A good compiler

C

Chris Hills

Richard Heathfield said:
Chris Hills said:


If they have a commercial interest in it,

No.... An interest (or bias) whether commercial or not.
they ought to declare that
when recommending it. If they're merely interested in it, that isn't
the same thing. But if they stand to gain money from its promotion,
then it is reasonable to deduce that their advice may not be impartial,
so yes, you're right, they should declare an interest.

The FOSS Devotees have an irrational, religious support for FOSS and
this should ALWAYS be declared weather or not money changes hands.
If they have a commercial interest in it, yes. Otherwise, why?

Because they are just as biased and Navia whose compiler is also free.
gcc is OT too. Occasionally, I answer questions about gcc (not often, of
course because it's OT) but then occasionally I answer questions about
Visual C (again, not often, because it's OT). The fact that we
occasionally see OT answers about gcc in here does not make gcc
topical.

I have noticed that whilst the vast majority of the time questions on
any compiler other than Gcc gets slamed for being OT it is not the same
for GCC. There are several Gcc threads running on clc now
Mr Navia has said, in a Usenet article which I'm sure you will have no
trouble in finding for yourself:

"Only if you want to include the compiler in your product you have to
pay.

Fair enough. It is the same for many tools. In fact both Linux and Gcc
are commercial as you can buy versions from several suppliers.....
(though thay way it is the "support and packaging" you are paying for
the effect is the same.

As usual FOSS Devotees are biased and manipulate the truth for a
distorted view of reality.

It is possible to get lcc-win32 and gcc for free or pay for both. YUet
you only attack lcc-win32. Typical of FOSS Devotees

I have some customers that pay for this (and I hope I get more) so
that I am able to finance the development of lcc-win32. I have been
financing this project for more than 6 years now. I have never made any
money with my work, but in the last time I have been able to recover
some of the money I spend developing this software. Free software must
have a way of earning money to finance development, I see no other
possibility. So I decided that companies that use my work in a
commercial environment should pay for the product they get. What is
wrong with this?" (This particular quote is taken from 17 Apr 2003.)

Nothing, of course, is wrong with that, but it certainly demonstrates
that he does sell lcc-win32,

Under some conditions just as Linux and Gcc is availible on commercial
packages. However you can obtain lcc-win32 for free and there is no
need for the vast majority of users to pay anything, just like FOSS
users
and therefore he has a financial interest
that he ought to have been declaring when touting his product.

The same as EVERYONE who touts GCC and Linux who has an interest in it.
As usual there is a lot of equivocation and splitting of hares. Just
because under certain circumstances you mighh pay for lcc-win32 just as
you might pay for Gcc yet you only want the declaration if interest to
go one way.
 
C

Chris Hills

It's blindingly obvious what his relationship is when you go to the
lcc-win32 website to get the FREE download of the compiler and other
tools.

It is defiantly NOT commercial but FREE software (if not open source)
As with Red hat and all the other commercially available versions of
Linux and Gcc there is a chargeable model but I bet more people make
more money selling, supporting and developing with FOSS than Navia does
(or is ever likely to ) with lc-win32

He shows EXACTLY what his relationship is to the compiler. However none
of the FOSS Devotees ever declare exactly what their interest/involvment
in FOSS is. Apparently there are one set of rules for FOSS Devotees and
one set of rules for everyone else. A bit like a fanatical religion
really. It has all the hall marks of one.

So if you are going to support FOSS you should declare your interests.
Do you make any money developing or using FOSS? Ie if you get any form
of compensation then you should declare it. I am sure there are many
here who develop with FOSS and earn a good living who don't admit this
when recommending it.

Those earning money using FOSS have a real commercial interest because
they need to keep the FOSS religion alive in order to have a enough
willing FOSS devotees to do the work to keep them in tools to earn
money...

Incidentally I have no idea how good or bad lcc-win32 is. That is not
what is being argued here. It is just being smeared by FOSS people
because it is not Gcc and you can under some circumstances (just like
most FOSS) pay for it.
 
