[sage-devel] Re: sage-1.8

[Iftikhar Burhanuddin]

On Mon, 22 Jan 2007, William Stein wrote:
> I've released SAGE-1.8.  This is mostly a bugfix release, but has a bunch of 
> code

sage-1.8 installed successfully [1] but there's junk [2] on each startup
and other weirdness [3].

Ifti.

+
[1]
===
Writing
/Users/weirdalerdos/Documents/sage-1.8/local/lib/python2.5/site-packages/pysqlite-2.3.2-py2.5.egg-info

real1m3.548s
user0m36.142s
sys 0m19.976s
Successfully installed sqlite-3.3.11
Now cleaning up tmp files.
Making SAGE/Python scripts relocatable...
Making script relocatable

real64m57.838s
user39m20.542s
sys 15m4.382s
To install gap, gp, singular, etc., scripts
in a standard bin directory, start sage and
type e.g., install_scripts('/usr/local/bin')
at the command prompt.

SAGE build/upgrade complete!
172:~/Documents/sage-1.8 weirdalerdos$ make
cd spkg && ./install all 2>&1 | tee -a ../install.log
make[1]: Nothing to be done for `all'.

real0m0.046s
user0m0.004s
sys 0m0.005s
To install gap, gp, singular, etc., scripts
in a standard bin directory, start sage and
type e.g., install_scripts('/usr/local/bin')
at the command prompt.

SAGE build/upgrade complete!
172:~/Documents/sage-1.8 weirdalerdos$

+
[2]
===
172:~/Documents/sage-1.8 weirdalerdos$ ./sage
--
| SAGE Version 1.8, Release Date: 2007-01-22 |
| Type notebook() for the GUI, and license() for information.|
--

---
   Traceback (most recent call
last)

/Users/weirdalerdos/Documents/sage-1.8/local/bin/ in
()

/Users/weirdalerdos/Documents/sage-1.8/local/lib/python2.5/site-packages/sage/all_cmdline.py
in ()
  1 """nodoctest"""
> 2 from sage.all import *
  3

/Users/weirdalerdos/Documents/sage-1.8/local/lib/python2.5/site-packages/sage/all.py
in ()
 66 from sage.algebras.all   import *
 67 from sage.modular.allimport *
---> 68 from sage.schemes.allimport *
 69 from sage.graphs.all import *
 70 from sage.groups.all import *

/Users/weirdalerdos/Documents/sage-1.8/local/lib/python2.5/site-packages/sage/schemes/all.py
in ()
 27 from plane_curves.all import *
 28
---> 29 from elliptic_curves.all import *
 30
 31 from plane_quartics.all import *

/Users/weirdalerdos/Documents/sage-1.8/local/lib/python2.5/site-packages/sage/schemes/elliptic_curves/all.py
in ()
 22
 23
---> 24 from ell_generic import is_EllipticCurve
 25
 26 from ell_rational_field import cremona_curves,
cremona_optimal_curves

/Users/weirdalerdos/Documents/sage-1.8/local/lib/python2.5/site-packages/sage/schemes/elliptic_curves/ell_generic.py
in ()
 32 import sage.misc.latex as latex
 33 import sage.modular.modform as modform
---> 34 import sage.functions.transcendental as transcendental
 35
 36 # Schemes

: No module named functions.transcendental
sage:

++

[3]
===
sage: E = EllipticCurve([1,0])
---
 Traceback (most recent call
last)

/Users/weirdalerdos/Documents/sage-1.8/local/bin/ in
()

: name 'EllipticCurve' is not defined
sage:





--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: Dlogs

[William Stein]

 From David Kohel:
> ... dlog in SAGE mod p is slow.

