GSOC 2017 project: Protocol Buffer Data Transport Layer


#1

Hello!

My name is Nikita, I’m a second-year master student from Moscow and looking for a project to participate in Google Summer of Code this year.

I’ve found the project: https://github.com/numfocus/gsoc/blob/master/2017/ideas-list-stan.md
about implementing protocol buffer support for Stan.
Could you say, please, is it still relevant?

There is a proposal about it on wiki, does it mean that it’s already in work?

If this project is available I could start to see into and write an application for GSOC.
If not, It would be also very interesting for me to be involved in related tasks.

A few words about me:
I have a good background in math, algorithms and advanced machine learning. Currently I’m most engaged and doing research in theoretical optimization, but also interesting in programming and technologies.
I’m new to open source community but have some industrial experience (mostly in C++ and Python).
I have a webpage with CV located there.

Thanks!


#2

Hi, the info on the Wiki is just to give people direction so yes the project is available. The way GSoC works is that you put in an application via GSoC and after the deadline period we look at the applications and decide if we have one we’re interested in working with. The deadline is April 3rd (not exactly sure about the cutoff, that’ll be on the GSoC website) so the only concern is that you’re cutting it close: the student’s portion of the application is really extensive (they basically want a design and implementation plan).

The eligibility requirements and application link are here: https://summerofcode.withgoogle.com/organizations/4968629817835520/

With your background it seems like it would be fairly straightforward and I’m happy to give feedback. They suggest using a Google doc for the application so it’s easy to share and get feeback.


#3

Ok, thank you!

Then I start to work on the plan.
I’ll share the draft as soon as it’ll be ready.


#4

That’s great, looking forward to reading it.


#5

Hello!

I’ve published a draft of proposal at GSOC site.
Here’s a link to google doc: https://docs.google.com/document/d/1DXq_wf1rZBalXc7PE_rCW5Yf7z8TDou32dv-nwFo1yI/edit?usp=sharing (you can comment by this link).
I’ve expressed how do I understand the project and a list of milestones.

Please, write any of your thoughts and remarks about it.
Thanks!


#6

I went through and added my comments and hints, overall the timeline looks reasonable and my major comment was to write tests first. I think today is the deadline so I hope you get the time to make some updates and submit.


#7

Many thanks for the comments!
I’ve tried to resolve most of them. The final submission is uploaded.


#8

The proposal looks good to me. I hope the GSoC folks approve it!


#9

I agree that the proposal is looking good. I wish we had gotten more time with it since we do want to be very specific about our plans but for the amount of time we had it looks good. We can improve during the pre-coding period.

I would still like to make sure we can all work together smoothly. One of the GSoC suggestions is to ask for some basic contributions to the project prior to the slot assignment period. Nikita, are you willing to pick two small code-related issues on stan-dev/stan or stan-dev/math and go through the process of creating a PR, getting it reviewed, getting it through tests? If you’re able to do that don’t be too ambitious about the issues you pick. I just want you to be familiar with the process we go through for code-related PR’s.


#10

Yes, I think it will be reasonable for me to make some contributions. I have already started to read the code, the development process description and so on, while the writing of proposal.

Then, for the beginning, I’ll start with doing some small issue at this week, closer to the weekends.

If I miss something important or just in case of useful suggestions, I’ll be glad to hear it.


#11

That sounds great, I’ll keep an eye out for your work on github.


#12

On second thought, here are some specific issues that might be good to start, any of these would be a good start.


#13

Hi, I’ve tried to resolve the first issue but got permission denied pushing new branch.

I ssh-cloned a copy of stan-dev/math:
git clone ssh://git@github.com/stan-dev/math.git

Just tried to create a new branch and to push it:
git checkout -b bugfix/issue-514-gamma-parameter-name
git push -u origin bugfix/issue-514-gamma-parameter-name

And got a permission error:
ERROR: Permission to stan-dev/math.git denied to doykov.
fatal: Could not read from remote repository.

Please make sure you have the correct access rights
and the repository exists.

The same story with https-access.

Is there a permission for the repository that needs to be set somewhere or am I doing something wrong?
If not, then maybe there is a problem with my ssh-key.
Thanks.


#14

Yeah there are permissions that need to be set up, I think you need to ask Daniel Lee or Sean Talts to do it. If they don’t respond here just open a new topic on discourse with the request.


#15

What is your GitHub ID? I can add you.

And why aren’t I seeing these Discourse topics as “new”? I think I may be missing a lot of the traffic going by.


#16

Thanks!
I’ve already got permissions by Daniel Lee and created a pull request for one simple issue: https://github.com/stan-dev/math/pull/530


#17

So you probably noticed but we didn’t get this project funded via GSoC. I think we had a good start but we didn’t have much time to clean up the proposal or work together and those are important pieces of the project application. I wanted to say thanks to Nikita for beating the submission deadline and I’m sure if we advertise earlier and start proposals at a more reasonable time next year we could put in something that gets a funding slot.


#18

Yeah, unfortunately we didn’t get it, but I would like to say Many thanks to everyone who helped me!