Re: JAVA and ADA JGNAT

Re: JAVA and ADA JGNAT

[David Starner]

On Tue, 18 Jan 2000 18:29:00 -0500, Mark Burge <mburge@acm.org> wrote:
>Does anyone know where to get a copy of JGNAT (www.gnat.org) which is
It's out there if you pay for it. So far they haven't made a public
release, and no one who has it wants to icurr the wrath (annoyance?)
of ACT enough to release a copy.

-- 
David Starner - dstarner98@aasaa.ofe.org
If you wish to strive for peace of soul then believe; 
if you wish to be a devotee of truth, then inquire.
   -- Friedrich Nietzsche

JAVA and ADA JGNAT

[Mark Burge]

Does anyone know where to get a copy of JGNAT (www.gnat.org) which is
described as: "The JGNAT system comprises a compiler generating Java
bytecode that is compatible with Java virtual machines conforming to Sun's
standard (JDK 1.1 and above), and a set of tools to aid in developing Ada
programs for the Java platform. JGNAT supports the development of both
applications and applets."  They say it is being distributed: "Following the
standard policy of Ada Core Technologies and ACT Europe, the entire JGNAT
system is being distributed under the GNU GPL licensing scheme." but I am
unable to find a copy anywhere.

Re: JAVA and ADA JGNAT

[Gautier]

Mark Burge wrote:

> Does anyone know where to get a copy of JGNAT (www.gnat.org) which 
                                                 ^^^^^^^^^^^^
Interesting: this address leads to GNAT, Inc.,
"Global Network of Astronomical Telescopes", a non-profit organization !

-- 
Gautier

_____\\________________\_______\
http://members.xoom.com/gdemont/

Re: JAVA and ADA JGNAT

[Ed Falis]

In article <862t3o$9aa1@news.cis.okstate.edu>,
  dstarner98@aasaa.ofe.org wrote:
> On Tue, 18 Jan 2000 18:29:00 -0500, Mark Burge <mburge@acm.org> wrote:
> >Does anyone know where to get a copy of JGNAT (www.gnat.org) which is
> It's out there if you pay for it. So far they haven't made a public
> release, and no one who has it wants to icurr the wrath (annoyance?)
> of ACT enough to release a copy.

So try friggin' appletmagic.  It's been available _free_ for a couple of
years now at www.appletmagic.com and at www.aonix.com as part of the windows
compiler download, and probably for the other platforms as well.  Words or
reality.  Take your pick.

- Ed


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[David Starner]

On Wed, 19 Jan 2000 05:00:41 GMT, Ed Falis <falis@ma.aonix.com> wrote:
>In article <862t3o$9aa1@news.cis.okstate.edu>,
>  dstarner98@aasaa.ofe.org wrote:
>> On Tue, 18 Jan 2000 18:29:00 -0500, Mark Burge <mburge@acm.org> wrote:
>> >Does anyone know where to get a copy of JGNAT (www.gnat.org) which is
>> It's out there if you pay for it. So far they haven't made a public
>> release, and no one who has it wants to icurr the wrath (annoyance?)
>> of ACT enough to release a copy.
>
>So try friggin' appletmagic.  It's been available _free_ for a couple of
>years now at www.appletmagic.com and at www.aonix.com as part of the windows
>compiler download, and probably for the other platforms as well.  Words or
>reality.  Take your pick.

Calm down. He asked about JGNAT, so I answered about JGNAT. It's not like
ACT has been spamming the newsgroup about this program; most of the noise
has been user made. And as for "Words or reality" - not only is AppletMagic
not free in the GNUdist sense, a time-limited trial version isn't free in
a useful sense either. 

-- 
David Starner - dstarner98@aasaa.ofe.org
If you wish to strive for peace of soul then believe; 
if you wish to be a devotee of truth, then inquire.
   -- Friedrich Nietzsche

Re: JAVA and ADA JGNAT

[Ed Falis]

In article <863hm1$8q61@news.cis.okstate.edu>,
  dstarner98@aasaa.ofe.org wrote:
> Calm down. He asked about JGNAT, so I answered about JGNAT. It's not like
> ACT has been spamming the newsgroup about this program; most of the noise
> has been user made. And as for "Words or reality" - not only is AppletMagic
> not free in the GNUdist sense, a time-limited trial version isn't free in
> a useful sense either.

Yes, I got carried away.  My apology to the group.

I didn't realize the Averstar version was time-limited.  The ObjectAda
version is not, but there's a limit of a couple hundred units per
application.

- Ed


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Preben Randhol]

Gautier <gautier.demontmollin@maths.unine.ch> writes:

| Mark Burge wrote:
| 
| > Does anyone know where to get a copy of JGNAT (www.gnat.org) which 
|                                                  ^^^^^^^^^^^^
| Interesting: this address leads to GNAT, Inc.,
| "Global Network of Astronomical Telescopes", a non-profit organization !
                                               ^^^^^^^^^^^^^^^^^^^^^^^^^

That's why I like .org over .com :-)

-- 
Preben Randhol -- [randhol@pvv.org] -- [http://www.pvv.org/~randhol/]     
         "Det eneste trygge stedet i verden er inne i en fortelling." 
                                                      -- Athol Fugard

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <862t3o$9aa1@news.cis.okstate.edu>,
  dstarner98@aasaa.ofe.org wrote:
> It's out there if you pay for it. So far they haven't made a
> public release, and no one who has it wants to icurr the wrath
> (annoyance?) of ACT enough to release a copy.

Actually the situation is as follows. JGNAT is not yet available
either in a public release or as a supported product (no one
has yet paid for it :-)

Rather it is in an intensive beta testing mode at the moment,
and a number of our customers and some other people who work
closely with us are helping us with this beta testing.

The testing is going well, but there are still some glitches
that must be fixed before a release. Probably the main one is
that applets can only be run under the applet viewer and not
from browsers, we really feel that this problem must be fixed
before a general release.

It would indeed be inappropriate to release the beta version
generally, and that is why it has not been done (by us or
anyone else).

It is never helpful to release software prematurely. The fact
that software is licensed under the GPL does not somehow change
this fundamental observation!

I know that it is frustrating to a lot of folks that JGNAT is
not out there yet, it's frustrating to us here at ACT here too,
but we never operate in the mode of releasing things before we
think they are ready.

The good news is that we are really VERY close now to the
release, which will include both a commercial version of
JGNAT Professional, and a public release of JGNAT. We will
post an announcement on CLA when it becomes available.

Robert Dewar
Ada Core Technologies

appropriately


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[David Starner]

On 25 Jan 2000 19:06:02 +0100, Preben Randhol <randhol@pvv.org> wrote:
>Robert Dewar <robert_dewar@my-deja.com> writes:
>
>| It is never helpful to release software prematurely. The fact
>| that software is licensed under the GPL does not somehow change
>| this fundamental observation!
>
>I agree wholeheartedly with this.

"It is never helpful to release software prematurely", i.e.
"it is never helpful to release software before it should be
released" is a tautology. So I'd have to agree with the statement
too.

But, in the context, I'd have to disagree. Many good products get
released early in the development cycle to the public, to no
harm to anyone. GCC, Linux, most open source projects, for
example. While ACT's lack of openness compared to those projects
means a beta release wouldn't help ACT as much as it does
Linus and the GCC team, I can't see why it would hurt more
then it does there. It will certainly work for some people,
and those who doesn't work for can read the beta label and
go on.

-- 
David Starner - dstarner98@aasaa.ofe.org
If you wish to strive for peace of soul then believe; 
if you wish to be a devotee of truth, then inquire.
   -- Friedrich Nietzsche

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <86kpbu$aik1@news.cis.okstate.edu>,
  dstarner98@aasaa.ofe.org wrote:
> While ACT's lack of openness compared to those projects
> means a beta release wouldn't help ACT as much as it does
> Linus and the GCC team, I can't see why it would hurt more
> then it does there.

Actually I think part of what goes on here is that ACT is
*more* open than a lot of the Linux and GCC development.
Take GCC, it is no secret that Cygnus does a LOT of GCC
development that is closely held before being made public
and often closely held for a long time. The same is true
of course for Linux developments at Redhat. The difference
is that we are quite open about the general status of our
internal developments, and we share roadmaps.

It's never helpful to have testing jump too far ahead of
development. In the case of JGNAT, the appropriate stage
for the last couple of months has been to have a selected
small number of beta testers kicking the tires.

The next step will be a general beta release, that corresponds
to the sort of thing David Starner is talking about. Note that
David has absolutely zero knowledge of the state of JGNAT right
now, so he is hardly in a position to make a judgment on the
right point at which to start general beta testing.

If David is saying that ALL developments should be made
completely open day by day, all I can say is that I don't know
of many open source or free software development projects that
work that way, with the possible exception of GNOME (and a
number of small scale projects). I definitely think that would
not be helpful to GNAT users. Yes, it might be fun for a few
enthusiasts and hobbyists, but the confusion of having lots
and lots of versions of GNAT around, most of them being works
in progress that were non-functional would not in our judgment
be helpful to the general Ada community, and that is our
primary constituency as far as the public release goes.

