> From: Erik Reuter <[EMAIL PROTECTED]>
>
> On Sat, Feb 16, 2002 at 12:28:16PM -0600, The Fool wrote:
> 
> > Apt-get only updates packages that come with debian. 
> 
> All 3950 of them.
> 
> > And people call M$ bad for bundling middle ware with windows, but
> > linux most linux distros do the same, with more middleware packages
> > than microsoft does.
> 
> You are mistating the argument. Most people I have heard complain about
> this are upset with Microsoft forcing certain software to be used, for
> example, resellers who want to include Windows on their computer also
> had to include other programs.

In what way are you 'forced' to use specific software?  There is almost
nothing that comes just with windows.  Computer vendors are a different
matter.  And I'm not argueing thaat point anyway.  Windows is a bare
bones distrubution, by itself.

> I don't know how familiar you are with Debian, but the Debian packages
> are totally separate. You can install or not install any one that you
> like. It is about choice, something that Microsoft likes to deny people
> whenever possible.

Yes but not all packages that can be used with debian come with debian. 
I had to hunt down my modem manufacturer (it had one, but compiled for
newer versions of the kernel than debian was using, which means you had
download the latest kernel, than compile it...)

> > > > BS.  For every program written for linux there are several
hundred
> > > > written for windows.  In particular things like 'autoCAD',
'WinAMP',
> > > > etc.
> > > 
> > > WinAMP is available for linux. It's called xmms.

Might I point out that this not actually WinAMP just a linux clone
program that mimics many of the features that WinAMP has?  It is not a
port of winamp itself?  Nor does it have any of the upcoming features of
winamp 3?
 
It look from reading that page that it doesn't have many of the features
that winamp 2 has.

Might I point out that WinAMP on windows has the largest user base of any
player?  That includes windows media player, real player, and other dreck
(the reason AOL paid $400 Million for it).  It also has the largest base
of compatible 'plugins' that extend the code.

> > But how many user made plugins are made for linux?
> 
> I've never counted. Here you go, knock yourself out, there are quite a
> few: http://www.xmms.org/plugins.html

Not really.  None of the ones I use, (and I use very many).

> > I used to think that too.  But it's not true.  These freeware 'CAD'
> > programs do not in any way compare to AutoCAD.
> 
> Have you used either one? I've used AutoCAD, and QCad. I can do
> everything I need to do in either one, but I prefer QCad.

It's not about, what I or You can do, but what the CAD software can do,
which puts AutoCAD out of "QCad's" league.  More specifically it's about
what these $150k+ engineers and designers need for their work.

"Free" software just doesn't compare.

> > There was a very convincing argument about this on rasfwr-j a few
months
> > ago.
> > 
> > Long url:
> > 
> >
http://groups.google.com/groups?hl=en&frame=right&th=cb57b0b655247da&seekm

> > =slrn9i5qma.hk3.jsn%40ts016d40.chi-il.concentric.net#link1
> > 
> > Read the entire thread.
> 
> I skimmed through it. There was nothing of substance there. A lot
> of Stallman politics and name-calling. A brief mention by someone
> who thought there were a few apps on Windows that didn't have exact
> alternatives on Linux. No proof or serious evidence was presented.
> Did you have some specific argument there that you thought was
> convincing? It would help if you would point it out if you want to
> discuss it.

?? You read not far enough.  Brief.  Here's where the long part of the
message comes in:

And I should point out that the person making this argument (novak) is
not exactly a microsoft lover.

-----
Message 7 in thread     
From: John S. Novak, III ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View this article only  
Date: 2001-06-10 09:53:33 PST   
On Sun, 10 Jun 2001 11:28:30 +0200, Rouk <[EMAIL PROTECTED]>
wrote:

>You still have to admit that, with Free Software, or, at least Open
Source
>software, are a good alternative to hardcoded links to Microsoft
websites. If
>you don't, well, i don't mind. I was just giving my opinion on the
subject.

It is _an_ alternative.

But most of the hard core Stallman freaks are either transplants from
an alien universe where everyone has the time, energy, inclination,
and expertise to go code diving into massive applications to remove
features they don't like, or, more likely, are people who've been
brought up using boring software to begin with and haven't the
faintest idea about the very messy real world outside of basic
operating systems, word processors, spreadsheets, and browsers
(which everyone knows how to design.)

When you show me Free Software that competes favorably with things
like AutoCad, HP ADS, Cadence, and other major specialized packages,
I'll start singing the praises of Free Software.  

But until you can eliminate the quarter million dollars in software
licenses (easily-- and much more if you take multiples into account) it 
takes my group to design a hunk of electronics, stop pretending that
Free Software is the universal panacaea.  It ain't.

--
John S. Novak, III              [EMAIL PROTECTED]
The Humblest Man on the Net

---
Message 15 in thread    
From: Michael Bruce ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View this article only  
Date: 2001-06-10 11:18:27 PST   
On 10 Jun 2001 16:53:30 GMT, John S. Novak, III <[EMAIL PROTECTED]>
wrote:
> On Sun, 10 Jun 2001 11:28:30 +0200, Rouk <[EMAIL PROTECTED]>
wrote:
> 
>>You still have to admit that, with Free Software, or, at least Open
Source
>>software, are a good alternative to hardcoded links to Microsoft
websites. If
>>you don't, well, i don't mind. I was just giving my opinion on the
subject.
> 
> It is _an_ alternative.
> 
> But most of the hard core Stallman freaks are either transplants from
> an alien universe where everyone has the time, energy, inclination,
> and expertise to go code diving into massive applications to remove
> features they don't like, or, more likely, are people who've been
> brought up using boring software to begin with and haven't the
> faintest idea about the very messy real world outside of basic
> operating systems, word processors, spreadsheets, and browsers
> (which everyone knows how to design.)

