Re: Magma

[Dima Pasechnik]

On 2013-09-03, rjf <fat...@gmail.com> wrote:
> Octave, Maxima, Sage....
> vs
> Matlab, Mathematica, Magma
>
> It seems that people want professionally supported products
> (whatever that might turn out to be, in reality). And they do not
> care so much about open source.
Who do you mean by "people"? These who have more than a couple of billions in the
bank?
Magma lobbists did a good job on Simons, indeed.
So with one hand the foundation publishes things like
https://www.simonsfoundation.org/quanta/20130222-in-computers-we-trust/
and with the other (which does control the money)
does a funding step so much out of line with this publication...
It seems like these two hands not only don't coordinate with each other,
but are actually unaware of each other's existence...

By the way, have you ever used Magma or Matlab yourself?
To save you trouble doing a reality check,
you might like reading this:
http://abandonmatlab.wordpress.com/

>
> RJF
>
>
>

[rjf]

On Monday, September 2, 2013 10:46:24 PM UTC-7, Dima Pasechnik wrote:

On 2013-09-03, rjf <fat...@gmail.com> wrote:
> Octave, Maxima, Sage....
> vs
> Matlab, Mathematica, Magma
>
>  It seems that people want professionally supported products
> (whatever that might turn out to be, in reality).  And they do not
> care so much about open source.
Who do you mean by "people"? These who have more than a couple of billions in the
bank?

I was thinking of the professors at colleges who use Matlab themselves and require
that their students use Matlab as well.  I was also thinking of the people who develop
and sell applications that use Matlab as a base.  I was thinking of research projects
like Chebfun which use Matlab,  and even make use of object-oriented programming
stuff.
 

Magma lobbists did a good job on Simons, indeed.
So with one hand the foundation publishes things like
https://www.simonsfoundation.org/quanta/20130222-in-computers-we-trust/

I just read this.  In my opinion, it shows some professors (Teleman at Berkeley,
for example) of being -- at best -- peculiar.   It shows Zeilberger using Maple
(not free)
 

and with the other (which does control the money)
does a funding step so much out of line with this publication...

 

It seems like these two hands not only don't coordinate with each other,
but are actually unaware of each other's existence...

By the way, have you ever used Magma or Matlab yourself?

I have used Matlab occasionally, beginning when it was just a (free, open source)
FORTRAN library.  At one point I wrote a Matlab parser (in Lisp), with the
intention of making a simulation in Lisp.  Macsyma has such a feature, though
Macsyma is not distributed, really.

 I do not recall using Magma, though I think  I used Cayley to try out.  I recall
being  quite
unfavorably impressed by the "language" part of Cayley, and the revision of
the language represented by Magma was not, in my opinion, a good one.

To save you trouble doing a reality check,
you might like reading this:
http://abandonmatlab.wordpress.com/

I think the person writing this really doesn't understand who is using Matlab
and its toolkits, and why.  There are many critiques of Matlab, and this one
is probably among the least relevant I have seen.  What about a critique
of Matlab from the perspective of (say) its eigenvalue routines?
Or its speed compared to LAPACK.

That would be relevant.

>
> RJF
>
>
>

[Dima Pasechnik]

On 2013-09-03, rjf <fat...@gmail.com> wrote:
>
>
> On Monday, September 2, 2013 10:46:24 PM UTC-7, Dima Pasechnik wrote:
>>

>> On 2013-09-03, rjf <fat...@gmail.com <javascript:>> wrote:
>> > Octave, Maxima, Sage....
>> > vs
>> > Matlab, Mathematica, Magma
>> >
>> > It seems that people want professionally supported products
>> > (whatever that might turn out to be, in reality). And they do not
>> > care so much about open source.
>> Who do you mean by "people"? These who have more than a couple of billions
>> in the
>> bank?
>
>
> I was thinking of the professors at colleges who use Matlab themselves
> and require that their students use Matlab as well.

I tend to question ethics of professors peddling products
like Matlab to kids, if it's not about research, but just a
teaching tool. Hell, where I work at the moment, they use Matlab for
an introductory programming course...

> I was also
> thinking of the people who develop and sell applications that use
> Matlab as a base.

Such people surely exist and have vested interests in supressing Matlab
competitors.

> I have used Matlab occasionally, beginning when it was just a (free, open
> source)
> FORTRAN library. At one point I wrote a Matlab parser (in Lisp), with the
> intention of making a simulation in Lisp. Macsyma has such a feature,
> though
> Macsyma is not distributed, really.

Then you at least know that in Matlab
0==[0], etc...

>
> I do not recall using Magma, though I think I used Cayley to try out. I
> recall
> being quite
> unfavorably impressed by the "language" part of Cayley, and the revision of
> the language represented by Magma was not, in my opinion, a good one.
>
> To save you trouble doing a reality check,
>> you might like reading this:
>> http://abandonmatlab.wordpress.com/
>>
>> I think the person writing this really doesn't understand who is using
> Matlab and its toolkits, and why.

Hmm, lots of well-qualified in CS people are forced to use Matlab for
one or another reason. And hate it, for reasons explained in the blog.
E.g. it has quite a bit of Windows-only functionality, and sane people
tend not to do big computations on Windows...

> There are many critiques of Matlab, and this one
> is probably among the least relevant I have seen. What about a critique
> of Matlab from the perspective of (say) its eigenvalue routines?
> Or its speed compared to LAPACK.
>
> That would be relevant.

it's well-known that Matlab incorporates some LAPACK version, which for
obvious reasons need not be properly tuned for the hardware you have.
Need I say more?


>
>>
>> > RJF
>> >
>> >
>> >
>>
>>
>

[rjf]

On Tuesday, September 3, 2013 10:34:07 AM UTC-7, Dima Pasechnik wrote:

On 2013-09-03, rjf <fat...@gmail.com> wrote:
>
>
> On Monday, September 2, 2013 10:46:24 PM UTC-7, Dima Pasechnik wrote:
>>
>> On 2013-09-03, rjf <fat...@gmail.com <javascript:>> wrote:
>> > Octave, Maxima, Sage....
>> > vs
>> > Matlab, Mathematica, Magma
>> >
>> >  It seems that people want professionally supported products
>> > (whatever that might turn out to be, in reality).  And they do not
>> > care so much about open source.
>> Who do you mean by "people"? These who have more than a couple of billions
>> in the
>> bank?
>
>
> I was thinking of the professors at colleges who use Matlab themselves
> and require that their students use Matlab as well.  

