Question : Participating in Google Code-In?

[Athitya Kumar]

Hey all.

I think the thread title says enough - Are we participating in Google Code-In 2017? Google Code-In 2017 just got officially released yesterday, and it seems like open-source organizations can apply to be mentor organizations from October 9 onwards according to the GCI timeline. I feel that participating in GCI 2017 would help us in improving the reach of Ruby and our projects to pre-college students.

Regards,

Athitya Kumar (Facebook | LinkedIn | Homepage | Blog | GitHub) 

4th year undergrad

IIT Kharagpur

[John Woods]

I think we should. We applied once before, but didn't get in that time. Someone would need to write the application.

--
You received this message because you are subscribed to the Google Groups "SciRuby Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sciruby-dev...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Athitya Kumar]

Sorry for the late reply, John.

 

I think we should.

Great. I just wanted to check if you (and other members) are okay with the idea of applying for GCI.

We applied once before, but didn't get in that time

Do we have the rejected application for reference? I had a look at the general GCI guide and maybe the tasks that were previously designed weren't bite-sized (doable in 3-5 hours) or easy enough for the participants.

Someone would need to write the application.

Now that we've a successful GSoC application, I think that the challenging part (probably) lies in designing bite-sized tasks (easily fixable issues). I'd definitely like to contribute in this process. :-)

[Athitya Kumar]

Hey all.

With the application portal for Google Code-In's mentoring organizations starting tomorrow, I just wanted to have a rough estimate of availability of potential mentors. This would definitely help while drafting the application and I've cc-ed some of the GSoC 2017 students and mentors along with this mail regarding the same.

For example, this would be my availability :

Timeline : Whole (Nov 28th 2017 - Jan 31st 2018)

Projects : daru, daru-io, daru-view

Kindly respond with your preferences too. Also, maybe it's time to collaborate on aggregating the bite-sized tasks? We can start drafting the application once we know the format of the application (after the portal opens tomorrow). :-)

Regards,

Athitya Kumar

[Shekhar Prasad Rajak]

Hello Athitya,

I just read about GCI. I am also available during the timeline  (Nov 28th 2017 - Jan 31st 2018) as mentor. Let me know about the application procedure.  

Thanks,

Regards,

Shekhar Prasad Rajak

[Athitya Kumar]

That sounds great, Shekhar. :-)

Maybe John / Sameer would be able to provide a more clear picture regarding the application procedure for the organization. But, this what I think will be the case - 

0. GCI portal for organizations opens tomorrow.

1. Org admins register themselves and organization on GCI portal.

2. Org admins share the format of application to be submitted in GCI portal.

3. Org members start collaborating on designing the tasks and drafting the application via Docs / GitHub Wiki.

P.S - We also have a collection of previous applications in the SciRuby Wiki, for reference.

[John Woods]

We've only applied the once, I think. You've already found the previous application in the wiki, which is what I was going to reply with.

As a side note, I haven't had the time to edit the blog for months now with my work commitments. It'd be really beneficial for applications like GCI (and GSOC too) if we could start regularly updating the blog. We really need someone with exceptionally good writing skills who is willing to volunteer for that if we want to start bringing in this kind of application.

John

--

[Victor Shepelev]

Well, neither I nor Sameer (who are the most active from "mentors" side currently) are native speakers (I can be wrong about Sameer, though, I am not absolutely sure). But at least Sameer as one of GSoC admins should at least have merging rights to blog, don't you think?..

To unsubscribe from this group and stop receiving emails from it, send an email to sciruby-dev+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to the Google Groups "SciRuby Development" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sciruby-dev+unsubscribe@googlegroups.com.

[Sameer Deshmukh]

Dear John,

I can update the blog if you give me push access and let me update to the latest Octopress. The current version breaks my personal blog setup, which I would rather avoid.

Also, I am available for GCI tasks, but Atithya should take care of applications etc. since I cannot commit enough time for both GSOC and GCI.

[Pjotr Prins]

I think we can be a bit more relaxed when it comes to non-native
English. As long as someone with reasonable English has seen it I
believe we can publish. English is a lingua franca in Science, we
don't have to be experts. As one Professor said, I write crap English,
but it still gets me published in Nature. It is more important to get
the message across.

Blogs can be modified. There is no reason not to publish early and
often. We edit it if and when we need to.

Nothing discourages people as badly as waiting for months to follow
up.