...poorly.  That is, "which everyone knows how to design poorly".

> When you show me Free Software that competes favorably with things
> like AutoCad, HP ADS, Cadence, and other major specialized packages,
> I'll start singing the praises of Free Software.  

I don't think that there are any limitations within the Free Software
ideology that prevent programs in this area from being created.

That is a very careful statement.

> But until you can eliminate the quarter million dollars in software
> licenses (easily-- and much more if you take multiples into account) it

> takes my group to design a hunk of electronics, stop pretending that
> Free Software is the universal panacaea.  It ain't.

Well, I do think that Free Software - as a way of thinking about software
-
is a near-universal panacaea.  I do not think that Free Software - as 
embodied by the existing body of Free code - is.

What I think you mean is that _right now_, you can't just throw out Free
Software as an answer for a large number of problems.  This is true.  I
look at that as a problem to be solved, rather than a fact to be
lamented.

-- 
Michael Bruce
[EMAIL PROTECTED]

---

Message 16 in thread    
From: John S. Novak, III ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View this article only  
Date: 2001-06-10 18:25:19 PST   
On Sun, 10 Jun 2001 18:18:26 -0000, Michael Bruce <[EMAIL PROTECTED]>
wrote:

>> But most of the hard core Stallman freaks are either transplants from
>> an alien universe where everyone has the time, energy, inclination,
>> and expertise to go code diving into massive applications to remove
>> features they don't like, or, more likely, are people who've been
>> brought up using boring software to begin with and haven't the
>> faintest idea about the very messy real world outside of basic
>> operating systems, word processors, spreadsheets, and browsers
>> (which everyone knows how to design.) 
>...poorly.  That is, "which everyone knows how to design poorly".

Irrelevant, really.

There are two broad classes of applications, here, Bruce:  Those which
are in the common experience, and those which are not.  Everyone of a
given skill level in software design can spit out (given time and
motivation) things in the rpevious category.  Everyone nows what a
browser is supposed to do, because everyone has used one.  Everyone
knows what a word processor is supposed to do, because everyone has
used one.  Everyone knows what an operating system is supposed to do
because not only has everyone used one, but it's a required course in
many curricula.  

CAD programs, for instance, are decidedly NOT in that former class,
but in the latter.  Having an ink-wet degree in computer science is
not even close to being enough to design a real CAD program.  Nor is
an ink-wet CS degree and an ink-wet EE degree.  Mainly, because the
requisite background knowledge of just what a CAD program is supposed
to do, along with all the sophisticated models involved, just ain't
there yet.

Occasionally you'll get a halfway decent application out of a
university project somewhere, but those tend not to be the ones
actually used in industry because there's no profit motivation to keep
the applications up to date in an industry that thrives on yearly
innovation.

>> When you show me Free Software that competes favorably with things
>> like AutoCad, HP ADS, Cadence, and other major specialized packages,
>> I'll start singing the praises of Free Software.   
>I don't think that there are any limitations within the Free Software
>ideology that prevent programs in this area from being created.

It's not an ideology thing, which is why you (and all Stallman's ilk)
will be forever blind to the real limitations-- there are not enough
talented programmers who feel like giving away their talent for free
to make it work.  If there were, it would staggeringly difficult to
keep the project coordinated enough to continue past one major
software revision, or properly hooked into the user community.  And be
under no illusion, here-- the CAD user community is a whole fucking
lot more demanding than the pisswits that Microsoft tries to please.

--
John S. Novak, III              [EMAIL PROTECTED]
The Humblest Man on the Net

---
        
From: Michael Bruce ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View: Complete Thread (270 articles) | Original Format  
Date: 2001-06-10 21:59:34 PST   
On 11 Jun 2001 01:25:17 GMT, John S. Novak, III <[EMAIL PROTECTED]>
wrote:
> On Sun, 10 Jun 2001 18:18:26, Michael Bruce <[EMAIL PROTECTED]>
wrote:
> 
>>> But most of the hard core Stallman freaks are either transplants from
>>> an alien universe where everyone has the time, energy, inclination,
>>> and expertise to go code diving into massive applications to remove
>>> features they don't like, or, more likely, are people who've been
>>> brought up using boring software to begin with and haven't the
>>> faintest idea about the very messy real world outside of basic
>>> operating systems, word processors, spreadsheets, and browsers
>>> (which everyone knows how to design.) 
>>...poorly.  That is, "which everyone knows how to design poorly".
> 
> Irrelevant, really.

Yes, it is.  I just felt like pointing it out.

This is part of a larger personal Issue with the state of programming
as a practice, in general.  Feel free to ignore it.

> There are two broad classes of applications, here, Bruce:  Those which
> are in the common experience, and those which are not.  Everyone of a
> given skill level in software design can spit out (given time and
> motivation) things in the rpevious category.  Everyone nows what a
> browser is supposed to do, because everyone has used one.  Everyone
> knows what a word processor is supposed to do, because everyone has
> used one.  Everyone knows what an operating system is supposed to do
> because not only has everyone used one, but it's a required course in
> many curricula.  

True statements.

> CAD programs, for instance, are decidedly NOT in that former class,
> but in the latter.  Having an ink-wet degree in computer science is
> not even close to being enough to design a real CAD program.  Nor is
> an ink-wet CS degree and an ink-wet EE degree.  Mainly, because the
> requisite background knowledge of just what a CAD program is supposed
> to do, along with all the sophisticated models involved, just ain't
> there yet.

Also true statements, so far as they go.

I've met, interacted with, and know personally quite a few people with -
as you say - ink-wet degrees in computer science.  I wouldn't personally
trust them to write much of anything.

