ibm-netrexx

Wishlist NetRexx New Release

Classic

List

Threaded

17 messages Options

rvjansen

Wishlist NetRexx New Release

Hi list,

well, the first wish would be a new release at all, with great stuff we
did no think of ourselves.

But apart from that, very nice would be:

0)An integrated compiler server? For the people who are not using mine.
1) trace i
(This is great in classic Rexx. We miss it)
2) a set of bitwise operators
(We are always confused about this and have to look up the manual. Ok,
options binary. Then someone forgets why or needs decimal stuff and it
breaks.)
3) Javadoc comments on indirect properties propagated to generated getter
and setter method
4) (optional) Generated propertychange event support on indirect properties
5) Java 1.5 generics (please)
6) Aspect Oriented NetRexx extensions?

best regards,

Rene Vincent Jansen.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

Patric Bechtel

Re: Wishlist NetRexx New Release

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wed, 21 Jan 2004 13:23:51 +0100 (CET), [hidden email] wrote:

Hi Rene,

I'm a fan of NetRexx, too, and we (that is, my company) built a rather huge
framework with it, with several 100.000 lines of code. So, we've a LOT
experience with NetRexx (and with Rexx in former times, too. I still use
it for small replacement for perl where a jvm and even a perl interpreter
is just too big).

>Hi list,

>well, the first wish would be a new release at all, with great stuff we
>did no think of ourselves.

>But apart from that, very nice would be:

>0)An integrated compiler server? For the people who are not using mine.

I use an ant script for all my builds. I somewhat improved the NetRexxC
task of ant (included in the 1.5 version), so that error messages can
be filtered and displayed in jEdit.
Second, I've improved the NetRexx syntax highlightning of jEdit (since 4.1,
I will publish an even more enhanced version for 4.2 soon).
So, Antfarm, ant and jEdit make quite a good ide for NetRexx which I'm
using for the last years already.

>1) trace i
>(This is great in classic Rexx. We miss it)

I don't miss that one too much.

>2) a set of bitwise operators
>(We are always confused about this and have to look up the manual. Ok,
>options binary. Then someone forgets why or needs decimal stuff and it
>breaks.)

yep. And probably an "and" keyword instead the ugly & and "or" instead of |.
NetRexx is wonderful wordy and easy to read, but this one time it's a bit
inconsequent... ;-)

>3) Javadoc comments on indirect properties propagated to generated getter
>and setter method

Cool idea. Probably not too important, but nice.

>4) (optional) Generated propertychange event support on indirect properties

Would be nice, too. But how to overload these then?

>5) Java 1.5 generics (please)

YES. Yes. Yes. PLEEEAAZE! I would really LOVE having this in NetRexx. At least
to a distinct level.

>6) Aspect Oriented NetRexx extensions?

Woah. THAT would be a bummer. Even better than generics. Alltogether this
would really be a killer. Like NetRexx was is the early days. Simple, easy
to learn, easy to read, the best for teams (IMHO).

7.) as a starting point: iterator syntax sugar, like the new 1.5 compiler does. :-)

But, after all, I'm not sure if Michael will have time for this. I wrote a
message to the list 6 weeks ago and still have no answer (regarding a name
resolution problem).
Maybe there's another solution: If we could get the beast to Open Source,
that could get a kick, like it is in the AS/400 toolkit or the icu project.

That's a decision Michael has to make...

cu,

Patric

>best regards,

>Rene Vincent Jansen.

>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
>[hidden email]
>with the following message in the body of the note
>unsubscribe ibm-netrexx <e-mail address>

-----BEGIN PGP SIGNATURE-----
Version: PGPsdk version 1.7.1 (C) 1997-1999 Network Associates, Inc. and its affiliated companies.

iQA/AwUBQA56MHxoBrvMu8qQEQIpoQCgvx+EUhy2dsbqv0cA2JZEspEVZ38AnRiw
7NT6WpoG2HK8q9ZlaxeL8TP9
=OGay
-----END PGP SIGNATURE-----

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

rvjansen

Re: Wishlist NetRexx New Release

> I'm a fan of NetRexx, too, and we (that is, my company) built a rather
> huge
> framework with it, with several 100.000 lines of code. So, we've a LOT
> experience with NetRexx (and with Rexx in former times, too. I still use
> it for small replacement for perl where a jvm and even a perl interpreter
> is just too big).
That goes without saying. It would have been very marketable to all old
mainframe hands and some amiga/os2 people as a painless way into Java.
Alas, probably the timing of the design of ORexx got in the way.

