1. Rob, please make EU uncrippled...
I' ve been eh.. "evaluating" Eu on and off since 2.0 now, mainly producing
silly, megalomanic
vaporware, while having a lot of fun. But the fun always ends when I reach the
300 statement
limit, and hence this suggestion:
Why not make the PD version uncrippled, that is, make your money selling the
shrouder,
binder, profiler and soon (we must hope) the compiler(!), and let us free-riders
enjoy the
interpreter hindered?
This way EU might get a much larger user base (and one could remove those
annoying little
'stamps'). I'm not sure whether you would make more or less money this way, it's
just that
crippleware is so depressing :\
Come on, Rob, be a pal! :)
Thanks,
Tor
2. Re: Rob, please make EU uncrippled...
Tor Gausen writes:
> Why not make the PD version uncrippled, that is, make
> your money selling the shrouder, binder, profiler and soon
> (we must hope) the compiler(!), and let us free-riders enjoy
> the interpreter hindered?
Thanks for the suggestion, but I think there would be
too many free-riders who could get along quite nicely
without the other options.
Deciding on the best level of
"registration incentives" (never say crippling!)
is not easy. You can give away a lot for free and hope
to get lots of users but with a low registration rate, or
you can be stingy and hope for a high registration rate
from fewer users. For now, I'm content the way things are.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
3. Re: Rob, please make EU uncrippled...
What "registration incentives" will the compiler have ? Will it just
give us the same Euphoria features as the interpeter.
I don't see any advantage a compiler will bring if the core langauge
is not improved. Speed is not an issue because the compiler will only
knock down the loading time and probably not increase the runtime any
large amount. With the compiler you lose interactive debuging. Also
you will have to support 6 versions of Euphoria and this will restrict
the core langauge growth even more.
4. Re: Rob, please make EU uncrippled...
> -----Original Message-----
> From: Bernie Ryan
> What "registration incentives" will the compiler have ?
One thing it will do is make Eu more flexible and useful. Now we can create
DLL's (for instance) to use with other languages! I think we all agree that
there are certain tasks which Eu is great for, and now we can take advantage
of this all over the place. I think that the compiler is a good step for
Eu, because now it can be used within projects that are based in other
languages.
> Will it just
> give us the same Euphoria features as the interpeter.
> I don't see any advantage a compiler will bring if the core langauge
> is not improved. Speed is not an issue because the compiler will only
> knock down the loading time and probably not increase the runtime any
> large amount. With the compiler you lose interactive debuging. Also
> you will have to support 6 versions of Euphoria and this
> will restrict the core langauge growth even more.
It may, but based on a couple of years of using Eu and observing its growth,
I don't think that the reason Rob hasn't grown the language a lot more is
because it's too big a task. It seems pretty clear that either Rob doesn't
want to do certain things, or he can't decide on what to do. Once his mind
is made up, things seem to move along pretty quickly.
Besides, he's found another revenue stream, and if that won't motivate him,
I don't know what would. ;)
Matt Lewis
5. Re: Rob, please make EU uncrippled...
On Tue, 16 May 2000 10:00:57 -0700, Matthew Lewis <MatthewL at KAPCOUSA.COM>
wrote:
>> What "registration incentives" will the compiler have ?
>
>One thing it will do is make Eu more flexible and useful. Now we can
create
>DLL's (for instance) to use with other languages! I think we all agree
that
>there are certain tasks which Eu is great for, and now we can take
advantage
>of this all over the place. I think that the compiler is a good step for
>Eu, because now it can be used within projects that are based in other
>languages.
Matt: DLL support can be added to the interpeter, It's being done on
other dos extenders. It seems to me it would be much easier to implement
DLL support than a compiler. It can then be used by anyone to extend the
langauge and allow the users to write their on features. It all comes
down to the only way to extend the langauge is convice Rob that you
need some feature that is missing from the core langauge.
and add the features that the user wants.
6. Re: Rob, please make EU uncrippled...
On 16 May 2000, at 11:34, Bernie Ryan wrote:
Date sent: Tue, 16 May 2000 11:34:06 -0400
Send reply to: Euphoria Programming for MS-DOS <EUPHORIA at
LISTSERV.MUOHIO.EDU>
From: Bernie Ryan <xotron at BUFFNET.NET>
Subject: Re: Rob, please make EU uncrippled...
To: EUPHORIA at LISTSERV.MUOHIO.EDU
> What "registration incentives" will the compiler have ? Will it just
>
> give us the same Euphoria features as the interpeter.
>
> I don't see any advantage a compiler will bring if the core langauge
>
> is not improved. Speed is not an issue because the compiler will only
>
> knock down the loading time and probably not increase the runtime any
>
> large amount.
Speed can and will be a problem in realtime interactivity, and in
active graphics. Did you mean "not DEcrease the runtime"?
>With the compiler you lose interactive debuging. Also
So debug in interpreted mode. Then compile and doublecheck, then
sell your application to the masses. Speaking of double checking the
compiled code, the first "service pack" (pronounced like "bug fix")
for win2k is out,, it's 190Megs big.
> you will have to support 6 versions of Euphoria and this will
> restrict
>
> the core langauge growth even more.
Not really, but what i prolly will do is make only one or two C
compilers "approved" by RDS, due to differences in compilers.
Kat
7. Re: Rob, please make EU uncrippled...
Bernie wrote.....