I tend to question ethics of professors peddling products
like Matlab to kids, if it's not about research, but just a
teaching tool. Hell, where I work at the moment, they use Matlab for
an introductory programming course...

At least at UC Berkeley, students get access free, while they are students.
And, while I am not sure about the current course, the engineers indeed
used it in an intro course.   (Not the computer science course...)
 

> I was also
> thinking of the people who develop and sell applications that use
> Matlab as a base.
Such people surely exist and have vested interests in supressing Matlab
competitors.

Not really,  If their code runs on Octave also, that makes their sales
base wider.  Though it is unclear why someone would be willing to
pay for some application toolkit (possibly big bucks) and try to
run it on a free platform simulation.

I think it is worth pointing out that to many people of certain limited
perspective, the alternative to Matlab is FORTRAN, and the major
difference is that Matlab is interactive and FORTRAN is batch.

Some people continue to use programming languages
that are essentially batch: edit/compile/link/load/execute  -- loop.

It strikes me as terrible.   Oh I forgot.  Python is not interactive.

But IPython uh, is interactive?  Do people always use IPython then?
Sort of like Matlab.  Interactive,.



 

> I have used Matlab occasionally, beginning when it was just a (free, open
> source)
> FORTRAN library.  At one point I wrote a Matlab parser (in Lisp), with the
> intention of making a simulation in Lisp.  Macsyma has such a feature,
> though
> Macsyma is not distributed, really.

Then you at least know that in Matlab
0==[0], etc...

I haven't thought about it, but if everything you know about is a matrix, then
why not identify a scalar as a 1x1 matrix.  It's an engineering issue not
a formal statement about mathematics.
 

>
>  I do not recall using Magma, though I think  I used Cayley to try out.  I
> recall
> being  quite
> unfavorably impressed by the "language" part of Cayley, and the revision of
> the language represented by Magma was not, in my opinion, a good one.
>
> To save you trouble doing a reality check,
>> you might like reading this:
>> http://abandonmatlab.wordpress.com/
>>
>> I think the person writing this really doesn't understand who is using
> Matlab  and its toolkits, and why.  
Hmm, lots of well-qualified in CS people are forced to use Matlab for
one or another reason. And hate it, for reasons explained in the blog.

I have some criticisms about Matlab, but I never encountered any of the blog
problems.
 

E.g. it has quite a bit of Windows-only functionality, and sane people
tend not to do big computations on Windows...

Gratuitous insults.  Since Sage apparently still does not run natively on Windows??
 

> There are many critiques of Matlab, and this one
> is probably among the least relevant I have seen.  What about a critique
> of Matlab from the perspective of (say) its eigenvalue routines?
> Or its speed compared to LAPACK.
>
> That would be relevant.

it's well-known that Matlab incorporates some LAPACK version, which for
obvious reasons need not be properly tuned for the hardware you have.
Need I say more?

It didn't always include LAPACK.

Frankly, having an interactive interface to an un-tuned version of LAPACK
might very well be worth the price of admission for some people.

I'm guessing there is an iPython interface to LAPACK.  But if Python is
so great, when do we see a version written in Python?  Or maybe there
is one but it is, um,  un-tuned??  Sorry.  I wandered off topic. Is Sage
rewriting Magma in Python?

RJF

[Tom Boothby]

On Tue, Sep 3, 2013 at 1:05 PM, rjf <fat...@gmail.com> wrote:

> It strikes me as terrible. Oh I forgot. Python is not interactive.

What crack do you smoke?

[Robert Bradshaw]

On Tue, Sep 3, 2013 at 1:05 PM, rjf <fat...@gmail.com> wrote:
>

> On Tuesday, September 3, 2013 10:34:07 AM UTC-7, Dima Pasechnik wrote:
>>
>> On 2013-09-03, rjf <fat...@gmail.com> wrote:
>> >
>> >
>> > On Monday, September 2, 2013 10:46:24 PM UTC-7, Dima Pasechnik wrote:
>> >>
>> >> On 2013-09-03, rjf <fat...@gmail.com <javascript:>> wrote:
>> >> > Octave, Maxima, Sage....
>> >> > vs
>> >> > Matlab, Mathematica, Magma
>> >> >
>> >> > It seems that people want professionally supported products
>> >> > (whatever that might turn out to be, in reality). And they do not
>> >> > care so much about open source.
>> >> Who do you mean by "people"? These who have more than a couple of
>> >> billions
>> >> in the
>> >> bank?
>> >
>> > I was thinking of the professors at colleges who use Matlab themselves
>> > and require that their students use Matlab as well.
>>
>> I tend to question ethics of professors peddling products
>> like Matlab to kids, if it's not about research, but just a
>> teaching tool. Hell, where I work at the moment, they use Matlab for
>> an introductory programming course...
>>
> At least at UC Berkeley, students get access free, while they are students.
> And, while I am not sure about the current course, the engineers indeed
> used it in an intro course. (Not the computer science course...)

The first hit is free. Once they graduate they (or someone else) must
fork out the money to keep up their habit. Similarly if they want to
share their work with others.

>> > I was also
>> > thinking of the people who develop and sell applications that use
>> > Matlab as a base.
>> Such people surely exist and have vested interests in supressing Matlab
>> competitors.
>>
> Not really, If their code runs on Octave also, that makes their sales
> base wider. Though it is unclear why someone would be willing to
> pay for some application toolkit (possibly big bucks) and try to
> run it on a free platform simulation.
>
> I think it is worth pointing out that to many people of certain limited
> perspective, the alternative to Matlab is FORTRAN, and the major
> difference is that Matlab is interactive and FORTRAN is batch.

Ignorance + what they "grew up with" (see above) is often a huge, if
not the dominant, factor in people's choice of tools.

> Some people continue to use programming languages
> that are essentially batch: edit/compile/link/load/execute -- loop.
>
> It strikes me as terrible. Oh I forgot. Python is not interactive.

You're ignorance is showing here; see above.