> I use an ant script for all my builds. I somewhat improved the NetRexxC
> task of ant (included in the 1.5 version), so that error messages can
> be filtered and displayed in jEdit.
> Second, I've improved the NetRexx syntax highlightning of jEdit (since
> 4.1,
> I will publish an even more enhanced version for 4.2 soon).
> So, Antfarm, ant and jEdit make quite a good ide for NetRexx which I'm
> using for the last years already.
>
As stated previously, our team is on Emacs. Arjan Bos made a great Emacs
mode for NetRexx, with syntax coloring and other important features. It is
an IDE in itself!

>4) (optional) Generated propertychange event support on indirect properties
> Would be nice, too. But how to overload these then?
>

well, that would be the optional part of it.

>
> But, after all, I'm not sure if Michael will have time for this. I wrote a
> message to the list 6 weeks ago and still have no answer (regarding a name
> resolution problem).
> Maybe there's another solution: If we could get the beast to Open Source,
> that could get a kick, like it is in the AS/400 toolkit or the icu
> project.
>
> That's a decision Michael has to make...
>

rvjansen

Re: Wishlist NetRexx New Release

In reply to this post by Patric Bechtel

Hi Patrick,

> I'm a fan of NetRexx, too, and we (that is, my company) built a rather
> huge
> framework with it, with several 100.000 lines of code. So, we've a LOT
> experience with NetRexx (and with Rexx in former times, too. I still use
> it for small replacement for perl where a jvm and even a perl interpreter
> is just too big).
That goes without saying. It would have been very marketable to all old
mainframe hands and some amiga/os2 people as a painless way into Java.
Alas, probably the timing of the design of ORexx got in the way.

> I use an ant script for all my builds. I somewhat improved the NetRexxC
> task of ant (included in the 1.5 version), so that error messages can
> be filtered and displayed in jEdit.
> Second, I've improved the NetRexx syntax highlightning of jEdit (since
> 4.1,
> I will publish an even more enhanced version for 4.2 soon).
> So, Antfarm, ant and jEdit make quite a good ide for NetRexx which I'm
> using for the last years already.
>
As stated previously, our team is on Emacs. Arjan Bos made a great Emacs
mode for NetRexx, with syntax coloring and other important features. It is
an IDE in itself!

>4) (optional) Generated propertychange event support on indirect properties
> Would be nice, too. But how to overload these then?
>

well, that would be the optional part of it.

Well, I have mentioned before that the closed source aspect of it worried
me a great deal; it is understandable though, seen its roots as EWP in
1995. But the times have changed, also for IBM, and they open up a lot of
source right now. It would be great to have community support for it, and
I would not mind (being not too principal in professional matters) if
someone made a C#, CLR mode for it, like the ANTLR people did.

best regards,

Rene Vincent Jansen.

> I'm a fan of NetRexx, too, and we (that is, my company) built a rather
> huge
> framework with it, with several 100.000 lines of code. So, we've a LOT
> experience with NetRexx (and with Rexx in former times, too. I still use
> it for small replacement for perl where a jvm and even a perl interpreter
> is just too big).
That goes without saying. It would have been very marketable to all old
mainframe hands and some amiga/os2 people as a painless way into Java.
Alas, probably the timing of the design of ORexx got in the way.

> I use an ant script for all my builds. I somewhat improved the NetRexxC
> task of ant (included in the 1.5 version), so that error messages can
> be filtered and displayed in jEdit.
> Second, I've improved the NetRexx syntax highlightning of jEdit (since
> 4.1,
> I will publish an even more enhanced version for 4.2 soon).
> So, Antfarm, ant and jEdit make quite a good ide for NetRexx which I'm
> using for the last years already.
>
As stated previously, our team is on Emacs. Arjan Bos made a great Emacs
mode for NetRexx, with syntax coloring and other important features. It is
an IDE in itself!

