GSoC application

[John Woods]

Hi all,

Here's what I have so far for our application:

https://github.com/SciRuby/sciruby/wiki/Google-Summer-of-Code-2015

And here's the proposed student app:

https://github.com/SciRuby/sciruby/wiki/GSoC-2015-Student-Application

We have some money, so I'd suggest we also apply for GNOME Outreach Program for Women. Thoughts?

Cheers,

John

[Timothy Gregory]

Hi all
yes you are right to help women pick up ruby needs visible uses
i know fine-arts degree students are often obliged to do 3D in Python

because open source 3D software often use scripts in Python
free ones like blender or free-cad and open-cad etc all have Python in common

--
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.

[Magdalen Berns]

I agree with the idea to apply for Outreach Program for Women, but it's worth noting in advance that it has changed its name to "Outreachy" and I believe they may have begun accepting applications for other minority groups now too. The outreach program also does not have to be a code project so it's more flexible if you want to someone to take a project focused on improving some aspect of documentation, marketing, graphic design etc I think they have a system now where if you pay for the first then you can get a few more funded internships, but I can't remember the specifics off hand so don't quote me on that!

As regards the GSoC application:

I think the given answer to "If you chose "veteran" in the checkbox, please summarize your involvement in Google Summer of Code and the successes and challenges of your participation. Please also list your pass/fail rate for each year." is really strong.

"How many potential mentors do you have for this year's program? What criteria did you use to select them?" should be revised (in my view). I'm a massive advocate of honesty, but being honest does not have to mean laying all the cards out on the table by default. Objectively, I reckon someone might read that and think there is a problem which could be down to one of two possibilities either

a) you are telling the truth and the female mentors dropped out because they were not qualified for the job in the first place (in which case they may wonder whether SciRuby is acting responsibly in giving out mentioning roles for code projects to people who might not have been up to the task)

b) your female mentors dropped out because the organisation was not inclusive after all and in that case pointing out their qualifications could come across as an underhand dig.

b) is the conclusion I am fairly certain a woman reading the application would arrive at, since that's the sort of thing women in tech are exposed to, but either way a) or b) are not likely to reflect well on the project. What are the actual statistics for female mentors who dropped out vs. the male ones? Is the data really that significant or would it be a reasonable things to not to comment on the pattern you have observed on the gender balance of mentor engagement in this section?

yes you are right to help women pick up ruby needs visible uses
i know fine-arts degree students are often obliged to do 3D in Python

because open source 3D software often use scripts in Python
free ones like blender or free-cad and open-cad etc all have Python in common

What does grabbing arts students obligated to pick up some python have to do with engaging women to code for SciRuby? We're a minority in STEM undergraduate degrees but we do exist there, believe it or not: The numbers start dropping off at undergrad but the drop off seems to become most significant when you start looking at post graduate and beyond statistics. That glass ceiling, I tell you.... 

Anyway, besides all that though, don't you reckon that anyone with a limited educational background in science and maths would be likely to struggle with a scientific programming project, even if they know how to code a little?

[John Woods]

Replies inline below. Thanks, Magdalen — I really appreciate your insight on this.

On Mon Feb 16 2015 at 2:46:14 PM Magdalen Berns <magdal...@gmail.com> wrote:

I agree with the idea to apply for Outreach Program for Women, but it's worth noting in advance that it has changed its name to "Outreachy" and I believe they may have begun accepting applications for other minority groups now too. The outreach program also does not have to be a code project so it's more flexible if you want to someone to take a project focused on improving some aspect of documentation, marketing, graphic design etc I think they have a system now where if you pay for the first then you can get a few more funded internships, but I can't remember the specifics off hand so don't quote me on that!

As regards the GSoC application:

I think the given answer to "If you chose "veteran" in the checkbox, please summarize your involvement in Google Summer of Code and the successes and challenges of your participation. Please also list your pass/fail rate for each year." is really strong.