We certainly are NOT waiting until JGNAT is 100% finished and
validated and bug free etc. Indeed the coming public release
will very definitely still be a beta version as far as we are
concerned.

Robert Dewar
Ada Core Technologies


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Preben Randhol]

Robert Dewar <robert_dewar@my-deja.com> writes:

| It is never helpful to release software prematurely. The fact
| that software is licensed under the GPL does not somehow change
| this fundamental observation!

I agree wholeheartedly with this.

-- 
Preben Randhol -- [randhol@pvv.org] -- [http://www.pvv.org/~randhol/]     
         "Det eneste trygge stedet i verden er inne i en fortelling." 
                                                      -- Athol Fugard

Re: JAVA and ADA JGNAT

[David Starner]

On Tue, 25 Jan 2000 23:06:06 GMT, Robert Dewar <robert_dewar@my-deja.com> wrote:
>Actually I think part of what goes on here is that ACT is
>*more* open than a lot of the Linux and GCC development.

No. ACT never releases beta versions to the general public, and 
does not have development lists open to the public, or even readable
to the public. I, as a developer, wouldn't work on GNAT because
of this. I, as a developer, have found and fixed a few minor bugs
in GCC, because I could.

>Take GCC, it is no secret that Cygnus does a LOT of GCC
>development that is closely held before being made public
>and often closely held for a long time. The same is true
>of course for Linux developments at Redhat. 

The way I heard it, any Cygnus-only developments have to be specially
marked in the internal tree, and they try to get new stuff in the 
public distribution as soon as it's paid for. Be that as it may,
Cygnus is not GCC's maintainer, and Red Hat not Linux's. For GCC, 
a steering commitee is in charge of it, and Cygnus' developments get
looked at the same way as anyone else's. No matter what they do inside
the company, the public GCC developments are public, and show no signs
of being a facade. Mark Mitchell, of CodeSourcery, is a prime example
that GCC is not controlled by Cygnus.

>It's never helpful to have testing jump too far ahead of
>development. In the case of JGNAT, the appropriate stage
>for the last couple of months has been to have a selected
>small number of beta testers kicking the tires.

"With enough eyes, all bugs become shallow." (Linus Torvald)
It's not improbable that the applet bug would have been fixed, had 
the right person checked out JGNAT. 

>The next step will be a general beta release, that corresponds
>to the sort of thing David Starner is talking about. 

No, I'm talking about development snapshots and stuff. 

>If David is saying that ALL developments should be made
>completely open day by day, all I can say is that I don't know
>of many open source or free software development projects that
>work that way, with the possible exception of GNOME (and a
>number of small scale projects).

Ironically, I think most small scale projects don't, because they
don't have the resources and interest to mount a CVS server and
mailing list. But the mainstream of many projects work that way, GCC
and Linux especially. The main development source is out there to 
study - sure, someone may have been developing this over here, and 
someone else this that hasn't been merged in yet, but the main flow
is there. And the head developers usually aren't happy with this side
developments, because they make merging in terrible.

>I definitely think that would
>not be helpful to GNAT users. Yes, it might be fun for a few
>enthusiasts and hobbyists, but the confusion of having lots
>and lots of versions of GNAT around, most of them being works
>in progress that were non-functional

Why is this a worry? It doesn't seem to have been a major problem
for Linux and GCC. ACT also has the advantage of a smaller community
that isn't as likely to abuse the privilage.

I'm not espoucing anything that radical, just basically the
whole Cathedral and Bazaar stuff.

-- 
David Starner - dstarner98@aasaa.ofe.org
If you wish to strive for peace of soul then believe; 
if you wish to be a devotee of truth, then inquire.
   -- Friedrich Nietzsche

Re: JAVA and ADA JGNAT

[Alfred Hilscher]

Robert Dewar wrote:
> now, so he is hardly in a position to make a judgment on the
> right point at which to start general beta testing.

The right point for releasing would be if you have coded:

procedure JGNAT is
begin
  null;         --  ;-)
end JGNAT;

By the way, will the compiler itself run on the JVM , too ?

Re: JAVA and ADA JGNAT

[Gautier]

Florian Weimer:
> XEmacs (as opposed to Emacs) and Mozilla are among the largest free
> software projects and are managed in the Bazaar style.

Good. But does it hold for a compiler, and for a language that should
care about safety ? The tolerance towards a browser crashing (for the
nth time) on a SGI station or the 567th Linux patch is not the same
as towards code generated by an Ada compiler or the compiler itself.
I can tell that people who have worked with high-end products (compilers,
debuggers) get strange colours (RGB values on demand) when an Ada
thing crashes. 

-- 
Gautier

_____\\________________\_______\
http://members.xoom.com/gdemont/

Re: JAVA and ADA JGNAT

[Jean-Marc Bourguet]

In article <877lgxuquu.fsf@deneb.cygnus.argh.org>,
  Florian Weimer <fw@s.netic.de> wrote:
> Robert Dewar <robert_dewar@my-deja.com> writes:
>
> > Actually I think part of what goes on here is that ACT is
> > *more* open than a lot of the Linux and GCC development.
>
> If you mean Linux = kernel, this is wrong, I think. Release cycles are
> very short, and if there's a fix for a particular problem, you can get
it
> at once and don't have to wait for the next release.

> In the past, I've struggled with several GNAT bugs although they had
> already been fixed in ACT's internal version.

And how many bugs have you not struggled with because the quality
control of ACT is better that what it is possible to do with a
more open development?

I've worked on compilers. The most important thing when you develop
such kind of program is the test suites, and the test suites for a
compiler can not be made public because they contain code coming from
the ACT customers and from DEC for which ACT do not have the right.
I'd not like do work on GNAT without beeing able to run those tests.
Yes, I may fix my problem, but sure I'll break a lot of other things.

Open development has a cost (beeing able to get the same quality
is one, there are other). I do not know if the assertion of ACT
that for GNAT the cost would outfit the benefice is valid. I only
know that I've no data to say otherwise.

The only people who seemed to be motivated enough to work on gnat
(and not only wanting access to the version of the day), are the
members of ALT. And I understand that they got support from ACT.

> Maybe you can obtain these fixes if you've got a support contract,
> but with other free software, this is not necessary.

From what I've gotten here and on chat, it is quite difficult to get
access to the internal version. You'd have to show a bug which stop
you (i.e. no workaround) which is fixed. I think the value of the
support contract is not the access to the internal version, but access
to R. Dewar to ask him questions and get an answer.

-- Jean-Marc


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[David Starner]

On 26 Jan 2000 11:25:27 +0100, Preben Randhol <randhol@pvv.org> wrote:
>dvdeug@x8b4e53cd.dhcp.okstate.edu (David Starner) writes:
>
>| But, in the context, I'd have to disagree. Many good products get
>| released early in the development cycle to the public, to no
>| harm to anyone. GCC, Linux, most open source projects, for
>                              ^^^^^^^^^^^^^^^^^^^^^^^^^
>
>Which is also a problem with a lot of open source software. That is a
>lot of them never makes it past the "I have a plan for a new app"
>announcement and then the project dies out. 

I'd disagree. The problem is not that they're open, but that nobody did
any work. It would have happened anyway, you just would never had heard
about it. Linux and Debian GNU/Linux are two examples of projects that
were open from the barest beginnings, and were successful partially
because of it.

>I think that one have overstressed the bazaar angle a bit to much. One
>needs to at least get to the stage where the program is reasonably
>usable to start attract users and then try to get more people to
>help. 

Usually. I think our definitions of reasonably usable are different,
though. Again, the Linux kernel was released at 0.0.1 and it worked
well enough to attract developers (which is what's important to the
success of a project, not users).

>As for compilers I think this is even more crucial. I would rather
>have a JGNAT that works well than an alpha product that makes the
>programs crash a lot and thus gets a bad reputation. 

Sure. But I'd rather have a JGNAT than no JGNAT. And I don't
find people getting much respect for complaining about a product
that was marked ALPHA. I don't think it will get a bad reputation
for alpha-class behavior as an alpha product.

>But it might be
>that I for general OS products expect them to be alpha or beta and
>crash a lot, I don't expect compilers to do the same. :-)

Then stop using the ones marked alpha or beta. I've seen very few
open source products not marked alpha or beta that were not stable,
and I've found many marked such that were. YMMV.

-- 
David Starner - dstarner98@aasaa.ofe.org
If you wish to strive for peace of soul then believe; 
if you wish to be a devotee of truth, then inquire.
   -- Friedrich Nietzsche

Re: JAVA and ADA JGNAT

[Ted Dennison]

In article <m3emb5gmeg.fsf@kiuk0156.chembio.ntnu.no>,
  Preben Randhol <randhol@pvv.org> wrote:
> As for compilers I think this is even more crucial. I would rather
> have a JGNAT that works well than an alpha product that makes the
> programs crash a lot and thus gets a bad reputation. But it might be
> that I for general OS products expect them to be alpha or beta and
> crash a lot, I don't expect compilers to do the same. :-)

When Gnat was first under development it was released early and often. I
remember trying out a version sometime around '94 that had no tasking
support and was riddled with bugs. Even in '95 it was not uncommon to
get bugs in simple assignments that involved no new Ada95 features. I
think Gnat *did* get somewhat of a bad reputation from those early
releases. So perhaps ACT's current policy is a reaction to that.