> But IPython uh, is interactive? Do people always use IPython then?
> Sort of like Matlab. Interactive,.

IPython is an even nicer interactive environment than the one provided
by Python.

> --
> You received this message because you are subscribed to the Google Groups
> "sage-flame" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-flame+...@googlegroups.com.
> To post to this group, send email to sage-...@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-flame.
> For more options, visit https://groups.google.com/groups/opt_out.

[rjf]

y'know, I thought python was interactive, and that was how I had used
it (very briefly).  But then just checking on google I found iPython.
so what the hell was that for??

So to reiterate the point I was trying to make, for some people the
choice is  interactive (Matlab)  vs FORTRAN (batch).
There are other programming language implementations in each
category.  Presumably including interactive FORTRAN and
batch Matlab.  No matter...

To some people, Matlab defines interactive, and they like that.
End of discussion.  Don't confuse them with facts.

 

[William Stein]

On Tue, Sep 3, 2013 at 7:08 PM, rjf <fat...@gmail.com> wrote:
>
>
> On Tuesday, September 3, 2013 1:13:57 PM UTC-7, Tom wrote:
>>
>> On Tue, Sep 3, 2013 at 1:05 PM, rjf <fat...@gmail.com> wrote:
>>
>> > It strikes me as terrible. Oh I forgot. Python is not interactive.
>>
>> What crack do you smoke?
>
>
> y'know, I thought python was interactive, and that was how I had used
> it (very briefly). But then just checking on google I found iPython.
> so what the hell was that for??

Have you ever used iPython or even Python?



--
William Stein
Professor of Mathematics
University of Washington
http://wstein.org

[Richard Fateman]

On 9/3/2013 7:31 PM, William Stein wrote:
> On Tue, Sep 3, 2013 at 7:08 PM, rjf <fat...@gmail.com> wrote:
>>
>> On Tuesday, September 3, 2013 1:13:57 PM UTC-7, Tom wrote:
>>> On Tue, Sep 3, 2013 at 1:05 PM, rjf <fat...@gmail.com> wrote:
>>>
>>>> It strikes me as terrible. Oh I forgot. Python is not interactive.
>>> What crack do you smoke?
>>
>> y'know, I thought python was interactive, and that was how I had used
>> it (very briefly). But then just checking on google I found iPython.
>> so what the hell was that for??
> Have you ever used iPython or even Python?

Never used iPython. I have looked at Python programs, e.g. the article
by Peter Norvig
comparing Lisp and Python. I may have run the beginning of some Python
tutorial
years ago, but I do not recall ever doing anything "useful" in Python.
I have at one
time used a library set up for Python, calling it from Lisp. It may
have be some relative
of GMP. I found the parts that were exposed in that library to be
unsatisfactory, probably
because of memory control.

I've also read some comments about Python in the context of Sage, e.g.
comments that
suggest that Python and Sage don't work together too well -- they have
different integer
types, for a start.

Does this make it plausible to pontificate about Python? Well, anyone
can pontificate. People writing
about Lisp sometimes amuse themselves by ranting about parentheses,
something
that is totally a non-issue when using an appropriate editor.
Are my views on Python meaningful? Eh, depends on whether you agree
with them or not,
I suppose.

RJF



>
>
>

[William Stein]

So just to clarify, in your entire life you have not spent even one
second using any of the following programs:

* Python
* IPython
* Sage
* Magma

[ ] Yes, I have never *used* any of the above.

[ ] No, I have actually used at least one of them for at least one
second (reading a paper with code doesn't count).


-- William

>
>
>
>>
>>
>>
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-flame" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-flame+...@googlegroups.com.
> To post to this group, send email to sage-...@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-flame.
> For more options, visit https://groups.google.com/groups/opt_out.

[Tom Boothby]

On Tue, Sep 3, 2013 at 7:08 PM, rjf <fat...@gmail.com> wrote:

> y'know, I thought python was interactive, and that was how I had used
> it (very briefly). But then just checking on google I found iPython.
> so what the hell was that for??

I applied your amazing algorithm, and have used it deduce that BSD
does not have a windowing system, because obviously Mac OSX is a
windowing system for BSD.

Gosh, this is some nice crack. Life is much better now!

[Dima Pasechnik]

On 2013-09-03, rjf <fat...@gmail.com> wrote:
>
>
> On Tuesday, September 3, 2013 10:34:07 AM UTC-7, Dima Pasechnik wrote:
>>

>> On 2013-09-03, rjf <fat...@gmail.com <javascript:>> wrote:
>> >
>> >
>> > On Monday, September 2, 2013 10:46:24 PM UTC-7, Dima Pasechnik wrote:
>> >>
>> >> On 2013-09-03, rjf <fat...@gmail.com <javascript:>> wrote:
>> >> > Octave, Maxima, Sage....
>> >> > vs
>> >> > Matlab, Mathematica, Magma
>> >> >
>> >> > It seems that people want professionally supported products
>> >> > (whatever that might turn out to be, in reality). And they do not
>> >> > care so much about open source.
>> >> Who do you mean by "people"? These who have more than a couple of
>> billions
>> >> in the
>> >> bank?
>> >
>> >
>> > I was thinking of the professors at colleges who use Matlab themselves
>> > and require that their students use Matlab as well.
>>
>> I tend to question ethics of professors peddling products
>> like Matlab to kids, if it's not about research, but just a
>> teaching tool. Hell, where I work at the moment, they use Matlab for
>> an introductory programming course...
>>
>> At least at UC Berkeley, students get access free, while they are students.

great, and so they create a life-time Matlab dependence habit.
Should Philip Morris supply free sigs to UC Berkeley undegrads?

> And, while I am not sure about the current course, the engineers indeed
> used it in an intro course. (Not the computer science course...)
>
>
>> > I was also
>> > thinking of the people who develop and sell applications that use
>> > Matlab as a base.
>> Such people surely exist and have vested interests in supressing Matlab
>> competitors.
>>
>> Not really, If their code runs on Octave also, that makes their sales
> base wider. Though it is unclear why someone would be willing to
> pay for some application toolkit (possibly big bucks) and try to
> run it on a free platform simulation.
>
> I think it is worth pointing out that to many people of certain limited
> perspective, the alternative to Matlab is FORTRAN, and the major
> difference is that Matlab is interactive and FORTRAN is batch.
>
> Some people continue to use programming languages
> that are essentially batch: edit/compile/link/load/execute -- loop.
>
> It strikes me as terrible. Oh I forgot. Python is not interactive.