>4) (optional) Generated propertychange event support on indirect properties
> Would be nice, too. But how to overload these then?
>

well, that would be the optional part of it.

smoerk

Re: Wishlist NetRexx New Release

In reply to this post by rvjansen

[hidden email] wrote:

> Hi list,
>
> well, the first wish would be a new release at all, with great stuff we
> did no think of ourselves.

as there is no new release for several years now, the only thing which
could give NetRexx a future would be an open source release of NetRexx.

did anyone ask Mike Cowlishaw, if we can hope for an open sourced
NetRexx? Mike are you still reading the list?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

JNiifen@Attglobal.net

Re: Wishlist NetRexx New Release

I know I would love to see NetRexx and Object Rexx on the Suse distribution.
Joe
smoerk wrote:

> [hidden email] wrote:
>
>> Hi list,
>>
>> well, the first wish would be a new release at all, with great stuff we
>> did no think of ourselves.
>
>
> as there is no new release for several years now, the only thing which
> could give NetRexx a future would be an open source release of NetRexx.
>
> did anyone ask Mike Cowlishaw, if we can hope for an open sourced
> NetRexx? Mike are you still reading the list?
>
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> To unsubscribe from this mailing list ( ibm-netrexx ), please send a
> note to
> [hidden email]
> with the following message in the body of the note
> unsubscribe ibm-netrexx <e-mail address>
>

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

Harijs Buss

Re: Wishlist NetRexx New Release

In reply to this post by smoerk

On Wednesday 21 January 2004 20:13, smoerk rakstija:

> as there is no new release for several years now, the only thing which
> could give NetRexx a future would be an open source release of NetRexx.

Right. And there are _lots_ of people now in Linux world which have OS/2 and
mainframe roots :-) We would gladly use NetREXX if it would be Open Source.

Harry

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

Mystery Guest

Re: Wishlist NetRexx New Release

In reply to this post by rvjansen

I guess my concern is what happens to NetRexx if Mike wins the lottery and
retires
or, heaven forbid, he gets hit by a beer truck. Is there a champion waiting
in the
wings at IBM to carry on the NetRexx torch?

./CK

>From: Harijs Buss <[hidden email]>
>To: [hidden email]
>Subject: Re: Wishlist NetRexx New Release
>Date: Fri, 23 Jan 2004 01:16:41 +0200
>
>On Wednesday 21 January 2004 20:13, smoerk rakstija:
>
> > as there is no new release for several years now, the only thing which
> > could give NetRexx a future would be an open source release of NetRexx.
>
>Right. And there are _lots_ of people now in Linux world which have OS/2
>and
>mainframe roots :-) We would gladly use NetREXX if it would be Open
>Source.
>
>Harry
>

_________________________________________________________________
MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*.
http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoin.msn.com%2f%3fpage%3dmisc%2fspecialoffers%26pgmarket%3den-ca

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

Kai Schmidt

Re: Wishlist NetRexx New Release

Am Freitag, 23. Januar 2004 06:33 schrieb Mystery Guest:
> I guess my concern is what happens to NetRexx if Mike wins the lottery and
> retires

> >Right. And there are _lots_ of people now in Linux world which have OS/2
> >and mainframe roots :-) We would gladly use NetREXX if it would be Open

Hi,

I agree concerning OpenSource-Thoughts this might help to push NetRexx.

After my experience with NetRexx (from the very first versions on):

a.) I use more and more python (which is almost as clean and intuitve as
NetRexx) when I'm free to choose the language for a given problem.

b.) I choose Netrexx in the Java-world, if I'm free to choose the syntax
dialect within JVM (java world). If JPython would work, that would be
something else...
My experiences of writing Netrexx Syntax (which I still like much better than
Java Syntax) in cooperation with pure Java-people is not satisfying. ["Java
is Java Syntax. Period."] Those people do not even know NetRexx. I don't know
why but in my (german) environment, hardly anyone knows NetRexx. Sad ...