Pj.

> the [2]SciRuby Wiki, for reference.

>
> --
> You received this message because you are subscribed to the Google
> Groups "SciRuby Development" group.
> To unsubscribe from this group and stop receiving emails from it,

> send an email to [3]sciruby-dev...@googlegroups.com.
> For more options, visit [4]https://groups.google.com/d/optout.

>
> --
> You received this message because you are subscribed to the Google
> Groups "SciRuby Development" group.
> To unsubscribe from this group and stop receiving emails from it, send

> an email to [5]sciruby-dev...@googlegroups.com.
> For more options, visit [6]https://groups.google.com/d/optout.
>
> References
>
> 1. mailto:athity...@gmail.com
> 2. https://mailtrack.io/trace/link/37278115cb63320f1a3824c91e5cf32675912887?url=https%3A%2F%2Fgithub.com%2FSciRuby%2Fsciruby%2Fwiki&userId=691437&signature=42e04b2d96f5391a
> 3. mailto:sciruby-dev...@googlegroups.com
> 4. https://groups.google.com/d/optout
> 5. mailto:sciruby-dev...@googlegroups.com
> 6. https://groups.google.com/d/optout

--

[John Woods]

Yes, Pjotr is right, I think. If people are super worried they can send stuff to me for editing and I'll try to reply pretty quickly. That'll be easier to do if someone else is responsible for actually publishing the site.

Sameer and Pjotr, I gave both of you access.

To unsubscribe from this group and stop receiving emails from it, send an email to sciruby-dev...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Arafat Khan]

Hi Sciruby Developers, 

I am Arafat from IIT Kharagpur. I have been involved with Scientific Ruby Gems for some time and am the lead developer of Tensorflow.rb. 

I have spent quite some time working with Sameer and Victor and learnt a lot from the community. 

I also recently spoke at Euruko 2017 about Tensorflow. 

I was wondering if I could also mentor for Google code in. My availability is  (Nov 28th 2017 - Jan 31st 2018) and I could mentor for NMatrix projects. 

I would also love to help out with the application in any way I can. 

Thanks

>    3. mailto:sciruby-dev+unsub...@googlegroups.com
>    4. https://groups.google.com/d/optout
>    5. mailto:sciruby-dev+unsub...@googlegroups.com

[Athitya Kumar]

Hey all.

I've just registered a sample organization (can be renamed to our org name) on the Code-In portal, to get to know the questionnaire. At least 1 more org admin for Code-In has to be added on the portal. We have 9 more days to submit the application. I've aggregated the question on our Wiki - please feel free to collaborate the application and tasks pages. :-)

[Athitya Kumar]

Hey all.

We have 4 more days to submit the application. I have registered SciRuby organization at the Code-In portal. Here are the bucket list of things that are to be done ASAP :

1. One more org admin has to be added at the Code-In portal. Who shall be that chosen one?

2. The application for the organization is almost ready for a review (except the question on unreponsive mentors). Feel free to review and collaborate.

3. We're supposed to list down at least 25 tasks, with the issue-trackers containing more. Kindly try to jot down issues of projects that you'd like to mentor.

The deadline for submission is 21:29 IST, 24th October 2017.

Peace!

[Victor Shepelev]

Heya. Finally dug my mail down to this thread.

Please add me as an admin (or tell me what I should do to be added as one).

I'll write my proposed list of tasks here today, we'll have some time to discuss it.

[Sameer Deshmukh]

Hello all,

I've registered myself as an admin. However, please note that it will be very difficult for me to participate in too many activities other than admin tasks and helping with some coding. University life is turning out to be very demanding! You all can please go ahead and write the byte-size tasks. I'll go over them soon and make some suggestions if I find anything missing.

I've gone through the application that has been made by Athitya, and will make some edits tomorrow morning (JST).

We can submit it by tomorrow night after someone else has gone through the edits.

Also, thank you very much Athitya for taking the initiative.

[Victor Shepelev]

OK, here are my considerations.

1. I am strictly against including Tensorflow in our application. I don't want to sound too harsh, but this project's affiliation with SciRuby looks less-than-obvious to me. We are not responsible for its design and code quality and evolution, and project's author is well aware of this. In his EuRuKo talk, he had never mentioned SciRuby ever had some role in a project.

