[brlcad-devel] GCI preparations

[Christopher Sean Morrison]

I’ve written up some of the main tasks needing attention before GCI begins in a couple weeks:

http://brlcad.org/wiki/Google_Code_In/2016/TODO <http://brlcad.org/wiki/Google_Code_In/2016/TODO>

Anyone care to lead any of those tasks, write your name onto the wiki entry. We have just a little under two weeks before GCI begins to get set up properly. Fortunately, we have something like 20+ mentors this year! With a little help from everyone, we should be able to get some really useful work done in a really short amount of time.

We need to get started on topic brainstorming. Ideas welcome. General rule of thumb for this year is that coding tasks need to constitute about 1-2 hours of work for a core dev. Other tasks should be in the 2-4 hour ballpark (for an experienced contributor). There is a special “beginner task” category to help participants where you don’t assume anything, and the tasks help them do the non-beginner tasks (e.g., download, compile, and install BRL-CAD is a canonical beginner task). We can start the brainstorming here and then migrate priority topics into a spreadsheet. Main trust areas should align with a specific area of work as previously mentioned: http://svn.code.sf.net/p/brlcad/code/brlcad/trunk/doc/STRATEGY <http://svn.code.sf.net/p/brlcad/code/brlcad/trunk/doc/STRATEGY>

Thanks Costa for OpenCL-related major tasks set (see #6 in the TODO). That’s a great idea and aligns with our STRATEGY.

Cheers!
Sean

[Oleksandr Dubenko]

I made a GCI viewer to showcase completed GCI tasks. It can be nice
addition to yash's gallery. It's not categorized but provides information
about all tasks.
http://gciviewer.surge.sh/
I'm going to tweak it a little later, but it's already functional.

Post by Christopher Sean Morrison
I’ve written up some of the main tasks needing attention before GCI begins
http://brlcad.org/wiki/Google_Code_In/2016/TODO
Anyone care to lead any of those tasks, write your name onto the wiki
entry. We have just a little under two weeks before GCI begins to get set
up properly. Fortunately, we have something like 20+ mentors this year!
With a little help from everyone, we should be able to get some really
useful work done in a really short amount of time.
We need to get started on topic brainstorming. Ideas welcome. General
rule of thumb for this year is that coding tasks need to constitute about
1-2 hours of work for a core dev. Other tasks should be in the 2-4 hour
ballpark (for an experienced contributor). There is a special “beginner
task” category to help participants where you don’t assume anything, and
the tasks help them do the non-beginner tasks (e.g., download, compile, and
install BRL-CAD is a canonical beginner task). We can start the
brainstorming here and then migrate priority topics into a spreadsheet.
Main trust areas should align with a specific area of work as previously
mentioned: http://svn.code.sf.net/p/brlcad/code/brlcad/trunk/doc/STRATEGY
Thanks Costa for OpenCL-related major tasks set (see #6 in the TODO).
That’s a great idea and aligns with our STRATEGY.
Cheers!
Sean
------------------------------------------------------------------------------
_______________________________________________
BRL-CAD Developer mailing list
https://lists.sourceforge.net/lists/listinfo/brlcad-devel

[Christopher Sean Morrison]

I made a GCI viewer to showcase completed GCI tasks. It can be nice addition to yash's gallery. It's not categorized but provides information about all tasks.
http://gciviewer.surge.sh/ <http://gciviewer.surge.sh/>
I'm going to tweak it a little later, but it's already functional.

Wow, this could be a nice front end for managing outstanding tasks. What tweaks do you have planned? We could use this for processing if you’re willing to keep working on it. Useful for browsing them regardless.

We could use a system for tracking/changing their state and “what’s next”. I was planning manually, but something like this could help. Also anticipate we’ll get 1-2 hundred more tasks this year.

Side note, idea isn’t to keep completed GCI tasks around. It’s to “process them”, which is leveraging their work and/or scheduling a subsequent action. Most require some follow-up effort or we just haven’t gotten to integrating the work yet.

Takes a lot of effort to leverage them, so much that a system for managing their integration status became necessary. Now they’re at least in a nice single catalog list (thanks to Peter!). Next step is to organize them into groups and set up workflows to process them.

For example, the various design tasks need to be grouped together. Among all the t-shirt designs, someone needs to look at each one to determine if any are something we want to use (directly or indirectly) and then actually work on getting t-shirts made. All the concepts then need to go into a gallery so we have visibility into what was considered and so any future designs can be made as good or better.

Some will turn into new (incomplete) tasks to use the work, for future GCI or motivated dev. Some can be integrated immediately. Some will get reviewed and discarded.

Cheers!
Sean

[Oleksandr Dubenko]

Yes, this can definetly grow into task managing platform. Working on it.

On Thu, Nov 17, 2016 at 5:06 AM Christopher Sean Morrison <***@mac.com>
wrote:

On Nov 16, 2016, at 3:39 PM, Oleksandr Dubenko <***@gmail.com> wrote:

I made a GCI viewer to showcase completed GCI tasks. It can be nice
addition to yash's gallery. It's not categorized but provides information
about all tasks.
http://gciviewer.surge.sh/
I'm going to tweak it a little later, but it's already functional.


Wow, this could be a nice front end for managing outstanding tasks. What
tweaks do you have planned? We could use this for processing if you’re
willing to keep working on it. Useful for browsing them regardless.

We could use a system for tracking/changing their state and “what’s next”.
I was planning manually, but something like this could help. Also
anticipate we’ll get 1-2 hundred more tasks this year.

Side note, idea isn’t to keep completed GCI tasks around. It’s to “process
them”, which is leveraging their work and/or scheduling a subsequent
action. Most require some follow-up effort or we just haven’t gotten to
integrating the work yet.

Takes a lot of effort to leverage them, so much that a system for managing
their integration status became necessary. Now they’re at least in a nice
single catalog list (thanks to Peter!). Next step is to organize them into
groups and set up workflows to process them.

For example, the various design tasks need to be grouped together. Among
all the t-shirt designs, someone needs to look at each one to determine if
any are something we want to use (directly or indirectly) and then actually
work on getting t-shirts made. All the concepts then need to go into a
gallery so we have visibility into what was considered and so any future
designs can be made as good or better.

Some will turn into new (incomplete) tasks to use the work, for future GCI
or motivated dev. Some can be integrated immediately. Some will get
reviewed and discarded.

Cheers!
Sean


------------------------------------------------------------------------------