--
T.E.D.

http://www.telepath.com/~dennison/Ted/TED.html


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Pascal Martin]

In article <86mqi6$6dd$1@nnrp1.deja.com>, Jean-Marc Bourguet <bourguet@my-deja.com> wrote:

> And how many bugs have you not struggled with because the quality
> control of ACT is better that what it is possible to do with a
> more open development?

This is the reverse, actually: GNAT is very difficult to build. IMHO GNAT
would have tremendeously benefited from an open development model,
as it would have forced GNAT developpers to fix their build process.
I used the first Gnome versions, and it was a nightmare because of
the build process: whatever program I downloaded would not work
because the third version digit of GTK did not match or because a very
obscure library was missing (no list available). Since then, they fixed it
and that makes Gnome usable & much more attracting (at least to me).

No such thing seems to happen to GNAT. I suspect ACT is protecting its
business using FUD and obscurity: "if you try to build your compiler yourself,
be warned". GNAT is complicated to build, and when you are in trouble,
Dewar put the sales hat on. And it is not cheap: the GNAT trap. Look like 
also they don't want anyone to compete with them.

There is nothing wrong with ACT doing business. But you have to realize
the not-so-open mindset: ACT is not a communauty, it is a for-profit
business. Don't idealize them. They are not the only ones: sendmail is 
moving toward this direction too.

ACT could actually work a different way, thanks to the Ada certification
process: certifying an Ada compiler version takes time and money.
Selling official Ada version seems a valid business model to me, even
if snapshot are released to the public.

If safety is your concerne (rightly !) you will, of course, only use a well 
identified and managed version of GNAT. Why in the world does that 
means that everyone else would be banned from using other versions ?

Linux version are very well identified. The odd release number strategy
is working nicely (it is simple) and the even version is both maintained
and managed. Nobody is going to use an "odd release" in a "mission
critical" application by mistake, except by stupidity, which I hope is not 
too common in the safety critical communauty.

I believe GNAT is Dewar's brainchild, and he has a parental crisis: he
do not want the kid to leave the house. Is he still reviewing all the
GNAT code that's checked in ?

------------------------------------------------------------------
Pascal F. Martin.

Re: JAVA and ADA JGNAT

[David Starner]

On Wed, 26 Jan 2000 17:17:43 GMT, Pascal Martin <pascal.martin@iname.com.nospam> wrote:
>This is the reverse, actually: GNAT is very difficult to build. IMHO GNAT
>would have tremendeously benefited from an open development model,
>as it would have forced GNAT developpers to fix their build process.

To be fair, the build process is that of GCC 2.8.1. When they upgrade
to 2.9x or 3.0, it will be much easier. My big complaint about GNAT is
that it builds on one version of one compiler, whereas GCC builds on
just about any Unix C compiler you can dig up, including things that
don't even try to conform to the 10 year old standard.

-- 
David Starner - dstarner98@aasaa.ofe.org
If you wish to strive for peace of soul then believe; 
if you wish to be a devotee of truth, then inquire.
   -- Friedrich Nietzsche

Re: JAVA and ADA JGNAT

[David Starner]

On Wed, 26 Jan 2000 11:49:59 +0000, Gautier <gautier.demontmollin@maths.unine.ch> wrote:
>Florian Weimer:
>> XEmacs (as opposed to Emacs) and Mozilla are among the largest free
>> software projects and are managed in the Bazaar style.
>
>But does it hold for a compiler, and for a language that should
>care about safety ? 

Look at GCC. It's going fairly well, and has fixed many serious bugs
in the 2.8.1 series of GCC that GNAT depends on.

-- 
David Starner - dstarner98@aasaa.ofe.org
If you wish to strive for peace of soul then believe; 
if you wish to be a devotee of truth, then inquire.
   -- Friedrich Nietzsche

Re: JAVA and ADA JGNAT

[David Starner]

On 26 Jan 2000 19:44:59 +0100, Preben Randhol <randhol@pvv.org> wrote:
>| Usually. I think our definitions of reasonably usable are different,
>| though. Again, the Linux kernel was released at 0.0.1 and it worked
>| well enough to attract developers (which is what's important to the
>| success of a project, not users).
>
>Well my point exactly, it worked well enough to attract developers. It
>was then not premature...

That's begging the point. I would argue that any program that does
something not done before in free software would be not premature
once it runs even the minimal stuff.

>| Then stop using the ones marked alpha or beta. I've seen very few
>| open source products not marked alpha or beta that were not stable,
>| and I've found many marked such that were. YMMV.
>
>One that was released prematurely _as stable_ twice, is the Gnome
>Project http://www.gnome.org. 

Fair enough. But that's the exception, not the rule. I would note
that the parts of GNOME marked stable are stable now.

-- 
David Starner - dstarner98@aasaa.ofe.org
If you wish to strive for peace of soul then believe; 
if you wish to be a devotee of truth, then inquire.
   -- Friedrich Nietzsche

Re: JAVA and ADA JGNAT

[Aidan Skinner]

On Tue, 25 Jan 2000 23:06:06 GMT, Robert Dewar
<robert_dewar@my-deja.com> wrote: 

>and often closely held for a long time. The same is true
>of course for Linux developments at Redhat. The difference

I've heard this a lot, and I've never been *entirely* convinced about
it, especially given that redhat make their in-development
distribution available via ftp... 

Of course, there's no way to be sure (without working for Redhat), but
I've yet to see any evidence that suppourts this.

Abscence of evidence is not evidence of abscence after all.

>development. In the case of JGNAT, the appropriate stage
>for the last couple of months has been to have a selected
>small number of beta testers kicking the tires.

OTOH there isn't anything inherently wrong with the cathedral
development model, and it's necessary in a lot of cases (eg. it is,
IMO, entirely justified in XFree86)

>of many open source or free software development projects that
>work that way, with the possible exception of GNOME (and a
>number of small scale projects). I definitely think that would

The Linux kernel is close to this, in that releases of the
development tree are fairly frequent.

>in progress that were non-functional would not in our judgment
>be helpful to the general Ada community, and that is our
>primary constituency as far as the public release goes.

In this case I would tend to think that this is the right approach,
simply because of the complexity of GNAT and the fact that it's a
relatively monolithic application.

I think that the structure of something is more important than it's
size, GNOME benefits from a bazaar model because it's a collection of
things which occasionally interoperate and don't depend on each other
and there's a clear distinction between gnome-core, gnome-libs
etc. and the various parts of gnome-libs aren't interdependent.

- Aidan
-- 
Little Willy was a chemist, Little Willy is no more,
What he thought was H2O, Was H2SO4.
http://www.skinner.demon.co.uk/aidan/

Re: JAVA and ADA JGNAT

[Aidan Skinner]

On Wed, 26 Jan 2000 17:17:43 GMT, Pascal Martin
<pascal.martin@iname.com.nospam> wrote: 

>This is the reverse, actually: GNAT is very difficult to build. IMHO GNAT
>would have tremendeously benefited from an open development model,

This is one of the things that the ALT project, Juergen in particular,
has been trying to fix. Installing gnat on a linux machine from
tarballs can be _painful_, from rpms/debs it's painless.

>No such thing seems to happen to GNAT. I suspect ACT is protecting its
>business using FUD and obscurity: "if you try to build your compiler yourself,

If you've got a real problem with this, and want to fix it, well, you
can get the source code (it's under the GPL remember) and fork it.

Whether or not this would be helpful is something that you've got to
consider.

- Aidan

-- 
Little Willy was a chemist, Little Willy is no more,
What he thought was H2O, Was H2SO4.
http://www.skinner.demon.co.uk/aidan/

Re: JAVA and ADA JGNAT

[Florian Weimer]

Robert Dewar <robert_dewar@my-deja.com> writes:

> Actually I think part of what goes on here is that ACT is
> *more* open than a lot of the Linux and GCC development.

If you mean Linux = kernel, this is wrong, I think. Release cycles are
very short, and if there's a fix for a particular problem, you can get it
at once and don't have to wait for the next release.  In the past, I've
struggled with several GNAT bugs although they had already been fixed in
ACT's internal version.  Maybe you can obtain these fixes if you've got
a support contract, but with other free software, this is not necessary.

> If David is saying that ALL developments should be made
> completely open day by day, all I can say is that I don't know
> of many open source or free software development projects that
> work that way, with the possible exception of GNOME (and a
> number of small scale projects). 

XEmacs (as opposed to Emacs) and Mozilla are among the largest free
software projects and are managed in the Bazaar style.  The Linux
kernel is similar, but in a way completely different, because there are
a couple of reviewers (with Linus at the top) who decide which goes in
to the standard distribution.  Nowadays, most free software projects of
a certain size offer public read-only access to a CVS repository, and
usually, it isn't a problem to get write access if you want to contribute
code (perhaps after signing a copyright assignment to the FSF).

