Re: Euphoria needs more popularity!
Robert Craig wrote:
>
> Derek Parnell wrote:
> > Version 2.0 Official Release March 25, 1998
> > + 12 months
> > Version 2.1 Official Release for WIN32 + DOS32 March 29, 1999
> > + 10 months
> > Version 2.2 Official Release for WIN32+DOS32 January 14, 2000
> > + 25 months
>
> During that time I spent a year designing, developing and porting the
> Euphoria to C Translator to 4 platforms and 7 different C compilers.
> I should have merged that information into the release notes.
Like I said, a shitload of work. Too much for a one-man team, I'd argue.
> > Version 2.3 Official Release February 11, 2002
> > + 18 months
> > Version 2.4 Official Release July 3, 2003
> > + 18 months?
> > Version 2.5 ?December, 2004?
> > That's right, we have moved from v2.0 to v2.5 in 6.5 years. Not Fast.
>
> Maybe I should call the next one 3.0. Will that help?
Not one bit.
> Isn't Win32Lib officially still an "alpha" release, 0.60.5,
> after those same 6.5 years? 
And I took in on around v0.50. How many releases does that mean? A least
15 in the same time as your 5. The literal version number is not important.
The number of releases is a more telling statistic.
> > RDS is slow to release anything, and they keep declining offers of help.
> > Could this be another reason for people being discouraged?
> >
> > Yes, I know there is a shitload of work to get out a new release, but
> > there is also many able people to help, if only you'd let them.
>
> Yeah, I heard you can produce a baby in 1 month by
> assigning 9 women to the job.
I know you are not serious about that statement, but that might be
a problem too. It seems that you are saying that your work can never
be helped by using multiple people. Prove it!
> > I would prefer a major release every 6 months with weekly minor (patch)
> > releases, instead of this cold molasses.
>
> That's what I keep telling myself.
> It just never works out that way.
How we we know - we've not seen you try it.
> There's just a lot of overhead in getting
> a new release out,
Sure is! You could do with a hand.
>plus when do I charge for an upgrade?
Whenever you wanted to.
> I'll give it some thought.
Yeah, you do that. So version 2.6 is scheduled for when ... February 2006?
> > Perfection is not so nearly
> > as important as improvement. Having a perfect product that doesn't meet
> > my needs is not much use. Having a nearly perfect product that I can still
> > use is a much better prospect.
What? No comment? This seems to be the area of greatest consensus.
> > Robert,
> > Are you using beta testers?
>
> Every official release comes after an alpha and a beta
> release. Earlier releases would waste my time on
> handling a lot of bug reports and questions.
Heaven forbid! We wouldn't want to find all those bugs so early would we?
Again, if this is a lot of work, then maybe you could do with help?
> I have tons of existing Euphoria code I can use as testing fodder.
I should hope so. This is great for regression testing, but what about
planned test cases, a test plan, a formal Euphoria test suite?
> > Has anyone impartially inspected or reviewed your code?
>
> Not besides Junko,
Then you are doing yourself a disservice. Inspection/Reviewing is the
most cost effective method of finding bugs. It finds more than testing
and finds them earlier.
>but very soon the entire world will
> be able to inspect the front-end of Euphoria, and everyone will
> own a complete Euphoria interpreter with 100% Euphoria source code,
> able to run all Euphoria programs on all platforms.
So you intended to have the work reviewed after it is released. Not
worlds best practice, I can assure you.
> > Do you have a formal(-ish) issue log that you are working through?
>
> Yes.
Great! Well done.
> > Do you need more man-hours in the day to work on Euphoria?
>
> Of course.
So get more people to give you those man-hours! I'd pay you
to be on your team.
> > Sorry to sound so frustrated, but I am. I love Euphoria and I continue
> > to champion it, but I also begin to tire.
>
> I thank you for your patience, and your tremendous
> efforts on Win32Lib. Euphoria would be in deep trouble without you.
Bullshit. Its a great product that could be well by itself,
but has been left behind by the competition.
> I'd like to speed up my progress too. I have been slacking
> off a bit, but you have to give me credit for sticking with
> this project for 15 years (since initial design - 1.0 was
> released 11 years ago).
I respect your committment and your efforts. You had created a
revolutionary product that deserved more air time. RDS's development
methodology is the major reason for Euphoria's lack of success - not
the product itself.
> But I would certainly turn it over to
> the masses if I couldn't or didn't want to continue.
How about loosening the apron-strings a bit earlier than that? Your
baby needs to grow up.
--
Derek Parnell
Melbourne, Australia
|
Not Categorized, Please Help
|
|
