An idea on how to manage project creation | General | Forum

Okay, so here are some ideas on how we can handle the creation/organisation of projects in future.  It was inspired by Rizwan's recent post here.

Basically, the idea is that projects come in three "flavours", which for the purposes of explaining my ideas I'll call "green", "yellow" and "red" – we can change the names to whatever we like if we adopt this system.

A project being green means: This is a real, official CSTART project.  It has its own section on the forums, the treasurer is allowed to give money to it, etc.

A project being yellow means: We would really like to do this in the future, so much so that we are making design decisions for green projects in such a way as to make sure the green technology leads up to these projects.  The only reason this isn't green right now is because we don't have enough manpower/experience/money.  This project will be green in the future unless something drastically changes.

A project being red means: Someone has suggested this project and we might do it one day depending on a variety of factors, but it's not really on the tables yet.  We're not really basing any design decisions on these projects.  There is no guarantee that this project will ever happen.

Now, as for how projects attain their colours.  I think we should create a web app (this shouldn't be too hard and I am sure we could find someone who wanted to do it).  Basically the web app would let *anybody* suggest a project, by giving it a name and a 250 word or less summary.  Obviously there should be a CAPTCHA to prevent spam.  All projects created in this way are red projects by default.  There should be an RSS feed for new red projects.

Anybody who is a member of CSTART (i.e. who has gone through the formal joining process that we will need to have once we incorporate, which may include paying dues or making a donation) can view all red projects and can give each one a "Yes" vote or "No" vote.  A red project becomes yellow as soon as (i) more than some threshold number of users have voted on it and (ii) more than some threshold percentage of those votes have been "Yes".  Red projects which do not turn yellow after a certain number of months are expired and do not show up in the listings.

The process of turning from yellow to green is a little more difficult.  Basically the bylaws should put a restriction on the maximum number of green projects allowed at any one time (2 or 3).  Each time a green project is completed and an empty space appears in the roster of green projects, a big vote is called, that all formal members of CSTART may participate in.  Basically, one project from the current list of yellow projects must be chosen to be upgraded to green, and the one which gets the most votes wins.

Now, an optional additional level of restrictions that we may like to impose: Each new project that is submitted (to become a red project) must be either "underneath" a current green project (i.e. must be a logical follow on from that project which involves mainly extending existing technology, not creating new technology from scratch.  The suborbital experiment rocket project we have discussed would be underneath OHKLA, for instance) or "beside" a current green project (i.e. must be such that considerable technology exchange can happen between the two projects.  The Project X satellite project that we discussed would be beside CLLARE, for instance).  This gives the projects a kind of tree structure.  If we went with this arrangement, the process of turning from yellow to green could be refined so that the yellow projects which are voted on to replace a finished green project must be projects underneath the completed project.

The biggest advantage of this system is that it is extremely open and also democratic.  Literally anybody can propose a new project, and all the members together decide which ones we actually want to implement.  Also, it lets us show a clear sense of future direction via the yellow projects without spreading ourselves too thin.

The biggest disadvantage I can think of is that if people aren't discriminating enough we may see too many yellow projects turning up.  Perhaps we can deal with this by putting a limit on the number of projects which can be yellow at any one time.

If people like this system, the image below shows how we could initiate it (since we currently have too few people to do meaningful votes and since currently the core 5 are effectively "benevolent dictators, I see no problem with us initiating the system in a certain state and then letting it progress from there according to the voting rules).

Basically, CLLARE and OHKLA are the only green projects (suggesting that 2 is the green project limit).

Our suborbital experiment rocket project (the follow on from OHKLA) is yellow and is underneath OHKLA.  The Project X project is also yellow and is beside CLLARE.  These are the only two yellow projects.

I've thrown in a bunch of red projects that are underneath others but if anybody really doesn't like any of them I won't be upset about getting rid of them.

I welcome any and all feedback on this idea, including proposed changes to the system.

Sure, it seems like a sensible plan to me to have projects move through a series of stages once they turn green.  The first stage should be a feasibility analysis, where we crunch the numbers (making approximations where we have to) and look for plausible off-the-shelf hardware and basically make sure it can be done.  If this stage passes we can go into a stage where we create a detailed list of requirements, figuring out what all the required subsystems are, etc. and roughly figuring out how things will work.  Then a "real engineering" phase where we start designing/building everything we decided we would need.  Maybe not these exact stages but something like this should work fine.