Thanks. I was particularly proud of this answer also. =)

 

"How many potential mentors do you have for this year's program? What criteria did you use to select them?" should be revised (in my view). I'm a massive advocate of honesty, but being honest does not have to mean laying all the cards out on the table by default. Objectively, I reckon someone might read that and think there is a problem which could be down to one of two possibilities either

a) you are telling the truth and the female mentors dropped out because they were not qualified for the job in the first place (in which case they may wonder whether SciRuby is acting responsibly in giving out mentioning roles for code projects to people who might not have been up to the task)

b) your female mentors dropped out because the organisation was not inclusive after all and in that case pointing out their qualifications could come across as an underhand dig.

b) is the conclusion I am fairly certain a woman reading the application would arrive at, since that's the sort of thing women in tech are exposed to, but either way a) or b) are not likely to reflect well on the project. What are the actual statistics for female mentors who dropped out vs. the male ones? Is the data really that significant or would it be a reasonable things to not to comment on the pattern you have observed on the gender balance of mentor engagement in this section?

You're doing a great job of highlighting my biggest concern with the answer I wrote.

Basically, I asked my friends Anna and Stacy to mentor. Stacy initially said yes and then decided she didn't have time, due to her dissertation semester being nigh upon her. Anna said yes, and even recruited her friend, Alex.

But then we didn't hear from any of them. They didn't review applications, they didn't reply to any emails. If they dropped out because of lack of inclusiveness, we have no way of knowing that. I can certainly ask Anna (she's one of my best friends), but she's never been the type to be deterred by a lack of inclusiveness.

I think it's more likely that the women who agreed to mentor dropped because they didn't have any real motivation for participating. All of our other mentors are Rubyists, so they stand to directly gain from more people using and developing SciRuby. (Yes, sure, everyone stands to directly gain from more women coding, but the incentives are less direct.)

Now, all of this is difficult to application-ize, so any suggestions on how to phrase this answer so that it doesn't sound like we've been terrible at inclusiveness? 

yes you are right to help women pick up ruby needs visible uses
i know fine-arts degree students are often obliged to do 3D in Python

because open source 3D software often use scripts in Python
free ones like blender or free-cad and open-cad etc all have Python in common

What does grabbing arts students obligated to pick up some python have to do with engaging women to code for SciRuby? We're a minority in STEM undergraduate degrees but we do exist there, believe it or not: The numbers start dropping off at undergrad but the drop off seems to become most significant when you start looking at post graduate and beyond statistics. That glass ceiling, I tell you.... 

Thanks for pointing this comment out. I missed it when I first read over the person's email, but I'll send him a quick message now about it.

 

Anyway, besides all that though, don't you reckon that anyone with a limited educational background in science and maths would be likely to struggle with a scientific programming project, even if they know how to code a little?

[Carlos Agarie]

I submitted the answers on melange, but we can still edit them.

Is there any organization that could vouch for us? I also don't know what to write in that last question. 

-----

Carlos Agarie

Software Engineer

+55 11 97320-3878 | @carlos_agarie

[John Woods]

We don't need/want a vouching org since we're a veteran organization. 

[Magdalen Berns]

Replies inline below. Thanks, Magdalen — I really appreciate your insight on this.

No problem :-) 

On Mon Feb 16 2015 at 2:46:14 PM Magdalen Berns <magdal...@gmail.com> wrote:

I agree with the idea to apply for Outreach Program for Women, but it's worth noting in advance that it has changed its name to "Outreachy" and I believe they may have begun accepting applications for other minority groups now too. The outreach program also does not have to be a code project so it's more flexible if you want to someone to take a project focused on improving some aspect of documentation, marketing, graphic design etc I think they have a system now where if you pay for the first then you can get a few more funded internships, but I can't remember the specifics off hand so don't quote me on that!

As regards the GSoC application:

I think the given answer to "If you chose "veteran" in the checkbox, please summarize your involvement in Google Summer of Code and the successes and challenges of your participation. Please also list your pass/fail rate for each year." is really strong.