Anyway, yes, PARI has a function znlog, which should be very fast:

gp.znlog

I'll have to add it to the PARI interface (libs/pari/gen.pyx) then that should
give a very fast implementation mod p for p a not-too-big prime.  E.g.,



sage: p = 2^32+61
sage: a = gp.znprimroot(p); a
Mod(2, 4294967357)
sage: time gp.znlog(97, a)
498735128
CPU time: 0.00 s,  Wall time: 0.05 s



I'm really glad I'm not writing everything in SAGE from scratch!

If somebody on sage-dev sends me a patch that does what you want (see email 
below)
instead of me having to do it, that would be nice...  I won't work
on this until probably a day from now...

  -- William


On Mon, 22 Jan 2007 21:51:52 -0800, David R. Kohel <[EMAIL PROTECTED]> wrote:

> Hi William, David,
>
> I produced the following baby examples in Magma for use in my course.
>
> First a prime for which p-1 is smooth:
>
> sage: p = 2^32+15
> sage: (p-1).factor()
> 2 * 3^2 * 5 * 131 * 364289
>
> Then one containing a "large" prime factor:
>
> sage: p = 2^32+61
> sage: (p-1).factor()
> 2^2 * 1073741839
>
> Both should take relatively trivial time to solve discrete logarithms,
> but the former "should" recognize by initial trial division that the
> problem is much easier.
>
> Unfortunately the problem in my book doesn't come back snappily:
>
> sage: FF = FiniteField(p)
> sage: c = FF(4294967356)
> sage: x = FF(2)
> sage: c.log(x)
>
> This gets down to the following function:
>
>  return arith.discrete_log_generic(self, a, n) # TODO update this function
>
> which does a BSGS (not even Pollard rho).
>
> Shouldn't Pari or gmp have a more reasonable implementation?
>
> This is at the level of IntegerMod_abstract rather than finite fields.
>
> --David
>
>



--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] SAGE Days 3 Funding

[SAGE Days]

Hello,

SAGE Days 3 will be February 17 - 21, 2007:

   http://sage.math.washington.edu/sage/days3/

We applied for and received $10K funding to cover participant expenses.
In order to help us decide whether we have enough money, or will have to
hunt for other sources of funding, please send the following information
to [EMAIL PROTECTED] right *NOW*:

  1. Your name
  2. Approximately how much money you will need from
 IPAM in order to be able to attend SAGE Days.

If the sum of the all the needed amounts is less than $10K, we're set (and
may even be able to invite more people).   If it's over $10K, we'll either
find more money from other sources, or have to reduce the funding of some
participants.   We'll know which is the case within a few days, and send
an email to you explaining the situation.

We will notify you about funding no later than January 26. Hotel
reservations must be made by 2/2/07, so once you know
you're funded, you should make haste to get your hotel room reserved.
Also note that several people have already reserved a room, and are
looking for a roommate.

Thanks for your help!!

   William, Craig, and Nathan

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] sage-1.8

[William Stein]

Hello,

I've released SAGE-1.8.  This is mostly a bugfix release, but has a bunch of 
code
people sent me too.   The main goals for sage-1.9 (to be released Thursday) are
optimization of linear algebra, a couple more bug fixes (most known bugs have 
now
been fixed), and improvement of the SAGE documentation and doctesting.

New Features:

  * Added sqlite.
  * New sage-cleaner: Greatly improved cleanup of tmp and killing
of orphaned zombie processes.
  * lots of updates to David Kohel's crypto package.
  * sagex-- Robert Bradshaw added support for list comprehension
and +=, -=, etc. to SageX.