And pigs fly...

>
> But IPython uh, is interactive? Do people always use IPython then?
> Sort of like Matlab. Interactive,.
>
>
>
>
>
>> > I have used Matlab occasionally, beginning when it was just a (free,
>> open
>> > source)
>> > FORTRAN library. At one point I wrote a Matlab parser (in Lisp), with
>> the
>> > intention of making a simulation in Lisp. Macsyma has such a feature,
>> > though
>> > Macsyma is not distributed, really.
>>
>> Then you at least know that in Matlab
>> 0==[0], etc...
>>
>> I haven't thought about it, but if everything you know about is a matrix,
> then
> why not identify a scalar as a 1x1 matrix. It's an engineering issue not
> a formal statement about mathematics.
>
>
>> >
>> > I do not recall using Magma, though I think I used Cayley to try out.
>> I
>> > recall
>> > being quite
>> > unfavorably impressed by the "language" part of Cayley, and the revision
>> of
>> > the language represented by Magma was not, in my opinion, a good one.

Cayley was the first computer algebra system I was using, back in 1992,
and I didn't like it at all. Any nontrivial data conversion had to be
done by printing data to a file, editing it, and reading it back again,
etc...

>> >
>> > To save you trouble doing a reality check,
>> >> you might like reading this:
>> >> http://abandonmatlab.wordpress.com/
>> >>
>> >> I think the person writing this really doesn't understand who is using
>> > Matlab and its toolkits, and why.
>> Hmm, lots of well-qualified in CS people are forced to use Matlab for
>> one or another reason. And hate it, for reasons explained in the blog.
>>
>
> I have some criticisms about Matlab, but I never encountered any of the blog
> problems.
>
>
>> E.g. it has quite a bit of Windows-only functionality, and sane people
>> tend not to do big computations on Windows...
>>
>
> Gratuitous insults.

Let me put it this way. You might like to go and dig up the stats on
usage of Windows for HPC things, and I bet you won't get much better
figures than for smartphones running Windows...
Needless to say, peddling Matlab as any sort of universal platform for
scientific computations is an insult to consumers intelligence, e.g. for
the fact that several things are Windows-only in Matlab.

Perhaps you do have a serious stake in Mathworks and Microsoft,
and then indeed not screaming "Gratuitous insults"
would be hurting your retirement...

> Since Sage apparently still does not run natively on
> Windows??
>

it runs if you want to fiddle with Cygwin.

>
>>
>> > There are many critiques of Matlab, and this one
>> > is probably among the least relevant I have seen. What about a critique
>> > of Matlab from the perspective of (say) its eigenvalue routines?
>> > Or its speed compared to LAPACK.
>> >
>> > That would be relevant.
>>
>> it's well-known that Matlab incorporates some LAPACK version, which for
>> obvious reasons need not be properly tuned for the hardware you have.
>> Need I say more?
>>
>
> It didn't always include LAPACK.
>
> Frankly, having an interactive interface to an un-tuned version of LAPACK
> might very well be worth the price of admission for some people.
>
> I'm guessing there is an iPython interface to LAPACK. But if Python is
> so great, when do we see a version written in Python?

You could have seen it already in 2005. It's called numpy (http://www.numpy.org/),
a python library that interfaces LAPACK.
You may build it with an optimized for the platform LAPACK, naturally.

> Or maybe there
> is one but it is, um, un-tuned?? Sorry. I wandered off topic. Is Sage
> rewriting Magma in Python?

Magma has a peculiar "business model". It invites researchers who
developed a highly sophisticated algorithm for problem X
<insert your favourite hard algebraic, say, problem here> to implement it
in Magma, hosts them for few months, if needed, and then as a result this algorithm
is only available in Magma.
Now, if you and your students and collaborators need to solve X in your research,
you have to pay 10K$ to get sufficently many copies of Magma licence.
Now, if there are 10 groups who would end up spending that much on
Magma, Magma would recover their costs, 10-fold at least...
So much for them being non-profit. They might be non-profit on average,
but locally it's quite profitable, and you feel in in your pocket.

I can myself name at least one such X I would dearly want to use, and
which is only available in Magma, following the scenario above, in fact.
(I certainly can afford 10 Magma licences at present, but I find this
largerly thrown away money - at the very least I'd need a perpetual
licence, not a fixed term, and something that I can give people to try).






>
> RJF
>
>
>
>
>>
>

[rjf]

On Tuesday, September 3, 2013 7:31:34 PM UTC-7, William Stein wrote:

On Tue, Sep 3, 2013 at 7:08 PM, rjf <fat...@gmail.com> wrote:
>
>
> On Tuesday, September 3, 2013 1:13:57 PM UTC-7, Tom wrote:
>>
>> On Tue, Sep 3, 2013 at 1:05 PM, rjf <fat...@gmail.com> wrote:
>>
>> > It strikes me as terrible.   Oh I forgot.  Python is not interactive.
>>
>> What crack do you smoke?
>
>
> y'know, I thought python was interactive, and that was how I had used
> it (very briefly).  But then just checking on google I found iPython.
> so what the hell was that for??

Have you ever used iPython or even Python?

So far as I know, I have never used iPython, unless it was part of some tutorial that
I might have briefly visited.  I have used Python or a simulation of Python in
some tutorial, and I believe I have downloaded a python system on some
computer. 
 I am pretty sure that I have not run a python program to compute anything
of any particular interest.  But then I often write programs when the objective
is to write/debug/time a procedure,  and not especially relish the returned
value  (e.g.  multiplying polynomials with random coefficients
).

RJF

[rjf]

[ X ] No, I have actually used at least one of them for at least one

second (reading a paper with code doesn't count).

As I've said.   I will even provide you more info.  I've never downloaded Sage.
I have probably never used an on-line version of Sage, but I'm not sure.  My probable
intent for trying Sage would be to find some more-or-less embarrassing glitch,
but  I'm not sure I ever did it. 

