[GEM Development] DayDream (was NextGem..)

[Ben A L Jemmett]

> I'm not in a huff at all.  I would describe my attitude as playful.

Oh, in that case I apologise.  It came across (to me) as anything but
playful.

> However, I am sensing some hostility here, and I'm not sure why.  After
all,
> I'm trying to work out ways to (a) continue to develop and promote GEM,
and
> (b) to begin the implimentation of other technological frameworks.

But a) releasing a Unix distribution which looks like GEM and can run GEM
apps using existing GEM under DOSemu isn't really developing and promoting
GEM as much as splitting that project in two (see below), and b) an entirely
new distribution seems rather like overkill if the aim is to implement a new
'technological framework' for GEM.

> Yes, I am planning on skinning KDE for a specific purpose.  As I was
saying
> to Liam a while back, I want to be able to use Unix in an easy way.

As far as I can see, that has nothing to do with GEM and as such really
isn't appropriate for this mailing list.  As I said, I don't mind the odd
off-topic discussion, but it sounds to me like this is the main thrust of
your new project -- in that case, it's likely to become more than the odd
discussion.  It's obviously a laudable aim, but I'm sure there are better
places to discuss it.

> Yes, I am planning to see about running GEM transparently in Unix so
people
> can run GEM apps.
> Yes, I am interested in seeing John's work, which would suggest we could
> even get Unix-native GEM.  That would be a wonder.

Both of these are entirely on-topic, as I said, and I'd be very interested
in seeing how that works out.  A few minutes' thought brings up a few ideas
for the first, which aren't very elegant but would work...  However, I do
have misgivings about making that a main selling point of an easy-to-use
distribution; for one, it's a tiny bit of functionality.  It also seems to
be an attempt to lump together two entirely disjoint sets of users -- those
interested in running GEM apps within Unix, and those interested in an
easy-to-use Unix distribution.

That's also why I don't think NextGEM is a good moniker for your new
project.  Those who want easy-to-use Unix are not likely to know what GEM
refers to -- compare this to Lindows, which is a lot more obvious to the
uninitiated.  Everyone's heard of Windows, after all!  On the other hand,
those who want to run GEM on Unix may well not want to install a new Unix
distribution to take advantage of that ability.

I'd suggest instead you split the project in two.  The bulk of it is the
easy-to-use Unix system, and if you do that right there'll be a lot of
interest in it as a standalone system.  The GEM-in-DOSemu half (or any
Unix-native GEM system) would surely be better as another standalone
product -- for a first release you could have your OpenGEM 3 (the version
you'll bundle FreeDOS with) preconfigured within DOSemu, with wrapper
scripts to start GEM or a GEM application from Unix.  Call it "OpenGEM 3 for
Unix" or something, and offer that separately from the easy-to-use Unix
system.  Offer source and binary tarballs and system-specific packages, and
you'll get more interest from GEM users I'll bet.

> I don't think that anything I said could be described as a huff.

What sounded like a huff to me was:

: So, Shane transitioning to Unix for his next project is not so far from
: OpenGEM as it seems.  Ahem.  Taps foot.  Lame indeed.

It's the last five words that did it.  But I've apologised if I read that
wrong...

> Yes, some of what I am doing is far removed from GEM, some of it is
> not.  On the whole however I am trying to do something that will allow GEM
> development to progress in a new direction.

Well, that may be how you see it, but from what you've said NextGEM won't be
a development of GEM; it sounds like it'll be a development of a Unix system
with GEM stuffed in in case anyone wants to use it.  *shrug*  Porting GEM to
Unix again doesn't seem to require a new Unix distribution, and in any case
I doubt it'll attract many developers.  More than GEM on DOS though, I
suppose...

> Ben, there is an implicit threat in your reply that you will throw me out
of
> the list because of what I am talking about.

Well, that implicit threat wasn't meant; the threat was more that I can
elect to moderate any/all posts to the list if the discussion moves towards
skinning KDE or making Unix easier to use for more than brief conversations,
because (as I said above) that's not what this list is about.

> I cannot understand this.  I
> don't think anyone could deny that I am after working hard on GEM, and I
> certainly have produced a lot of GEM work.  Indeed, in the last year I'm
the
> only person who has released new GEM material.  I've been working hard to
> get GEM and FreeDOS connected both in actuality and in people's minds.

Of course I'm not denying that you've done a lot of work on GEM.  You're not
the *only* person to have worked on GEM in the last year though -- I've
spent the best part of three weeks this summer on trying to get GEM/XM
building, for instance, and one of my aims for the coming months is to
develop a 'common controls' library for GEM apps (things like combo
boxes/listboxes, maybe tree views and whatever -- this is something I'd be
writing if I target GEM for my final year project and would be released
under GPL).

> I am disappointed with what you said.  I thought our work was not about
> tolerating or annoying people, I though it was about trying to make
software.
> That's what I'm doing.

Ah, well, you're slightly off there.  This list is about GEM development
(both the system and software for it).  It's not about software in general.
I think I've already made the distinction above; work on getting GEM to run
on Unix is entirely on topic, but the development of new Unix systems that
just happen to include GEM isn't.

Regards,
Ben A L Jemmett.
(http://web.ukonline.co.uk/ben.jemmett/, http://www.deltasoft.com/)