Featured GSOC Student Applications are Open!

Skaldarnar

Badges badges badges badges mushroom mushroom!
Contributor
Art
World
SpecOps
#1
Time is running, we have already entered the student application phase of Google Summer of Code 2018! If you are interested in a summer project with us, you have time until end of March (27th) to work on your proposal, get in contact with the community, and hand in your first PRs. Have a look at our project ideas to get some inspiration on what to work on.

If you want to learn more about GSOC and how you can improve your chances on getting in, have a read of these blog posts I wrote up last year:
We are happy to welcome anyone to the project, and we are really curious about your ideas for the summer projects. If you have any questions, just say hi in the forum or on our chat and ask whatever you want to know ;) Chances are high that there will be someone around able to help you quickly.
 

Cervator

Org Co-Founder & Project Lead
Contributor
Design
Logistics
SpecOps
#2
For some more details on the current state of proposals:

As expected we've had solid interest in the new/expanded gameplay template "generic" idea, just about covering all the basic suggestions (Light & Shadow, Master of Oreon, the brand new GooKeeper, GooeysQuests, Lost, and even a tower defense setup. Metal Renegades hasn't been spoken for yet as far as I know, but would be a pretty easy item to pivot to from any other such item as there is a bit of overlap here and there with multiple interested students in some places.

Those items are the more approachable ones for both students and mentors, as their scope tend to be pretty open. Go for an initial minimal viable product then iterate iterate iterate. There are usually more possible extensions than it is doable to do in one project but you can pretty much pick and choose, even swap out mid-project if problems are encountered. For the same reason it tends to be pretty easy to help out on mentoring these items, providing code review and play testing. We could use more such mentors, even right away to help with proposal review. If you have an interest and have made a good impression on the community so far you're up for consideration, just let us know here or on chat sometime. If shy you can PM me or somebody else who looks like they know what they're doing to ask, either on chat or here in the forum

More technically challenging ideas approached so far include cubic planets, multiworld, shaderland, improved automated testing, dest sol entity system integration, Android admin app, and a few more well less defined topics (along with anything I've missed). Those tend to be harder to mentor for and can sometimes be more difficult to scope right up front.

One such example is cubic planets which has two prospective students who might want to really consider the following: this idea may be too large to readily approach for this coming GSOC as-written. I'm surprised that multi-world is appearing simpler, I thought it was going to be the other way around. The main problem seems to be how multi-disciplinary the idea is turning out to be:
  • World shape including how the current chunk size is 32x64x32 with the greater dimension being "up" in part to take sunlight propagation into consideration (which gets tricky when "up" isn't always the same direction anymore)
  • Facets for world gen when suddenly everything doesn't start out flat, forever. There are existing suggestions for how to improve the facet system already, and that doesn't even begin to deal with being directionally challenged (what way is up again, and can sea level be the perimeter of a cube?)
  • Physics - gravity is no longer a universal constant in the same direction everywhere? Uh oh, how will we ever walk around the edges and so. And physics is also going through potential changes.
  • Light calculations (may get into rendering some) from a sun going around the planet, meaning the direction of light changes
In this case I'd feel bad if we end up leading students astray into a doomed project, which almost happened in our first GSOC when we had underscoped the idea @tdgunes started on, which as a result had to be shrunk over the summer. So I'd suggest for that idea (and maybe similar ones) to consider whether a subset would be a better option. In this case for instance how about improving physics with a focus on gravity manipulation where you could place a large cube (or a floating island) in an existing world but give it its own gravity so the player can walk about it fully? That way you suddenly avoid any concerns about world shape, world gen, and light calcs, but there'll probably still be enough for a GSOC project plus the other approaches could still be stretched for if time allows.