Re: JAVA and ADA JGNAT

[Preben Randhol]

dvdeug@x8b4e53cd.dhcp.okstate.edu (David Starner) writes:

| But, in the context, I'd have to disagree. Many good products get
| released early in the development cycle to the public, to no
| harm to anyone. GCC, Linux, most open source projects, for
                              ^^^^^^^^^^^^^^^^^^^^^^^^^

Which is also a problem with a lot of open source software. That is a
lot of them never makes it past the "I have a plan for a new app"
announcement and then the project dies out. 

I think that one have overstressed the bazaar angle a bit to much. One
needs to at least get to the stage where the program is reasonably
usable to start attract users and then try to get more people to
help. 

As for compilers I think this is even more crucial. I would rather
have a JGNAT that works well than an alpha product that makes the
programs crash a lot and thus gets a bad reputation. But it might be
that I for general OS products expect them to be alpha or beta and
crash a lot, I don't expect compilers to do the same. :-)

-- 
Preben Randhol -- [randhol@pvv.org] -- [http://www.pvv.org/~randhol/]     
         "Det eneste trygge stedet i verden er inne i en fortelling." 
                                                      -- Athol Fugard

Re: JAVA and ADA JGNAT

[Preben Randhol]

dvdeug@x8b4e53cd.dhcp.okstate.edu (David Starner) writes:

| I'd disagree. The problem is not that they're open, but that nobody did
| any work. It would have happened anyway, you just would never had heard
| about it. Linux and Debian GNU/Linux are two examples of projects that
| were open from the barest beginnings, and were successful partially
| because of it.

1. I didn't say that open software is bad, I prefer that very much to
   closed software, but I said that premature releases are bad.

2. I don't think a project that releases the source at say version
   0.30 is any worse off than one that releases it at 0.0.1 where
   nothing usually works at all. 

| Usually. I think our definitions of reasonably usable are different,
| though. Again, the Linux kernel was released at 0.0.1 and it worked
| well enough to attract developers (which is what's important to the
| success of a project, not users).

Well my point exactly, it worked well enough to attract developers. It
was then not premature...

| Sure. But I'd rather have a JGNAT than no JGNAT. And I don't

As I see it there is no black and white situation here, only that you
have to wait a bit longer for the product.

| Then stop using the ones marked alpha or beta. I've seen very few
| open source products not marked alpha or beta that were not stable,
| and I've found many marked such that were. YMMV.

One that was released prematurely _as stable_ twice, is the Gnome
Project http://www.gnome.org. 

One thing Open Source Projects and deadlines do not mix. :-) 

-- 
Preben Randhol -- [randhol@pvv.org] -- [http://www.pvv.org/~randhol/]     
         "Det eneste trygge stedet i verden er inne i en fortelling." 
                                                      -- Athol Fugard

Re: JAVA and ADA JGNAT

[Florian Weimer]

Gautier <gautier.demontmollin@maths.unine.ch> writes:

> > XEmacs (as opposed to Emacs) and Mozilla are among the largest free
> > software projects and are managed in the Bazaar style.
> 
> Good. But does it hold for a compiler, and for a language that should
> care about safety ? 

I don't know.  There is the argument that the more people actively work
with the code, the more bugs are found.  And I think some of the free
*BSD variants have got quite liberal policies regarding CVS access,
but the OS is considered to be extremly stable.

Of course, when it comes to safety-critical software (in the Ada sense),
the synergy effects of the bazaar won't be very effective.  Looking at
code of very different sources, most free software hackers seem to
prefer quick over clean solutions. ;)

> The tolerance towards a browser crashing (for the
> nth time) on a SGI station or the 567th Linux patch is not the same
> as towards code generated by an Ada compiler or the compiler itself.

I don't think that less open (or more closed ;) software development
presults in higher software quality per se.

Re: JAVA and ADA JGNAT

[Florian Weimer]

Jean-Marc Bourguet <bourguet@my-deja.com> writes:

> In article <877lgxuquu.fsf@deneb.cygnus.argh.org>,
>   Florian Weimer <fw@s.netic.de> wrote:
> > Robert Dewar <robert_dewar@my-deja.com> writes:
> >
> > > Actually I think part of what goes on here is that ACT is
> > > *more* open than a lot of the Linux and GCC development.

I've misread this statement, I think.  The explanation by Robert Dewar
that followed but which I neglected to quote made it quite clear that
he meant that most of the code Cygnus develops for their customers
never makes it into the public GCC version, and this seems to be true.

On the other hand, the development process of the *public* GCC version
hosted by Cygnus is much more open than the GNAT development process.
When I posted my reply, I was focused on this one, I think.

> > In the past, I've struggled with several GNAT bugs although they had
> > already been fixed in ACT's internal version.
> 
> And how many bugs have you not struggled with because the quality
> control of ACT is better that what it is possible to do with a
> more open development?

To be honest: I don't know.  That's the reason why I was very careful
not to give any advice to anyone on how to structure their development
process.  I only was irritated by Robert Dewar's statement of the
openness of ACT because I confused the development process with the
result.

> Open development has a cost (beeing able to get the same quality
> is one, there are other). 

Yes, that's certainly true.  Given the size of the free software Ada
community, it's probably better to invest manpower in actual GNAT
development than to support a potential bazaar directly at ACT by
maintaining a CVS repository and related things.

Re: JAVA and ADA JGNAT

[Jean-Marc Bourguet]

In article <X_Fj4.425$0o4.13328@cmnws01.we.mediaone.net>,
  Pascal Martin <pascal.martin@iname.com.nospam> wrote:
> In article <86mqi6$6dd$1@nnrp1.deja.com>, Jean-Marc Bourguet
<bourguet@my-deja.com> wrote:
>
> > And how many bugs have you not struggled with because the quality
> > control of ACT is better that what it is possible to do with a
> > more open development?
>
> This is the reverse, actually: GNAT is very difficult to build.

First, I see little correlation between the quality of a program and
the difficulty to build it, especially when like GNAT parts of the
program is written in different languages.

Secondly, the only time I tried to compile GNAT, I succeeded without
problem, so I can't even agree with the fact the GNAT is difficult
to build.

> IMHO GNAT would have tremendeously benefited from an open development
> model, as it would have forced GNAT developpers to fix their build
> process.

What would have be the benefit of a simpler build process? Until now
people complaining on the "closed" gnat development are complaining not
because they can't work on gnat (ALT proove that it is possible) but
that they can't get the very latest version.

[...]
> I believe GNAT is Dewar's brainchild, and he has a parental crisis: he
> do not want the kid to leave the house. Is he still reviewing all the
> GNAT code that's checked in ?

I sure hope all the code checked in GNAT is reviewed. I sure hope that
the new feature are designed and then coded (by the way, if I wanted
to work on gnat, it is more for the lack of avaibillity of design
documents that I'd be complaining).

Yours,

-- Jean-Marc


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Ted Dennison]

Jean-Marc Bourguet wrote:

> First, I see little correlation between the quality of a program and
> the difficulty to build it, especially when like GNAT parts of the
> program is written in different languages.

Good point. I consider Emacs a quality product, but have you ever tried to
compile it from sources? I've done it 3 times now, and I'd take putting
together a 50 pice swing-set over that any day.

--
T.E.D.

Home - mailto:dennison@telepath.com  Work - mailto:dennison@ssd.fsi.com
WWW  - http://www.telepath.com/dennison/Ted/TED.html  ICQ  - 10545591

Re: JAVA and ADA JGNAT

[Pascal Obry]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1521 bytes --]


Pascal Martin <pascal.martin@iname.com.nospam> a �crit dans le message :
X_Fj4.425$0o4.13328@cmnws01.we.mediaone.net...
> No such thing seems to happen to GNAT. I suspect ACT is protecting its
> business using FUD and obscurity: "if you try to build your compiler
yourself,
> be warned". GNAT is complicated to build, and when you are in trouble,
> Dewar put the sales hat on. And it is not cheap: the GNAT trap. Look like
> also they don't want anyone to compete with them.

This is plain wrong.

Pascal, you should consider taking the sources and forking them. You can
then attrack
many developers and make a "more open" GNAT version.

Why don't you do that ?

I found that it is always too easy to complain and harder to fix things :) !

Pascal.

--

--|------------------------------------------------------------
--| Pascal Obry                               Team-Ada Member |
--|                                                           |
--| EDF-DER-IPN-SID- T T I                                    |
--|                       Intranet: http://cln46gb            |
--| Bureau N-023            e-mail: p.obry@der.edf.fr         |
--| 1 Av G�n�ral de Gaulle  voice : +33-1-47.65.50.91         |
--| 92141 Clamart CEDEX     fax   : +33-1-47.65.50.07         |
--| FRANCE                                                    |
--|------------------------------------------------------------
--|
--|         http://perso.wanadoo.fr/pascal.obry
--|
--|   "The best way to travel is by means of imagination"

Re: JAVA and ADA JGNAT

[Florian Weimer]

Pascal Martin <pascal.martin@iname.com.nospam> writes:

> This is the reverse, actually: GNAT is very difficult to build.

Only if there's no prebuilt binary version for the host platform.

Otherwise, it's not a simple `./configure; make; make install', but only
a few additional commands are required, and they are well-documented.
In particular, you don't have to build seven additional libraries first
which hardly anyone else uses, and edit the source code to reflect
recent changes in GCC's C and/or C++ implementation.  This isn't very
difficult either, but it's much more time-consuming.