> Occasionally you'll get a halfway decent application out of a
> university project somewhere, but those tend not to be the ones
> actually used in industry because there's no profit motivation to keep
> the applications up to date in an industry that thrives on yearly
> innovation.

You would know more than I would.

>>> When you show me Free Software that competes favorably with things
>>> like AutoCad, HP ADS, Cadence, and other major specialized packages,
>>> I'll start singing the praises of Free Software.   
>>I don't think that there are any limitations within the Free Software
>>ideology that prevent programs in this area from being created.
> 
> It's not an ideology thing, which is why you (and all Stallman's ilk)
> will be forever blind to the real limitations-- there are not enough
> talented programmers who feel like giving away their talent for free
> to make it work.  If there were, it would staggeringly difficult to
> keep the project coordinated enough to continue past one major
> software revision, or properly hooked into the user community.  And be
> under no illusion, here-- the CAD user community is a whole fucking
> lot more demanding than the pisswits that Microsoft tries to please.

As a first cut, I'm going to declare that there shouldn't be such a
separation between the user community and the people producing the
software.

Take all that money that you spend on software licenses and employ
a programmer to help keep the parts of the program you care about
up to date, and doing exactly what you want.

There are some implementational problems with that approach, in that
for it to work effectively, competitors would need to cooperate to
a degree, and there would be the temptation of not hiring the programmer
and riding on to coattails of those who did, etc, etc.  

Since no large corporations have yet asked me to coordinate their
industry-wide Free Software plan, I have not given this enough serious
thought to come up with more than a sketch of an idea.  I am not trying
to represent it as more than that.

I think the more interesting shade of your argument is that of
coordination.  I do not think there are any really good methods
for efficiently developing software for complicated applications.
I don't think this is a problem just for Free Software, but I think
it affects Free Software more noticeably, for reasons that might
or might not be obvious, but which it's too late to get into right
now.

-- 
Michael Bruce
[EMAIL PROTECTED]

---
Message 18 in thread    
From: John S. Novak, III ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View this article only  
Date: 2001-06-10 22:21:42 PST   
On Mon, 11 Jun 2001 04:59:34 -0000, Michael Bruce <[EMAIL PROTECTED]>
wrote:

>> It's not an ideology thing, which is why you (and all Stallman's ilk)
>> will be forever blind to the real limitations-- there are not enough
>> talented programmers who feel like giving away their talent for free
>> to make it work.  If there were, it would staggeringly difficult to
>> keep the project coordinated enough to continue past one major
>> software revision, or properly hooked into the user community.  And be
>> under no illusion, here-- the CAD user community is a whole fucking
>> lot more demanding than the pisswits that Microsoft tries to please. 
>As a first cut, I'm going to declare that there shouldn't be such a
>separation between the user community and the people producing the
>software.

You'll have to define your terms, here.  
Especially 'separation.'

If you're implying that all users of CAD software should be capable
of programming CAD software (ie, that the user and programmer
communities should be in a large sense, the same) then you're
completely deranged in the same fashion as the original poster in
this subthread was.

>Take all that money that you spend on software licenses and employ
>a programmer to help keep the parts of the program you care about
>up to date, and doing exactly what you want.

'A programmer'?
You really have no concept of the scale of modern CAD systems, do you?

Nor have you really fully grasped the fact that CAD software really is
_not_ the same as banging out a few reams of garbage code in the
pursuit of some chicken-shit browser.  The models are extremely
complex.  They are derived, many times, from the careful, painstaking
observations and research of extremely sharp and extremely dedicated
academicians, high power researchers, and so forth.  When they are
published openly (which is not often) they are published in academic
journals, or PhD dissertations.  It takes an extremely proficient and
cross-disciplined programming TEAM to create good CAD software and keep
it current.

Then we get into the issue of testing and verification, which is a
whole 'nother nightmare.

The minimum number of people required to create good CAD software is,
at a rough estimate, the number of programmers, testers, management,
and other support staff as employed by Cadence, the ADS division of
Agilent, etc.  For perspective, here, Cadence employs some 5500
people.  And that's just one CAD software company.

What you're suggesting is that companies should somehow absorb the
cost of 5500 full time employees (many of who are going to be 100k/yr
hotshot programmers) for the benefit of not paying one software
license.

This is, quite simply, absurd.

>There are some implementational problems with that approach, in that
>for it to work effectively, competitors would need to cooperate to
>a degree, and there would be the temptation of not hiring the programmer
>and riding on to coattails of those who did, etc, etc.  

This is not an "implementational problem."
This is part of what makes the plan "completely unworkable."

--
John S. Novak, III              [EMAIL PROTECTED]
The Humblest Man on the Net

---
From: Jamie Bowden ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View: Complete Thread (270 articles) | Original Format  
Date: 2001-06-11 04:52:57 PST   
On 11 Jun 2001, John S. Novak, III wrote:

> On Mon, 11 Jun 2001 04:59:34 -0000, Michael Bruce
<[EMAIL PROTECTED]> wrote:
> 
> >> It's not an ideology thing, which is why you (and all Stallman's
ilk)
> >> will be forever blind to the real limitations-- there are not enough
> >> talented programmers who feel like giving away their talent for free
> >> to make it work.  If there were, it would staggeringly difficult to
> >> keep the project coordinated enough to continue past one major
> >> software revision, or properly hooked into the user community.  And
be
> >> under no illusion, here-- the CAD user community is a whole fucking
> >> lot more demanding than the pisswits that Microsoft tries to please.

