It has been a while since I have been on an SSI email list or actively
thinking about SSI other than remembering to pay dues every year. For
the last several years I have been creating web services as at home
projects and when I rereading some of the SSI site I realized that
something that I always wanted to do may be possible now, "Actually
participating in some other way then just paying dues". So here is the
body of an email message that I would like to distribute to high
visibility technocrats. Think of this one going to somebody like Vice
President Gore. You can get the concept from it. I would like to start
laying the foundation for this site soonest and would like everybody's
opinion.

Mitchell

In a message dated 1/7/01 7:22:05 AM Eastern Standard Time,
mejames@... writes:

Hi, I think I can save you a lot of time:

I looked at the site. It looks a lot like egroups.com: a discussion
(email list) area, a place to store documents, etc.

> I looked at the site. It looks a lot like egroups.com: a
discussion
> (email list) area, a place to store documents, etc.
>
> The site that we need requires the ability to do the standard
> engineering process of
> - creating requirements,
> - tracking requirements to design,
> - tracking derived requirements to lower and lower levels of design,
> - tracking design to implementation,
> - tracking design to test,
> - tracking test results back to requirement and design changes,
> - while keeping the "whys" linked to the appropriate design.
>
> This is called "Requirements Management" in industry. See
> http://www.incose.org/tools/reqsmgmt.html for a brief introduction
to
> the subject.
>
> A program commonly used for this in the aerospace industry is DOORS
> (http://www2.telelogic.com/doors/index.cfm) which is a product that
I
> have used extensively as a project lead for an American aerospace
> company. It costs about $10,000 US a seat and does not really do
the
> job. It is sort of a raw tool like a spreadsheet that your process
> has to be adapted to including limitations. At the most basic
level
> these programs are database servers with user front ends. Sounds a
> lot like a dhtml web site.
>
> Sites like egroups.com, sourceforge.com, and intranets.com may be
used
> by development teams but it is impossible to do requirements
> management at these sites and requirements management is the key to
> succesfully completeing a project.
>
> Mitchell James
>
> > Hi, I think I can save you a lot of time:
> >
> > http://spacestudiesinst.intranets.com/login.asp
> >
> > :) ed

Nope, that does not exist either....

> Sites like egroups.com, sourceforge.com, and intranets.com may be
used
> by development teams but it is impossible to do requirements
> management at these sites and requirements management is the key to
> succesfully completeing a project.
>

It is always hard to separate personal likes and dislikes from decisions and observations. In this case I do have a strong desire to solve a perceived problem that not all of us perceive as a problem. I work with engineers who consider vi the only tool needed for programing and go out of their way to avoid simple productivity improvement tools such as Delphi and MS Visual Studio. For them productivity improvement is not an issue.
Many great engineering projects were done long before we all became addicted to word processors, spreadsheets, and scheduling software. So it is certainly true that the right people can always get the job done with whatever rock they have at hand.
However, the whole concept behind modern management approaches is to be able to do the job when you don't have the perfect people on it. This means procedures, practices, and standards that are used across the board. What you are describing would be a SEI level 1 organization http://www.sei.cmu.edu/>.
I think that we should try for a little bit more.
Mitchell James

All I'm worried about is that the tools being used will become the
reason for the project existence, instead iof the project goals. I
have seen many projects, open source, and commercial where people
spend so much time acquiring, updating, and customizing the tools that
they never get near to the project goals.

Good point!
Reminds me of an IR&D project I had the opportunity to shut down ten years ago. This IR&D group had somehow talked management into funding a 10 manyear effort to prepare for the next contract based on the latest OOD tools. They didn't want to let previous experience get in the way of a resume building learning experience so they started with a clean slate and empty minds using a tool that was picked because it had a nice visual interface. When I was put on the project to find out what was going on they had compiled a three inch thick volume of circle and line drawings representing their work. The first question I asked after some study was, "Now what do you do with it?". The answer was that this was design documentation that would be placed on the shelf and left there because there was no way to get from their design to code. When I noticed that each object breakdown starting with the very top had no rhyme or reason, I asked what rules they used to define the relationship between objects the answer was that the only rule was there could not be more than nine objects or less than five. Why? Because that was a tool limitation on the maximum number of nodes in a drawing. And their response to anybody questioning the whys of their design was that they had spent too much time argueing over it to change it and that we obviously didn't understand cool OOD. Luckily we lost the bid for the contract this project was suppposed to help and the IR&D group had all found other companies to work for within several months. My review of the whole project was that the IR&D group wanted to play with an open ended tool without being encumbered by any extranous rules. The fact that the project could never have been finished using their methodology didn't matter.
And the same has you, I have seen and been forced to use tools upon which millions of dollars had been spent and being spent that solved a problem that didn't exist in the first place and that there certainly were much cheaper and better alternatives for by the time the tool was usable years later.
On the other hand, I have used tools both simple and complex that provided capabilities or solved problems that would have just been too difficult to do otherwise.
Hard to tell if InnerTransit.org is going to be useful or not at this time. It contains the best aspects of all the tools that I have ever used. The only way to tell if this is the tool for the job is to use it. InnerTransit.org is almost ready. I inserted the last major section yesterday, CVS source file control. I need to finish off source file control and then clean-up and implement the help function. It should be ready by the end of the month. Then we can start the projects that I am really interested in.
What projects is everybody interested in starting? I would like to start on a moon based manufacturing cell capable of partial self replication.
Mitchell James
All I'm worried about is that the tools being used will become the
reason for the project existence, instead iof the project goals. I
have seen many projects, open source, and commercial where people
spend so much time acquiring, updating, and customizing the tools that
they never get near to the project goals.
Bill

>
> What projects is everybody interested in starting? I would like to
> start on a moon based manufacturing cell capable of partial self
> replication.
>

No I haven't. Is there a URL.
Mitchell James
Have you seen the Rice University proposal for a robot which makes solar
cells from regolith ?

http://www.spacedaily.com/news/robot-00l.html

Interesting but not much information. Kind of hard to envision submicron positioning accuracy in a vehicle rolling accross rough grade. This one is more interesting
http://www.ennex.com/publish/199608-MB-SpaceFab.sht
Just noticed the stuff you had at the bottom. Will have to look at that later.
Mitchell James
http://www.spacedaily.com/news/robot-00l.html