Thanks. I was particularly proud of this answer also. =)

 

"How many potential mentors do you have for this year's program? What criteria did you use to select them?" should be revised (in my view). I'm a massive advocate of honesty, but being honest does not have to mean laying all the cards out on the table by default. Objectively, I reckon someone might read that and think there is a problem which could be down to one of two possibilities either

a) you are telling the truth and the female mentors dropped out because they were not qualified for the job in the first place (in which case they may wonder whether SciRuby is acting responsibly in giving out mentioning roles for code projects to people who might not have been up to the task)

b) your female mentors dropped out because the organisation was not inclusive after all and in that case pointing out their qualifications could come across as an underhand dig.

b) is the conclusion I am fairly certain a woman reading the application would arrive at, since that's the sort of thing women in tech are exposed to, but either way a) or b) are not likely to reflect well on the project. What are the actual statistics for female mentors who dropped out vs. the male ones? Is the data really that significant or would it be a reasonable things to not to comment on the pattern you have observed on the gender balance of mentor engagement in this section?

You're doing a great job of highlighting my biggest concern with the answer I wrote.

Basically, I asked my friends Anna and Stacy to mentor. Stacy initially said yes and then decided she didn't have time, due to her dissertation semester being nigh upon her. Anna said yes, and even recruited her friend, Alex.

But then we didn't hear from any of them. They didn't review applications, they didn't reply to any emails. If they dropped out because of lack of inclusiveness, we have no way of knowing that. I can certainly ask Anna (she's one of my best friends), but she's never been the type to be deterred by a lack of inclusiveness.

I think it's more likely that the women who agreed to mentor dropped because they didn't have any real motivation for participating. All of our other mentors are Rubyists, so they stand to directly gain from more people using and developing SciRuby. (Yes, sure, everyone stands to directly gain from more women coding, but the incentives are less direct.)

I think one of the women you mentioned was a supporting mentor for my project last year but I don't know about the others: Were they primary mentors or "just" supporting mentors?  It seems like less of a surprise/concern for a supporting mentor to find they are busier than they predicted half way through a project than a primary mentor, because the student is still in a situation where they are still technically being supported, in that case.

 

Now, all of this is difficult to application-ize, so any suggestions on how to phrase this answer so that it doesn't sound like we've been terrible at inclusiveness? 

I think the fact you have already been quite concious to reflect on inclusiveness in answers to questions gives some slack for answering a question like this more directly. Recalling the question, with that in mind i.e. "How many potential mentors do you have for this year's program? What criteria did you use to select them?" and it does seem to be prompting SciRuby to take a forward look rather than reflect. :D 

Perhaps if you do feel it is necessary to reflect though, you could say something about how you tend to put the mentor who is most keen on the project in charge but have supporting mentors on each project too who follow the students progress on projects, just in case mentors have unforeseen priorities which take over their time? That might give you an opportunity to indicate where other priorities got in the way of things for some mentors but ultimately that there was someone else in place to support the students so the support they received on their project was not adversely affected?  

Magdalen

[John Woods]

Excellent. Thanks, Magdalen. I've updated the application to clarify some of what you've said. I think the key here is that we don't *let* people be primary mentors unless they go put up an idea or stick their name in the hat; and in the case of our female mentors last year, *I* had to put their names in the wiki.

So — minimum level of investment required to be a mentor, whatever your gender.

--

[Magdalen Berns]

Excellent. Thanks, Magdalen. I've updated the application to clarify some of what you've said. I think the key here is that we don't *let* people be primary mentors unless they go put up an idea or stick their name in the hat; and in the case of our female mentors last year, *I* had to put their names in the wiki.

So — minimum level of investment required to be a mentor, whatever your gender.

Looks good.

[Sameer Deshmukh]

Congratulations SciRuby! We're in!

[John Woods]

Hah. Just saw this. Thanks for the notification. =D

--

[Karthikeyan A K]

Refreshing to know Ruby is just not limited to Rails. Hope there will something like http://www.sagemath.org/ in Ruby.

[John Woods]

It's been proposed before. We'd need someone to mentor for it. Would you be interested?