> >As a first cut, I'm going to declare that there shouldn't be such a
> >separation between the user community and the people producing the
> >software.
> 
> You'll have to define your terms, here.  
> Especially 'separation.'
> 
> If you're implying that all users of CAD software should be capable
> of programming CAD software (ie, that the user and programmer
> communities should be in a large sense, the same) then you're
> completely deranged in the same fashion as the original poster in
> this subthread was.
> 
> >Take all that money that you spend on software licenses and employ
> >a programmer to help keep the parts of the program you care about
> >up to date, and doing exactly what you want.
> 
> 'A programmer'?
> You really have no concept of the scale of modern CAD systems, do you?
> 
> Nor have you really fully grasped the fact that CAD software really is
> _not_ the same as banging out a few reams of garbage code in the
> pursuit of some chicken-shit browser.  The models are extremely
> complex.  They are derived, many times, from the careful, painstaking
> observations and research of extremely sharp and extremely dedicated
> academicians, high power researchers, and so forth.  When they are
> published openly (which is not often) they are published in academic
> journals, or PhD dissertations.  It takes an extremely proficient and
> cross-disciplined programming TEAM to create good CAD software and keep
> it current.

Not just CAD, but any modelling software.  We have several pieces of
software that model everything from a single sensor to multiple platforms
in space, on the ground, and in the air.  I know what sort of effort is
required to get that right, and we're not even doing what CAD systems do.
Sensors and radar are cool and all, but we don't even deal with the
physical aspects, just modelling the outputs of the various systems based
on user supplied input.  We employ quite a few people who have advanced
degrees in physics and math who do nothing but testing and verification
of
the models.  These people are not programmers, yet the software could not
exist at the levels of accuracy currently achieved without them.  They
require an infrastructure to do all that testing on (hey, look, I'm not
directly responsible for developement, yet I am an essential piece after
all).  What's really cool is when they take real sensor data (someone
paid
for that hardware to be built and deployed, and for the data collection)
and run comparisons to do the ultimate verification.  None of the above
is
programmer specific or intensive.  CAD software has to be worse, you're
modelling physical systems that are in motion.

> Then we get into the issue of testing and verification, which is a
> whole 'nother nightmare.

See above.

> The minimum number of people required to create good CAD software is,
> at a rough estimate, the number of programmers, testers, management,
> and other support staff as employed by Cadence, the ADS division of
> Agilent, etc.  For perspective, here, Cadence employs some 5500
> people.  And that's just one CAD software company.

And many of them are not programmers hopefully.  I see a lot of division
between programmers and scientists here, which is unfortunate.  Some of
our code was written by guys with a PhD in physics, and unfortunately, it
shows.  Nonportable crap that's impossible to deal with if you didn't
write it, no real interface, etc.  Our best efforts come from programmers
and scientists working together to get the math and overall coding where
the need to be so the code is more than a gross hack implemented either
by
someone who knew what they were doing, but doesn't program, or by a
programmer trying to implement math they don't really get.

> What you're suggesting is that companies should somehow absorb the
> cost of 5500 full time employees (many of who are going to be 100k/yr
> hotshot programmers) for the benefit of not paying one software
> license.
> 
> This is, quite simply, absurd.

For many reasons.

> >There are some implementational problems with that approach, in that
> >for it to work effectively, competitors would need to cooperate to
> >a degree, and there would be the temptation of not hiring the
programmer
> >and riding on to coattails of those who did, etc, etc.  
> 
> This is not an "implementational problem."
> This is part of what makes the plan "completely unworkable."

But dude, it works on Star Trek!

Jamie Bowden
-- 
"It was half way to Rivendell when the drugs began to take hold"
Hunter S Tolkien "Fear and Loathing in Barad Dur"
Iain Bowen <[EMAIL PROTECTED]>

---     
From: Michael Bruce ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View: Complete Thread (270 articles) | Original Format  
Date: 2001-06-11 05:52:29 PST   
On 11 Jun 2001 05:21:41 GMT, John S. Novak, III <[EMAIL PROTECTED]>
wrote:
> On Mon, 11 Jun 2001 04:59:34, Michael Bruce <[EMAIL PROTECTED]>
wrote:
> 
>>> It's not an ideology thing, which is why you (and all Stallman's ilk)
>>> will be forever blind to the real limitations-- there are not enough
>>> talented programmers who feel like giving away their talent for free
>>> to make it work.  If there were, it would staggeringly difficult to
>>> keep the project coordinated enough to continue past one major
>>> software revision, or properly hooked into the user community.  And
be
>>> under no illusion, here-- the CAD user community is a whole fucking
>>> lot more demanding than the pisswits that Microsoft tries to please. 
>>As a first cut, I'm going to declare that there shouldn't be such a
>>separation between the user community and the people producing the
>>software.
> 
> You'll have to define your terms, here.  
> Especially 'separation.'

We're moving in to waters that have gotten me in to trouble before...

> If you're implying that all users of CAD software should be capable
> of programming CAD software (ie, that the user and programmer
> communities should be in a large sense, the same) then you're
> completely deranged in the same fashion as the original poster in
> this subthread was.

No.  I just mean that the strict producer/consumer separation should be
got rid of, replaced with a more cooperative relationship.  It would
be good if some of the users did some of the programming, but I in
no way expect all of the users to do all of the programming.

>>Take all that money that you spend on software licenses and employ
>>a programmer to help keep the parts of the program you care about
>>up to date, and doing exactly what you want.
> 
> 'A programmer'?
> You really have no concept of the scale of modern CAD systems, do you?

Having never worked with one, not exactly.  I'm assuming they're very
complex, though.

