Enginering Process planning | General | Forum

I am largely okay with what we are doing here.

As for the hosting constraints, anything we do in php and mysql shouldn't have any problems.

For the personal blogs/tweets, I recommended elgg.org in another thread.

Post edited 4:36 pm – March 3, 2010 by Luke Maurits

Rizwan said:

That wont be hosted on their servers, it will be hosted on ours, under cstart.org/intranet or whatever name we choose. And it is open source too.

Ah, that's good.  Sorry, I just assumed without reason that elgg.org was a service rather than a product.  I'll give it a quick look now before I hhead off to University.

EDIT: Wow, elgg looks absolutely perfect.

Denis, a quick question for you re: requirements management.  Does most requirement management software, as far as you know, allow one to keep track of requirements implemented as numerical restrictions?  What I mean by this is can we, say, set a requirement that the complete lunar landing hardware configuration have a mass not exceeding 1000 kg, and have people be able to individually store the mass of various components they work on, with the system keeping a running total and raising a red flag if someone changes the mass of something and it pushes us over the limit?

First of all, I have ammended the Engineering Process wiki page to have a section on Software Engineering, since we are going to have to develop some extremely rigorous protocols for that part of our work.  The software that runs onboard spacecraft will need to be perfect, it cannot be developed in the usual ad hoc manner for open source software.  We will need planning, unit tests, etc. from the very beginning.

Secondly, I like the idea of advertising for help (and appreciate the How I Met Your Mother reference :), but I think we should wait a bit and do a big burst of advertising.  In particular I think that, in addition to advertising for web devs as outlined above, we should solicit help from (i) software engineering experts to help with the above, (ii) 3D animation people to get some higher quality conceptual renders of our hardware and also some videos.  Something close in quality to this NASA video:

We might as well make a big list of very precise, targeted people we need help from and post it all in one big batch.  What do people think?

Post edited 3:40 am – March 6, 2010 by Luke Maurits

Good to hear from you again,  brmj, looking forward to you be being fully back online, and good luck sourcing a new laptop!

As for the Engineering Process: inspired by the recent work on a new CLLARE plan, and the idea that it's worthwhile to develop a few proposals in moderate detail simultaneously and then choose the most promising, I propose the following: for large projects (like CLLARE), there is an initial proposal selection phase, in which the process described so far (with trees and decision voting, etc.) isn't started yet, and instead different teams (I use "teams" here in quite an informal sense: disorganised subsets of people) work to prepare proposal documents for different ideas: something kind of like the CLLARE Overview Document, but perhaps not as detailed (although there would need to be some degree of actual number work going on).  After a predefined period (3 months?  6 months?), there is a large vote on which of the proposals (I guess there might be around 5 or so) seems the most promising.  The most popular proposal then becomes the official one, and then we activate the formal process, with trees and voting, etc., to refine that one proposal.

Under this scheme, the "current" CLLARE idea (conical CM, open cab LL, etc) and my new idea (spherical CM which also acts as a lander) would both be "competing" proposals (but not fiercely competing, i.e. people could (and I suspect would) work on both), and other groups could come along and start their own proposals.  Sometime around July or something we could have a vote and pick one.  Before this point, things would be fairly informal.  The different teams could present whatever info they wanted in their proposals to whatever level of detail they feel like doing: those who slack off on and analysis and rely mostly on handwaving will simply not get many votes from people because their case will not be very convincing.

If people like this idea, we would have to come up with a few basic requirements to constrain the proposals (e.g. crew of one, must fit on Falcon 9) and a few clear judging criteria (simplicity, low cost), but otherwise we would let people come up with whatever they can imagine that fits the bill.  Possibly we could even have a small cash prize?  What do people think?

EDIT: This sort of environment, with lots of separate groups, working on a wide range of things, would be a great use for our elgg (or similar) system.