----- Original Message -----
From: "Wolfgang Lenerz" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, March 15, 2002 6:48 AM
Subject: Re: [ql-users] Open source
> On 14 Mar 2002, at 19:59, Marcel Kilgus wrote:
>
>
> > Ah, interesting. What means "in principle"? And what is the commercial
> > status supposed to be?
>
>
> If itis Open Source, there will be no more commercial status.
> The in principle refers to the fact that Tony said he would do it if we
find a
> suitable person. He did suggest you.... He is concerned about his
> code being savaged if let around freely.
>
A very reasonable requirement by Tony I think: nothing is more aggravating
than seeing one's beautifully crafted code massacred by someone who as much
idea about programming as they do about wlking on the moon.
> We didn't really talk about the legal status of the code (yet). I can't
> see, however, how Tony could retain copyright if many people
> worked on it. I'm sure that we will work out something suitable to
> most (propably not all, see below).
>
This is wrong. Tony can retain copyright if he wishes and, provided that
the code is managed out in portions to developers, then no loss of copyright
entails. It's just that those developers would have to agree that Tony
retained copyright: in effect they are his agents while working on the code.
The essential bit of course is _managed_ - it means that a) some form of
configuration management will be necessary, b) someone will have to make
decisions about who gets what part of the code to work on usw., and c)
everyone involved in development will have to agree to a defined process for
updating the configuration.
> > Well, I suppose the amount of people actually doing something with the
> > code will be quite limited. It's no easy stuff to deal with.
>
>
> Yes, that's true. There are some, however (you, Thierry, Jochen (?)
> etc come to mind).
>
> > So I might be able to do the job if necessary.
>
> Glad you volunteered! (I fyou ahdn't, I would have...)
> > But there's a lot to talk about first...
>
> I agree. Tony told me that he had started to integrate most of the
> different versions (one for each machine on which SMSQ/E runs),
> so that there is only one source code, with, I presume, different
> modules.
>
> IUf you allow me to, I'll tell him that you volunteered (or you can tell
> him yourself), to get the source code transfer organised.
>
Like Wolfgang I think it's unreasonable to expect the "prime developer" to
also be the manager of the project. I have frequently been in such a
position as a software developer for my employer and it is in no way ideal.
Essentially what is required is for someone, who does not need to be highly technical
in the sense of "an expert in C, 68k assembler, etc.", to undertake to be the project
owner on something like SourceForge. In this way only those who are given access are
permitted to update the source tree and distribute the product as "SMS/whatever."
> Now for some more personal notes from me:
> I'm, of course quite wiling to help in any way I can, even with the
> actual coding. I do suggest, however, that the "registrar" (for want
> of a better word(, keep a pretty tight rein over the way things are
> handled (sorry Phoebus, no soundforge...in my opinion - which is
> why the 'most' and not "all" above...).. I know that this will enrage
> the proponents of totally free sources, with which you can do
> whatever you want. However, we should consider that our
> resources are limited, and we will all be better off if we share them
> in an intelligent (and that means managed) manner. That doesn't
> mean that if somebody absolutely wants some feature, this feature
> can't be parcelled out to him/her (I'm being optimistic here).
>
An alternative to SourceForge would be for someone with a suitable platform to store
the sources in, say, a webCVS tree, and to give access to the designated developers.
This, of course, requires a platform with a permanent hig speed internet connection -
I'd . Perhaps it would make sense for QUANTA to provide such a platform - obviously
dependent upon costs.
> I personally also find it very important that, if we do some
> development on this, we do it for ALL machines that run SMSQ/E. I
> was a bit worried about your earlier proposalsn because you might
> have the (totally natural!) tendency to give a priority to QPC( I'm
> ONLY speculating here, NOT accusing you of anything!) which i
> something I personally would prefer to avoir (even if QPC is what I
> use most)....
>
Indeed there is little point in doing this for onyl a sub-set of the OSs that run SMSQ
for many reasons such as funding. Also, if I were Tony Tebby, I would want to be sure
that all likely users were being suppoerted.
> As to you being the registrar, I'm at once for and against it - here's
> why: I think that you are one of the few people capable of really
> understanding what Tony has done. As such, you are, of course
> ideally suited as registrar. On the other hand, since you are one of
> those rare persons, you are one of those few who can actually be
> doing some real coding work and quite selfishly, I'd rather see you
> do that...
>
As I said earlier it isn't a good idea for the registrar (project manager, whatever)
to be a developer. No criticism of anyone, especially Marcel, here, but as a
developer who has worked in both small and large development teams it is all too easy
to follow the wrong route when you are the only developer _and_ manager of the
project. Believe me, I've been there!
My full-time job is as a developer supporting an in-house configuration management
tool for mainframe computers, and, were it not for that fact that I am curretly
undergoing some intensive hospital treatment, I would have volunteered to act as
"registrar" for several reasons, not least of which is that I would like SMSQ to be
used more widely.
> The questionis : can you manage both.?
>
> Wolfgang
>
Peter
--
Peter S Tillier [EMAIL PROTECTED]
[EMAIL PROTECTED]
Opinions expressed are my own and not necessarily
those of my employer.
