In about 1988 I suggested that there was a need for a standards body to
take over the protection of Informix-4GL syntax. At that time we were
faced with 4Js, Qerix, Aubit, and Informix, to name 4, all producing 4GL
compilers that could drift in syntax. Nobody was interested. Is there
any interest now? And would it be a good idea for IIUG to co-ordinate
this effort, if not becoming the standards body itself?

Any thoughts would be much appreciated

Malcolm

-----Original Message-----
From: owner-informix-list@iiug.org [mailtowner-informix-list@iiug.org]
On Behalf Of sergio@moredata.pt
Sent: 07 April 2004 11:19
To: Andrew Hamm
Cc: informix-list@iiug.org
Subject: Re: An interesting meeting with IBM



> ** WebSphere
>
> - v5.1.2 includes a 95% port of 4GL syntax into WebSphere.

I remember a similar thing from the passed.

Ca someone define 95 % ?
- 95 % of the 4gl syntax ?
- 95 % of the average 4gl code will work ?


> - v5.1.2 is not considered quite complete enough to go into production

> with real 4GL code. v6 is expected to be complete enough for
> deployment.

What hapenned to EGL ?

> - WebSphere 4GL will be an enhanced 4GL that adds user interface
> features.

New Era and windows 4gl was enhanced too :-(

> - It's not clear, but it appears that the user interface statements of

> 4GL have been replaced, and some sort of migration process must be
> applied to the code to get it across.

And after the migration we develop in the new 4gl (95% compatible) or an
IBM Graphic Environment ?

This means that the path of 4gl(s) will be syntax divergence (between
Querix, 4js and IBM) ? And IBM commitement to open standards ?

4gl is a great language. Why destroy it making more diferent
(incompatible) flavours ? Just make some quick money and get the users
more tied to IBM ?

> All-up, I'm quite optimistic about directions for Informix users. 4GL
> users will probably be very happy with WebSphere.

Well it is an improvement, but knowing this history of 4gl i need to see
more in order to have even a litle smile. sending to informix-list

sending to informix-list
sending to informix-list

malcolm weallans wrote:
> In about 1988 I suggested that there was a need for a standards body
> to take over the protection of Informix-4GL syntax. At that time we
> were faced with 4Js, Qerix, Aubit, and Informix, to name 4, all
> producing 4GL compilers that could drift in syntax. Nobody was
> interested. Is there any interest now? And would it be a good idea
> for IIUG to co-ordinate this effort, if not becoming the standards
> body itself?
>
> Any thoughts would be much appreciated

Errrrrrmmmmmm - i don't think there's really any will to make an ISO
standard 4GL. In the grand scheme of things, 4GL is a small language market.

4JS is leading the charge, and IBM with WebSphere claims to be keen on ideas
from customers. I hope that 4JS and websphere will work together to stay
fairly close. Querix are trying hard but I think they'll be forced to follow
or risk going out on a limb. You've probably seen that Querix have worked to
make themselves compatible with D4GL/BDL. I don't think they've started on
Genero compatibility. Aubit has a bit of interesting freaky stuff, but then,
Andre has to earn a living too, so it doesn't seem to be racing ahead. Aubit
seems to be for people who don't want to pay licence fees, but are willing
to spend many hours working in C(++?) to fix or add features they need. That
doesn't include many business programmers from the looks of it.

Personally I think the debate on 4gl can be summed up in one "word":

> malcolm weallans wrote:
> > ... 1988 ...

4gl is old hat. Sweet yes, but old hat.

Informix missed the boat on development tools.

Next.

It is precisely because this is such a small volume market that we
should think about the potential confusion if IBM go one-way which is
incompatible with Qerix and 4Js. Suddenly we will have many
proprietary 4GLs out there, if we haven't already. (In my original
mail I set I floated the idea in 1988, that should have read 1998.)
The concept is that the IIUG (or other standards body) would certify
that versions of 4GL agreed to the standard language. Obviously this
would need funding but the compiler writers would pay to have their
syntax certified and the monies raised would be used to employ the
people to do the certification. I just don't fancy the problems when
IBM have changed screen handling of migration of 4GL written aeons ago
to the nes IBM dialect, to the Qerix dialect, and to the 4Js dialect.
So, if mehdi and peter collett are listening, and If IBM are
listening, how about solving this before it becomes a problem.