> I don't see any advantage a compiler will bring if the core langauge
> is not improved. Speed is not an issue because the compiler will only
> knock down the loading time and probably not increase the runtime any
> large amount.
This isn't to create a war in the speed versus features camps but I do
support Bernie's comments here. I will probably re-register for the
compiler but I certainly would re-register to get some of the additional
features that Bernie has suggested
> With the compiler you lose interactive debuging. Also
> you will have to support 6 versions of Euphoria and this will restrict
> the core langauge growth even more.
The compiler guys need for speed looks like it will be fulfilled soon
(which is cool) but I'm still going to be left wanting some features.
All the best
Mark
8. Re: Rob, please make EU uncrippled...
Bernie Ryan writes:
> I don't see any advantage a compiler will bring
> if the core langauge is not improved.
You make it sound like you are "dead in the water"
because you are lacking your wishlist items of the other day.
Most of those items struck me as mere conveniences,
not critical issues. (line continuation character, ...)
> Speed is not an issue because the compiler will only
> knock down the loading time and probably not increase
> the runtime any large amount.
In some cases the compiler will give you a much greater
speed of execution. e.g. 5x or more.
I'm hoping that the speed gap between Euphoria and C will
be narrowed enough that a certain class of users will say
"Hmmm... why am I beating my head over a brick wall
programming in C, when Euphoria is almost as fast?"
They'll also say, "Hmmm, suppose RDS went out of business,
I could convert my Euphoria code to C at anytime, and my boss
wouldn't shoot me."
> With the compiler you lose interactive debuging.
Develop with the interpreter.
Compile for speed.
> Also you will have to support 6 versions of Euphoria
> and this will restrict the core langauge growth
> even more.
I don't measure Euphoria's progress by the "growth"
in the core language each year. Does an airplane
designer pat himself on the back for adding 5 tons
to the weight of an airplane?
The compiler uses the same scanner, parser and run-time
routines as the interpreter. There will be a high degree of
code re-use, and dual-maintenance will be kept to a minimum.
For instance, almost all of your wish list items could be
implemented in the interpreter, and they would automatically
become part of the compiler on the next build.
I believe that this compiler will do more to stimulate the
creativity of the Euphoria world, than will a few more incremental
features delivered a few months earlier.
The ability to generate C might lead to a lot of possibilities,
such as: writing a DLL in Euphoria, not just accessing some else's
DLL, inserting C code into a program, rather than assembly code,
and all sorts of sleazy stuff that people might dream up.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
9. Re: Rob, please make EU uncrippled...
Robert Craig <rds at ATTCANADA.NET> wrote:
<snip>
>...inserting C code into a program...
Hmm... interesting comment. Do you mean inserting C code into a "Euphoria"
program -OR- the C program generated by Euphoria? If the former... then,
are you saying *GULP* that you will be adding a C interpreter to Euphoria?
If that was the case, then we could add our own language extensions as we
see fit (eg. record structures, etc.).
Sorry... I must have been daydreaming!
Lee.
10. Re: Rob, please make EU uncrippled...
On Wed, 17 May 2000 14:34:41 -0400, Lee West <leewest at ALTAVISTA.COM> wrote:
>see fit (eg. record structures, etc.).
Lee:
My mixedlib supports record structures.
Be sure to get the latest updated version when it is added to
contribution files but mean while you can look at the older version
that is already there.
Bernie
11. Re: Rob, please make EU uncrippled...
- Posted by Robert Craig <rds at ATTCANADA.NET>
May 17, 2000
-
Last edited May 18, 2000
Lee West writes:
> Hmm... interesting comment. Do you mean inserting
> C code into a "Euphoria" program -OR- the C program
> generated by Euphoria?
Either way would be possible.
I'd like to build the compiler and play with it,
before deciding things like this.
> If the former... then,
> are you saying *GULP* that you will be adding a
> C interpreter to Euphoria?
No.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
12. Re: Rob, please make EU uncrippled...
Rob,
> Thanks for the suggestion, but I think there would be
> too many free-riders who could get along quite nicely
> without the other options.
Sigh. So you are not nearly as dumb as I hoped... :)
> Deciding on the best level of "registration incentives"
> (never say crippling!)
IMHO, there must be two kinds of "registration incentives":
the positive and the negative. A compiler that speeds your
code up to about the level of C is a positive incentive,
while an intentional damage to an interpreter so that
after some time or in certain conditions it won't work
is called... CRIPPLING, CRIPPLING, CRIPPLING!!! :)
I still love this little programming language of yours though...
Tor
13. Re: Rob, please make EU uncrippled...
Robert Craig wrote:
>> If the former... then,
>> are you saying *GULP* that you will be adding a
>> C interpreter to Euphoria?
>
>No.
>
Then, an Euphoria program containing C code would not work properly (ie.
the C code would be considered comments and NOT be executed?) when
interpreted? The C code would ONLY be passed through (along with the
translated Eu2C code) to the final C output?
Lee.
14. Re: Rob, please make EU uncrippled...
Lee West writes:
> Then, an Euphoria program containing C code would
> not work properly (ie. the C code would be considered
> comments and NOT be executed?) when interpreted?
> The C code would ONLY be passed through (along with the
> translated Eu2C code) to the final C output?
Yes, assuming I support this feature.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