Re: JAVA and ADA JGNAT

[Florian Weimer]

aidan@skinner.demon.co.uk (Aidan Skinner) writes:

> On Tue, 25 Jan 2000 23:06:06 GMT, Robert Dewar
> <robert_dewar@my-deja.com> wrote: 
> 
> >and often closely held for a long time. The same is true
> >of course for Linux developments at Redhat. The difference

> Of course, there's no way to be sure (without working for Redhat), but
> I've yet to see any evidence that suppourts this.

When some Cygnus customer donates code to the GCC project, it almost
always requires a considerable amount of work to merge it with the
current GCC tree.  This means that the GCC versions used by customers
probably differ a lot from the public version, and I think this is
a hint that a lot of work is done which never finds its way into the
public GCC tree.

Re: JAVA and ADA JGNAT

[Preben Randhol]

dvdeug@x8b4e53cd.dhcp.okstate.edu (David Starner) writes:

| That's begging the point. I would argue that any program that does
| something not done before in free software would be not premature
| once it runs even the minimal stuff.

No, but as I said I think that general Open Source projects are
different from Open Source compiler projects. I need to trust the
compiler, not necessarily the CD-player software I use.

An app that I used, but don't anymore until it gets stable, is
gnomecal.

It is nice and it seems to work, but I don't trust it. The reason is
that an earlier version suddenly stopped alerting me about my
appointments. It went OK as I remembered the appointment myself, but I
cannot say that I trust the software enough to rely on it now. Though
I think the problem has been fixed in the October Gnome release. 

If I write some code and the compiler complains or the program
crashes, I do not want to be unsure whether it is _me_, most likely,
that has done something wrong, or if this is a bug in the compiler.

| >One that was released prematurely _as stable_ twice, is the Gnome
| >Project http://www.gnome.org. 
| 
| Fair enough. But that's the exception, not the rule. I would note
| that the parts of GNOME marked stable are stable now.

It will be the rule if deadlines are put on OS projects I fear. That
is one of the advantages with none commercial products, one can choose
when to release it.

-- 
Preben Randhol -- [randhol@pvv.org] -- [http://www.pvv.org/~randhol/]     
         "Det eneste trygge stedet i verden er inne i en fortelling." 
                                                      -- Athol Fugard

Re: JAVA and ADA JGNAT

[Chris Morgan]

Ted Dennison <dennison@telepath.com> writes:

> Good point. I consider Emacs a quality product, but have you ever tried to
> compile it from sources? I've done it 3 times now, and I'd take putting
> together a 50 pice swing-set over that any day.

It depends on the tools you have around. Compiling it on
out-of-the-box Solaris is probably still not that nice, but once I put
the stuff onto Solaris I know I'm going to need anyway, it's as easy
as falling off a log. Honestly I think it takes about 10 minutes to go
from thinking "hmmm, wonder if there is a new version of Emacs out
yet" to kicking off the build, and then less than five once it
finishes to install it, clean up the build area etc.
-- 
Chris Morgan <cm at mihalis.net>                  http://mihalis.net

Re: JAVA and ADA JGNAT

[Ted Dennison]

Chris Morgan wrote:

> Ted Dennison <dennison@telepath.com> writes:
>
> > Good point. I consider Emacs a quality product, but have you ever tried to
> > compile it from sources? I've done it 3 times now, and I'd take putting
> > together a 50 pice swing-set over that any day.
>
> It depends on the tools you have around. Compiling it on
> out-of-the-box Solaris is probably still not that nice, but once I put
> the stuff onto Solaris I know I'm going to need anyway, it's as easy

Yeah. As I remember the last time I had to do it for a SunOS box (admittedly
over 5 years ago), I first ended up having to compile gcc from sources. And of
course there's always the usual assortment of #ifdefs and #defines that aren't
quite right and have to be hacked. Yuk.

--
T.E.D.

Home - mailto:dennison@telepath.com  Work - mailto:dennison@ssd.fsi.com
WWW  - http://www.telepath.com/dennison/Ted/TED.html  ICQ  - 10545591

Re: JAVA and ADA JGNAT

[Florian Weimer]

Chris Morgan <cm@mihalis.net> writes:

> Ted Dennison <dennison@telepath.com> writes:
> 
> > Good point. I consider Emacs a quality product, but have you ever tried to
> > compile it from sources? I've done it 3 times now, and I'd take putting
> > together a 50 pice swing-set over that any day.
> 
> It depends on the tools you have around. Compiling it on
> out-of-the-box Solaris is probably still not that nice, 

Especially because it's lacking a C compiler, isn't it? ;)

Emacs is great to use, but the C sources are a mess: K&R compatible
(at least they that way), tons of #ifdef's in the actual code, no
clear separation between OS-specific and generic routines, colored
pointers, the undump feature---just to name the most awkward things.
I only tried once to built it on a not-yet supported system (some
bleeding-edge Linux variant), but quickly gave up.

Re: JAVA and ADA JGNAT

[Stefan Skoglund]

Ted Dennison wrote:
> Yeah. As I remember the last time I had to do it for a SunOS box (admittedly
> over 5 years ago), I first ended up having to compile gcc from sources. And of
> course there's always the usual assortment of #ifdefs and #defines that aren't
> quite right and have to be hacked. Yuk.

If you had to hack the defines configure was probably broken.
The #ifdefs and so on shouldn't need to be touched.

I had to change the pool size.

Re: JAVA and ADA JGNAT

[Pascal F. Martin]

In article <86la8r$519$1@nnrp1.deja.com>,
	Robert Dewar <robert_dewar@my-deja.com> writes:
> Actually I think part of what goes on here is that ACT is
> *more* open than a lot of the Linux and GCC development.
> [...] The same is true
> of course for Linux developments at Redhat.

RedHat is _NOT_ developping Linux. Linux is being
developped by a bunch of individuals and integrated by
Linus Torvald, an engineer currently employed by Transmeta Inc.

RedHat develop a Linux distribution: this is kind of product
integration work (akin to software editor).

It is incredible how even computer science legends like Robert
can be victim of marketing image and branding. I have to work
with someone who also believed Linux was a product of RedHat
and that the HardHat distribution was the "official" version
of Linux for the embedded market, just because of the name
match (how clever !).

------------------------------------------------------------------
Pascal F. Martin.

Re: JAVA and ADA JGNAT

[reason67]

In article <au9l4.3239$0o4.37770@cmnws01.we.mediaone.net>,
  pmartin@mail.earthlink.net (Pascal F. Martin) wrote:
> > The same is true
> > of course for Linux developments at Redhat.
>
> RedHat is _NOT_ developping Linux. Linux is being
> developped by a bunch of individuals and integrated by
> Linus Torvald, an engineer currently employed by Transmeta
> Inc

Read the sentence again. He did not say that Linux was developed at Red
Hat. He said Linux developments at RedHat. Red Hat does develop software
that is used in Linux systems (RPM would be the most notable)
---
Jeffrey S. Blatt


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Ted Dennison]

In article <3893777B.9874AB19@ebox.tninet.se>,
  Stefan Skoglund <stetson@ebox.tninet.se> wrote:
> If you had to hack the defines configure was probably broken.
> The #ifdefs and so on shouldn't need to be touched.

It depends on what you consider "broken". As I remember a lot of
problems were caused by the fact that we had X and Motif installed in
odd places.

--
T.E.D.

http://www.telepath.com/~dennison/Ted/TED.html


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Ted Dennison]

In article <87d7qm7yxi.fsf@deneb.cygnus.argh.org>,
  Florian Weimer <fw@s.netic.de> wrote:
> Chris Morgan <cm@mihalis.net> writes:
>
> > It depends on the tools you have around. Compiling it on
> > out-of-the-box Solaris is probably still not that nice,
>
> Especially because it's lacking a C compiler, isn't it? ;)

Now, yes. But that wasn't always the case. I think I would have been
better off if it hadn't come with a C compiler. Things worked way better
once I installed gcc. I think a lot of it was library issues. We had a
rather incompetent sysadmin who didn't install all the C libraries I
think. Some of the ones he did install were in wierd places.

--
T.E.D.

http://www.telepath.com/~dennison/Ted/TED.html


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Preben Randhol]

pmartin@mail.earthlink.net (Pascal F. Martin) writes:

| In article <86la8r$519$1@nnrp1.deja.com>,
| 	Robert Dewar <robert_dewar@my-deja.com> writes:
| > Actually I think part of what goes on here is that ACT is
| > *more* open than a lot of the Linux and GCC development.
| > [...] The same is true
| > of course for Linux developments at Redhat.
| 
| RedHat is _NOT_ developping Linux. Linux is being
| developped by a bunch of individuals and integrated by
| Linus Torvald, an engineer currently employed by Transmeta Inc.

I think he was referring to development of Linux software and not the
kernel.

-- 
Preben Randhol -- [randhol@pvv.org] -- [http://www.pvv.org/~randhol/]     
         "Det eneste trygge stedet i verden er inne i en fortelling." 
                                                      -- Athol Fugard

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <86np5s$baq1@news.cis.okstate.edu>,
  dstarner98@aasaa.ofe.org wrote:

> Look at GCC. It's going fairly well, and has fixed many
> serious bugs in the 2.8.1 series of GCC that GNAT depends on.