C

Chris Hills

Richard Heathfield said:
Chris Hills said:



How so? I put C in, and object code comes out. It walks like a C
compiler, swims like a C compiler, and quacks like a C compiler.

So it is fully C99 compliant?

It is no more compliant than all the other "c-like" compilers that get
slated here for having extensions and omissions to C99. Gcc is no more
a C compiler than lcc-win32 is. You can't have it both ways.
 
R

Richard

user923005 said:
I like GCC with Eclipse and the CDT. I also prefer DDD to GDB.
GCC 4.x has profile guided optimization and other advanced features.

ddd is just a front end to gdb.

ddd sucks. Its one of the ugliest guis ever made IMO.
 
R

Richard

Mark McIntyre said:
You have a different experience to me. We use gcc for a fairly
enormous set of systems at work which we compile for Windows, Solaris
and RHEL, and gcc not noticeably slower than MSVC on Linux (on Solaris
its a dog but thats a /solaris/ problem, not a gcc one...).


Hardly. gdb simply isn't a visual gui debugger. If you want that,
there are tools for linux.

Nearly all of which are rubbish. Eclipse might change that. DDD is a joke.
 
S

santosh

Richard said:
ddd is just a front end to gdb.

ddd sucks. Its one of the ugliest guis ever made IMO.

I'd assume that one would look for functionality and ease of use of the UI,
in a debugger if nowhere else, not it's eye-candy effect. Where do you find
DDD lacking, as far as ease of use is concerned?
 
B

Ben Bacarisse

Chris Hills said:
No.... An interest (or bias) whether commercial or not.

That is not practical. We cannot declare all our biases. Most will
be unconscious biases anyway. Careful reading of the arguments made
in a posting (and in the poster's history) are usually a better
indication than requiring public declarations.
The FOSS Devotees have an irrational, religious support for FOSS and
this should ALWAYS be declared weather or not money changes hands.

I would have thought that it would be obvious from the postings when
an irrational, religious, support is being expressed -- just as it will
be obvious when rational and well-argued support is presented.
The same as EVERYONE who touts GCC and Linux who has an interest in
it.

Of course, and that is obvious from the touting is it not? What more
could you want?

If I say "X is a great supplier of Y" I am obviously biased (I am
offering not evidence or argument) but I think it makes a difference
to the way one reads this bias if I then say "BTW, I get 10% of every
new sale of X makes". If am obliged, as you would have me be, to add
"BTW, I am an irrational devotee of the way X does things" I don't
think readers get any more than they had already.
 
R

Richard Heathfield

Chris Hills said:
No.... An interest (or bias) whether commercial or not.

Fine by me. I now state, for the record, that I currently have no
commercial interest in any C compiler, nor have I ever had such an
interest. Neither do I have a "religious" interest in any compiler. I
have used, and enjoyed using, gcc, Visual C, Borland C (and Turbo C),
LE370, C/370, Digital Mars, and one or two others that I don't recall.
I don't *care* what compiler I use, most of the time. It's of no
interest whatsoever to me. Yes, I like the idea of FOSS (if that means
what I think it means), but I also recognise that people have the right
to charge for their software if they wish, and I have no problem with
that. I am not "religiously" devoted to FOSS. I have written software
to sell, and I've written software to give away, and I don't have any
particular axe to grind in either direction.

The FOSS Devotees have an irrational, religious support for FOSS and
this should ALWAYS be declared weather or not money changes hands.

Fine - by all means point this out if you think it relevant in the
various threads where the issue may arise.

I have noticed that whilst the vast majority of the time questions on
any compiler other than Gcc gets slamed for being OT it is not the
same for GCC. There are several Gcc threads running on clc now

You have every right to point out that such threads are off-topic. Do
you? If not, you are part of the precipitate. If you care to do a
minimal amount of research, you will note that I have pointed out the
non-topicality of gcc threads on a number of occasions.

As usual FOSS Devotees are biased and manipulate the truth for a
distorted view of reality.