So, if NetRexx could give out better (more "normal", more human readable) java
source code). Could be an separate post-processor. The "real" Java people
should think, that the java-source comes from a human. [I have no clue if
thats achievable]

My Conclusion: NetRexx as Open Source as the "better Java" with human,
productive Syntax and the ability to (optional) deliver "readable"
Java-Source.

All the best

Kai

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

mark_cathcart

Re: Wishlist NetRexx New Release

I checked, Mike has been away from Jan 12 and won't be back until next
Monday earlist... ++Mark.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

rvjansen

Re: Wishlist NetRexx New Release

In reply to this post by Kai Schmidt

Hi Kai,

Interestingly you mention Python which I looked at and which seems to
have the wide acclaim that NetRexx deserves. As part of the "fallback
planning" I looked at Jython and I found the Java integration next to
unusable. In my opinion Sun blundered a bit by stressing the tight
coupling between the VM and the language syntax. I you have a look at
http://www.robert-tolksdorf.de/vmlanguages you may wonder why this is
Sun's (or IBM's, for that matter) official stance. Microsoft does this,
alas if I may add, a lot better by giving a broad choice of languages
for their new, Java VM (and IBM's Language Environment?) inspired
Common Language Runtime - which conveniently obfuscates the
multiplatform aspect for the management types. For professional reasons
I really, really would like to see a CLR version of NetRexx, preferably
also running on Mono.

Your remark on cooperation with pure Java-only programmers is right on
the mark. I experienced much easier cooperation with people coming from
other languages, with Classic Rexx of course as a natural fit. One
project I did in Netrexx in 1998 was even re-written by the Java-heads
who took it over within that company, only for them to notice that it
was a costly excercise (rewriting the string methods of type Rexx, the
parse statements I had and replacing asscociative arrays back woth
hashmaps) and did not attain the performance gains they sold it to
management with. While I think your "human Java" output option is a
good idea, I tend no to worry about it too much. The thing you could
achive this goal with, a parser/lexer combo producing NetRexx, would be
on my wish list. I am integrating Antlr grammars with NetRexx generated
classes, but I could do for more readability in the generated
lexer/parser source. The good thing about Antlr being open source is
that I could, given enough knowledge and time, could try to add the
NetRexx mode myself.

Anyone (except Thomas) have any other ideas of starting open source
NetRexx-related projects?

best regards,

Rene Vincent Jansen.

On Jan 23, 2004, at 12:05 PM, Kai Schmidt wrote:

> Am Freitag, 23. Januar 2004 06:33 schrieb Mystery Guest:
>> I guess my concern is what happens to NetRexx if Mike wins the
>> lottery and
>> retires
>
>>> Right. And there are _lots_ of people now in Linux world which have
>>> OS/2
>>> and mainframe roots :-) We would gladly use NetREXX if it would be
>>> Open
>
> Hi,
>
> I agree concerning OpenSource-Thoughts this might help to push NetRexx.
>
> After my experience with NetRexx (from the very first versions on):
>
> a.) I use more and more python (which is almost as clean and intuitve
> as
> NetRexx) when I'm free to choose the language for a given problem.
>
> b.) I choose Netrexx in the Java-world, if I'm free to choose the
> syntax
> dialect within JVM (java world). If JPython would work, that would be
> something else...
> My experiences of writing Netrexx Syntax (which I still like much
> better than
> Java Syntax) in cooperation with pure Java-people is not satisfying.
> ["Java
> is Java Syntax. Period."] Those people do not even know NetRexx. I
> don't know
> why but in my (german) environment, hardly anyone knows NetRexx. Sad
> ...
>
> So, if NetRexx could give out better (more "normal", more human
> readable) java
> source code). Could be an separate post-processor. The "real" Java
> people
> should think, that the java-source comes from a human. [I have no clue
> if
> thats achievable]
>
> My Conclusion: NetRexx as Open Source as the "better Java" with human,
> productive Syntax and the ability to (optional) deliver "readable"
> Java-Source.
>
>
> All the best
>
> Kai
>
>
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ~~~~~~
> To unsubscribe from this mailing list ( ibm-netrexx ), please send a
> note to
> [hidden email]
> with the following message in the body of the note
> unsubscribe ibm-netrexx <e-mail address>
>

rvjansen

Re: Wishlist NetRexx New Release

Hi Chad,

this is good news. I surely would be interested. Why not upload it to
antl.org?