Well I am not sure what bugs you are talking about. Actually
it has introduced a number of bugs that cause GNAT to
malfunction. We are working on repairing this now. The first
step was to take all the fixes for critical bugs that had been
done in the 2.8.1 thread, but NOT incorporated into gcc 2.95.
Without these fixes, there were indeed really serious problems
that would make GNAT unusable (many of these were problems that
could also show up in C code).

We are now working on merging the two technologies, and the
first step (merging the 2.8.1 changes into the 2.95 sources
has been completed by Richard Kenner). We are now working on
fixing the remaining bugs that have been introduced that
blow GNAT out of the water, and also on adapting GNAT to the
(in some cases unncessary) changes in the interface between the
front end and back end.

It is being a bigger job that we expected to get GCC 2.9x into
good enough shape to support GNAT, but this is definitely an
important goal.

By the way, we are not aware of any "serious bugs" in 2.8.1
that have been fixed in 2.95 that would affect GNAT. Yes, there
are lots of fixes for g++, but that's another story entirely.

Robert Dewar
Ada Core Technologies





Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <X_Fj4.425$0o4.13328@cmnws01.we.mediaone.net>,
  Pascal Martin <pascal.martin@iname.com.nospam> wrote:
> No such thing seems to happen to GNAT. I suspect ACT is
protecting its
> business using FUD and obscurity: "if you try to build your
compiler yourself,
> be warned". GNAT is complicated to build, and when you are in
trouble,
> Dewar put the sales hat on. And it is not cheap: the GNAT
trap. Look like
> also they don't want anyone to compete with them.

First, it is quite straightforward to build GNAT, lots of people
do it, the makefiles and procedures are well documented, and
anyone reasonably familiar with GCC can easily do these builds,
and many, many people build GNAT from sources. We don't work
on making this build "simpler", because it's simply not an
issue, I am not quite sure why Pascal Martin has had trouble,
but generally our experience has been that anyone with
reasonable gcc experience has no trouble, so it is not something
that is worth putting work into.

Actually, a lot of the users of the public version of GNAT
use versions not built by us, and that seems fine. We are
happy to cooperate with anyone doing such builds. For example,
ACT cooperates closely with the GNAT/Linux group that builds
RPM's for popular versions of GNU/Linux.

The policy of ACT is the same it has always been, all our
work is placed in the public tree as soon as it is in
releasable shape (we actually will go to more frequent
releases in the next year, as a result of the new contract
with SGI, which calls for quarterly releases -- we will try
to syncrhonize the public releases with these more frequent
GNAT Professional releases).

No company that I know of works by having its internal
development sources and development activity day by day
open, I think that would be chaotic. In particular, it is
quite right that not having access to the ACT test suite
makes it risky to make changes. We certainly never allow
even the most minor of changes to be made without running
the test suite, and this avoids many false steps that would
cause regressions.

It would be nice if there were more activity in contributions
to the public tree, but in practice that does not happen so
much for complex beasts like compilers. If you look at the
gcc changes, a great majority are made by Cygnus or full
time development folks at other companies.

And, despite "belief" sited earlier in this thread, I can
promise you that Cygnus does not work in bazarre mode
internally (anyone really seriously think that the new
Itanium compiler was developed that way? -- you probably
don't even know about it, even though the project has being
going on for quite a while now under non-disclosure -- and
that of course is the point).

Similarly, there are major developments that have not
seen the light of day yet from major Linux companies (and
I think I will use Linux there and not GNU/Linux quite
deliberately :-)

Does this mean that these evil companies are secretly doing
stuff that they should be doing openly? Not at all, it simply
means that these projects are under development, by a well
defined development team, using traditional high quality
software development techniques, but they are not ready for
any kind of distribution yet.

As for public beta tests, I know that this is a common concept
particularly from Microsoft, who even charges hundreds of
thousands of people to be beta testers.

But it's not the way we work for several reasons:

1. We think that beta testers need to be in close contact with
the developers, so that useful input is obtained in a systematic
way. Far too often "public beta" versions are just a way of
getting pre-releases of software that is not really in good
enough shape to release. Hobbyists, magazine reviewers, and
enthusiasts enjoy being able to get their hands on these early
versions, but they don't do much in terms of systematic input.

2. We think it is important that people experimenting with
Ada not find a confusing mess of different versions, many of
which are in some kind of beta stage, or otherwise not suitable
for general release.

3. So far, for the public releases of GNAT, we have been able
to make pretty frequent releases even without wide spread beta
programs, because of the internal field testing procedures that
we use.

Finally, we don't mind at all if other companies get into the
GNAT business. There are lots of gaps to fill in here, so most
likely such companies won't choose to compete in exactly the
same market as ACT, but that's up to them. For example, there
is a company in England that provides support for safety-
critical applications using the ERC32, an environment that
ACT does not support currently. We are working with them to
make sure that their business is successful. Similarly we worked
with Labtek when they were marketing a low cost supported
version of GNAT for NT.

It's certainly much easier to compete with ACT given that
you have our complete sources as a starting point than to
compete with any of the other Ada vendors who continue to
keep their sources highly proprietary!

Robert Dewar
Ada Core Technologies


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <86p6c1$vo5$1@nnrp1.deja.com>,
  Jean-Marc Bourguet <bourguet@my-deja.com> wrote:
> In article <X_Fj4.425$0o4.13328@cmnws01.we.mediaone.net>,
>   Pascal Martin <pascal.martin@iname.com.nospam> wrote:
> [...]
> > I believe GNAT is Dewar's brainchild, and he has a parental
> > crisis: he do not want the kid to leave the house. Is he
> > still reviewing all the GNAT code that's checked in ?

GNAT never was a one person project, anyone who thinks this
simply is misinformed. I still do quite a bit of technical
work on the system (mostly in the compiler itself), and sure
I review code checked in to this part, but other people review
code checkins in other sections. The GNAT system and all its
tools is a pretty large program now, it is quite beyond any one
person to be familiar with all sections of it. We do try to
ensure that for any particular section several people know the
code well, and can review checkins.

We also try to keep the code extremely well documented, so that
it is relatively easy for people new to the code to read it,
understand it, and, if so inclined, modify it :-), we have
received many useful contributions from outside ACT.

It's quite reasonable to have a public tree of GNAT for people
to work with. That's a goal of the GNAT/Linux group [it seems
much more appropriate to have this done outside ACT than inside
it]. Indeed the plan of action that we worked out with this
group is to build that tree and then integrate it into the
main GCC tree [the GCC council is perfectly happy to see that
integration occur, it is just a matter of getting a versoin
of GNAT that can integrate into this tree].

Once this is setup, it should be relatively easy to maintain.
Unlike the case with Cygnus, there are never any very major
discrepancies between the internal tree and the public version,
because we completely synchronize frequently (this is not true
at Cygnus, I gave the Itanium gcc as an example, another one
is version 5 of GDB). In fact syncrhonization of the two trees
is a major task for Cygnus, one on which they have several full
time people working.

This by the way does not reflect any nefarious goings on at
Cygnus, it merely reflects the fact that they have a large group
doing many internal development projects on many branches of
the tree, and keeping it all synchronized is not easy.

By comparison the GNAT development is much more focussed, and
there is for example only ONE internal tree at ACT.

Robert Dewar
Ada Core Technologies

Robert Dewar
Ada Core Technologies


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <3890DC4E.581F6A5@telepath.com>,
  Ted Dennison <dennison@telepath.com> wrote:
> Jean-Marc Bourguet wrote:
>
> Good point. I consider Emacs a quality product, but have you
ever tried to
> compile it from sources? I've done it 3 times now, and I'd
take putting
> together a 50 pice swing-set over that any day.


Well maybe that just means you know more about swing-sets
than gcc. In fact building emacs from sources is quite easy.
Until recently all our customers did this routinely, since
that's the way emacs is distributed. We have recently started
to provide binary builds of EMACS, and will eventually provide
binary builds of GLIDE. This is not so much because they are
hard to build, as to make it more convenient to be absolutely
sure you are working with the official supported version.


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <38912124.FE6B6AEF@telepath.com>,
  Ted Dennison <dennison@telepath.com> wrote:
bv> Yeah. As I remember the last time I had to do it for a SunOS
box (admittedly
> over 5 years ago), I first ended up having to compile gcc from
sources. And of
> course there's always the usual assortment of #ifdefs and
#defines that aren't
> quite right and have to be hacked. Yuk.


If you had to "hack" #ifdefs, then you were doing something
wrong!


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <87n1pswjz0.fsf@deneb.cygnus.argh.org>,
  Florian Weimer <fw@s.netic.de> wrote:
> I've misread this statement, I think.  The explanation by
Robert Dewar
> that followed but which I neglected to quote made it quite
clear that
> he meant that most of the code Cygnus develops for their
customers
> never makes it into the public GCC version, and this seems to
be true.
>
> On the other hand, the development process of the *public* GCC
version
> hosted by Cygnus is much more open than the GNAT development
process.