As for Magma, I may have been offered a free trial at some point ("beta test",
probably before the term beta test became so common).  Since there is nothing
in my research that particularly requires computations done by Magma, my
concern was primarily in the language design, which I already commented on.
I recall thinking it was a poor design.  I doubt that I used it, even if I had a free
copy.

And as for not counting "reading a paper with code"... That is a stupid
restriction.  In computer science papers there may be discussions of language
features -- sometimes quite heated discussions -- which could not be
"used"  because they were not yet (or ever) implemented.  Sometimes whole
languages are described but never implemented.  

Two examples that come to mind in the Lisp family -- Scheme and NIL 
("new implementation of lisp") in which the implementation followed the
presentation of the language in papers.


RJF

[rjf]

On Tuesday, September 3, 2013 11:28:08 PM UTC-7, Tom wrote:

On Tue, Sep 3, 2013 at 7:08 PM, rjf <fat...@gmail.com> wrote:

> y'know, I thought python was interactive, and that was how I had used
> it (very briefly).  But then just checking on google I found iPython.
> so what the hell was that for??

I applied your amazing algorithm, and have used it deduce that BSD
does not have a windowing system, because obviously Mac OSX is a
windowing system for BSD.

Your algorithm, such as it is, got you the right answer.  Bitmap displays
were not in evidence when BSD was written and first distributed, circa 1979
I think.  The BSD development was done with character display terminals;
generally ADM-3a terminals with (I think) 80 character lines.  Maybe
40 lines? Some higher character-count displays were also used -- I
can visualize hem -- white plastic with a blue raised stripe around the sides,
but I don't recall the brand.

There must have been some graphics displays since we also had a
graphics (xerographic) printer.  But there was no window system.

Later there were several competing graphics systems that provided
for "terminal" subwindows and more.  The ones that come to mind are
the X window system (from MIT, not Berkeley), Sun graphics
(from Sun or maybe Stanford ).  There was something from Bell
Labs on an unusual terminal  "BIT", and there was the Andrew window
system  from CMU.
Also Xerox, but that wasn't UNIX-based.
Maybe others.  Eventually HP, DEC, .... joined the game.  X-windows
seems to have survived as X-11, overcoming some severe issues including
pretty much total lack of security.  As I recall people could read from
or write to each others' screens on X systems.

But BSD from Berkeley initially had no windowing, though it presumably
had cursor positioning primitives for character displays like ADM3a
pretty early on.  Bill Joy wrote the VI editor early on.

Gosh, this is some nice crack.  Life is much better now!

Your ignorance never disappoints.
 

[rjf]

On Wednesday, September 4, 2013 12:38:46 AM UTC-7, Dima Pasechnik wrote:

>> I tend to question ethics of professors peddling products
>> like Matlab to kids, if it's not about research, but just a
>> teaching tool. Hell, where I work at the moment, they use Matlab for
>> an introductory programming course...
>>

>> (RJF) At least at UC Berkeley, students get access free, while they are students.

great, and so they create a life-time Matlab dependence habit.
Should Philip Morris supply free sigs to UC Berkeley undegrads?

The designation of Matlab was done by faculty in mechanical and civil engineering,
not computer scientists. 

I think there is also an issue that vast numbers of donated computers running the
Intel instruction set, with donated Windows software...tends to get students
"hooked" on commercially available hardware and software;  this is presumably
one reason for the donations.

Requiring students to build their own hardware and software starting with
sand scooped up from the beach is not a practical alternative. 

There is some intermediate stage though.  I taught a course (intro
to architecture) where students, in the course had to use
lthe assembly language of MIPS, not Intel.  They used a MIPS
simulator on intel..

...

 

>>
>> Not really,  If their code runs on Octave also, that makes their sales
> base wider.  Though it is unclear why someone would be willing to

..

Let me put it this way. You might like to go and dig up the stats on
usage of Windows for HPC things, and I bet you won't get much better
figures than for smartphones running Windows...

How would I go about telling the statisticians that I am using a Windows
system for scientific computing?

High performance these days comes with banks of computers running
special software.  Is most numerical computing done in HPC frameworks,
or is it just done by thousands of engineering firms checking out the
strength of bridges or roads or buildings?

 

Needless to say, peddling Matlab as any sort of universal platform for
scientific computations is an insult to consumers intelligence, e.g. for
the fact that several things are Windows-only in Matlab.

Perhaps you do have a serious stake in Mathworks and Microsoft,
and then indeed not screaming "Gratuitous insults"
would be hurting your retirement...

I do not own stock in Mathworks or Microsoft or Apple  directly. 
 I would not be surprised
(at least regarding Microsoft and Apple) if I have an indirect
stake through some pension fund.  I doubt that it would qualify
as "serious stake".

 

> Since Sage apparently still does not run natively on
> Windows??
>

it runs if you want to fiddle with Cygwin.

It also runs if I have a dual-boot computer.
Reminds me of an old riddle.

How many legs does a sheep have if you call its tail a leg?

(answer at extreme bottom..)

 

>
> I'm guessing there is an iPython interface to LAPACK.  But if Python is
> so great, when do we see a version written in Python?  

You could have seen it already in 2005. It's called numpy (http://www.numpy.org/),
a python library that interfaces LAPACK.

If I understand what you saying, LAPACK is not written in Python, as I said.
LAPACK is available as a library, but it is written in  FORTRAN or C.
 

You may build it with an optimized for the platform LAPACK, naturally.

Sure.

> Or maybe there
> is one but it is, um,  un-tuned??  Sorry.  I wandered off topic. Is Sage
> rewriting Magma in Python?

Magma has a peculiar "business model". It invites researchers who
developed a highly sophisticated algorithm for problem X
<insert your favourite hard algebraic, say, problem here> to implement it
in Magma, hosts them for few months, if needed, and then as a result this algorithm
is only available in Magma.
Now, if you and your students and collaborators need to solve X in your research,
you have to pay 10K$ to get sufficently many copies of Magma licence.

Can't you buy just one?  Or use someone else's copy off-site?
 

Now, if there are 10 groups who would end up spending that much on
Magma, Magma would recover their costs, 10-fold at least...
So much for them being non-profit. They might be non-profit on average,
but locally it's quite profitable, and you feel in in your pocket.

