Status Update

41 views
Skip to first unread message

Alan Moore

unread,
Jul 19, 2015, 7:16:02 PM7/19/15
to coops...@googlegroups.com
All,

Here is the most recent status and news update!

My primary collaborator, Bill G., is engaged in his new startup (check him out at http://www.brightcanopy.com) and another dev is coming online shortly.
  1. Membership Bylaws: Obtained samples from similar cooperatives and some close variants from existing organisations. Needs consolidation and drafting.
  2. SW License status: Prototype terms awaiting presentation to and feedback from membership. Blocked by Web site/App.
  3. Web site status: The placeholder site has some pending changes to better explain our principles, goals and strategies for getting there. Eval of Om Next in process, including integration with Clara and Server.
  4. Electron App: Prototype started but having issues with React not being defined.
  5. Clara rule engine: New/better support for ClojureScript completed. Clarity Electron App prototype started.
  6. Database: Schema for Datomic started. Work in progress. Datomic -> Prismatic Schema translator needs to be written.
  7. Servers/Clusters Onyx, Pedestal and others - evaluations under way.
  8. DevOps: Evaluating Ansible and Puppet.
  9. Legal: Need decision on structure and status (not-for-profit vs LLC vs B-Corp vs ?)
  10. Finance: None, I'm buying ;-)
  11. Research: TLA+ vs Alloy, trust networks, security.
  12. Project ideas: Datomic C++ clone, Embedded systems projects (config, msg), others.
Lots to do but I've been making good progress on many fronts! Right now there isn't a membership fee so I'm paying for hosting and other expenses. At some point we will need to get the charter solidified and incorporation of some sort so that the existing legal/financial frameworks can recognize us and we can accept membership fees. That will allow us to start a Beta with our rough prototype and get feedback.

Even though he is otherwise preoccupied, Bill G. always has great perspectives on our challenges and opportunities. Now and again he scans HN for related posts and forwards them my way, thanks Bill!

Take care.

Alan
-- 
"Whatever you can do, or dream you can do, begin it. Boldness has genius, power, and magic in it. Begin it now." - Goethe

Daniel Compton

unread,
Jul 19, 2015, 7:43:25 PM7/19/15
to coops...@googlegroups.com
Hi Alan

I'm a bit confused, how are Datomic, Pedestal, Om.Next, Onyx, Electron, TLA+ e.t.c. related to the work you're doing with Coopsource?

--
You received this message because you are subscribed to the Google Groups "Co-op Source" group.
To unsubscribe from this group and stop receiving emails from it, send an email to coopsource+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
--
Daniel

Alan Moore

unread,
Jul 19, 2015, 8:32:57 PM7/19/15
to coops...@googlegroups.com
Just evaluating them for the site, membership, project meta data, etc.

These will be used for building a MVP (Platform) for hosting co-op membership and projects. Once those are working other features such as consensus building, voting, record keeping, transparent accounting, dynamic equity splits, governance and related concerns will be addressed.

Given that time is limited for everyone, I want to take the drudgery out of the process by automating as much as possible. The exact policies that drive it all will be data and rule driven so they can be adapted. I'm anticipating that not every Co-op project will want the same policies, etc. so these features need to be flexible (dare I say plugins?)

There are lots of domain specific projects, products and services we can build together if the friction of organizing and decision making is reduced... Things like Slack only go so far.

TLA+ could be helpful in the long term for specifying and generating systems. In the short term those tools can help ensure our own internal systems are robust.

I'm sorry if my comments seem out of context or confusing. Let me know how this meets (or not) your expectations, ideas. This is a vast and varied subject and I've been thinking about it for some 15 years now so I'm apt to talk with others without setting context first.

I will push my edits to the site tonight so that will help in outlining my current thinking. I recognize that this might not match what others envision. Please share your thoughts, experience to help us come up with a common understanding.

Thanks!

Alan
-- 
"Whatever you can do, or dream you can do, begin it. Boldness has genius, power, and magic in it. Begin it now." - Goethe

Daniel Compton

unread,
Jul 20, 2015, 2:31:40 AM7/20/15
to coops...@googlegroups.com
To be clear, you're looking at building an app for coopsource with all those technologies, not evaluating them as projects that could be coop supported?

Alan Moore

unread,
Jul 20, 2015, 4:14:36 AM7/20/15
to coops...@googlegroups.com
Yes, exactly. I'm building a core set of functionality for the Co-op Source Foundation's site that can be used to bootstrap the remaining features/MVP as sub-projects. I expect this to take something on the order of four months with several (2-3) people working part-time, e.g. 5-10 hours/week. I've been retooling my development skills to include Clojure and ClojureScript and now I am productive enough that my prototypes are mostly working. Any Clojure developer is welcome to help get this core design completed and working.

Finally, when the system is useful enough, members can join to start proposing and working on revenue generating projects. Members can use ranked choice voting and other organizing/decision making tools provided by the site to move their projects forward with help from other members. Crowdfunding provided by the site can be used to raise funds for the individual Co-op projects.

I have a day job as do most of us working on side projects so patience is a virtue. In my experience, startups (or new ventures) are typically too early to market. Assuming projects will take some time to reach revenue we can use our time to do rapid prototyping and get user feedback early to ensure we are building the right thing and actually have a market for it. Of course, this is just what I've been envisioning and I'm open to alternate suggestions.

Re: applying the license (still being defined.) My opinion is that OSS licenses works best for things like operating systems, languages, containers, editors, network stacks, security infrastructure and the like. Co-op Source seems to fit best in vertical markets or more specialized domains where the end product is not necessarily of use to a large number/types of users. The lack of visibility/interest to the OSS community and larger commercial ventures means these vertical markets are probably underserved and potentially profitable. If we can address them very efficiently, without all the trappings of a "startup", including debt, VC control and short runways we stand a chance of providing a quality product or service, at an honest price and with reasonable profits. None of us is going to become unicorn-rich but I would venture a guess that by working on this project we can produce life changing results with monetary, professional and technical advancements.

IMHO, the Co-op Source license is best applied to greenfield projects - existing projects have built-in expectations and licenses that would make its introduction problematic. In new projects, members will agree to the terms up front and this will attract like minded developers.

As Co-op members we maintain agency in the code and we own it collectively. The end users get access to the source code for systems/apps they purchase and host. They can customize, fix and enhance for their own purposes but may not resell it or compete with the Co-op project. It is not clear what to do about source code access in the case of projects that produce services. There could be two or more variants of the license to cover the major use cases. The critical common factor is that only members can use/modify the code and they retain agency in their work product and share the burden and income by producing and supporting a commercial success with very happy customers.

By organizing as a Co-op, members have access to a larger pool of talent and collectively we can build things we could not tackle alone. We should be able to scale this quite large - I could see this operating somewhat like Valve but physically distributed and in many more markets. One project I have been thinking about is an app platform similar to Steam that could be used for many different kinds of businesses verticals with "downloadable features/plugins" provided by the community, third party integrators or other end users. A marketplace for apps built by Co-op projects/members. I have dozens of ideas for products or services but only a limited time to work on them... Now, if I only had a few other developers to help build them out :-)

Thoughts, comments.

Alan
Reply all
Reply to author
Forward
0 new messages