The GNAT development process inside ACT clearly corresponds to
the internal development at Cygnus, with one important
difference which is that EVERYTHING we do at ACT makes it into
the next public release of GNAT. That's always been true, and
will continue to be true.

Now what is missing is a public tree for GNAT that people could
play with. Note that until very recently, the same was true
for GDB, and it is definitely a disadvantage for GDB, why?
Because a substantial amount of GDB development is going on
outside Cygnus, and there needs to be a public tree which can
act as the focus for this distributed development. It is also
no secret that the intention is to establish a council for
GDB, similar to the council for GCC (in fact I acted as convenor
for the first couple of organizational meetings).

With GNAT, the development so far has gone on pretty much
entirely within ACT, and no other companies until quite
recently have got involved. This means that there has not
been the push to get a public tree going that existed with
GDB (a group of nine or so full time people working on GDB
at a company like HP is a considerably different story from
a couple of hobbyists who would like to hack around with GNAT
on a part time basis).

Nevertheless, it certainly seems desirable to have a public
tree, and as part of the GNU project, we have decided to center
this effort around GNU/Linux, which is why the GNAT/Linux team
seems the right organization to do this. It's making slow
progress partly because we are talking about people working in
their spare time, and partly because it won't really work nicely
until GNAT is fully merged into the new FSF version of GCC,
something that, as I said, we are working on!

Robert Dewar
Ada Core Technologies


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <slrn88v013.22t.aidan@skinner.demon.co.uk>,
  aidan@skinner.demon.co.uk wrote:
> On Tue, 25 Jan 2000 23:06:06 GMT, Robert Dewar
> <robert_dewar@my-deja.com> wrote:
>
> >and often closely held for a long time. The same is true
> >of course for Linux developments at Redhat. The difference
>
> I've heard this a lot, and I've never been *entirely*
convinced about
> it, especially given that redhat make their in-development
> distribution available via ftp...

Redhat keeps their internal development to themselves, but
not THAT much to themselves, the industry is quite aware
of important development initiatives at Redhat, because
they issue press releases :-)


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <au9l4.3239$0o4.37770@cmnws01.we.mediaone.net>,
  pmartin@mail.earthlink.net (Pascal F. Martin) wrote:
> In article <86la8r$519$1@nnrp1.deja.com>,
> 	Robert Dewar <robert_dewar@my-deja.com> writes:
> > Actually I think part of what goes on here is that ACT is
> > *more* open than a lot of the Linux and GCC development.
> > [...] The same is true
> > of course for Linux developments at Redhat.
>
> RedHat is _NOT_ developping Linux. Linux is being
> developped by a bunch of individuals and integrated by
> Linus Torvald, an engineer currently employed by Transmeta
> Inc.

Linux is not one product, it is a collection of more or less
compatible products, with a considerable commonality. Redhat
is not developing Linux, they are developing the Redhat version
of Linux, and they work hard to distinguish this so that you
will buy from them and not another Linux developer. If you
really think that a multi-billion dollar company like Redhat
relies entirely on a competitor (Transmeta) to control their
internal development, then you really have a reality model
quite divorced from any reality :-)

Note that for me Linux [more properly GNU/Linux, but perhaps
when talking about Redhat the term Linux is more appropriate]
development is not just about the kernel, it is about the
entire operating environment complete with both free software,
open source, and proprietary tools that are part of this
environment.

If you really think that Redhat is doing no development, then
it is simply because you are unaware of the major development
projects going on at Redhat. All though these are not yet all
open in the sense you mean when you talk about open development,
they are certainly not secret, a visit to the Redhat booth at
the recent Linux Expo in NY is quite enlightening in this
respect.

> RedHat develop a Linux distribution: this is kind of product
> integration work (akin to software editor).

No it is far more than that -- far more.

> It is incredible how even computer science legends like Robert
> can be victim of marketing image and branding.

Actually it is more the case that you are the victim of some
fairy tale view of the world that unfortunately does not
reflect reality. IPO's in the billion dollar range have quite
an interesting effect on the bazarre :-)

> I have to work
> with someone who also believed Linux was a product of RedHat
> and that the HardHat distribution was the "official" version
> of Linux for the embedded market, just because of the name
> match (how clever !).

Well it's *a* version of Linux for the embedded market. Everyone
selling *a* version would like people to believe they have *the*
version (for example, ACT is quite happy if you think that GNAT
is *the* version of Ada 95 :-)


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <m3emb5gmeg.fsf@kiuk0156.chembio.ntnu.no>,
  Preben Randhol <randhol@pvv.org> wrote:

> As for compilers I think this is even more crucial. I would
rather
> have a JGNAT that works well than an alpha product that makes
the
> programs crash a lot and thus gets a bad reputation. But it
might be
> that I for general OS products expect them to be alpha or beta
and
> crash a lot, I don't expect compilers to do the same. :-)


By the way, returning to the subject implied by the line
marked SUBJECT above :-) we do have some good news, the
initial beta testing of JGNAT has been going well, and we
need relatively few tweaks before we have a distributable
version, so it should not be too long now before you can
kick the tires.

(and I must say that I am NOT hoping that someone will say
 "great, this doesn't work at all, ACT has finally seen the
  light and is not waiting till software works before it
  is distributed for us to mess with." :-)

Robert Dewar
Ada Core Technologies


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <m3g0vk65as.fsf@kiuk0156.chembio.ntnu.no>,
  Preben Randhol <randhol@pvv.org> wrote:

> One thing Open Source Projects and deadlines do not mix. :-)

I completely disagree. Once again, just because a project takes
an open source view, or preferably a Free Software view [there
is a difference!] of licensing and distribution, does not mean
that somehow half baked development techniques have to be used.
There is no reason why an open source operation should not
achieve CMM level 5 and ISO 9000 status and be brilliant at
meeting deadlines.

Of course it is very difficult to be realiable at meeting
deadlines in any software work, as we all know, but open
source is simply a completely orthogonal concept that has
to do with what people can do with the product once they get
it, it does not in anyway constrain you from doing a good
job on software development!

Robert Dewar
Ada Core Technologies


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <86ncqt$l0p$1@nnrp1.deja.com>,
  Ted Dennison <dennison@telepath.com> wrote:
> When Gnat was first under development it was released early
and often. I
> remember trying out a version sometime around '94 that had no
> tasking support and was riddled with bugs.

If you had a version of GNAT in 94 that had no tasking, it
did not come from the GNAT project or ACT, and of course that's
something that we cannot (and do not attempt to prevent) -- the
proliferation of junk versions which are nothing to do with us.

But once the product existed, with proper support, that
precisely defines what we provide (I believe that Ted has
always used the public version, and complained furiously
about it -- his company pays for another compiler, and uses
GNAT, but won't pay for it -- that's quite fine, but it
means he is talking about the public version always without
support, and we have often advised him that we don't recommend
using the public version without support. If he was using a
version without tasking, that's truly a mess.

> Even in '95 it was not uncommon to
> get bugs in simple assignments that involved no new Ada95
> features. I think Gnat *did* get somewhat of a bad reputation
> from those early releases. So perhaps ACT's current policy is
> a reaction to that.

Even as early as 95, many serious users were writing and porting
large applications using GNAT. Yes, there were some bugs, and
most certainly GNAT is more stable now than then, as was true
for all Ada 95 products (well in 95, there were not too many
Ada 95 compilers around, stable or otherwise :-)

In any case there has been no change in ACT policy which is
that whenever we make a release of our commercial product,
we follow it with a public version that is essentially
identical technically (it has a different version number).

Certainly there is always a balance, we don't wait to issue
a public version of GNAT until we are sure it was perfect,
or you would never see even one public release. On the other
hand we try to get things into reasonable stable shape. We
usually get shot at from both sides ["those guys at ACT won't
release things, and keep things secret", and, as I remember
from T.E.D. himself "Let the Moaning begin" when something
in the public version does not work right :-)

The phrase "riddled" with bugs is a bit odd. It likely
reflected the fact that the public version that T.E.D.
was using was flawed, corrupted, put together by someone
other than us, or simply being misused
(you would be amazed at how many bug reports we get [some
of them from T.E.D.] that are simply cases of misunderstanding
Ada 95 or GNAT, and not actually bugs at all :-)

Robert Dewar
Ada Core Technologies


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Chris Morgan]

Robert Dewar <robert_dewar@my-deja.com> writes:

> Actually it is more the case that you are the victim of some
> fairy tale view of the world that unfortunately does not
> reflect reality. IPO's in the billion dollar range have quite
> an interesting effect on the bazarre :-)

                               ^^^^^^^

Greate typo. As much as ESR likes to think the Linux community is
"just" a bunch of hackers at a code bazaar, the amount of money
sloshing around this scene makes it more than a little bazaar, if not
more than a little bizarre, hence bazarre is _perfect_

Regards,

Chris

-- 
Chris Morgan <cm at mihalis.net>                  http://mihalis.net
	     http://www.mustbedestroyed.com

Re: JAVA and ADA JGNAT

[Aidan Skinner]

Robert Dewar <robert_dewar@my-deja.com> writes:

> what you are referring to when you talk about Microsoft
> technologies developed outside of Microsoft. Sure the original
> version of DOS was purchased, but even todays DOS (let alone

I believe IE was purchased from spyglass, as was one of the other
things their pushing heavily (media player?) IIRC...