I think the notion of non-profit has to do with taxes.  There are many
non-profit hospitals that collect huge sums of money and pay out huge
sums of money to administrators, drug suppliers, etc.  They are
non-profit and pay (no?) taxes because they have no stock holders to
whom they pay dividends.

Schools are non-profit, often but not always.

I can myself name at least one such X I would dearly want to use, and
which is only available in Magma, following the scenario above, in fact.
(I certainly can afford 10 Magma licences at present, but I find this
largerly thrown away money - at the very least I'd need a perpetual
licence, not a fixed term, and something that I can give people to try).

That is too bad. maybe you should see if they will offer you a free
license in exchange for your supplying them with some code?

RJF

>

answer to riddle.
4 legs.  Calling its tail a leg does not make it a leg.
 

[Richard Fateman]

On 9/4/2013 7:41 AM, rjf wrote:
(stuff snipped, re BSD windows)
The current article on Wikipedia X11 looks pretty good; certainly
better than my off-the-cuff comments. That Bell Labs display was BLIT
not BIT. It also puts the release date for BSD at 1978,
which predates any window system for BSD, but not
the ground-breaking work from Xerox, including Altos and then Stars.

Berkeley had a bunch of Stars, but I think they arrived after 1978.

RJF

[Tom Boothby]

> On Tuesday, September 3, 2013 11:28:08 PM UTC-7, Tom wrote:
>>
>> On Tue, Sep 3, 2013 at 7:08 PM, rjf <fat...@gmail.com> wrote:
>>
>> I applied your amazing algorithm, and have used it deduce that BSD
>> does not have a windowing system, because obviously Mac OSX is a
>> windowing system for BSD.
>
> Your algorithm, such as it is, got you the right answer. Bitmap displays
> were not in evidence when BSD was written and first distributed, circa 1979
> I think.

Man, this is some great shit, man. Now, cars don't have cupholders
'cause I couldn't find one in a Model T!

>> Gosh, this is some nice crack. Life is much better now!
>
> Your ignorance never disappoints.

And it never will. I live to serve. So long as you don't bogart the crack.

[Richard Fateman]

On 9/4/2013 12:02 PM, Tom Boothby wrote:
> Man, this is some great shit, man. Now, cars don't have cupholders
> 'cause I couldn't find one in a Model T!
>

Sarcasm fails you.

Again


I suppose if you visited the American Museum of Natural History you would
conclude that dinosaurs had metal rods up their butts.

RJF

[Tom Boothby]

On Wed, Sep 4, 2013 at 12:46 PM, Richard Fateman <fat...@gmail.com> wrote:
> On 9/4/2013 12:02 PM, Tom Boothby wrote:
>>
>> Man, this is some great shit, man. Now, cars don't have cupholders
>> 'cause I couldn't find one in a Model T!
>>
> Sarcasm fails you.

Oh, that old chestnut... but isn't your typical complaint that my
"attempts" at sarcasm are undetectable? Here you've registered a
false positive, for this was pure and unadulterated mockery. So the
failure is yours.

Again.

> I suppose if you visited the American Museum of Natural History you would
> conclude that dinosaurs had metal rods up their butts.

I submit that if I visited the Emeritus office of UC Berkeley's CS
department, I would conclude that dinosaurs have metal rods up their
butts.

[Richard Fateman]

On 9/4/2013 1:28 PM, Tom Boothby wrote:

On Wed, Sep 4, 2013 at 12:46 PM, Richard Fateman <fat...@gmail.com> wrote:

On 9/4/2013 12:02 PM, Tom Boothby wrote:

Man, this is some great shit, man.  Now, cars don't have cupholders
'cause I couldn't find one in a Model T!

Sarcasm fails you.

Oh, that old chestnut... but isn't your typical complaint that my
"attempts" at sarcasm are undetectable?  Here you've registered a
false positive, for this was pure and unadulterated mockery.  So the
failure is yours.

Again.

no, I think you were attempting to employ sarcasm. See the current wikipedia article..

The use of strategies which, on the surface appear to be appropriate to the situation, but are meant to be taken as meaning the opposite in terms of face management.
  "some great shit".


Now the second part is not sarcasm but irony  (again, wikipedia.)
one meaning is stated and a different, usually antithetical, meaning is intended.

"cars don't have cupholders"

I suppose if you visited the American Museum of Natural History you would
conclude that dinosaurs had metal rods up their butts.

I submit that if I visited the Emeritus office of UC Berkeley's CS
department, I would conclude that dinosaurs have metal rods up their
butts.

I don't see this as fitting into a standard category.   Maybe pathetic jibe?

RJF

[Tom Boothby]

On Wed, Sep 4, 2013 at 2:04 PM, Richard Fateman <fat...@gmail.com> wrote:

>> Oh, that old chestnut... but isn't your typical complaint that my
>> "attempts" at sarcasm are undetectable? Here you've registered a
>> false positive, for this was pure and unadulterated mockery. So the
>> failure is yours.
>>
>> Again.
>>
> no, I think you were attempting to employ sarcasm. See the current wikipedia
> article..

Fuck, man. Repeating a false statement does not make it true.

"some great shit" was a reference to your crack which I'm getting so
much enjoyment out of -- this is genuine. I love it when you say
stupid shit. Not sarcasm, but an honest statement of opinion.

The cupholder analogy may be too subtle for the advanced state of your
advanced state of your advanced state of your dementia. You declared
that the initial release of BSD had no windowing system, and concluded
that I was ignorant because (implied) BSD has no windowing system.
That is, you selected a single (obsolete) example BSD which is
relatively uncommon in that most BSD variants come with X or some
other windowing system. You applied this atypical example to the
whole, implying that indeed, BSD does not have a windowing system.

By way of analogy, I selected a single obsolete example car which is
relatively uncommon in that most cars have cupholders. I applied this
atypical example to the whole, stating that cars do not have
cupholders. This absurd conclusion, analogous to yours, was made as
"an impertinent imitation". See definition 3b:
http://www.merriam-webster.com/dictionary/mockery

>> I submit that if I visited the Emeritus office of UC Berkeley's CS
>> department, I would conclude that dinosaurs have metal rods up their
>> butts.
>
> I don't see this as fitting into a standard category. Maybe pathetic jibe?

