[GSoC] Contributor candidate introduction

Hello,

I'm interested in participating as contributor for the project "Reachability bitmap improvements". The area I'm interested in, the most, is the alternate compression scheme e.g. Roaring+Run, however I'm ecstatic about any bitmap performance improvement. Expected project size I'm targeting for is large (350 hours). I've already completed my micro-project and I will submit it in a few hours. I wanted to be sure I could submit v1 of it before committing to introducing myself.

I'm Plato Kiorpelidis, undergraduate student from Greece, studying Computer Science. I've been developing software systems for academic purposes and Greek universities for the previous 1.5 year inside my university department as full time job.

The idea of reading and implementing ideas proposed in academic papers, then running performance tests to compare and evaluate them, thrills me and peaks my interest. I've been browsing the git source code on and off for two years as an undergraduate, thinking it would be awesome if I could contribute, influence and become actively involved in a widely used tool such as git. The opportunity given by GSoC provides the environment and support for it to actually happen.

I've gone through the mailing list and looked for other candidates that could
also be interested in this project. Shubham Mishra is also interested.
- Could we collaborate on this project considering how broad it is or only one
can be selected?
He/she already has experience in open source and has participated in
GSoC before.
- Does that make me a better fit considering GSoC is for introducing someone to
open source and their communities or should I strive for a different
project instead?
I should note that the best case scenario for me is if we are both selected,
probably on different bitmap performance areas under "Reachability bitmap
improvements" project, however I don't know if it's possible. It probably
depends on the mentors listed for this project and the work load involved.
- Has this happened in a previous GSoC? Did it work?

Thanks, Plato Kiorpelidis

Hi Plato,

On Wed, Apr 06, 2022 at 12:43:59AM +0300, Plato Kiorpelidis wrote:

Hooray! I'm delighted to hear that you're interested.

I think there are a couple of options here, since we have a handful of bitmap-related projects that are all up for grabs. If Shubham is also interested in working on implementing Roaring+Run compression, then I would be happy for the two of you to work together.

Alternatively, if you each want to focus on different sub-projects, I would be happy to work with multiple students on different areas within the reachability bitmaps subsystem.

Yeah. Even though I'd be happy to work with as many students as I have time for, I'm not sure whether it's possible within GSoC's own rules, much less whether or not it's been done before.

Christian (cc'd) may have more information. Christian: is it possible to accept multiple students for the same GSoC project?

Thanks, Taylor

Hi Taylor, Plato and all,

On Tue, Apr 5, 2022 at 11:59 PM Taylor Blau <me@ttaylorr.com> wrote:

Though we appreciate it, we don't require introducing yourself. And if you choose to do it, you can do it whenever you want.

Unfortunately, GSoC rules forbid contributors from working together.

I think this would require creating different independent project ideas.

Unfortunately, I think there needs to be a different project for each GSoC contributor and GSoC rules ask for each contributor to work independently from other GSoC contributors. It's possible to have the same mentors for different projects though, so if the mentors for a project are OK with splitting it into different independent projects and mentoring all the independent projects, I think it could work.

Best, Christian.

Hi,

As the topic begans, I think I should also give my updates. Last week I spend time around "Croaring" library and right now, I start working on the initial version of proposal which I am planning to share by end of this week. I think if it's possible to split it into two, We can look forward to it with I am okay with picking other project involving bitmap too but only issue, I have to start again from scratch :)

Thanks, Shubham

On Wed, Apr 6, 2022 at 11:55 AM Christian Couder <christian.couder@gmail.com> wrote:

On Wed, Apr 06, 2022 at 07:41:35PM +0530, Shubham Mishra wrote:

Sounds great; I think that having a proposal of how/if we should use Roaring+Run bitmaps in Git will be useful in its own right.

I expect that it will generate some useful discussion, and provide a good roadmap for any contributor to get started working on it. Let's see how the proposal develops and adjust, split or not split the GSoC project as necessary.

Thanks, Taylor

On Wed, Apr 06, 2022 at 08:25:29AM +0200, Christian Couder wrote:

Thanks for this information, it's exactly the sort of thing that I was hoping for from your response ;-).

Like I said in my other message, let's play it by ear whether or not we need to split the project. If we get closer to the next deadline and we have >1 potential GSoC students interested, then I'm happy to split it up.

One thing that I worry about a little bit is that I think a new on-disk .bitmap format will be necessary for both of the projects that Plato and Shubham are interested in (though I could be wrong). Since GSoC forbids students from working together, it may be difficult to feasibly split the bitmaps project into multiple, smaller ones.

That said, if we do end up splitting the bitmaps GSoC project in two, when would we need to do it by?

Taylor

On 06/04/22 21:13, Taylor Blau wrote:

To the best of my knowledge, Google does not require the proposal ideas to be laid out by organization before the proposal deadline (which is April 19). This gives the flexibility for mentors and GSoC contributors to collaborate and come up with a project ideas that are not present in the initial ideas list and write proposals for them.