regards

malcolm

"Andrew Hamm" <ahamm@mail.com> wrote in message news:<c52h6c$2lcnlp$1@ID-79573.news.uni-berlin.de>...
> malcolm weallans wrote:
> > In about 1988 I suggested that there was a need for a standards body
> > to take over the protection of Informix-4GL syntax. At that time we
> > were faced with 4Js, Qerix, Aubit, and Informix, to name 4, all
> > producing 4GL compilers that could drift in syntax. Nobody was
> > interested. Is there any interest now? And would it be a good idea
> > for IIUG to co-ordinate this effort, if not becoming the standards
> > body itself?
> >
> > Any thoughts would be much appreciated
>
> Errrrrrmmmmmm - i don't think there's really any will to make an ISO
> standard 4GL. In the grand scheme of things, 4GL is a small language market.
>
> 4JS is leading the charge, and IBM with WebSphere claims to be keen on ideas
> from customers. I hope that 4JS and websphere will work together to stay
> fairly close. Querix are trying hard but I think they'll be forced to follow
> or risk going out on a limb. You've probably seen that Querix have worked to
> make themselves compatible with D4GL/BDL. I don't think they've started on
> Genero compatibility. Aubit has a bit of interesting freaky stuff, but then,
> Andre has to earn a living too, so it doesn't seem to be racing ahead. Aubit
> seems to be for people who don't want to pay licence fees, but are willing
> to spend many hours working in C(++?) to fix or add features they need. That
> doesn't include many business programmers from the looks of it.

malcolm.weallans@btopenworld.com (malcolm) writes:
> It is precisely because this is such a small volume market that we
> should think about the potential confusion if IBM go one-way which is
> incompatible with Qerix and 4Js. Suddenly we will have many
> proprietary 4GLs out there, if we haven't already. (In my original
> mail I set I floated the idea in 1988, that should have read 1998.)
> The concept is that the IIUG (or other standards body) would certify
> that versions of 4GL agreed to the standard language. Obviously this
> would need funding but the compiler writers would pay to have their
> syntax certified and the monies raised would be used to employ the
> people to do the certification. I just don't fancy the problems when
> IBM have changed screen handling of migration of 4GL written aeons ago
> to the nes IBM dialect, to the Qerix dialect, and to the 4Js dialect.
> So, if mehdi and peter collett are listening, and If IBM are
> listening, how about solving this before it becomes a problem.

I'm still evaluating Java/Eclipse/Ant for a 4GL replacement for all
these reasons. I'm currently forced to use Sun's JVM because IBM just
can't seem to release a JVM that works with Red Hat Enterprise Linux 3
and their gcj port of Eclipse.

--
Forte International, P.O. Box 1412, Ridgecrest, CA 93556-1412
Ronald Cole <ronald@forte-intl.com> Phone: (760) 499-9142
President, CEO Fax: (760) 499-9152
My GPG fingerprint: C3AF 4BE9 BEA6 F1C2 B084 4A88 8851 E6C8 69E3 B00B

malcolm wrote:
> It is precisely because this is such a small volume market that we
> should think about the potential confusion if IBM go one-way which is
> incompatible with Qerix and 4Js. Suddenly we will have many
> proprietary 4GLs out there, if we haven't already. (In my original
> mail I set I floated the idea in 1988, that should have read 1998.)

Welllll, 4JS are listening very well to the Genero newsgroup. It's been very
productive on both sides. IBM *claims* to want to work closely with users to
add features to the Informix product line (from my recent meeting) so
hopefully that applies to WebSphere 4GL.

Sergio has indicated that the wild boys from Aubit are willing to develop
and join the fun, and Querix has shown that they'll track 4JS (aka D4GL).

No offence to Aubit or Querix, but I think they'll need to mostly "follow".
Of course, if they have brilliant ideas then we can present them to the 2
big players and request, nay demand their inclusion.

The real issue will probably be getting some sort of consistency between 4JS
and IBM. Right now we have no idea what we're going to get in WebSphere.
Perhaps 4JS saw something during their recent meetings with IBM. It may be
that the runtime scheme of WebSphere mandates some sort of change; will we
see a return to something like the NewEra event driven model? If so then
there's a vast schism between the two, since 4JS have maintained the flow of
4GL classique.

I'm applying a sprinkling of patience until I see WebSphere's effort, and
read their TODO list.