2. As, unlike GSoC, Code-In explicitly allows and promotes "not-only-coding"/"not-only-new-features" tasks, let's design (I'll detail it at evening) "SciRuby Demo" tasks, like "Take data from StarWars API, do a pretty site of it and that would be it". Bite-size yet pretty looking demos could be invented easily.

3. Another promising engaging activity for some kind of people could be "comparison tables" (with pandas & R dplyr), it doesn't require any programming yet instead could be advertised as a cool way to understand data science tools and how they are similar and different.

I'll think a bit more about this throught the day, and will provide my final list in the evening.

Please discuss in the meantime :)

[Lokesh Sharma]

Sounds good. 

[Sameer Deshmukh]

Ok I don't mind if we exclude Tensorflow.rb this time. It is true that the code is not our responsibility, and there is also the fact I think the Ruby version is not upto date with the current Tensorflow version (we should make a better port soon though its an important library).

I think Arafat can still contribute to other projects like NMatrix, subject to him having knowledge of the subject.

[Sameer Deshmukh]

I have just updated and submitted the Code-in application on the Google page. John can you please skim through it once?

Also Athitya, next time you write an application please check the character limit for each answer. I had to truncate most questions.

The application has been submitted. Let me know if there are any last minute changes that need to be done.

[Sameer Deshmukh]

Who has added the nmatrix tasks in the task list? They are so complicated! Not byte sized at all. Please keep in mind that the mentor also need to effectively communicate the problem to the student. Something like 'Work on failing JRuby-NMatrix tests' is very far reaching. However, is responsible for NMatrix, please update the task list. I have removed the irrelevant tasks.

On Monday 23 October 2017 02:41 PM, Lokesh Sharma wrote:

[Prasun Anand]

Hi,

Thanks to Athitya for his efforts.

I think "Work on JRuby-NMatrix failing tests" would be good as GSoC task as it needs extensive ground work

like adding support for Complex dtype.

For, Google Code In, I would like to add tasks for writing ArrayFire examples.

e.g. monte-carlo estimation of PI

http://arrayfire.org/docs/benchmarks_2pi_8cpp-example.htm

Regards,

Prasun

[Sameer Deshmukh]

Yes Prasun, please go ahead with that. Also write an intro to array fire and explain to students how they start working on it.

[Victor Shepelev]

OK, guys, I've made my mind.

I should really ask everybody (especially Athitya, who spent so much effort on this) to forgive me, but we obviously don't have enough resources this year to make the list of tasks, apply properly, and most importantly, mentor people properly.

Being a mentor (both volunteer and paid) for a long time, I am seriously aware of the responsibility it implies, and prefer to miss the opportunity instead of doing crappy job on it.

I propose to skip this year and focus our efforts on Daru 1.0, proper publicity and infrastructure around it, and refreshing SciRuby.com, and this way to achieve some sustainability around the project.

V.

[Pjotr Prins]

On Mon, Oct 23, 2017 at 11:04:37PM +0300, Victor Shepelev wrote:
> OK, guys, I've made my mind.
> I should really ask everybody (especially Athitya, who spent so much
> effort on this) to forgive me, but we obviously don't have enough
> resources this year to make the list of tasks, apply properly, and most
> importantly, mentor people properly.
> Being a mentor (both volunteer and paid) for a long time, I am
> seriously aware of the responsibility it implies, and prefer to miss

> the opportunity instead of doing crappy job on it.

> I propose to skip this year and focus our efforts on Daru 1.0, proper
> publicity and infrastructure around it, and refreshing SciRuby.com, and
> this way to achieve some sustainability around the project.

It is very important to do things properly. Both for mentors and
students. We learn by doing, so sometimes you just have to jump in. We
did that the first GSoC :)

But I agree with Victor that for this thing we should have resources
lined up properly. It is pretty intensive to do this right. Main reason
I have not gotten involved.

Pj.

[Athitya Kumar]

Hey all.

I understand both the perspectives here. So yes, we had started off quite late and are running short of time. As a result, we somewhat have the resources (tasks) jotted down, but not with such great clarity as compared to what I saw in the Wiki while applying for GSoC 2017 as a student. I agree with this.

