GSOC 2016 - DONE!

[Cervator]

After years of applying we finally got picked, woohoo!

First thank you to all those who have helped over the years, and of course Google for running the program in the first place

For mentoring @Skaldarnar has already volunteered to be a primary mentor earlier and is our backup admin, while I'm primary admin and a backup mentor. The guideline is to have two mentors per student slot and as a new organization this year we are likely limited to 1-3 students (or 1-2 depending on what doc you're looking at).

Estimate for time involvement for a serious mentor is about 10 hours a week or more, which might be time you'd be coding anyway and just make yourself more available for questions.

We have lots of "light" mentors available on IRC like we always do, with several having reaffirmed their willingness to help out students even without being official mentors.

If we get more formal mentors we can request 2 or even 3 slots and see what GSOC comes back with. We already have several promising students Please let me know if anybody is interested, I need to send you an email invite if so, PM me one or post it here if you don't mind.

Mentors:

• Skaldarnar (primary, fully committed)
• @Florian (primary, fully committed)
• @msteiger (primary, fully committed)
  
• @manu3d (backup, rendering)
• @Marcin Sciesinski (backup, architecture)
• @SkySom (backup, content)
  
• Cervator (backup, organizing)

Some handy links:

• Timeline (official one, see below for copy paste and slightly more details)
  
• Mentoring manual
  
• Organization list - look at the company we're in! Only game project, so many big names!
  
• Us! Yay! - Maybe we got picked because of how perfectly "Dark Blue Gooey" fits the new color scheme

Timeline expanded a bit - copy paste so may not catch any changes (unlikely)

• March 14 - 25: Mentors and Org Admins review student draft proposals and give students feedback on their proposals.
• March 26 - April 10: Review all submitted student proposals with your org and consider how many you want to select and how many you can handle.
• April 11: Deadline for Org Admins to submit slot requests.
• April 13: Slot allocations are announced.
• April 14 - 20: Org Admins select the proposals to become student projects. At least 1 mentor must be assigned to each project before it can be selected.
• April 22: Accepted GSoC students/projects are announced.
• April 23 - May 22: Community Bonding Period.
• May 23: Coding begins.
• June 20 - 27: Midterm evaluations of students.
• August 6th: Saturday play test and GSOC review / demos
• August 23: Student deadline - final evaluations begin
• August 29th: Mentors submit final project evaluations
  
• August 30th: GSOC 2016 results announced!
  
• October 28-30: Mentor Summit in Sunnyvale California

 

[Cervator]

We've already met a surprising amount of students, huzzah!

Adding some student related info as well:

• Next timeline event is when apps/proposals start getting accepted in the system March 14th
  
• Student manual - probably important!
• Proposal writing - maybe the most interesting section right now. As a project we (MovingBlocks) have no additional guidelines, since we're as new to the program if not newer to it than most the students Feel free to discuss your ideas here in the forum, like in your contributor intro thread or a Suggestions post.
  
• GSOC FAQ - handy for specific questions asked before. Again we're just about as new at this as the rest of you, so lots of stuff we haven't a clue about either

As mentioned above we're likely only going to be able to have one or two students officially added through GSOC We're a first year org after all. But we're always super happy to help anybody learn! That has been a recurring topic through our multi-year existence, and some of my favorite stories come from when we met somebody unlikely who turned into a huge success story

For students hoping to land one of the GSOC spot my top advice would be to clone the code off GitHub, fix something, and send us a pull request! You can do that right now and that makes a great impression. Proposals come later

Communicate with us here in the forum and on IRC (or Slack, if you want an invite - let me know). Private messages and email aren't nearly as effective at building a community. You're also less likely to build good relations with potential mentors if you end up just talking to a single person privately.

And most importantly: enjoy yourself, learn some code, and meet interesting people!

 

[Cervator]

Another quick update and message for the students

First, a new GSOC item was suggested: enhancing our particle system. This was suggested by @Florian who might also become an official mentor (pending confirmation) and could mentor that item himself. So a good option and a cool one for those interested in rendering.

