gauge difficulty or usefulness.

@JeremyRand It seems like something that could be done, yes. Although to be honest, I'm personally not very thrilled about #382; to me it seems like just adding complexity for minor benefits / a very specific usecase. But that may be just my opinion.

@JeremyRand It seems like something that could be done, yes. Although to be honest, I'm personally not very thrilled about #382; to me it seems like just adding complexity for minor benefits / a very specific usecase. But that may be just my opinion.

That was kind of my opinion on it too. It seems like something that could introduce unforeseen problems for little benefit. But I'm not an expert on name_scan or how people are using it or anything.

Thanks @domob1812 @brandonrobertz . I won't label it for Hacktoberfest then, since contentious issues are generally not a good fit for Hacktoberfest.

Is https://github.com/namecoin/namecoin-core/issues/519 a good candidate?

Hacktoberfest is approaching (less than a month away).

According to the issue list, we have 103 issues available for Hacktoberfest participants, with the following language breakdown:

    41 Go
    26 Python
    16 C++
    8 NSIS
    7 JavaScript
    5 Shell

@domob1812 @brandonrobertz If you have time, feel free to glance at the issue list and let me know if anything looks obviously wrong. If you don't have time, no worries, it seems to be in pretty good shape.

@domob1812 OK to add https://github.com/namecoin/namecoin-core/issues/519 to the list, now that there appears to be consensus that it would be accepted?

Yes, ACK from me for that.

A brief belated retrospective: Hacktoberfest was hugely helpful to Namecoin, but in an indirect way. We did not receive any PR's from Hacktoberfest participants (perhaps we need to market better?). However, our Hacktoberfest issue list was dual-use; it was also reused by Outreachy, and we got a really good number of contributions there (and successfully got a high-quality intern). So, I think we should continue participating in Hacktoberfest going forward, since even if it doesn't directly result in PR's, it is a very effective mechanism to get an issue list that we can use for programs like Outreachy.

namecoin / meta

Participate in Hacktoberfest? #71