The only significant requirement is that once a proposal has been selected, the corresponding GSoC contributor should not be made to work on drastically different project.

I hope Chrisitan will correct me in case I missed something.

On Wed, Apr 6, 2022 at 10:12 PM Kaartic Sivaraam <kaartic.sivaraam@gmail.com> wrote:

;-) Note that my answers are based on what I know from previous years. Rules sometimes change from year to year though, so everyone is welcome to check.

I agree with this, but please keep in mind that in https://git.github.io/General-Application-Information/ we ask applicants to "Discuss their project by posting drafts of their application on the mailing-list long before the deadline." So splitting the bitmap project should be done soon, so that applicants have time to prepare and discuss their proposals based on the splitted project.

See also the "Note about getting involved early" section in the document, which has information about what applicants should do when they get involved quite late.

On https://google.github.io/gsocguides/mentor/selecting-a-student.html there is also:

"Don’t Be That Person: Don’t even think about selecting a GSoC contributor with whom you’ve had no contact. You should establish an active back-and-forth prior to making a decision. If you or your GSoC contributor have failed to make this happen, do not proceed."

Best, Christian.

Plato Kiorpelidis <kioplato@gmail.com>  wrote:

Hello Plato, nice to meet you :) :)

I am also interested in it and has already submitted a proposal[1] on it (For some network related issue, I mistakenly sent three mails of the same proposal; please ignore the other two)

I saw Taylor is willing to split the project into sub parts. I have no problem with it. I will change the proposal accordingly.

Thanks :)

[1] https://lore.kernel.org/git/20220406200440.27010-1-chakrabortyabhradeep79@gmail.com/T/#u

Thanks :)

On Thu, Apr 07, 2022 at 09:16:44AM +0200, Christian Couder wrote:

I've thought a little bit more about this, and I think we should leave the project in one piece (i.e., avoid splitting it up).

My thinking is roughly that of the sub-projects listed:

  - investigate replacing EWAH with Roaring+Run
  - add a "table of contents" to the .bitmap file
  - append-only bitmap generation
  - amorphous / larger related projects

...the first two both require serious (re-)consideration of the .bitmap format, and I don't think it's possible to split that up among multiple GSoC projects.

The "append-only" project is interesting, but probably not large enough for a GSoC student to occupy their time with completely. I think the last bullet point has the opposite problem, which is that there is _too_ much to do there, and that it isn't well defined enough.

So we _could_ split the two projects up, having one work on any preliminary format changes, and the other stay busy with a smaller bitmap project, but I have significant concerns about that setup being successful.

Instead, let's keep the project as-is, and have the accepted student focus on one of the first two bullet points, taking into account that we'll first want to spend some time thinking about whether or not the current .bitmap format is appropriate given the proposed changes.

Thanks, Taylor

Taylor Blau <me@ttaylorr.com> writes:

Yeah, also I doubt that bitmap stakeholders want to assign such a centrally important design task to an apprentice, even with a competent mentor or two.

I have to preface this with the fact I always am the one that is the least optimistic about the timeline, but "append-only" somehow sounds like a project with an interesting enough size and complexity to me.

Just my two cents.

    (me goes back to preparation of -rc1)

Hi Plato,

On Fri, Apr 08, 2022 at 04:34:11PM -0400, Taylor Blau wrote:

If this sounds interesting to you, I would encourage you to submit a proposal similar to the ones in

    https://lore.kernel.org/git/20220406200440.27010-1-chakrabortyabhradeep79@gmail.com/
    https://lore.kernel.org/git/20220409184350.1124-3-shivam828787@gmail.com/

Feel free to take a look at mine and Kaartic's responses to each of those, two, since they may influence your planning / which sub-project sounds most interesting / etc.

Thanks, Taylor

On Wed, Apr 13, 2022 at 11:55 PM Taylor Blau <me@ttaylorr.com> wrote:

Hey Taylor,

Yes, it does sound interesting to me because it's an experimental project. I'm working exclusively on my proposal for the past week reading papers and benchmarks, while writing digests about them to include in my proposal, in favor of roaring method. As you also said, having a proposal with references and digests about previous work, discussions and papers does help regardless of being selected.

Technical question: will roaring bitmaps' random access feature help with introducing the "table of contents" feature? Or is it bitmap compression agnostic?

Thanks, I've already read both of them and used them as guideline for my proposal.

I saw your and Kaartic's reviews and incorporated them into my proposal.

Thanks, Plato

Hello all,

Just a heads up that the proposal deadline (19Apr2022 18:00 UTC) is approaching shortly. I noticed that Abhradeep has submitted his proposal so far. Others, do make sure you submit your proposals in the GSoC website before the deadline. Only then can we consider your proposals for GSoC [ref].

Thanks.

[ref]: https://developers.google.com/open-source/gsoc/faq#should_i_send_proposals_directly_to_the_mentoring_organizations

-- Sivaraam

Kaartic Sivaraam <kaartic.sivaraam@gmail.com> wrote:

That is the initial version of my proposal. I will submit the final version soon.