At this point we're past 20 students which is amazing! I wish we could get all of you official GSOC slots but as mentioned above we'll likely only get one or two slots. So that leaves us with a few recommendations:

• If you really want to focus on a slot here for Terasology we already have several students looking great with lots of communication here in the forum, on IRC, and with pull requests already merged with bug fixes and other enhancements. That should be your goal - before proposals even start in a week it looks like we'll have a good idea who will be active and able to stick with the program over summer.
• If the stipend from Google is important for you financially and you're not already at the head of the pack in activity and coding you might want to also prepare proposals for other organizations. I don't know if all organizations have this many students competing for slots, but I do know some other orgs have more slots than we'll have so your odds may be better elsewhere.
  
  • If you think you'll have trouble but still want to try making a little money then let me know. We don't have enough donations to get anywhere near a Google-sized stipend but we might be able to make some extra code bounties available if it'll help get some more quality code written and some skills learned
• If you primarily just want to learn then we're here for you in any case! Help on IRC has been more active than ever, and we're thrilled to teach and learn together with all of you. Even students here for GSOC are already helping others!

I hope as many of you will stick around as possible! Community here has always been top priority and it is awesome to see people learn and succeed

 

[Cervator]

Mentor list updated woo! We now in theory have enough to cover two student slots, with @Florian stepping up as another primary and responding to students already

Myself and @manu3d will be backups for specific topics when available and needed. Might find a few more like that still, along with those already helping in general.

Several new potential topics have been raised, noted in various contributor intro threads. Keep in mind that proposals aren't necessarily tied to one of the existing items on the ideas list. If somebody shows up with a unique skill and a great idea we won't turn it away just to make the person work on some less fitting item!

I checked in #gsoc on Freenode IRC and it does look like our student interest (past 30 students easily at this point) is above average while our likely slots are way below average since we're a new org. Again this unfortunately means the odds of getting an official GSOC slot + stipend with us are lower than elsewhere If the stipend is critical to you and you don't think you're near the top here please check out some of the other orgs and submit proposals there (as well). There are veteran orgs with more slots yet less interest than we've gotten. Non-game projects need love too!

Still, if the stipend isn't critical then as mentioned we're thrilled to help every last one of you learn stuff! And there are always code bounties to fall back on, with several hundred dollars & euros of un-allocated funds looking for a good home if it helps get some good work done.

Also, I'm aiming for us to hold a dev meeting on Saturday evening (European time - afternoon in the US and early morning Sunday in Australia). More details and a timezone calculator in a separate thread soon. Primary topic is future architecture beyond our engine v1.0.0 release (scheduled for Sunday) but we might get into specific GSOC proposals that could relate to the discussed architecture.

 

[Cervator]

We've entered the proposal period and already have a few submitted - nice! Yet we still have more students showing up. Getting a bit late to start and the bar is already pretty high. Anything is possible but if you hope to get a GSOC spot you probably should be coding easy (or harder!) topics by now for the experience and to learn the code plus be talking to us regularly about your proposal

I've copy pasted a slightly more detailed timeline in the top post. Next deadline is proposals submitted March 25th

Again we would be thrilled to help any and all of you learn even outside GSOC, before, during, and after the project summer. Just let us know your preference - if you don't think you'll submit and would prefer an informal learning experience please tell us so we can focus accordingly

We are up to 7 mentors, 3 of them primary, so maybe we even have a chance to land a third slot We won't know until April 13th and according to the documentation we aren't supposed to reveal the count either, just pick our preferred proposals and let them be announced April 22nd. Here's for hoping for the best experience for everybody involved!

Edit: Also new potential GSOC item - character model customization & animation

 

[Cervator]

Reminder: Proposals are due Friday! Not actually sure when during the day they're due.

If you submit a final proposal in PDF format we can't actually see it until after the deadline is over. But you can apparently still edit it. So if you keep making revisions to your draft we can comment there and you can adjust the PDF as needed. Better submit a PDF early than forget and miss out!

Final stretch - make it count!

And remember: you're more than welcome to contribute and learn along with us outside of GSOC. Anybody arriving around this time will have an extremely difficult time catching up with code, community, and proposal. But we'll be here a long time and love helping!

 

[Cervator]

Exact deadline for final proposals is March 25 19:00 UTC - that's 14 hours from now!

Submit a final proposal even if you're not totally done yet, you can re-upload more changes up until the deadline if you need to. You don't want to wait till the last minute and lose power or something!

As of this writing we have 16 final proposals and 11 drafts, so probably there are still a few more final ones coming. Don't give up if you don't think you're near the top, at the very least this'll be a learning experience in how to write proposals and introducing yourself to a new community - you'll be doing that again whether it is for education or just life

Yet again a reminder: even if we can only pick even just one or two of you we'd be thrilled to help as many of you as possible learn things and have fun! If you don't make it this year spend some time practicing and maybe getting used to our project if you want to try again next year. If we get in we'll likely have a better chance for more slots, woo!

