Paul Surgeon wrote:
> On Saturday 17 December 2005 13:40, Erik Hofman wrote:
>>
>> What's this, now you want releases that have missing options for some
>> OS'es??
> 
> No, that's not what Melchior implied or said at all.
> What he said is that he would prefer if we hold off with ANY 1.0 release
> until 
> FG is ready to be called version 1.0.
> He didn't mention any OS's in an exclusion list did he?
> 
>> Give me a patch and I'll commit it, until then there's little room for
>> discussion.
> 
> So in other words "f*ck all the developers, we're going to release 1.0
> whether 
> they like it or not" because you or Curt say so?!

No, that's not what *Erik* implied or said at all.  Come on, Paul.


> I thought this was a community project but I was clearly wrong.
> It would appear that we have zero say over this matter and even less
> chance of 
> getting any honest answers from either you or Curt.

This just seems way over the top to me.

Here you write:

> I would also like to know what this urgency for a 1.0 release is but all we 
> get are answers that skirt the issue or no answers at all.

and in the other post you write:

> No what would make us more happy is to know why there is such an urgency to
> have two FG releases in the space of a couple of months when up till now
> we've been releasing about once per year.

I haven't been able to be as involved as I'd like, and maybe by virtue
of missing a post here and there I've somehow missed out on the key post
that explains this -- but I don't have any sense of this "urgency to
release 1.0" at all.

I *have* seen posts to the effect that either Curt or Erik or both (can't
remember) would like to see the next release be a 1.0 release, and to that
effect would like to hold off on new features, instead polishing/fixing
what's present.  And I remember seeing Curt express an "unofficial" desire
to do a v1.0 release sometime in early 2006.  But I haven't seen a post
saying ". . .and that release will occur early in January" or something
like that, and I haven't seen posts to the effect of "please hurry up
and identify bugs/get those bugfix patches in so we can do the 1.0
release."  Have there been such posts?

Maybe I've just missed the posts in question.  But I don't see this
"urgency to release 1.0 in a hurry" that you guys are concerned
about.  Yet obviously it seems very clearly there to the two of you
-- so much so that it must be because of a "secret agenda."  I just
spent 15 minutes googling the mailing list archives and can't find
anything implying that there's a hurry to release 1.0.  What am I
missing?

Right now, the infrastructure of the project is such that Curt has to
handle the release pretty much solo, I guess.  My understanding
(perhaps wrong) is that Curt's time constraints mean that if a
`launch window' gets missed, the next opportunity to release may be
many months later.  That's not the worst thing to happen, of course,
if there's some sort of showstopper bug.  And even if there isn't,
I agree with you that calling a release "1.0" means something, and
a release should be held up or called something else if it's not
ready to be called that.  The problem is that I don't think everyone
agrees what "1.0" should mean -- that is, what the absolutely
necessary qualities of the 1.0 release are.

For instance, from my perspective, the absolute most important thing
to have in the 1.0 release -- more important than any features of the
simulator (including that the simulated aircraft even fly!) -- is solid
accompanying documentation along with very clearly described channels
for seeking help, reporting bugs, etc.  But I'm probably in the extreme
minority in that regard.  Melchior thinks without landing lights, we're
effectively saying that FG is a daytime simulator only, since nighttime
landings are too unrealistic otherwise; there are other people here who
disagree.  Back in January, these were some of the things that various
individuals suggested were necessary before a 1.0 release:

- improving the Getting Started manual;
- concentrating on making the c172 as nice as it can be;
- fixing the JSBSim issue with gear handing when the aircraft is at stop;
- polishing fgrun;
- being able to switch aircraft from within FG;
- being able to switch airports from within FG, and to choose airports
by colloquial location name;
- significant improvement in fps;
- ability to save both options/configuration and flight status for later
resumption;
- a formalized procedure for collecting/handling/tracking bug reports;

I'm sure there were more I'm not catching.  Anyway, some of these have
happened and some haven't; given that not everyone agreed with all of
these things, what does the fact that some are missing mean for a
a v1.0 release (if anything at all)?  Some developers are bound to feel
that some of these suggestions are absolutely not necessary for 1.0;
others feel very strongly about including them.

And we can set up some kind of "voting procedure" or something like
that to create a requirements list for 1.0 -- but if that's too
ambitious, or if nobody knows how to add those features, or nobody
has the time, then we've just renamed FlightGear to "Duke Nukem Forever"
(for those unfamiliar with the reference, see
http://en.wikipedia.org/wiki/Duke_Nukem_Forever ).

I don't know.  I too would like to see v1.0 be as good as possible.
But I don't understand getting very very upset about this, especially
when 1) I don't see the "we must get 1.0 out the door right away"
stuff going on (maybe I'm just missing it), and 2) there are a
kajillion things that at least one developer thinks should be in
1.0, and obviously it isn't all going to happen.

-c





_______________________________________________
Flightgear-devel mailing list
[email protected]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
2f585eeea02e2c79d7b1d8c4963bae2d

Reply via email to