> Nor have you really fully grasped the fact that CAD software really is
> _not_ the same as banging out a few reams of garbage code in the
> pursuit of some chicken-shit browser.  The models are extremely
> complex.  They are derived, many times, from the careful, painstaking
> observations and research of extremely sharp and extremely dedicated
> academicians, high power researchers, and so forth.  When they are
> published openly (which is not often) they are published in academic
> journals, or PhD dissertations.  It takes an extremely proficient and
> cross-disciplined programming TEAM to create good CAD software and keep
> it current.

Never said that it didn't.

> Then we get into the issue of testing and verification, which is a
> whole 'nother nightmare.
> 
> The minimum number of people required to create good CAD software is,
> at a rough estimate, the number of programmers, testers, management,
> and other support staff as employed by Cadence, the ADS division of
> Agilent, etc.  For perspective, here, Cadence employs some 5500
> people.  And that's just one CAD software company.
> 
> What you're suggesting is that companies should somehow absorb the
> cost of 5500 full time employees (many of who are going to be 100k/yr
> hotshot programmers) for the benefit of not paying one software
> license.
> 
> This is, quite simply, absurd.

I think your extrapolation is a little over the top.  I also think your
conclusions are a little off.  Cadence makes money, yes?  In other words,
Cadence's customers are already paying enough money for Cadence to pay
its programmers, pay its non programmers, and still have a little
left over.

I guess my main contention is that that money should change hands on a
different basis.  Cadence as a company can still be doing most of the
development, as far as I'm concerned.  I just want a move away from
selling licenses, and restricting the use and modification of the
software.

I think that more people would probably be able to write (or at least
develop parts of) CAD software if the source was publically available.

>>There are some implementational problems with that approach, in that
>>for it to work effectively, competitors would need to cooperate to
>>a degree, and there would be the temptation of not hiring the
programmer
>>and riding on to coattails of those who did, etc, etc.  
> 
> This is not an "implementational problem."
> This is part of what makes the plan "completely unworkable."

With the current state of programming practice, I would say that the
development of high-quality software for specialized application
areas is highly problematic.

-- 
Michael Bruce
[EMAIL PROTECTED]

---
From: Pat O'Connell ([EMAIL PROTECTED]) Subject: Re: Address Change Advisory 
Newsgroups: rec.arts.sf.written.robert-jordan   
View: Complete Thread (270 articles) | Original Format  
Date: 2001-06-11 20:19:37 PST   
Michael Bruce wrote:
> 
> On 11 Jun 2001 05:21:41 GMT, John S. Novak, III <[EMAIL PROTECTED]>
wrote:
> > On Mon, 11 Jun 2001 04:59:34, Michael Bruce <[EMAIL PROTECTED]>
wrote: [snip] 
> > If you're implying that all users of CAD software should be capable
> > of programming CAD software (ie, that the user and programmer
> > communities should be in a large sense, the same) then you're
> > completely deranged in the same fashion as the original poster in
> > this subthread was.
> 
> No.  I just mean that the strict producer/consumer separation should be
> got rid of, replaced with a more cooperative relationship.  It would
> be good if some of the users did some of the programming, but I in
> no way expect all of the users to do all of the programming.

Users != programmers, in the vast majority of cases, especially in
CAD. There is a user programmable language in AutoCAD, called
AutoLISP, which can automate certain parts of designing using AutoCAD.
Engineering firms hire *programmers* that understand LISP, and they
pay them well.

> >>Take all that money that you spend on software licenses and employ
> >>a programmer to help keep the parts of the program you care about
> >>up to date, and doing exactly what you want.
> >
> > 'A programmer'?
> > You really have no concept of the scale of modern CAD systems, do
you?
> 
> Having never worked with one, not exactly.  I'm assuming they're very
> complex, though.
> 
> > Nor have you really fully grasped the fact that CAD software really
is
> > _not_ the same as banging out a few reams of garbage code in the
> > pursuit of some chicken-shit browser.  The models are extremely
> > complex.  They are derived, many times, from the careful, painstaking
> > observations and research of extremely sharp and extremely dedicated
> > academicians, high power researchers, and so forth.  When they are
> > published openly (which is not often) they are published in academic
> > journals, or PhD dissertations.  It takes an extremely proficient and
> > cross-disciplined programming TEAM to create good CAD software and
keep
> > it current.
> 
> Never said that it didn't.
> > Then we get into the issue of testing and verification, which is a
> > whole 'nother nightmare.
> >
> > The minimum number of people required to create good CAD software is,
> > at a rough estimate, the number of programmers, testers, management,
> > and other support staff as employed by Cadence, the ADS division of
> > Agilent, etc.  For perspective, here, Cadence employs some 5500
> > people.  And that's just one CAD software company.
> >
> > What you're suggesting is that companies should somehow absorb the
> > cost of 5500 full time employees (many of who are going to be 100k/yr
> > hotshot programmers) for the benefit of not paying one software
> > license.
> >
> > This is, quite simply, absurd.
> 
> I think your extrapolation is a little over the top.  I also think your
> conclusions are a little off.  Cadence makes money, yes?  In other
words,
> Cadence's customers are already paying enough money for Cadence to pay
> its programmers, pay its non programmers, and still have a little
> left over.

So that the user won't have to.

> I guess my main contention is that that money should change hands on a
> different basis.  Cadence as a company can still be doing most of the
> development, as far as I'm concerned.  I just want a move away from
> selling licenses, and restricting the use and modification of the
> software.

Then how would Cadence make its money, if they can't license their
software?

> I think that more people would probably be able to write (or at least
> develop parts of) CAD software if the source was publically available.

Doubtful. Most engineers aren't programmers (not all of course; I'm an
engineer and a programmer).

With the release of the source code, development of Cadence software
would then stop, and the company would go out of business, because no
one would now buy the software; they'd attempt to "grow their own"
changes, probably without success.