Since I'm not a "FOSS Devotee", I will leave it to those who are to
defend themselves against that charge, should they wish so to do.
It is possible to get lcc-win32 and gcc for free or pay for both.
YUet
you only attack lcc-win32. Typical of FOSS Devotees

Check your facts. Yes, I have praised Linux and dissed Windows on
occasion, but on the other hand I have also pointed out that
Microsoft's C compiler is actually very good, and that Windows
programming is very enjoyable. FOSS Devotee? Sure about that?

[...] So I decided that companies that use my work
in a commercial environment should pay for the product they get. What
is wrong with this?" (This particular quote is taken from 17 Apr
2003.)

Nothing, of course, is wrong with that, but it certainly demonstrates
that he does sell lcc-win32,

Under some conditions just as Linux and Gcc is availible on commercial
packages. However you can obtain lcc-win32 for free and there is no
need for the vast majority of users to pay anything, just like FOSS
users

If those who stand to make money from gcc touted it here, they'd be
acting in just as bad a way as Mr Navia. As far as I'm aware, however,
they don't. Do you have evidence to the contrary?

<snip>
 
R

Richard

santosh said:
I'd assume that one would look for functionality and ease of use of the UI,
in a debugger if nowhere else, not it's eye-candy effect. Where do you find
DDD lacking, as far as ease of use is concerned?

Everything it does. From its disgusting non standard widgets to its non
standard menus, its inability to bring up any documentation during
debug, to its horrible resizing. What were they thinking?

Yes, the graphical representation of structures can be nice. I do like
the way you can string different displays of a variable together,

But then I am happy to use gdb/gud now. But I wouldn't recommend it for a
noob.
 
R

Richard Heathfield

Chris Hills said:

Incidentally I have no idea how good or bad lcc-win32 is. That is not
what is being argued here. It is just being smeared by FOSS people

No, it isn't. The criticism is not against lcc-win32 but against its
promotion in a technical newsgroup by the one person who has most to
gain from its widespread use. And it isn't just "FOSS people" who are
making this criticism. I am not a "FOSS people", for instance. I am a
programmer who *sometimes* sells software and *sometimes* gives it
away.

Please check your facts before tarring all of Mr Navia's many critics
with the same brush.
 
R

Richard Heathfield

Chris Hills said:
So it is fully C99 compliant?

No, you know it isn't, but it does conform to C99 *to the extent that I
need it to* - in other words, it conforms to C90. Since I write in the
common subset of C90 and C99, it is as conforming as I could possibly
wish for.
 
J

jacob navia

Richard said:
Nearly all of which are rubbish. Eclipse might change that. DDD is a joke.

I would underscore that sentence.

I have used them all (kdevelop debugger, ddd, etc etc) and they all are
just GUI front ends to gdb, presenting you with a SUBSET of the
capabilities of gdb.

For instance, some of them will not allow you to view the registers,
or some of them will not allow you to see the disassembly.

Most of them do not allow "Set next statement", even if gdb can
do that with some obscure command.

They add some bugs to gdb, that even if it is a shit, it has only
few bugs.

And please, do not tell me that "I can report bugs" to the gdb
people.

I found a bug in gdb, fixed it, and send them the fix in the
corresponding gnu "gdb" group.

I never received even an acknowledgment that my fix was received.

jacob
 
J

jacob navia

santosh said:
I'd assume that one would look for functionality and ease of use of the UI,
in a debugger if nowhere else, not it's eye-candy effect. Where do you find
DDD lacking, as far as ease of use is concerned?

ddd is not able to do anything better than gdb. It is dead since
5-6 years (i.e. no longer maintained) and it has the ability of
crashing an X server, what is already a feat for a user program.

But it has some things ok:

o it displays the program text.
o it can display disassembly
o it displays the registers
o it gives you access to gdb, so you can use it for the
commands the GUI doesn't implement.

