How should new members apply to the TSC and what will we value in applications?

[bradcray]

Chapel's TSC is designed to be open to members of the community after a majority vote of the current TSC members. This issue asks: what materials would we like to request from those applying to join the TSC? For example:

• a paragraph or two detailing the candidate's history and experience with the Chapel language and/or project
• an explanation of why they would like to be on the TSC
• …?

In addition, it asks what factors we should list as things that would strengthen an application, such as potentially:

• uses of Chapel in applications
• contributions to the Chapel community in the form of PRs (code, doc changes)
• involvement in Chapel discussions and issues
• organizing of, or participation in, Chapel community events (such as ChapelCon)
• having publications, blog posts, or social media activity related to Chapel
• providing funding to the Chapel project or other Chapel-related development efforts
• unique perspectives, strengths, or experiences that you'd bring
• other potential benefits to the project
• …?

Andy suggests we also include statements qualifying the above, such as:

• note that we don't expect potential applicants to meet all of the above criteria and will consider applications holistically
• that the above are examples of things we will consider but not consider ourselves limited by

Please post your thoughts about things we should request and value in the application process.

---

edit: Here's an attempt at drafting questions for a TSC application form based on the discussion here and in TSC meetings so far:

---

(Draft) TSC Application

For a non-exhaustive list of things we value in TSC applications please see link (where this link would ultimately point to a document, but for now relies on the list in the OP here)

Name:
GitHub username:
Affiliation(s): (e.g., workplace(s), school(s), relevant professional groups)
Notable projects/products you work on or have worked on: (list names, provide links if available)
History and experience with the Chapel language, project, and community: (please provide a paragraph or two summarizing your past involvement with Chapel)
Motivations for applying to the TSC: (please provide a paragraph or two explaining what you hope to bring to the TSC and what you hope to get out of it)
Advocate/Sponsor (optional): (Is there someone on the TSC or in the Chapel community who knows you and is willing to advocate on your behalf?)
Disclosure of conflicts that could prevent you from meeting the expectations for TSC members: Please provide anything you think we ought to know

---

Currently, I think our model for the application process is:

• we'll solicit these applications by opening a GitHub issue using a template on this repository
• at that point, the TSC will:
  • review the written application
  • vote on moving forward with it or not
    • if so, invite the individual to introduce themselves and answer questions at a TSC meeting
    • if not, provide feedback to the candidate on why and what more they might do to address the concerns in cases where it's possible (e.g., not enough investment in Chapel to date, not bringing enough to the table)
• at that point, the TSC will vote on whether to add the individual to the TSC or not

[bradcray]

Tagging @lydia-duncan , @stonea , and @DanilaFe who had things to say on this topic when we discussed in the last TSC meeting.

[stonea]

Your list of asks and factors looks good to me.

A couple other things to think about are:

• if we should come up with a list of TSC member responsibilities. It's something we might want to include in a "join the TSC" application and it might help us come up with other factors to add to our list.
• how often will we consider applicants? Is it something we do as applicants stream in or do we do this as some kind of periodic process (e.g. once a year or once a release).

Other factors I can think of adding to your list:

• "participation in Chapel community events (such as ChapelCon)"
• "having publications, blog posts, or social media activity related to Chapel"

In the wording on our application I would make it clear that:

• our list of factors is something we "consider but aren't limited by", and that
• we don't expect successful applicants to meet all these factors and that we'll consider applicants holistically

[bradcray]

Great feedback Andy, thanks! I've incorporated the factors and qualifiers you listed in the OP, and have fired off #23 to start trying to capture the responsibilities/expectations of TSC members.

My thought on how often we consider applicants is to figure out how often we're going to meet first and see whether that seems like an appropriate interval for considering new members (specifically, I haven't fired off a new issue for that myself).

[bradcray]

I also find myself wondering: Should we add "providing funding" to the list above? On one hand it seems a bit mercenary. On the other hand, if someone provided significant funding to the project and wanted a seat on the TSC, I think that should probably carry some weight.

[lydia-duncan]

Probably a no brainer, but a prospective TSC member should be in good standing w.r.t the Code of Conduct.

If we accept "providing funding" members, I might be inclined to set a limit on the percentage of TSC members that can be in that category. There will hopefully be many people in our community that can't provide funding for whatever reason and I wouldn't want them to feel as though their voice was less valued.

[karlonw]

Will there be (should there be) a soft (or hard) limit to the number of TSC members? At what point is there so many members that gathering a quorum to make decisions becomes difficult? Do other HPSF projects have established limits?

[e-kayrakli]

I also find myself wondering: Should we add "providing funding" to the list above? On one hand it seems a bit mercenary. On the other hand, if someone provided significant funding to the project and wanted a seat on the TSC, I think that should probably carry some weight.

@bradcray is there a guidance from LF/HPSF on this topic? For open source project such things matter, and I don't think we are the first to think about it at the very least.