The now laid-off programmers who knew Cadence software would go
elsewhere, and not necesssarily to users of Cadence, who would maybe
hire someone to "do this one tweak" then lay them off again.

-- 
Pat O'Connell
Take nothing but pictures, Leave nothing but footprints, 
Kill nothing but vandals...

---
From: Michael Bruce ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View: Complete Thread (270 articles) | Original Format  
Date: 2001-06-11 20:51:24 PST   
On Mon, 11 Jun 2001 21:18:40 -0600, Pat O'Connell <[EMAIL PROTECTED]> wrote:
> Michael Bruce wrote:
>> 
>> On 11 Jun 2001 05:21:41 GMT, John S. Novak, III <[EMAIL PROTECTED]>
wrote:
>> > On Mon, 11 Jun 2001 04:59:34, Michael Bruce <[EMAIL PROTECTED]>
wrote: [snip] 
>> > If you're implying that all users of CAD software should be capable
>> > of programming CAD software (ie, that the user and programmer
>> > communities should be in a large sense, the same) then you're
>> > completely deranged in the same fashion as the original poster in
>> > this subthread was.
>> 
>> No.  I just mean that the strict producer/consumer separation should
be
>> got rid of, replaced with a more cooperative relationship.  It would
>> be good if some of the users did some of the programming, but I in
>> no way expect all of the users to do all of the programming.
> 
> Users != programmers, in the vast majority of cases, especially in
> CAD. There is a user programmable language in AutoCAD, called
> AutoLISP, which can automate certain parts of designing using AutoCAD.
> Engineering firms hire *programmers* that understand LISP, and they
> pay them well.

Was I hallucinating, or did I acknowledge that that is the way things
stand today?

[...]

>> > What you're suggesting is that companies should somehow absorb the
>> > cost of 5500 full time employees (many of who are going to be
100k/yr
>> > hotshot programmers) for the benefit of not paying one software
>> > license.
>> >
>> > This is, quite simply, absurd.
>> 
>> I think your extrapolation is a little over the top.  I also think
your
>> conclusions are a little off.  Cadence makes money, yes?  In other
words,
>> Cadence's customers are already paying enough money for Cadence to pay
>> its programmers, pay its non programmers, and still have a little
>> left over.
> 
> So that the user won't have to.

Point?

>> I guess my main contention is that that money should change hands on a
>> different basis.  Cadence as a company can still be doing most of the
>> development, as far as I'm concerned.  I just want a move away from
>> selling licenses, and restricting the use and modification of the
>> software.
> 
> Then how would Cadence make its money, if they can't license their
> software?

Pay them to work on whatever they're working on now, rather than paying
them for something that's already done.

Move to a model for paying for software that better fits the intrinsic
realities of software, not the intrinsic realities of selling shoes
or bicycles or pocket calculators.

>> I think that more people would probably be able to write (or at least
>> develop parts of) CAD software if the source was publically available.
> 
> Doubtful. Most engineers aren't programmers (not all of course; I'm an
> engineer and a programmer).

I don't see how this follows.

> With the release of the source code, development of Cadence software
> would then stop, and the company would go out of business, because no
> one would now buy the software; they'd attempt to "grow their own"
> changes, probably without success.
> 
> The now laid-off programmers who knew Cadence software would go
> elsewhere, and not necesssarily to users of Cadence, who would maybe
> hire someone to "do this one tweak" then lay them off again.

Then we enter a dark age, where corporate greed has sacrificed long
term prosperity for short term gain and left a world without any
significant ability to develop technology.  Efficiency plummets,
the stock market implodes, stock brokers and housewives with Datek
accounts are jumping out of windows.  Within a few years, we're all
dressed in leather, and belong to gangs cruising the interstates in
beat-up yet fetchingly customized old cars, looking for rich folk
to hijack.  If only we hadn't shared so much, all of this could've
been avoided!

Tragic, but post-apocalyptic life sounds fun anyway.

-- 
Michael Bruce
[EMAIL PROTECTED]

---
From: Pat O'Connell ([EMAIL PROTECTED]) Subject: Re: Address Change Advisory 
Newsgroups: rec.arts.sf.written.robert-jordan   
View: Complete Thread (270 articles) | Original Format  
Date: 2001-06-11 21:58:48 PST   
Michael Bruce wrote:
> 
> On Mon, 11 Jun 2001 21:18:40 -0600, Pat O'Connell <[EMAIL PROTECTED]>
wrote:
> > Michael Bruce wrote:
> >>
> >> On 11 Jun 2001 05:21:41 GMT, John S. Novak, III <[EMAIL PROTECTED]>
wrote:
> >> > On Mon, 11 Jun 2001 04:59:34, Michael Bruce
<[EMAIL PROTECTED]> wrote: [snip] 
> >> > If you're implying that all users of CAD software should be
capable
> >> > of programming CAD software (ie, that the user and programmer
> >> > communities should be in a large sense, the same) then you're
> >> > completely deranged in the same fashion as the original poster in
> >> > this subthread was.
> >>
> >> No.  I just mean that the strict producer/consumer separation should
be
> >> got rid of, replaced with a more cooperative relationship.  It would
> >> be good if some of the users did some of the programming, but I in
> >> no way expect all of the users to do all of the programming.
> >
> > Users != programmers, in the vast majority of cases, especially in
> > CAD. There is a user programmable language in AutoCAD, called
> > AutoLISP, which can automate certain parts of designing using
AutoCAD.
> > Engineering firms hire *programmers* that understand LISP, and they
> > pay them well.
> 
> Was I hallucinating, or did I acknowledge that that is the way things
> stand today?

And it'll probably stay that way. Not every techie wants to program.

