From: Jim Peters (jim_at_uazu.net)
Date: 2002-03-08 12:37:44
Doug Sutherland wrote:
> > THIS IS A discussion forum so WHY do you what to stop
> > the discussion and sharing of valuable information?
>
> Good question. I'm frustrated too.
I have an idea. Why don't we seriously create a [talkcheapeeg] group
to have all these discussions in ? Then members of that group can
come back to the [buildcheapeeg] list with a finished proposal and/or
code once they've thrashed it all out. (Or maybe I'm just trying to
push the problem elsewhere, where I don't have to read it).
The problem for me is that if someone comes to the group saying "Tell
me all about this subject I know nothing about" or "We *have* to
discuss this very important aspect of the project which I have done no
research on", then everyone has to spend a lot of time explaining
things. Even worse, if someone posts to the list proposing a
half-baked and badly-researched solution, everyone has to jump in and
correct it for fear of anyone actually agreeing to it. This wastes a
lot of time for people who have other things to be getting on with.
It would be much better for people to do a little bit of thinking or
research first. Alternatively, [talkcheapeeg] is available for
endless thrashing out of badly thought-out proposals.
I seem to remember I came to this group with some bright idea that I
wanted everyone to agree with, and after some flame war, I decided it
would be better to go it alone, and simply prove it by doing it.
John, how does that sound as an approach ? It would mean actually
doing some coding, though.
I appreciate the part that Jim-M has been playing here, because
although he explodes in flame sometimes, he does seem to consistently
encourage people to get on with what they are good at, and keep on
working towards a solution. He also fails to believe in anything he
can't see for himself, which is good as all the design in the world is
worth nothing if it never reaches hard code.
> I want more than that. I want to build a house, and a flexible and
> solid one. Forget elegant and best, as Rob Sachs says we need a
> solid foundation. I'll go back to what I said here:
>
> http://groups.yahoo.com/group/buildcheapeeg/message/2298
>
> A program is not a system. There is some value to just
> starting coding, but it almost never evolves into a
> good architecture. Spoken from "real world" like Jim
> always mentions. In the "real world" of software you
> don't go very far withot anwsering the above questions
> first. It's importance increases in magnitudes as you
> add more developers. Yadda Yadda Yadda. Perhaps talk
> is cheap but so is "just coding". BTDT. If we want a
> system that compares with some of the commercial ones,
> we need to do some planning. If planning will be
> discouranged, then I don't think I will waste my time.
In my opinion, we are talking about two extremes. The pure coding
version, and the pure design version. In reality I would hope we
would use a combination.
Coming from a pure coding angle requires rewrite after rewrite to get
a nice system. However, if you are willing to see code as a liquid
malleable substance, this might be do-able. If you hate writing or
modifying code, you are obviously not going to be happy with this
approach. However, it does have the advantage that you find out
straight away if you had some misconception of the dynamics of the
system.
Coming from a pure design angle means knowing everything before you
start coding, and coding only once. This is perfect for those who
don't like to code, or who hope someone else will do the coding. This
might be do-able for projects that have been done 100 times before and
have been well documented and studied, like writing a database system
perhaps, but if the terrain is not well known, then working from pure
design seems to me to be a catastrophe in the making.
I personally don't trust any designer who is not willing to test his
ideas in code, or back them up with decent research or experience. It
really worries me when John says that he doesn't intend to code all
this. Who does he think is going to code it, then, if he won't ?
We are all volunteers here, and no-one is getting paid to be anyone
else's code monkey. If it doesn't fit with someone's personal
priorities, it simply won't happen, no matter how perfect the design.
Do you think we can get back from the two extremes ? Designers must
be willing to test their ideas in code (or else through research), and
coders must be willing to structure their code to enable
'house-building'.
My personal approach just recently is to ignore any nonsense proposals
arriving on the list, not even to correct glaring inadequacies or
omissions, on the basis that it is simply more efficient to code it
all myself than to try to adapt the 'list' solution into something
usable. There is no point wasting energy persuading someone of
something that they don't want to believe when you know for certain
they are wrong. Let them live in their illusory world, and hope they
leave you alone to get on with your work. This is the "walled-off"
developer scenario.
I really want to make progress with this project -- it seems like a
very worthwhile goal to me. If I wasn't so interested in reaching the
goal, I might well have drifted off by now. I think we have a good
team coming together here, with experience in lots of areas, all from
quite different backgrounds. With some respect for one another's
abilities, and acceptance of our own individual lack of ability in
certain areas, I really hope we can work out these difficulties.
Jim
-- Jim Peters (_)/=\~/_(_) jim_at_uazu.net (_) /=\ ~/_ (_) Uazú (_) /=\ ~/_ (_) http:// B'ham, UK (_) ____ /=\ ____ ~/_ ____ (_) uazu.net
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:40 BST