After I sent this, I was a little worried that in your advanced state
of your advanced state of dementia, you might have interpreted this as
a threat. Californians can get so touchy.

[rjf]

On Wednesday, September 4, 2013 3:55:11 PM UTC-7, Tom wrote:

On Wed, Sep 4, 2013 at 2:04 PM, Richard Fateman <fat...@gmail.com> wrote:

>> Oh, that old chestnut... but isn't your typical complaint that my
>> "attempts" at sarcasm are undetectable?  Here you've registered a
>> false positive, for this was pure and unadulterated mockery.  So the
>> failure is yours.
>>
>> Again.
>>
> no, I think you were attempting to employ sarcasm. See the current wikipedia
> article..

Fuck, man.  Repeating a false statement does not make it true.

Obscenity rarely clarifies a statement.  You seem to be tone-deaf in
terms of its use.  As well as sarcasm.
 

"some great shit" was a reference to your crack which I'm getting so
much enjoyment out of -- this is genuine.

 

 I love it when you say
stupid shit.

You find enjoyment in simple things.
 

 Not sarcasm, but an honest statement of opinion.

That's part of your problem too.  You simply don't understand humor.  You
also don't understand insult.  Or rhetoric.  Probably part of your cognitive
disability.
 

The cupholder analogy may be too subtle for the advanced state of your
advanced state of your advanced state of your dementia.  You declared
that the initial release of BSD had no windowing system, and concluded
that I was ignorant because (implied) BSD has no windowing system.

Actually, there's BSD, which you can read about, yet again, conveniently,
in wikipedia
http://en.wikipedia.org/wiki/Berkeley_Software_Distribution
and you can look for the word "windows" in it.  It occurs twice,
each time in the phrase "Microsoft Windows".

Does BSD "have a windowing system"?
In no essential or inevitable way does BSD have a windowing system.

Do Ford F150 pickup trucks have car deodorizer trees hanging from
the rear view mirror?  How about fuzzy dice?

That is, you selected a single (obsolete) example BSD which is
relatively uncommon in that most BSD variants come with X or some
other windowing system.

and some come with fuzzy dice?  Therefore one can conclude...?
 

 You applied this atypical example to the
whole, implying that indeed, BSD does not have a windowing system.

You imply that because some people later also loaded / distributed some window system
along with a BSD system that it is part of BSD (always).
 

By way of analogy, I selected a single obsolete example car which is
relatively uncommon in that most cars have cupholders.

Typical cars of today have built-in cupholders.  Different concept from
add-on accessory cupholders, which I suppose might be used on a Model T.

 I applied this
atypical example to the whole, stating that cars do not have
cupholders.  

Some don't, of course.
 

This absurd conclusion, analogous to yours, was made as
"an impertinent imitation".  See definition 3b:
http://www.merriam-webster.com/dictionary/mockery

Of course you are free to believe whatever you wish. You are
also pretty much free to post what you want.

If there are any others reading this, they may agree with
you, or with my characterization, which was "pathetic jibe".
 

>> I submit that if I visited the Emeritus office of UC Berkeley's CS
>> department, I would conclude that dinosaurs have metal rods up their
>> butts.
>

> I don't see this as fitting into a standard category.   Maybe pathetic jibe (PJ) ?

After I sent this, I was a little worried that in your advanced state
of your advanced state of dementia, you might have interpreted this as
a threat.

Sure. Jokes about dementia fit right in there. PJ.

 

 Californians can get so touchy.

I don't know where that came from, but why not go further off topic.?

[Tom Boothby]

On Wed, Sep 4, 2013 at 4:45 PM, rjf <fat...@gmail.com> wrote:
>
>> Fuck, man. Repeating a false statement does not make it true.
>>
> Obscenity rarely clarifies a statement. You seem to be tone-deaf in
> terms of its use. As well as sarcasm.

The word "fuck" has many uses, see Carlin for some examples. That one
wasn't meant for clarification, it was more of an expression of
amazement.

> Does BSD "have a windowing system"?
> In no essential or inevitable way does BSD have a windowing system.

Quoth wikipedia 'Today the term "BSD" is often used non-specifically
to refer to any of the BSD descendants which together form a branch of
the family of Unix-like operating systems.' Literally the second
sentence of the article.

>> I applied this
>> atypical example to the whole, stating that cars do not have
>> cupholders.
>
>
> Some don't, of course.

Stop the presses! He actually understands that sets can have multiple objects!

>> This absurd conclusion, analogous to yours, was made as
>> "an impertinent imitation". See definition 3b:
>> http://www.merriam-webster.com/dictionary/mockery
>
> Of course you are free to believe whatever you wish. You are
> also pretty much free to post what you want.

This is beyond absurd. You know that, right? We are directly
discussing the intent behind what I wrote a handful of hours ago. You
can interpret it however you like, but this is not a "belief" of mine.
This is fact: I was mocking you, not being sarcastic.

>> Californians can get so touchy.
>
> I don't know where that came from, but why not go further off topic.?

Yes, let's. Have some selected quotations from timecube.com

EARTH HAS 4 CORNER SIMULTANEOUS 4-DAY TIME CUBE WITHIN SINGLE
ROTATION. 4 CORNER DAYS PROVES 1 DAY 1 GOD IS TAUGHT EVIL. IGNORANCE
OF TIMECUBE4 SIMPLE MATH IS RETARDATION AND EVIL EDUCATION DAMNATION.
CUBELESS AMERICANS DESERVE - AND SHALL BE EXTERMINATED. I am a Knower
of 4 corner simultaneous 24 hour Days that occur within a single 4
corner rotation of Earth. Until you can tear and burn the bible to
escape the EVIL ONE, it will be impossible for your educated stupid
brain to know that 4 different corner harmonic 24 hour Days rotate
simultaneously within a single 4 quadrant rotation of a squared
equator and cubed Earth. The Solar system, the Universe, the Earth and
all humans are composed of + 0 - antipodes, and equal to nothing if
added as a ONE or Entity. All Creation occurs between Opposites.
Academic ONEism destroys +0- brain. If you would acknowledge simple
existing math proof that 4 harmonic corner days rotate simultaneously
around squared equator and cubed Earth, proving 4 Days, Not
1Day,1Self,1Earth or 1God that exists only as anti-side. This page you
see - cannot exist without its anti-side existence, as +0- antipodes.
Add +0- as One = nothing. Seek Awesome Lectures, MY WISDOM DEBUNKS
GODS OF ALL RELIGIONS AND ACADEMIA.