[bradcray]

Capturing some notes from today's TSC meeting along with some of my reactions to them (some stated during the meeting, others not):

My starting proposal based on discussion on this issue so far was that we ask applicants to:

• submit 1–3 paragraphs detailing:
  • experience with the Chapel language and/or project
  • why they would like to be on the TSC and what they think they would bring to it
• give a live presentation at the meeting as well?

I later proposed that applications be done by opening a GItHub issue with their paragraphs as their application; we could have an issue template supporting these applications.

My take on the discussion was that people generally:

• liked keeping the amount of work relatively low for applicants

• liked the idea of having the person do something live in the meeting to establish rapport / prove that they were not an AI

• proposed potentially accepting written proposals first, giving them a thumbs-up or down, and only inviting those who passed to present at the meeting to save time

• A question came up as to whether we should fast-track known people to save them the trouble of applying. Most statements in response to this suggested that we should treat everybody equitably.

  • It was also proposed that we retroactively write the application paragraphs ourselves to establish who were are and set an example for others (but, presumably, with no intention of using them to vote anyone out of the TSC)

• There was a suggestion that applicants have a sponsor or advocate which in some ways sounds appealing, but it may raise barriers for people who don't know the community very well (but maybe that's appropriate?).

  • At the very least, giving the applicant the opportunity to name people who they think would advocate for them seems attractive as "something else that will be considered", but not a requirement

• There was a suggestion that the application was perhaps too short, and that we should ask for 1–3 paragraphs for each of the criteria that we were evaluating them against to keep from being overwhelmed by too many applications

  • My personal take (unstated at the meeting I think) is that we should keep it short initially, and raise the bar if/when this becomes a problem
  • A counterpoint was that if we're going to do it eventually, we might as well do it now

• It was suggested indicate their timezone and ability to attend TSC meetings. I feel like this is something we should deal with when it becomes an issue for fear of suggesting a bias toward our timezones. It is also likely to come up organically when candidates have to do their live introduction at a meeting.

• There were questions about how we weed out people who are unqualified / what metric we use for evaluating people

  • My take is that we should not shy away from turning down people who don't have a track record with the project or technology, asking them to establish more of a relationship before applying again. It's difficult for me to imagine someone with no Chapel experience or background somehow sneaking onto the TSC without our being aware.
  • I tend to think we don't need a formula for getting onto the TSC or not, but that we'll make the decisions through informal conversation amongst the TSC (likely in private) followed by a formal anonymous vote. But this is just my opinion and would need to be decided.

• We had some discussion about people who had needs or preferences for anonymity, etc. We also touched on how it will be important for us to be able to interact with members of the TSC in order for it to be a functional body and how these two things could be in tension with one another.

  • A good response to that is that we need to be up-front about our expectations for TSC members (What are the responsibilities and expectations of TSC members? #23)

[ShreyasKhandekar]

One thing I'm not sure about but could there be conflicts of interest for people on the TSC? Should those be disclosed on any application?

[bradcray]

@ShreyasKhandekar : Can you say more about what would constitute a conflict of interest for TSC candidacy, and what the disclosure would be designed to prevent?

I generally think of conflicts of interest as being about not having someone give you a free ride because they're your friend/acquaintance/colleague/etc. But in this case, being known by a TSC member seems like a plus (and aligned with the notion of having a sponsor or advocate, as proposed in the last meeting), in terms of having that person serve as someone who could advocate for the candidate based on personal experience.

I suppose another form of conflict of interest would be "I work on competing technology XYZ", but I don't really think of that as a conflict of interest for an open-source project, assuming the candidate has qualities that make them attractive to the Chapel project and community.

[lydia-duncan]

I think being a colleague in a professional setting seems fine but populating the TSC with people that are just our friends would be more suspect (e.g., is this just a way to stack the votes in your favor or do they also actually care about the language?)

[e-kayrakli]

is this just a way to stack the votes in your favor or do they also actually care about the language?

The second part of this question is something that we should ask about any candidate in any case. The first part is an interesting question, though if put the candidates through the filter for the second question, that may be enough, I think?

I think there is always a question about nepotism in scenarios like this, but I am also worried about adding barriers or over-legislating candidacy/membership. I am not dismissive of some CoI management, but I am not convinced yet that we need to worry about it too much as long as we focus on merit. A realistic take on this is that we have a relatively small community today, and I suspect there will be CoIs with candidates for some definition of the term.

[bradcray]

Generally agreeing with Engin, I agree that populating the TSC with just our friends wouldn't be good for the project, but am not clear on how listing conflicts of interest would help with that? (in that, usually a conflict of interest indicates that the person with the conflict should somehow be removed from the decision-making process due to the conflict, at least in my experience).

Is the proposal to have candidates list all of the people on the TSC that they know/are friends with, or the nature of their relationship with all existing TSC members? What would we do with that information, practically speaking?

Related, are we expecting that TSC members would not be forthcoming about such friendships and relationships when we're discussing a candidate?