OO & Standard Libraries

OO & Standard Libraries

[Martin Dowie]

Just stumbled across

http://www.jot.fm/

which is published by one "Bertrand Meyer".

In its first issue it is discusses standard libraries!

Worth a read.

Perhaps Ehud and/or John could contribute something once
Ada-Europe is over (or even drop a letter in prior to raise
awareness)?

Re: OO & Standard Libraries

[Marin David Condic]

I particularly liked:

"Bundling of libraries with IDE, compiler or VM creates defacto standards
and control points for language/platform vendors."

I think the article has some interesting things to say on the subject and
ought to be a good read for those of us interested in getting a "standard"
component library into Ada.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com


"Martin Dowie" <martin.dowie@nospam.baesystems.com> wrote in message
news:3cd93d95$1@pull.gecm.com...
> Just stumbled across
>
> http://www.jot.fm/
>

Re: OO & Standard Libraries

[Ted Dennison]

"Martin Dowie" <martin.dowie@nospam.baesystems.com> wrote in message news:<3cd93d95$1@pull.gecm.com>...
> http://www.jot.fm/
...
> In its first issue it is discusses standard libraries!
> 
> Worth a read.

If its the one I think it is (there isn't anything there by Meyer),
then its interesting, but there's nothing earth-shattering in there to
anyone who's followed our List discussions.

The bit about test suites did remind me that I need to submit the List
package test suite to CVS though...

Re: OO & Standard Libraries

[martin.m.dowie]

"Ted Dennison" <dennison@telepath.com> wrote in message
news:4519e058.0205081043.405a719f@posting.google.com...
[snip]
> If its the one I think it is (there isn't anything there by Meyer),
> then its interesting, but there's nothing earth-shattering in there to
> anyone who's followed our List discussions.
>
> The bit about test suites did remind me that I need to submit the List
> package test suite to CVS though...

All true, but it could be another avenue for marketing Ada and getting
people talking about it.

Re: OO & Standard Libraries

[Eric G. Miller]

In <abbgk3$bc9$1@nh.pace.co.uk>, Marin David Condic wrote:

> I particularly liked:
> 
> "Bundling of libraries with IDE, compiler or VM creates defacto standards
> and control points for language/platform vendors."
> 
> I think the article has some interesting things to say on the subject and
> ought to be a good read for those of us interested in getting a "standard"
> component library into Ada.

I wish he'd back up statements like the following:
"""
While there are many outstanding open source contributions there is 
naturally a much larger number of poor quality implementations. The
Open Source GUI libraries such as Motif and GTK are both problematic
and are in large part responsible for the lack of Linux use on the desktop.
"""

The author doesn't provide any basis for any of the assertions in
this paragraph.  Does the word "monopoly" mean anything?  Besides,
Motif is only recently "open source"; way too late for consideration
by "open source" folks switching to it from KDE/QT or GNOME/GTK.

The author is apparently arguing for a consortium type organization
with license agreements, etc...  This is exactly what kept Linux
folks from using Motif (and why GTK was created).  Funny, he seems
to be in the business of being a middleware middleman...

Anyway, I don't much care for journal articles that make lots of
unsubstantiated arguments and assertions; the author fails to 
provide even one reference to back up any of his arguments.  As 
such, it's just some guy's opinion. Granted, it's a guest column
and not a "research ariticle" -- kind of stretching the idea of
a "journal" publication by having guest columns... just call it
a magazine!

Re: OO & Standard Libraries

[Marin David Condic]

Well, it borders on a "Puff Piece" - a thinly veiled advertisement to
promote the organization the guy is involved with. As such, it doesn't
really have the credibility of a truly scientific research paper, but that
doesn't mean it doesn't have some valid observations.

As for a "consortium type organization" - I don't really have a problem with
that. It is an idea that has some merit. Programmers develop things that are
made widely available under some kind of easy terms and then share in the
rewards if/when any arise. Its a sort of "partnership" or "shared ownership
interest" that could stimulate development activity. It can be a good thing
to offer someone some financial reward for putting in the sweat equity to
develop a product.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com


"Eric G. Miller" <egm2@jps-nospam.net> wrote in message
news:pan.2002.05.08.20.46.44.866374.2597@jps-nospam.net...
>
> The author is apparently arguing for a consortium type organization
> with license agreements, etc...  This is exactly what kept Linux
> folks from using Motif (and why GTK was created).  Funny, he seems
> to be in the business of being a middleware middleman...
>