A few of you didn't find us (or even GSOC) till near the proposal deadline. With a proposal submitted or not you might not have had a chance to meet a lot of people or even check out the code. After the deadline there's still almost a month (April 22nd) till it'll be announced whose proposals will be turned into projects. So if you've missed out you actually have more time still, with the pressure of the deadline off. Not that you should then stress out to cram code for weeks just to better your chances to get picked - enjoy yourself and learn things instead

To celebrate the deadline passing, our v1.0.0 release itself, and all our GSOC students, we'll run a hackathon over the weekend, actually starting Friday since that's a holiday in many places plus @Skaldarnar came up with the idea and volunteered to start hacking code and helping people early Friday! Nothing official till the deadline passes (proposals come first) but lots of code and fun times so come join us!

 

[Cervator]

Update: Proposals are done! Everybody take a collective deep breath and let it out again. Time to relax a little bit

We got 35 final proposals submitted. Amazing! Probably way above average for a new org like ours.

I've posted a new thread considering what to do with our substantial student overflow - please read if interested: http://forum.terasology.org/threads/continuing-mentoring-beyond-gsoc-and-new-issue-markings.1522

Also don't forget the weekend hackathon, which might be able to pick up more steam Saturday now that the deadline is passed. This is a great way to make a good impression, especially if you arrived fairly close to the deadline

 

[Cervator]

Projects have been announced by Google and are now official/public!

https://summerofcode.withgoogle.com/organizations/4668766554161152/

Choosing was hard but made a little less bad by managing to land a full 3 slots despite only expecting 1-2

We had just shy of 10 or so final proposals we were considering and more than 3 of them were perfectly valid and desirable to work over the summer, but we did have to pick just 3

Our three picks:

• @Rostyslav Zatserkovnyi with NUI. This one was pretty much a givenn since he's been going nuts with good contributions since getting here. Heck he's even trying to apply some Gestalt stuff to Destination Sol. Crazy
• @tdgunes with DAG pipelines. Good proposal, good community impression (some of it on Slack which is less visible than forum/IRC), good activity with code and we need more 3d wizardry
• @Cpt. Crispy Crunchy with City Simulation, which is very central to livening up our worlds and aims for the very creature society focus the game name was developed for. He's also been going nuts with code and the very impressive Shattered Worlds module

Congratulations!

For the rest of our students thank you again so much for your proposals, comments, contributions, interest, and so on. We wish we could've picked all of you! Proposal or not we'd be happy to help any of you learn and would love to have more of your contributions. As the dust settles I hope we can talk about the initiative for more mentorship in general and include several of you and your proposals in that setup.

I am also more than happy to increase the code bounties on the big issues, we both have enough community funds for that and I'll add extras personally if it helps! Won't be a full GSOC stipend by any means, but it could end up a few hundred dollars for doing interesting work, at less overall scope than a full GSOC project. Please let me know if any of you are interested, here, on IRC, via convo, or anywhere you can reach me

This will be a super cool summer!

 

[Cervator]

We've discussed the possibility of adopting a few more ideas as unofficial projects, maybe with up to a $500 code bounty. See this update in the "Continuing Mentoring" thread for details.

As for what comes next: we need to get organized for the summer both for official and unofficial projects

This is where us being new to this isn't super helpful as we don't have a nice framework for how to structure this work, and all previous work has been full volunteer "it'll be finished when it is done" style. We have the proposals and most the final ones have good timelines, which was a key deciding factor. If any students who participated in prior GSOCs have any pointers or ideas feel free to chime in!

My personal recommendation (please provide feedback!) would go something like the following:

• Consider the timeline in the proposal or consider the proposal on a week-to-week or feature-to-feature basis if you haven't already
• For each week / feature (the GSOC work period goes from May 23rd through August 23rd, so about 13 weeks) name the primary goal
• Make a checklist with 13 items in it - for instance in a forum work thread or the central issue on GitHub. Or make 13 cards in Trello or 13 issues in a GitHub repo (more on that later)
  • 13 items is not a final number. Maybe you have 10, maybe you have 20. Just need a number
• Add sub-items if appropriate (nested bullets, a Trello card checklist, checkboxes on a GitHub issue..)
• As the work period progresses check off completed work so you have a visual on your progress. Submit PRs to have work merged early if appropriate (if there's something to merge - helps avoid stale code)

For unofficial projects (without a formal GSOC slot) you aren't bound to the GSOC work period, but maybe you would still want to set yourself milestones to better emulate the "project feel"

I'm not sure which tool would be preferred. Individual preference?