Of course there is no go to definition, no automatic variable
display, no restart, etc etc. But this is obvious. It is just
gdb.

jacob
 
K

Keith Thompson

Chris Hills said:
There are many FOSS Detotees on here who push FOSS without declaring
an interest. If you want to stop Navia then all those recommending
all the other free SW should stop as well.

I note that in this NG if it is not GCC then it is OT. Despite GCC
is not C but a C-like language compiler.

I think you overestimate the extent to which gcc-specific discussions
are tolerated here.

Detailed discussions of gcc-specific externsions are clearly off-topic
(unless, say, we're talking about whether they're valid extensions as
permitted by C99 4p6). Discussions of how to persuade gcc to behave
as a conforming C90 or C95 compiler, or as a partially conforming C99
compiler, are probably close to the edge, but they don't bother me; to
the extent that such knowledge helps programmers who happen to use gcc
understand what is and isn't standard C, it's a good thing.

Just one example: gcc allows pointer arithmetic on void*. I've
mentioned this here a number of times, but only to point out that it's
non-standard, or to explain why gcc isn't diagnosing some invalid
code, or (sometimes) to complain that it's a bad idea.

On the other hand, lengthy advocacy of gcc's non-standard "qfloat"
type, or of its support for operator overloading, is clearly
off-topic. Oh, wait, it's not gcc that has those features, it's
lcc-win32, isn't it?

If one of the authors of gcc spent time here pushing gcc's statement
expressions, local labels, labels as values, nested functions, etc.,
without regard to the fact that they are not standard C, I believe
that person would receive as much criticism as jacob navia does here
for his similar advocacy of lcc-win32's extensions. But they don't do
that.

I have no objection to lcc-win32 as a C compiler. I have no objection
to lcc-win32 as a C-with-extensions compiler. I object to its
developer's behavior in this newsgroup, particularly to his stubborn
refusal to understand just what we're objecting to. I suggest you may
be viewing these disputes through the filter of your own personal
biases (as, no doubt, we all do).
 
J

jacob navia

Keith said:
I think you overestimate the extent to which gcc-specific discussions
are tolerated here.

We had a recent mega-thread about how to avoid a gcc warning.
After dozens of messages I mentioned lcc-win32 and it was immediately
answered with "off topic off topic" even if the whole thread
was off topic but none of the traffic police would care to notice.
Detailed discussions of gcc-specific externsions are clearly off-topic
(unless, say, we're talking about whether they're valid extensions as
permitted by C99 4p6). Discussions of how to persuade gcc to behave
as a conforming C90 or C95 compiler, or as a partially conforming C99
compiler, are probably close to the edge, but they don't bother me; to
the extent that such knowledge helps programmers who happen to use gcc
understand what is and isn't standard C, it's a good thing.

We have had this discussion a number of times. In short:
I see this group as a formu to discuss C, and the possible evolution of
C.

You, the "regulars" and all the traffic police have no right to say
to me if that is ok or not.

Just one example: gcc allows pointer arithmetic on void*. I've
mentioned this here a number of times, but only to point out that it's
non-standard, or to explain why gcc isn't diagnosing some invalid
code, or (sometimes) to complain that it's a bad idea.

On the other hand, lengthy advocacy of gcc's non-standard "qfloat"
type, or of its support for operator overloading, is clearly
off-topic. Oh, wait, it's not gcc that has those features, it's
lcc-win32, isn't it?

Yes, and I think they show a way of enhancing C, and if you dislike it
tell people your opinion. But I will go on telling MY opinion of it
even if you do not agree with it.
If one of the authors of gcc spent time here pushing gcc's statement
expressions, local labels, labels as values, nested functions, etc.,
without regard to the fact that they are not standard C, I believe
that person would receive as much criticism as jacob navia does here
for his similar advocacy of lcc-win32's extensions. But they don't do
that.

Yes, they do not give a damm about C. They are C++ only, and C is
not very high in their priority list. They get paid for other types
of development, and the evolution of C is for them indifferent.
I have no objection to lcc-win32 as a C compiler. I have no objection
to lcc-win32 as a C-with-extensions compiler. I object to its
developer's behavior in this newsgroup, particularly to his stubborn
refusal to understand just what we're objecting to.