I agree that it would be better to have other implementations, but I'm
not seeing that happen for a number of reasons, one of those being that
the logical consequence of syntax being easy for humans seems to be
that it must be hard on compilers. The reference implementation should
be a great start in writing an alternative.
If I would be without work for a longer period I certainly would like
to try; perhaps we could start a repository of working pieces if you
would be willing to release your trials.
If we could have a working antlr grammar for NetRexx (perhaps with the
assistance of Terence Parr, or one of the other antlr people) we could
make a better planning of what is required. I'd suggest going for the
compiler first and leaving the interpreter as a separate effort.

Thomas Schneider reminded me that RexxLA is in Boeblingen this year. I
surely will come over if we can round up some people interested in
discussing this and other NetRexx related stuff.

best regards,

Rene Vincent Jansen.

On Jan 24, 2004, at 3:10 AM, Chad Slaughter wrote:

> On Fri, Jan 23, 2004 at 02:34:44PM +0100, René Jansen wrote:
>> achive this goal with, a parser/lexer combo producing NetRexx, would
>> be
>> on my wish list. I am integrating Antlr grammars with NetRexx
>> generated
>> classes, but I could do for more readability in the generated
>> lexer/parser source. The good thing about Antlr being open source is
>> that I could, given enough knowledge and time, could try to add the
>> NetRexx mode myself.
>
> I added support to antlr to generate NetRexx a couple years ago for my
> own purposes. I'm not sure where the code is but if you're interested
> I could try and dig it up.
>
>> Anyone (except Thomas) have any other ideas of starting open source
>> NetRexx-related projects?
>
> Instead of IBM releasing their netrexx compiler, the community could
> build there own. I've attempted this myself off and on with some
> success. The blank operator always kills me for some reason.
>
>
>
> --
> Chad Slaughter
> slaught at lusars dot net
>

Patric Bechtel

Re: Wishlist NetRexx New Release

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sat, 24 Jan 2004 14:51:40 +0100, René Jansen wrote:

Hi Chad, hi René,

that's really good news. I already had SOME mails with Michael regarding
opening the source code. Sure, it will be a good starting point for some
implementation details. But, as I now believe, the source code itself
was first developed in another direction, so it's not that streamlined
as we expect or (or as Michael wished it to be before opening it...).
At least, that's my impression so far... But, after all, I'm very
dedicated to the 'readable source code' idea, so having an Open Source
NetRexx version would be wonderful. Being able to extend it, fix
bugs on my own, fixing some annoyances and so forth.
So, I will be glad to supply some work to the project.
Another thing still missing is a 'real' script version of NetRexx... ;-)

Patric

>Hi Chad,

>this is good news. I surely would be interested. Why not upload it to
>antl.org?

>I agree that it would be better to have other implementations, but I'm
>not seeing that happen for a number of reasons, one of those being that
>the logical consequence of syntax being easy for humans seems to be
>that it must be hard on compilers. The reference implementation should
>be a great start in writing an alternative.
>If I would be without work for a longer period I certainly would like
>to try; perhaps we could start a repository of working pieces if you
>would be willing to release your trials.
>If we could have a working antlr grammar for NetRexx (perhaps with the
>assistance of Terence Parr, or one of the other antlr people) we could
>make a better planning of what is required. I'd suggest going for the
>compiler first and leaving the interpreter as a separate effort.

>Thomas Schneider reminded me that RexxLA is in Boeblingen this year. I
>surely will come over if we can round up some people interested in
>discussing this and other NetRexx related stuff.

>best regards,

>Rene Vincent Jansen.