OTOH this is so totally off-topic for cla it's not true... ;)

- Aidan
-- 
http://www.skinner.demon.co.uk/aidan/
Before asking a tech a question, think: "Does this person care? Is
this in anyway meaningful to their existence?". If the answer is "No", 
please read the documentation supplied, specifically Chapter 9: Suicide.

Re: JAVA and ADA JGNAT

[Pascal Martin]

In article <87i8h0$lki$1@nnrp1.deja.com>, Robert Dewar <robert_dewar@my-deja.com> wrote:

> If you really think that a multi-billion dollar company like Redhat
> relies entirely on a competitor (Transmeta) to control their
> internal development, then you really have a reality model
> quite divorced from any reality :-)

Well.. Transmeta is not in competition with RedHat:

- RedHat sells a Linux-based OS, as software product,
   Transmeta sells chips and releases a Linux kernel and a
   limited set of tools for free (a meta-distribution, they say)
   to OEMs.

- RedHat's stated goal is to provide an OS for Internet servers
   and to cash in support services, Transmeta is selling a chip
   for thin Internet clients and laptops.

> If you really think that Redhat is doing no development, then
> it is simply because you are unaware of the major development

I am aware of it: I use Gnome :-), in which RedHat participated a lot.
But they do not develop Linux, the kernel. And probably 90% of the software
they deliver is not from RedHat, so they are hardly in control: the Ethernet 
drivers and Beowulf (NASA), X11 (Open Group), Apache (The Apache Group), 
the BSD tools, Emacs (FSF), Mozilla (Netscape), KDE (KDE Consortium), 
AbiWord (AbiSource), XV (John Bradley), sendmail (Sendmail, Inc), etc... 

Add to that the fact that many developments made in RedHat labs are 
copyrighted by the Author (not RedHat) and follow this one when he leaves 
(Rasterman and Enlightment comes to mind).

Nothing in common with ACT, which I believe develops most of the software
it delivers. I see ACT as a technology provider, RedHat more as a software editor.

Considering Microsoft, software editors may have a bright future (a lot of
Microsoft technologies have been developped outside of Microsoft..).

> (for example, ACT is quite happy if you think that GNAT
> is *the* version of Ada 95 :-)

Well.. having worked for Alsys/TSP/Aonix/<whatever..>, I do not buy in
such a thing. Imagine: I once though AdaMagic was The Ada95 Reference
Implementation 8-}

------------------------------------------------------------------
Pascal F. Martin.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <DAan4.552$pE4.8687@cmnws01.we.mediaone.net>,
  Pascal Martin <pascal.martin@iname.com.nospam> wrote:
> In article <87i8h0$lki$1@nnrp1.deja.com>, Robert Dewar
<robert_dewar@my-deja.com> wrote:

> Well.. Transmeta is not in competition with RedHat:

Of course Transmeta is in competition with Redhat, sure they
do not currently compete head on, but they are both aiming
at the commercial Linux market. Right now, Redhat is not
focussed on the same area (Transmeta's aim is to produce
a mobile version of Linux, which is stripped down, and it
will not be "just a kernel". By the way, this version of
Linux is not yet available from TM, so saying that
Transmeta "releases a Linux Kernel" is wrong. If you
are talking about the general work TL does with the Linux
kernel, I would not regard that as a TM product at all
(it was not mentioned in the big press do last week for
example, except in passing). It is Mobile-Linux (tm, or
at least I think it is trademarked) that is the product.

Redhat is not currently in the mobile Linux market, but
that does not mean they are not both competitors.

Take Tivo as an example. Currently this consumer product
uses a version of Linux developed specifically for Tivo,
but I can see this kind of market as a target for both
Redhat Linux and Mobile-Linux.

> > If you really think that Redhat is doing no development,
then
> > it is simply because you are unaware of the major
development
>
> I am aware of it: I use Gnome :-), in which RedHat
participated a lot.

No, no, I am not talking about things like Gnome, which was
developed openly, I am talking about the internal development
projects which are well known in the valley (as I said, this
is not just a matter of secrets leaking out, though everyone
in the valley always knows what is going on except in very
unusual cases like Transmeta), but also because of the press
releases Redhat has made.

> Add to that the fact that many developments made in RedHat
> labs are  copyrighted by the Author (not RedHat) and follow
> this one when he leaves (Rasterman and Enlightment comes to
> mind).

The only special power held by the copyright holder for GPL'ed
software is the ability to subsequently release a version under
some other license (e.g. a fully proprietary product). So this
is not really an issue (I am not sure why you raised it).

> Considering Microsoft, software editors may have a bright
> future (a lot of Microsoft technologies have been developped
> outside of Microsoft..).

Well if you think that Microsoft is simply a software editor
and does no internal development, then your viewpoint is even
more remote from reality than I thought. I am not even sure
what you are referring to when you talk about Microsoft
technologies developed outside of Microsoft. Sure the original
version of DOS was purchased, but even todays DOS (let alone
todays Windows and NT) contains little of that original code.



Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Pascal Martin]

In article <87jrv3$n7r$1@nnrp1.deja.com>, Robert Dewar <robert_dewar@my-deja.com> wrote:
>> Well.. Transmeta is not in competition with RedHat:
> 
> Of course Transmeta is in competition with Redhat, sure they
> do not currently compete head on, but they are both aiming
> at the commercial Linux market. Right now, Redhat is not
> focussed on the same area (Transmeta's aim is to produce
> a mobile version of Linux, which is stripped down, and it

I stand by my opinion :-): the Transmeta sales material is full
of CPU chip description. They are competing head on with
Intel and AMD, not with RedHat. They provide Mobile-Linux
in order to sell chips: you cannot sell CPU chips without software.
I have seen no Transmeta sales brochures on business support,
nor any adds for Linux CDs.

Transmeta is aiming to sell its chips to PC and gizmo vendors.
RedHat is aiming to sell the OS to these and support to the
customers of these. The two are going to live happily together
for years to come (at least I hope so..)

For an authorized view on Transmeta "software business", here
is the Linus testimony: "Actually, Mobile Linux came about because 
we had customers that wanted to run Linux on those smaller
devices like handhelds, so it was on their demand really.It isn't
really a new version of Linux, mind you, [...]" (URL:
http://gnet.dhs.org/linus/)

> Redhat is not currently in the mobile Linux market, but
> that does not mean they are not both competitors.

Even when RedHat will be in the mobile market, that will be
to Transmeta advantage: they might sell more chips.

> Take Tivo as an example. Currently this consumer product
> uses a version of Linux developed specifically for Tivo,

They use, but they don't sell. If you buy a computer, you do
not become an IBM competitor, even if you bought a Sun..
If you use Linux, you do not become a RedHat competitor
automatically.

I guess Tivo is busy enough trying to convince us couch
potatoes that they can enhance our TV experience. They
don't have any time left for entering the Linux distribution
business, a crowded one if you want my opinion (I know,
Robert, you did not ask :-).

Is the same fear (customers becoming competitors) driving
ACT's strategy ?

> The only special power held by the copyright holder for GPL'ed
> software is the ability to subsequently release a version under
> some other license (e.g. a fully proprietary product). So this
> is not really an issue (I am not sure why you raised it).

You misunderstood: these are two separate points: 1) the copyright
is not RedHat, so they cannot even argue about it, 2) it happens
that when the developper leaves, the software often follows, which 
never happens in the proprietary world (because of the copyright).

> Well if you think that Microsoft is simply a software editor
> and does no internal development, then your viewpoint is even
> more remote from reality than I thought. I am not even sure

I did not say such a thing. I just meant that Microsoft does a lot 
of money as a software editor. I read somewhere, about 5 years 
ago, that a significant part of the Microsoft income was made 
of ouside software sales. 

Now you can call me fuzzy ! :-))

------------------------------------------------------------------
Pascal F. Martin.

Re: JAVA and ADA JGNAT

[Robert Dewar]

In article <XlDn4.1$Qn4.22@cmnws01.we.mediaone.net>,
  Pascal Martin <pascal.martin@iname.com.nospam> wrote:
> I guess Tivo is busy enough trying to convince us couch
> potatoes that they can enhance our TV experience. They
> don't have any time left for entering the Linux distribution
> business, a crowded one if you want my opinion (I know,
> Robert, you did not ask :-).

Please reread my message, you completely misunderstood it.
Of course Tivo is not in the Linux business. My point was
that they are a typical customer who might go to TM or to RH.

> Is the same fear (customers becoming competitors) driving
> ACT's strategy ?

Bizarre non-sequitur, since nothing I said was about this
in the first place, but the answer is, of coures not, this
is a silly idea! Our strategy is drive entirely by what we
think is best for our customers. This surely is true for
any company, or at least any successful company.



Sent via Deja.com http://www.deja.com/
Before you buy.

Re: JAVA and ADA JGNAT

[Wes Groleau]

> ....have quite
> an interesting effect on the bazarre :-)

Is that a cross between bizarre and bazaar ?

:-)