I DO understand what you say. I just do NOT AGREE WITH IT!


I suggest you may
be viewing these disputes through the filter of your own personal
biases (as, no doubt, we all do).

Exactly
 
R

Richard

jacob navia said:
We had a recent mega-thread about how to avoid a gcc warning.
After dozens of messages I mentioned lcc-win32 and it was immediately
answered with "off topic off topic" even if the whole thread
was off topic but none of the traffic police would care to notice.


We have had this discussion a number of times. In short:
I see this group as a formu to discuss C, and the possible evolution of
C.

Gotta say, I do too. There is no other C group for less "restricted"
programmers of C to go. Having said that, I just mean slightly OT should
be tolerated to a degree. We don't need Santosh racing Default user to
see who can post "OT" warnings first - it's really rather childish and
makes the signal to noise ratio intolerable at times. I realise a couple
of the usual pedantic usenet warriors have me labelled as a "troll"
already - but that simply is not true. It's just that I'll be damned to
see them constantly bullying and poking when its not necessary - and
they do need to be reminded that they are in a public forum and that
their behaviour and the tone of their replies sets the standards by
which others should be judged.
You, the "regulars" and all the traffic police have no right to say
to me if that is ok or not.

Regulars do have a right to try and keep things reasonably on track -
but a diverting thread should be fine - if you dont like it, killfile
it.
Yes, and I think they show a way of enhancing C, and if you dislike it
tell people your opinion. But I will go on telling MY opinion of it
even if you do not agree with it.


Yes, they do not give a damm about C. They are C++ only, and C is
not very high in their priority list. They get paid for other types
of development, and the evolution of C is for them indifferent.

And pretty much complete.
I DO understand what you say. I just do NOT AGREE WITH IT!

Good man. Stick to your guns.
 
R

Richard Heathfield

jacob navia said:
Keith Thompson wrote:


We had a recent mega-thread about how to avoid a gcc warning.

Such threads are off-topic. Did you point out that the gcc discussion
was off-topic? If not, you can hardly complain that nobody else did.
After dozens of messages I mentioned lcc-win32 and it was immediately
answered with "off topic off topic"

And so it is.
even if the whole thread
was off topic but none of the traffic police would care to notice.

Did *you* point out that it was off-topic? If not, you can hardly
complain that nobody else did.

In short:
I see this group as a formu to discuss C,

It is.
and the possible evolution of C.

No, that's what comp.std.c is for.
You, the "regulars" and all the traffic police have no right to say
to me if that is ok or not.

Yes, they do, and you have the right not to pay any attention, and they
have the right to point out that you are misusing the newsgroup by
pushing your product here, and you have the right not to agree, and
they have the right to treat you like a pariah and recommend that
people avoid your compiler, and you have the right not to be happy
about that, and so on. Add them up and take them away, and you'll come
out with a positive, or a negative, value to advertising your product's
misfeatures in this group. Presumably you think it's a positive value
despite the constant criticisms that you receive as a result, or you
wouldn't do it. Spammers use the same reasoning, of course. Stuff
community relations - you have a *product* to ship.
Yes, and I think they show a way of enhancing C, and if you dislike it
tell people your opinion. But I will go on telling MY opinion of it
even if you do not agree with it.

And I'll go on telling mine, which is that you are abusing this
newsgroup with your constant off-topic discussions of compiler
extensions. Your activities draw much criticism here, as do your
misunderstandings about C, which inspire no confidence that you are
competent to maintain a C compiler, let alone write one.

In short, your posting style here makes you your own worst enemy. Still,
it's a big foot and you seem to have plenty of ammunition, so fire away
if you must.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

No members online now.

Forum statistics

Threads
473,780
Messages
2,569,611
Members
45,280
Latest member
BGBBrock56

Latest Threads

Top