>On Jan 24, 2004, at 3:10 AM, Chad Slaughter wrote:

>> On Fri, Jan 23, 2004 at 02:34:44PM +0100, René Jansen wrote:
>>> achive this goal with, a parser/lexer combo producing NetRexx, would
>>> be
>>> on my wish list. I am integrating Antlr grammars with NetRexx
>>> generated
>>> classes, but I could do for more readability in the generated
>>> lexer/parser source. The good thing about Antlr being open source is
>>> that I could, given enough knowledge and time, could try to add the
>>> NetRexx mode myself.
>>
>> I added support to antlr to generate NetRexx a couple years ago for my
>> own purposes. I'm not sure where the code is but if you're interested
>> I could try and dig it up.
>>
>>> Anyone (except Thomas) have any other ideas of starting open source
>>> NetRexx-related projects?
>>
>> Instead of IBM releasing their netrexx compiler, the community could
>> build there own. I've attempted this myself off and on with some
>> success. The blank operator always kills me for some reason.
>>
>>
>>
>> --
>> Chad Slaughter
>> slaught at lusars dot net
>>

>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
>[hidden email]
>with the following message in the body of the note
>unsubscribe ibm-netrexx <e-mail address>

-----BEGIN PGP SIGNATURE-----
Version: PGPsdk version 1.7.1 (C) 1997-1999 Network Associates, Inc. and its affiliated companies.

iQA/AwUBQBKGnXxoBrvMu8qQEQLPkgCeJIhN+XpBCny6VcwL/VjsYNkA3GgAniPI
5J5s716Xffc+dDYrSkHKLOmB
=B9Gh
-----END PGP SIGNATURE-----

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

Mike Cowlishaw-2

Re: Wishlist NetRexx New Release

In reply to this post by Mystery Guest

> did anyone ask Mike Cowlishaw, if we can hope for an open sourced
> NetRexx? Mike are you still reading the list?

In principle I would like to do this, but the code would need a lot of
work before it would be suitable as a OS project; it is very much a
'research vehicle' rather than 'production code'. I'm working about
150% of my time on decimal things at present, so I do not expect to be
able to work on NetRexx in the near future -- sorry.

Mike

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Mike Cowlishaw, IBM Fellow
IBM UK (MP5), PO Box 31, Birmingham Road, Warwick, CV34 5JL
mailto:[hidden email] -- http://www2.hursley.ibm.com/decimal

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

Mike Cowlishaw-2

Re: Wishlist NetRexx New Release

In reply to this post by Mystery Guest

If anyone does start a NetRexx open source project, by the way, I would
certainly support that -- any questions, etc. I could also probably make
the testcases available as 'sample code'. Writing the testcases is the
most tedious part of compiler development....

(I would also recommend that an implmentation be pure compiler .. the
interpreter option is 'hard', and perhaps not terribly useful.)

Mike

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Mike Cowlishaw, IBM Fellow
IBM UK (MP5), PO Box 31, Birmingham Road, Warwick, CV34 5JL
mailto:[hidden email] -- http://www2.hursley.ibm.com/decimal

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

Chad Slaughter-2

Re: Wishlist NetRexx New Release

On Sun, Jan 25, 2004 at 07:47:52AM +0000, Mike Cowlishaw wrote:
> the testcases available as 'sample code'. Writing the testcases is the
> most tedious part of compiler development....

That would be a wonderful gift. As you say, it is very tedious and time
consuming but so very necessary. I think that alone would would cut
a compiler project down to a nice manageable size.

--
Chad Slaughter
slaught at lusars dot net

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>

Ed Tomlinson

Re: Wishlist NetRexx New Release

In reply to this post by Mike Cowlishaw-2

On January 25, 2004 02:39 am, Mike Cowlishaw wrote:
> > did anyone ask Mike Cowlishaw, if we can hope for an open sourced
> > NetRexx? Mike are you still reading the list?
>
> In principle I would like to do this, but the code would need a lot of
> work before it would be suitable as a OS project; it is very much a
> 'research vehicle' rather than 'production code'. I'm working about
> 150% of my time on decimal things at present, so I do not expect to be
> able to work on NetRexx in the near future -- sorry.

Mike is there a person you would trust to morph the code? If not,
how do you define near future?

TIA
Ed Tomlinson

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>