> >> I think your extrapolation is a little over the top.  I also think
your
> >> conclusions are a little off.  Cadence makes money, yes?  In other
words,
> >> Cadence's customers are already paying enough money for Cadence to
pay
> >> its programmers, pay its non programmers, and still have a little
> >> left over.
> >
> > So that the user won't have to.
> 
> Point?

Most engineering firms don't want to hire programmers; they hire
engineers, designers, and drafters instead. That's why they buy
Cadence, Autocad, or whatever CAD program.
 
> >> I guess my main contention is that that money should change hands on
a
> >> different basis.  Cadence as a company can still be doing most of
the
> >> development, as far as I'm concerned.  I just want a move away from
> >> selling licenses, and restricting the use and modification of the
> >> software.
> >
> > Then how would Cadence make its money, if they can't license their
> > software?
> 
> Pay them to work on whatever they're working on now, rather than paying
> them for something that's already done.

Cadence is, like most programs, a work in progress.
> 
> Move to a model for paying for software that better fits the intrinsic
> realities of software, not the intrinsic realities of selling shoes
> or bicycles or pocket calculators.

Reality: CAD programs are purchased to design shoes, bicycles, or
pocket calculators. That's what an engineer or drafter wants to do
with it. 

Similarly, when I use Winword, Word Perfect, or Star Office, it's
because I want to prepare a document, not to monkey with the internal
code.
 
> >> I think that more people would probably be able to write (or at
least
> >> develop parts of) CAD software if the source was publically
available.
> >
> > Doubtful. Most engineers aren't programmers (not all of course; I'm
an
> > engineer and a programmer).
> 
> I don't see how this follows.

You're not working in the real world, are you?

Most engineers would rather pay a competent software firm to deliver a
finished software product (in this case, a CAD program) that helps
them get their job done faster and more accurately. They don't want to
monkey with the code themselves; they just want to get their jobs
done. They're happy to leave coding to people who understand it.

> > With the release of the source code, development of Cadence software
> > would then stop, and the company would go out of business, because no
> > one would now buy the software; they'd attempt to "grow their own"
> > changes, probably without success.
> >
> > The now laid-off programmers who knew Cadence software would go
> > elsewhere, and not necesssarily to users of Cadence, who would maybe
> > hire someone to "do this one tweak" then lay them off again.
> 
> Then we enter a dark age, where corporate greed 

YM "bankruptcy"

> has sacrificed long
> term prosperity for short term gain and left a world without any
> significant ability to develop technology.  

Yup--because you took away Cadence's means to stay in business.

-- 
Pat O'Connell
Take nothing but pictures, Leave nothing but footprints, 
Kill nothing but vandals...

---
        
From: Michael Bruce ([EMAIL PROTECTED]) Subject: Re: Address Change
Advisory  Newsgroups: rec.arts.sf.written.robert-jordan 
View: Complete Thread (270 articles) | Original Format  
Date: 2001-06-12 05:54:40 PST   
More history kept than strictly necessary, for clarity.

On Mon, 11 Jun 2001 22:58:39 -0600, Pat O'Connell <[EMAIL PROTECTED]> wrote:
> Michael Bruce wrote:
>> 
>> On Mon, 11 Jun 2001 21:18:40 -0600, Pat O'Connell <[EMAIL PROTECTED]>
wrote:
>> > Michael Bruce wrote:
>> >>
>> >> On 11 Jun 2001 , John S. Novak, III <[EMAIL PROTECTED]> wrote:
>> >> > If you're implying that all users of CAD software should be
capable
>> >> > of programming CAD software (ie, that the user and programmer
>> >> > communities should be in a large sense, the same) then you're
>> >> > completely deranged in the same fashion as the original poster in
>> >> > this subthread was.
>> >>
>> >> No.  I just mean that the strict producer/consumer separation
should be
>> >> got rid of, replaced with a more cooperative relationship.  It
would
>> >> be good if some of the users did some of the programming, but I in
>> >> no way expect all of the users to do all of the programming.
>> >
>> > Users != programmers, in the vast majority of cases, especially in
>> > CAD. There is a user programmable language in AutoCAD, called
>> > AutoLISP, which can automate certain parts of designing using
AutoCAD.
>> > Engineering firms hire *programmers* that understand LISP, and they
>> > pay them well.
>> 
>> Was I hallucinating, or did I acknowledge that that is the way things
>> stand today?
> 
> And it'll probably stay that way. Not every techie wants to program.

And I explicitly _said_ that I wouldn't want all users to be programmers.

>> >> I think your extrapolation is a little over the top.  I also think
your
>> >> conclusions are a little off.  Cadence makes money, yes?  In other
words,
>> >> Cadence's customers are already paying enough money for Cadence to
pay
>> >> its programmers, pay its non programmers, and still have a little
>> >> left over.
>> >
>> > So that the user won't have to.
>> 
>> Point?
> 
> Most engineering firms don't want to hire programmers; they hire
> engineers, designers, and drafters instead. That's why they buy
> Cadence, Autocad, or whatever CAD program.

That's fine.  That's why I wrote the paragraph below.

That still seems only vaguely connected to the paragraph you were
responding to.

>> >> I guess my main contention is that that money should change hands
on a
>> >> different basis.  Cadence as a company can still be doing most of
the
>> >> development, as far as I'm concerned.  I just want a move away from
>> >> selling licenses, and restricting the use and modification of the
>> >> software.
>> >
>> > Then how would Cadence make its money, if they can't license their
>> > software?
>> 
>> Pay them to work on whatever they're working on now, rather than
paying
>> them for something that's already done.
> 
> Cadence is, like most programs, a work in progress.

Well, duh.

What does this have to do with what I said?