[Richard Fateman]

On 9/4/2013 6:03 PM, Tom Boothby wrote:
> On Wed, Sep 4, 2013 at 4:45 PM, rjf <fat...@gmail.com> wrote:
>>> Fuck, man. Repeating a false statement does not make it true.
>>>
>> Obscenity rarely clarifies a statement. You seem to be tone-deaf in
>> terms of its use. As well as sarcasm.
> The word "fuck" has many uses, see Carlin for some examples. That one
> wasn't meant for clarification, it was more of an expression of
> amazement.

Treated as an expression of amazement, it doesn't clarify.

>
>> Does BSD "have a windowing system"?
>> In no essential or inevitable way does BSD have a windowing system.
> Quoth wikipedia 'Today the term "BSD" is often used non-specifically
> to refer to any of the BSD descendants which together form a branch of
> the family of Unix-like operating systems.' Literally the second
> sentence of the article.

BSD can support a windowing interface, but it doesn't require it, nor
does BSD define
a windowing interface. Obstinacy buys you very little. Talk to someone,
maybe computer
science faculty member close to your home.

Of course you are free to believe whatever you wish. You are
also pretty much free to post what you want.

> This is beyond absurd.

That's not particularly helpful.
....

> This is fact: I was mocking you, not being sarcastic.

Eh, some people think they can carry a tune, but they can't.

RJF

[Tom Boothby]

On Wed, Sep 4, 2013 at 8:56 PM, Richard Fateman <fat...@gmail.com> wrote:

>> The word "fuck" has many uses, see Carlin for some examples. That one
>> wasn't meant for clarification, it was more of an expression of
>> amazement.
>
> Treated as an expression of amazement, it doesn't clarify.

Treated as a hammer, a water balloon doesn't drive nails, either. I
was amazed (though I cannot say why) at your stupidity. If you can't
fucking handle a little goddamned profanity, get the hell off of
sage-flame, jackass. Ultraconservative parents of 6 year-olds give me
less shit about the occasional "f-bomb" than you do. If you may
recall, this mailing list was created for the express purpose of
maintaining an outlet for pottymouths fighting over bullshit
tangentially related to sage while keeping it off sage-devel.

>
>>
>>> Does BSD "have a windowing system"?
>>> In no essential or inevitable way does BSD have a windowing system.
>>
>> Quoth wikipedia 'Today the term "BSD" is often used non-specifically
>> to refer to any of the BSD descendants which together form a branch of
>> the family of Unix-like operating systems.' Literally the second
>> sentence of the article.
>
> BSD can support a windowing interface, but it doesn't require it, nor does
> BSD define
> a windowing interface. Obstinacy buys you very little. Talk to someone,
> maybe computer
> science faculty member close to your home.

I do talk to CS faculty rather often, actually, and they're civil,
informative, and respectable. In fact, those traits are so common
among CS faculty, in my experience, that I'd say "CS faculty are
civil, informative, and respectable". Of course, if somebody pressed
me for details, I'd say, "well except for this one dumbfuck, Richard
Fateman. But don't worry about him, he's got one foot in the grave
and obviously suffers from dementia".

>> This is fact: I was mocking you, not being sarcastic.
>
> Eh, some people think they can carry a tune, but they can't.

Trying for some PJ yourself, I see. But no matter how poorly I sing,
when I sing, it is still singing.

[Richard Fateman]

On 9/4/2013 9:49 PM, Tom Boothby wrote:
> On Wed, Sep 4, 2013 at 8:56 PM, Richard Fateman <fat...@gmail.com> wrote:
>
>>> The word "fuck" has many uses, see Carlin for some examples. That one
>>> wasn't meant for clarification, it was more of an expression of
>>> amazement.
>> Treated as an expression of amazement, it doesn't clarify.
> Treated as a hammer, a water balloon doesn't drive nails, either. I
> was amazed (though I cannot say why) at your stupidity. If you can't
> fucking handle a little goddamned profanity, get the hell off of
> sage-flame, jackass.

My suggestion is you find someone nearby who has read your comments,
if there are any, and have a chat. You might also wonder how potential
employers
might react to Googling your name, say... tom boothby potty mouth.

> Ultraconservative parents of 6 year-olds give me
> less shit about the occasional "f-bomb" than you do. If you may
> recall, this mailing list was created for the express purpose of
> maintaining an outlet for pottymouths fighting over bullshit
> tangentially related to sage while keeping it off sage-devel.

No, support of pottymouths was not the purpose. It was established for
criticism of Sage. I suspect that you are the only one who has soiled
the nest.

>
>>>> Does BSD "have a windowing system"?
>>>> In no essential or inevitable way does BSD have a windowing system.
>>> Quoth wikipedia 'Today the term "BSD" is often used non-specifically
>>> to refer to any of the BSD descendants which together form a branch of
>>> the family of Unix-like operating systems.' Literally the second
>>> sentence of the article.
>> BSD can support a windowing interface, but it doesn't require it, nor does
>> BSD define
>> a windowing interface. Obstinacy buys you very little. Talk to someone,
>> maybe computer
>> science faculty member close to your home.
> I do talk to CS faculty rather often, actually, and they're civil,
> informative, and respectable.

You probably mean "respectful". You might ask them if BSD is a
windowing system,
and see what they say.

> In fact, those traits are so common
> among CS faculty, in my experience, that I'd say "CS faculty are
> civil, informative, and respectable".

The ones at UW that I know are reasonable people, certainly.

> Of course, if somebody pressed
> me for details, I'd say, "well except for this one dumbfuck, Richard
> Fateman. But don't worry about him, he's got one foot in the grave
> and obviously suffers from dementia".

And if someone asked me for my opinion of you, I would simply suggest
that they
look at what you've written.

> Trying for some PJ yourself, I see. But no matter how poorly I sing,
> when I sing, it is still singing.

Metaphorically speaking then, you can't carry a tune.




>