But I'd also like to see from the perspective of what Pjotr mentioned about the first GSoC application. We don't have the resources, at this point of time. Maybe we can take a leap of faith, and then make things right if given the opportunity? That is, students start working on Code-In tasks from Nov 28th onwards. I think that this would be enough time (until Nov 28th) to gather all resources, get the tasks aligned with as much details and clarity as possible. If we're to apply and get selected, we could definitely get this done before Nov 28th, and do justice to our role of mentoring students. :-)

I know that's a couple of if-s over there. But, I guess that's the thing with taking a leap of faith. We still have 19 hours to go until the Code-In portal closes. We probably have to reach a consensus before that, on whether to skip this year or not. What are your opinions, Sameer and John?

And no issues, Victor. Even if we're to skip this year, the efforts put into the drafting the application would hopefully come in handy as reference material if and when SciRuby applies for future versions of Code-In. :-)

Regards,

[Sameer Deshmukh]

Hmmm Victor has a point.

On a side note, I've already stated I will not be too involved in direct mentoring of code-in participants this year because of University and research. I also want to dedicate more time to coding than communication.

We need to consider the fact that we're dealing with high school students and that might require a lot more spoon feeding than the average GSOC participant.

The final decision should be taken by people who will be directly interacting with students the most (Victor and Athitya - but mostly Victor since Athitya has not mentored before and will require guidance himself).

I would still be open to participation if Athitya and Shekhar come up with a very detailed list of tasks (the current one is falls short considerably) and dedicate more time to mentoring. Victor is busy with new releases.

[Pjotr Prins]

Code-in is very competitive. If I remember correctly we failed to get
in once already.

One thing to note is that if we mess up we may have decreased our next
GSoC chances.

I agree more time should be invested by mentors.

It is probably wise to wait till next year and prepare more. Why do we
want to do this? To get tasks done or to get people involved in
SciRuby?

Pj.

> Athitya Kumar ([1]Facebook | [2]LinkedIn | [3]Homepage | [4]Blog |
> [5]GitHub)Ā

> 4th year undergrad
> IIT Kharagpur
>

> References
>
> 1. https://mailtrack.io/trace/link/35d49a3c1709868f4b3604bdaf6702a3c808e38c?url=https%3A%2F%2Fwww.facebook.com%2Fathitya.kumar&userId=691437&signature=f837428da5d4a783
> 2. https://mailtrack.io/trace/link/54aea42d47c6017d10456f7b418e33dc51c67b9b?url=https%3A%2F%2Fwww.linkedin.com%2Fin%2Fathitya-kumar-a152a8a9&userId=691437&signature=d808d34af7a3e5d0
> 3. https://mailtrack.io/trace/link/49810121b062483c139eb425825c0506970d8b99?url=https%3A%2F%2Fathityakumar.github.io%2F&userId=691437&signature=e12cd733c54920ef
> 4. https://mailtrack.io/trace/link/3a6e61b185ecfdb6e8f713196c04dbd37de6e6a0?url=https%3A%2F%2Fathityakumar.github.io%2Fblog%2Fpage1%2F&userId=691437&signature=818939d95824ef93
> 5. https://mailtrack.io/trace/link/5d1cc4d9515d529803365c91bf0cbf195dd96c34?url=https%3A%2F%2Fgithub.com%2Fathityakumar&userId=691437&signature=83501981477ffbe0

--

[Athitya Kumar]

Yes, I understand that Code-In is quite competitive. But I did not know that getting rejected in Code-In would affect our GSoC chances. I thought they were exclusive open-source programs that wouldn't influence each other. But if that's the case, then I wouldn't want to risk our GSoC chances by applying for Code-In. Let's buckle up for daru-v1.0 release and clean-ups / publicity of daru-io and daru-view. :-)

Psst psst, 6 hours to go before the Code-In application closes. If this is the consensus, Sameer / Victor - feel free to discard the application in the Code-In portal.

Regards,

Athitya Kumar (Facebook | LinkedIn | Homepage | Blog | GitHub) 

[Pjotr Prins]

On Tue, Oct 24, 2017 at 03:16:32PM +0530, Athitya Kumar wrote:
> Yes, I understand that Code-In is quite competitive. But I did not

> know that getting rejected in Code-In would affect our GSoC chances. I

> thought they were exclusive open-source programs that wouldn't
> influence each other. But if that's the case, then I wouldn't want to

They *should* not influence each other. But they are the same
organizers - or close to each other... If we mess up they will know.

Pj.

[Sameer Deshmukh]

Hmmm good thought. I’m pulling our application in that case.

Regards,
Sameer Deshmukh