>> Move to a model for paying for software that better fits the intrinsic
>> realities of software, not the intrinsic realities of selling shoes
>> or bicycles or pocket calculators.
> 
> Reality: CAD programs are purchased to design shoes, bicycles, or
> pocket calculators. That's what an engineer or drafter wants to do
> with it. 

Well, duh.

What does this have to do with what I said?

> Similarly, when I use Winword, Word Perfect, or Star Office, it's
> because I want to prepare a document, not to monkey with the internal
> code.

What's your point?

>> >> I think that more people would probably be able to write (or at
least
>> >> develop parts of) CAD software if the source was publically
available.
>> >
>> > Doubtful. Most engineers aren't programmers (not all of course; I'm
an
>> > engineer and a programmer).
>> 
>> I don't see how this follows.
> 
> You're not working in the real world, are you?

Is a large steel company "the real world"?

> Most engineers would rather pay a competent software firm to deliver a
> finished software product (in this case, a CAD program) that helps
> them get their job done faster and more accurately. They don't want to
> monkey with the code themselves; they just want to get their jobs
> done. They're happy to leave coding to people who understand it.

Where in the sentance you quoted did I mentioned engineers?

I was, in fact, talking about professional programmers.  I was responding
to Novak's assertion that writing CAD software was something that it took
special experience and knowledge to do.  I agree with his statement, and
was pointing out that opening the code would probably increase the number
of people that could successfully write CAD and other specialized
technical software.
 
>> > With the release of the source code, development of Cadence software
>> > would then stop, and the company would go out of business, because
no
>> > one would now buy the software; they'd attempt to "grow their own"
>> > changes, probably without success.
>> >
>> > The now laid-off programmers who knew Cadence software would go
>> > elsewhere, and not necesssarily to users of Cadence, who would maybe
>> > hire someone to "do this one tweak" then lay them off again.
>> 
>> Then we enter a dark age, where corporate greed 
> 
> YM "bankruptcy"
> 
>> has sacrificed long
>> term prosperity for short term gain and left a world without any
>> significant ability to develop technology.  
> 
> Yup--because you took away Cadence's means to stay in business.

Actually, the corporations who rely on Cadence software to design their 
products would have taken away Cadence's means to stay in business, by
refusing to pay for software unless it had a draconian license attached
to it.

-- 
Michael Bruce
[EMAIL PROTECTED]

---
Message 32 in thread    
From: Alaric Fox ([EMAIL PROTECTED]) Subject: Re: Address Change Advisory 
Newsgroups: rec.arts.sf.written.robert-jordan   
View this article only  
Date: 2001-06-13 04:17:22 PST   
In article <[EMAIL PROTECTED]>, 
[EMAIL PROTECTED] says...
> On Sun, 10 Jun 2001 18:18:26 -0000, Michael Bruce
<[EMAIL PROTECTED]> wrote:
> 
> >> But most of the hard core Stallman freaks are either transplants
from
> >> an alien universe where everyone has the time, energy, inclination,
> >> and expertise to go code diving into massive applications to remove
> >> features they don't like, or, more likely, are people who've been
> >> brought up using boring software to begin with and haven't the
> >> faintest idea about the very messy real world outside of basic
> >> operating systems, word processors, spreadsheets, and browsers
> >> (which everyone knows how to design.) 
> >...poorly.  That is, "which everyone knows how to design poorly".
> 
> Irrelevant, really.
> 
> There are two broad classes of applications, here, Bruce:  Those which
> are in the common experience, and those which are not.  Everyone of a
> given skill level in software design can spit out (given time and
> motivation) things in the rpevious category.  Everyone nows what a
> browser is supposed to do, because everyone has used one.  Everyone
> knows what a word processor is supposed to do, because everyone has
> used one.  Everyone knows what an operating system is supposed to do
> because not only has everyone used one, but it's a required course in
> many curricula.  
> 
I'd say that the same "everyone" that knows what a browser or a word 
processor is supposed to do is not the same "everyone" that knows what an

operating system *is*, much less "is supposed to do".  As far as being 
able to *write* one based off of maybe one 400-level course requiring a 
'C' or better, I'd say no way.

> CAD programs, for instance, are decidedly NOT in that former class,
> but in the latter.  Having an ink-wet degree in computer science is
> not even close to being enough to design a real CAD program.  Nor is
> an ink-wet CS degree and an ink-wet EE degree.  Mainly, because the
> requisite background knowledge of just what a CAD program is supposed
> to do, along with all the sophisticated models involved, just ain't
> there yet.
> 
The vast majority of programmers have little to no knowledge of what they

are programming -- they just know how.  Generally, architects (who may 
have no programming knowledge) define the algorithms and functions that a

programmer must implement.  In the CAD and similar fields, the 
programmers themselves probably don't know the basics of what they're 
doing.  If Arizona State University is typical, software engineers by 
design have less hard science and math requirements than the real 
engineering programs.

-- 
Alaric

-------

end long bit

> 
> > The problem is, linux emulation etc. doesn't keep up with product
> > revisions.  I.e. by the time they get it right and working a new
> > version (and all the associated games / software) have moved on to
the
> > next version.
> 
> You may be right. As I said, I wasn't making any claims about games,
and
> I know very little about gaming on Linux.

It's true for a number of things.  The .net clone 'Mono' will probably
never catch up in functionality with .net.

> > Notepad is better in every way.
> 
> Oh. I use Notepad sometimes too. What is the file size limit that it
can
> open? Anyway, I'm still not sure what we are discussing here. My point
> was simply that lots of people exchange MS Office document files, and
> that they can be read and written with various Linux software programs
> if necessary. I wasn't recommending it, only saying it was possible if
> needed.

In Windows NT or above?  How much memory + Page file ya got?

In Win 9x use Edit.

Reply via email to