One particular tool I'm personally curious about is https://www.zenhub.io which is a browser plugin that adds a kanban board, issue estimating (using story points), and a burndown chart right to the GitHub site. So if a GSOC project has 13 items you make 13 issues, estimate their comparative complexity, and complete them over time, giving you a nice graph of your process. I need to play with this some myself soon. Could be 13 issues in the engine repo (could get a bit spammy) or maybe 13 issues in the student's fork (which can have its own issue tracker)

I would suggest for our primary students (@Rostyslav Zatserkovnyi, @tdgunes and @Cpt. Crispy Crunchy) to consider their checklisting preferences and details over the next couple weeks in the bonding period, then we can see if we should pick something consistent across the board and apply it in the next two weeks before the work period begins

Anybody interested in working a GSOC or similar item in a likewise structured manner please let us know both your preferences on work style and which item you would be interested in (see the mentoring thread for a nice list)

 

[Cervator]

Oh, and office hours with mentors also came up. This is a bit tricky since we're all somewhat irregularly available. Most the main mentors are in Europe, so I imagine weekday evenings European time will be the most ideal times, as are likely some weekends with perhaps expanded hours.

Saturdays have traditionally been our "project office hours" but without much need for them or consistency we haven't used that concept much. We might need to change that and try to make sure one or more mentors will be around at certain times (but not necessarily a specific mentor every time)

Another idea brought up was to have a per-project channel on Slack to leave questions and have them answered, which has potential. But then that's not actually much different than just using a forum thread, unless a particular student + mentor are both online at the same time. This is another thing to look into figuring out over the next couple weeks

 

[Viveret]

I think having per-student forks with their own issues would individualise each project, and then using that tool to schedule issue fix deadlines would be very helpful, especially since its integrated with GitHub.

 

[Rostyslav Zatserkovnyi]

Here's my take on the project structure:

• Core forum thread - contains periodical reports and demo screenshots/videos; aggregates external feature requests.
• Summary GitHub issue - in my opinion, should be separated from the actual issue describing the project, but this is up for discussion. Contains proposal info (student/mentor details?, abstract, proposal link?, issue of origin) and a checklist with links to...
• Individual GitHub issues - located in an individual fork to avoid massive issue spam (right now plugging these into the main repo would result in ~30-40 additional issues!) These can then be plugged into ZenHub / Trello / a physical Kanban board / anything that provides a fancy way of tracking progress. Here's an example of this in action (found in a GitHub search for "gsoc"): https://github.com/mitmproxy/mitmweb/issues

 

[tdgunes]

Personally, I think both cervator and rzats recommendations are very nice. But more particularly:

• Core forum thread - I was thinking about writing periodical reports especially on my blog. But it would be great to keep all relevant stuff at the same place. For keeping a personal record, I want to post my reports also to blog, if that is ok.
• Using ZenHub, I am kind of biased and also curious when it comes to a new tool. I'll study it more, and write my feedback here.
• Using forks for storing GSOC project issues, It might be better to not spam main repository and post issues and progress into our own forks. For not individualising, we can put a summary issue on the main repository. This would not be beneficial just for this, but also for keeping track of the students and their progress for mentors.
• Making checklists, it's a good idea, would be better if these checklists are prepared before that week starts. It would be really hard predict and estimate whole GSOC period. Also always nice to have sub-items in a checklist!

 

[Cpt. Crispy Crunchy]

My approach would be as follows:

• Module forum thread - I want to keep this in the style of the other module threads and post current work progress and screenshots etc. there and to have people comment on it in a general fashion
• Per week planning - I'm inclined to use Trello, although I have no experience with it yet, I heard good things from friends about it and I definitely need such checklists. If the rest plans to go along with Zenhub which seems to do the same thing I would be open to it too as I like it to be consistent.
  
• Github Issues - I don't think it would make sense to repost everything which is already in Trello/Zenhub. This allows for posting problems/bugs which one can't solve immediately to be easier to find. It could however make sense to have one Issue where the progress of all individual steps is tracked. Just to see which ones are merged already or need reworking etc. But then again this would be unnecessary as long as it is synchronized with Trello/Zenhub.