In closing:

If you're a prospective student it is now or never (until next year!) if you want to participate, but don't be afraid to change an item while talking with somebody related to the item. It is still possible to submit a great proposal if you've already made past efforts to show your skill and interact with the community (harder to do all that from scratch with a week left)

If you're a prospective mentor please let us know right away so we can get your feedback on the pending proposals, and can get a better feel for how many may be interested in helping mentor the different items. This will heavily affect how many projects we can support over the summer

And to all: may the odds be ever in your favor! No, seriously, you don't have to murder the other students, you could also pivot :D
 

Skaldarnar

Badges badges badges badges mushroom mushroom!
Contributor
Art
World
SpecOps
#3
Thank you for the additional information! Good to have you help out on some details, leaves me some time to actually look at proposals and feedback. Looking at the activity we already have by prospective students I can only second the call for mentors. :)

If you're a prospective mentor please let us know right away so we can get your feedback on the pending proposals, and can get a better feel for how many may be interested in helping mentor the different items. This will heavily affect how many projects we can support over the summer
If you want to get involved in helping students to master their summer project just let us know (as said, either here, on chat, or via PM).
 

Cervator

Org Co-Founder & Project Lead
Contributor
Design
Logistics
SpecOps
#4
Here are a few more notes:
  • Student app deadline is in a few days. I'll be around most of Saturday and Sunday and there'll be others around as well to get feedback from. Students please do take advantage of this and start upgrading your proposals to the final version! If you don't submit a final version we cannot select you at all
  • There is another nice blog post out at https://medium.com/@owtf/google-summer-of-code-writing-a-good-proposal-141b1376f076 about writing good proposals for any last minute tips
  • GitHub Universe is now selling super early bird tickets. Last year this was timed perfectly with the GSOC Mentor Summit allowing two of us to both go to the summit as well as GitHub Universe. I don't know if the timing will work out as well this year but it might, and we expect to be able to send 3+ mentors (last year we sent 4). But you have to actually sign up as mentor first to have a chance to make the summit and then maybe also GHU, but if it works out you could go there same trip with at least partial reimbursement :)
  • Did I mention we could always use MOAR MENTORS!?!?! ;)
 

Cervator

Org Co-Founder & Project Lead
Contributor
Design
Logistics
SpecOps
#5
Student application deadline has now passed!

... and I want to accept like 15 projects :p (out of 29 final proposals)

Yeah we're going to need some more mentors, we'll be capped by mentor capacity far before we run out of interesting proposals. Guess I know what I'll be doing for the next two weeks! Best hide under a rock if you don't want to be pestered about mentoring

Next up:
  • By April 7th we have to submit how many slots we'd like from Google. So we don't have to pick the specific projects by then, just have our math done. That does mean we'll have a good idea who to pick and we'll need to have maxed out our mentor reach by then
    • If we are waffling on whether a proposal is good enough to pick this is the time you have to turn it up to 11 by slinging some code at GitHub to prove you can do what your proposal suggests you can
  • By April 17th we actually have to pick the specific projects and sort out any conflicts (if two orgs pick the same student)
  • By April 23rd Google will announce the actual projects. We can't tell anybody until then. Which kinda sucks! ;)
For anybody we can't pick I can just about guarantee we'd still love to work with you, we just won't have the capacity to guarantee you'll have enough mentor assistance to succeed with the GSOC timeline. For the few places where there's more than one student on the same proposal we occasionally ask if a non-picked student would be interested in helping out with the item as a mentor of sorts. That worked well last year as one student opted out for unrelated reasons but then stepped in to help mentor the item instead (at a lower time demand - I suggest to expect a weekly meeting for an hour plus occasional PR review, testing, and/or brainstorming).

We've toyed with the idea of running our own unofficial GSOC-but-not-GSOC projects at a more casual pace with a smaller stipend, so that's another option for students on an item that doesn't turn into a project at all. That could be structured the same way, but with floating/expanding time boxes. Which at that point I guess are more like plain boxes with questionable structural integrity.
 

manu3d

Active Member
Contributor
Architecture
#6
Cervator wrote:
... and I want to accept like 15 projects :p (out of 29 final proposals)

Just so we are clear with everybody, Cervator was probably on mushrooms when he wrote this. We are in the process of tracking his dealer to make sure next time he sells him some lighter stuff. ;)