Thanks :)

hi,

I submitted my proposal after fine-tuning it more according to the reviews :)

Thanks, Shubham

On Tue, Apr 19, 2022 at 5:06 AM Abhradeep Chakraborty <chakrabortyabhradeep79@gmail.com> wrote:

Abhradeep Chakraborty <chakrabortyabhradeep79@gmail.com> wrote:

I have submitted the final version in the GSoC portal.

Thanks :)

Adding other potential contributors to notify them too.

See mail with details below.

On 18 April 2022 11:24:34 pm IST, Kaartic Sivaraam <kaartic.sivaraam@gmail.com> wrote:

Due to time constraints I shall be unable to participate in GSOC this year, but I'll continue to contribute to Git and invest time in understanding its codebase and hopefully apply next year as a GSOC applicant.

Thanks and Regards, Jayati

On Mon, Apr 18, 2022 at 11:54 PM Kaartic Sivaraam <kaartic.sivaraam@gmail.com> wrote:

Hi Jayati,

On 19-04-2022 18:46, Jayati Shrivastava wrote:

That's understandable. Thanks for letting us know. Hoping to hear from you on the mailing list :)

Others,

Thanks for submitting the proposals on time.

-- Sivaraam

Hello,

I shall not be able to participate in GSoC this year. I'll try to contribute and understand the codebase more instead and try to participate next year hopefully.

Regards, Khalid Masum

On Tue, Apr 19, 2022 at 11:40 PM Kaartic Sivaraam <kaartic.sivaraam@gmail.com> wrote:

Hi Khalid Masum,

On 20-04-2022 18:46, Labnan Khalid Masum wrote:

That's understandable. Thanks for letting us know. Hoping to hear from you on the mailing list :)

-- Sivaraam

Hi,

On Wed, Apr 20, 2022 at 8:31 PM Labnan Khalid Masum <khalid.masum.92@gmail.com> wrote:

Thanks for letting us know! You are very welcome to participate next year and contribute in the meantime.

Chances are high to get accepted for those who start getting involved early.

Best, Christian.

On Mon, Apr 18, 2022 at 8:54 PM Kaartic Sivaraam <kaartic.sivaraam@gmail.com> wrote:

Hey Kaartic,

Thanks for the gentle reminder :) My proposal is nearly complete. I'll sent it for review in the mailing list and If the time allows you I'll be grateful if you review it.

Thanks, Plato

Hello all,

We were reviewing the proposals and noticed some missing information. Hence we wanted to gather the information now.

Shubham, You have mentioned that you'll be able to dedicate 35-40 hours per week. We just confirm if that's feasible despite you having the commitment of a full-time job.

Also, do let us know if you would have any other commitments during the GSoC period that you're already aware of.

Plato, Do leṭ us know if you would have any other commitments during the GSoC period that you're already aware of.

Apologies for reaching out late. Since the project selection deadline is approaching soon, do respond as soon as it is feasible for you.

-- Sivaraam

Hi Kaartic,

At my job, my timings are flexible. I already discussed about GSoC with my team and Microsoft is supportive for open source contributions. I have done side projects and open source contributions in past also, so this shouldn't be a problem. I also planned to spend weekends on this (I usually spend weekends on my desk, so this is fine for me).

I do not have any other commitments or planned vacations throughout the GSoC timeline. I would be able to dedicate 35-40 hours per week :)

Thanks, Shubham

On Mon, May 9, 2022 at 12:14 PM Kaartic Sivaraam <kaartic.sivaraam@gmail.com> wrote:

On Mon, May 09, 2022 at 08:42:38PM +0530, Shubham Mishra wrote:

Be careful to not over-extend yourself. I think it's great that you're excited about GSoC, but in order for it to be the most successful for you, you should make sure that your work on Git happens during the times that work best for you.

I'm not familiar with your exact working arrangement, but I would caution you that participating in GSoC (if your GSoC project expects 35-40 hours a week from you) _is_ equivalent to a full-time job.

If your existing full-time job allows you to spend time working on GSoC, that's great. But if your normal work is separate from GSoC, having two full-time commitments seems like a recipe for disaster.

Likewise, GSoC shouldn't get in the way of your personal commitments. Taking breaks is part of doing productive work ;-).

Thanks, Taylor

On 22/05/09 12:14PM, Kaartic Sivaraam wrote:

Hey Kaartic,

GSoC is my priority and I already spend most of my time studying Git and the project's related material. I've got some university classes and assignments until early June, but it's something I've already been doing in parallel with GSoC and my work on dir-iterator [1]. I don't have something major planned for the summer since I would prefer investing most of my time into Git and GSoC. After this semester ends in early June I'll be completely free to focus on GSoC.

[1]: https://lore.kernel.org/git/20220509175159.2948802-1-kioplato@gmail.com/

Thanks, Plato