So all in all I would agree with rzats, tdgunes and Cervator .

That leaves only the question if we want to use Trello or Zenhub or dismiss consistency.

 

[Cervator]

Okay thanks for all the feedback

I did a bit of a write-up of my recommendations in a different thread just so we can discuss it separately: http://forum.terasology.org/threads/organizing-via-forum-on-github-with-zenhub-waffle-io-replacement-and-in-trello.1544/

It is a bit sizable, but then I tried to cover a lot including the stuff mentioned above! And yes updates in a topic thread (or even contributor thread) then also shared via blog or social media is perfectly fine

 

[Cervator]

Just past a week till the coding period begins!

Also found out today the mentor summit will be October 28-30th in Sunnyvale California. Up to two mentors can go on at least partial travel stipends. I'd like to go and check it out, anybody else interested?

Tomorrow (Saturday) I'm putting aside the entire afternoon (US east) for Terasology, DestSol, and particularly GSOC stuff. Have mentioned it in a couple places as an option to examine potential ZenHub approaches and related tools & organizing. Pinging @Rostyslav Zatserkovnyi @tdgunes @Cpt. Crispy Crunchy for reference

We've also made channels in Slack for the individual projects:

• #citysimulation for @Cpt. Crispy Crunchy's topic
  
• #nui-editor for @Rostyslav Zatserkovnyi and NUI
  
• #rendering for @tdgunes with the DAG pipelines
  
• And a bonus #combat-and-friends for @Anthodeus and friends for unofficial projects since several of them seem to group well: combat, anatomy, customizable models with limbs, etc. @korp and @Kartikey Agrawal have been in on some discussion, maybe we can drag @Viveret into it too for putting a particle system to work on combat-related effects? Or various status effects like a poisoned aura of something

The session tomorrow (actually starting around 10-11 hours after this post, sorry about the short notice!) is meant to be more about tooling and process, then after maybe later some assorted discussion or release prep. Maybe we can do some multiplayer testing - @minnesotags you up for updating the test server to latest dev build running JS + Caves? Thinking we might be able to release engine v1.1 / Alpha 2 if it tests out OK. If that crashes too hard maybe we can play something else for the fun of it, I still have an ARK server laying around that could use a reset and some fresh action.

@msteiger on the other hand has been looking at maybe doing a more intense design session Monday, maybe using https://appear.in/terasology - this would be brainstorming, technical details, and so on. That might just be for the main students and related primary mentors, particularly as appear.in is limited to 8 people max (voice/video chat), at least for the free version. Probably ping him on Slack if interested

If anybody needs a Slack invite just let one of us know.

Bonus: Like dark themes? I came across https://github.com/StylishThemes/GitHub-Dark and it includes support for ZenHub!

 

[Cervator]

Bump! We officially went live with the first coding period less than a week ago, and progress is being made

All three official projects have forum threads, GitHub issues, and even a ZenHub board if I recall correctly. More to organize and stay on top of. For now I noticed three neat bits of new stuff this week:

@tdgunes has an initial DAG PR on GitHub with some code that does stuff. What kind of stuff? Stuffy stuff? I don't know! But it looks like 3d wizardry of the highest order for sure. Or at least some structure to play with later

@Cpt. Crispy Crunchy is working on the Dynamic Cities module, dealing with the world and landscape. Caught an interesting screenie on Slack of things to come

Finally @Rostyslav Zatserkovnyi put together a nice tree view for our UI, with a little demo video

Beyond the official projects we also have a few students looking at unofficial projects, mostly surrounding combat and related topics like anatomy. That all seems to be coming along quite nicely as well.

More to come!

 

[Viveret]

Unfortunately I won't be able to participate and work on the particle system as I landed an internship that's 40 hrs a week the entire summer. I'll be sure to stick around and check things out however, and perhaps after the internship I can work a little bit on particle systems or anything else.

 

[Cervator]

Unfortunately I won't be able to participate and work on the particle system as I landed an internship that's 40 hrs a week the entire summer. I'll be sure to stick around and check things out however, and perhaps after the internship I can work a little bit on particle systems or anything else.

That's not unfortunate at all! That's awesome. Congratulations

Keep in touch and let us know if you get bored sometime!