Fri 31 Oct 2014 6:29PM

Realistic Feature Proposals

PC Pablo Cúbico Public Seen by 105


What if we could have any "can I haz" feature proposal, just as an idea, and then that idea could evolve into a project?

I've been using HackDash, an open platform from Hacks/Hackers Buenos Aires, which is a dashboard that allows you to post ideas, so other people can join, and then discuss somewhere else (IRC, etc). If some are for real, those ideas can grow into projects and even have github repos associated.

I've found it is pretty engaging, some "can I haz" ideas suddenly get a few people on board and turn into realities. And the rest are simply ignored.

Check this sample HackDash I've made.



Pablo Cúbico Fri 31 Oct 2014 6:30PM

Just found out you can discuss the ideas right there.


Shmerl Fri 31 Oct 2014 6:42PM

Well, Loomio is used for that purpose already to some degree.


Pablo Cúbico Fri 31 Oct 2014 7:07PM

I think it's a bit closed and flooded, I've found that one, being a dashboard, is easier to externalize (sharing, etc) and get a better visual "map" of the features proposed, and see who liked, joined or proposed each one at a glance.


goob Fri 31 Oct 2014 7:12PM

Well, I think it would be great if Loomio wasn't used for 'can I haz' feature requests, so if you want to set up a new platform for those, that's fine with me - as long as there's no suggestion that such feature requests will be implemented.


Pablo Cúbico Sat 1 Nov 2014 1:31AM

Well... if it's going to end in nothing (as no projection of anything getting implemented), then there's no point.

However, I think there should be a way to distinguish realistic from unrealistic future requests, besides endless discussions. The dash enables you to evolve an idea into shorter stages, as "brainstorming", "researching"; etc. In that sense, I feel that feature requests should follow some path, maybe publishing enhancing the guidelines for making realistic feature requests could do some good instead of having another platform.

Maybe not for "realistic" feature requests, but a guide to more "complete" feature requests, trying to answer some specific questions: how should it be implemented? what's the impact on performance? what would the UX be? what's the impact on infrastructure? etc.


Pablo Cúbico Sat 1 Nov 2014 1:31AM


Jason Robinson Sat 1 Nov 2014 8:28AM

I think the point Goob was making was that even if everybody wants something, someone still has to write the code. The fact that everyone wants something and it isn't done doesn't mean that it will not be implemented, it just means no one has written the code for it.

We're community devs only, no one works on this full time :)

Personally I think it would be awesome to filter "I want" type of feature requests out of Loomio, so feel free to set it up! :)


Pablo Cúbico Tue 4 Nov 2014 6:14AM

I think as a primary filter we can propose some kind of structure for creating more complete feature proposals, guiding people with some questions as:

  • What's the core functionality of this feature?
  • Can you line up the basic architecture?
  • What would be it's impact on performance?
  • What are any possible security issues?
  • How does it align with Diaspora's purpose/mission/philosophy, etc?
  • Can you estimate (at least, roughly) a timeframe for getting this done?
  • Think as a recruiter: do you have someone on board to implement this? What skills are needed?

That's a basic list, I think someone with more experience hacking on Diaspora's codebase could come up with some better or more specific questions.