Bug fixes include the following:

 * trac #151 -- clisp.run / maples left running -- it's still happening 
though many cases have been fixed. SOLVE this frickin' zombie problem somehow 
once for all...
 * trac #149 -- failure in E.sha_an()
 * trac #63 -- in safari notebook, tab key inserts tab AND moves to 
beginning of cell
 * trac #148 -- links in saved notebook worksheets
 * trac #45 - ntl modulus reset correclty.
 * trac #192 -- polynomial arithmetic bug
 * trac #196 -- _sig_on/_sig_off doesn't work at all -- this will fix trac 
#192. (BUT -- double check that this doesn't break in cygwin).
 * trac #39 -- implement equality testing for ZZ[x,y,...] using M2 (or give 
error message)
 * trac #189 -- minor sage notebook formating bug
 * trac #130 - problem with multiline history recall in IPython
 * trac #168 -- Plot bounds ignored when frame=True
 * trac #32 -- missing coercion of polynomials between different bases 
functionality
 * trac #35 -- load "adosfile.sage" breaks
 * trac #154 -- gfan changes broke interface slightly.
 * trac #96 -- time command doesn't work in .sage files !?
 * trac #156 -- freeze of gfan (actually singular -- not gfan's fault)
 * trac #195 -- Can't compile spyx files with hyphens in names



Changelog (there are nearly a hundred entries):

changeset:   2571:784aeecd826b
tag: tip
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 19:04:42 2007 -0800
summary: new ver

changeset:   2570:a1cac63e2166
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 18:55:25 2007 -0800
summary: allow blank space at beginning of multiline input.

changeset:   2569:95b2f648ac89
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 18:52:52 2007 -0800
summary: Doctest fixes for 64-bit linux.

changeset:   2568:e3983766a089
parent:  2564:0a0b3eea1ed7
parent:  2567:48523109f5a5
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 18:34:41 2007 -0800
summary: merge

changeset:   2567:48523109f5a5
user:Bobby Moretti <[EMAIL PROTECTED]>
date:Mon Jan 22 18:32:20 2007 -0800
summary: Made lost of progress implementing the calculus SEP.

changeset:   2566:0d79b540694e
parent:  2347:7b3f809fc092
parent:  2565:e8cdebc57afc
user:Bobby Moretti <[EMAIL PROTECTED]>
date:Thu Jan 11 19:03:03 2007 -0800
summary: Merged william's calc code.

changeset:   2565:e8cdebc57afc
parent:  2327:c7196faad3a6
user:Bobby Moretti <[EMAIL PROTECTED]>
date:Thu Jan 11 18:46:58 2007 -0800
summary: Applied william's patch

changeset:   2564:0a0b3eea1ed7
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 18:30:15 2007 -0800
summary: Fix bug where M2 used always for polynomial ideal reduce.

changeset:   2563:26f24fb0ff24
parent:  2559:eac08cc900d0
parent:  2562:69d70870d0e8
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 18:10:18 2007 -0800
summary: Fixed doctests for switching rounding mode to RNDN for complex 
numbers.

changeset:   2562:69d70870d0e8
parent:  2560:f8802e318ba8
parent:  2561:517646505c87
user:Yi Qiang <[EMAIL PROTECTED]>
date:Mon Jan 22 16:36:08 2007 -0800
summary: improved povray interface

changeset:   2561:517646505c87
parent:  2542:16cb93aecfa5
user:Yi Qiang <[EMAIL PROTECTED]>
date:Mon Jan 22 16:34:59 2007 -0800
summary: Made povray more flexible by allowing you to pass in kwargs

changeset:   2560:f8802e318ba8
parent:  2549:a8269163193c
user:Yi Qiang <[EMAIL PROTECTED]>
date:Mon Jan 22 16:34:18 2007 -0800
summary: Added docstrings to dsage

changeset:   2559:eac08cc900d0
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 17:30:36 2007 -0800
summary: change complex rounding mode.

changeset:   2558:bfc3f5f377eb
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 14:13:13 2007 -0800
summary: new ver

changeset:   2557:364ab36bdf3c
user:William Stein <[EMAIL PROTECTED]>
date:Mon Jan 22 14:03:57 2007 -0800
summary: fix doctests.

changeset:   2556:18ac6acde340
parent:  2554:42c0c088f7c0
parent:  2555:98bce602990d

[sage-devel] Re: sqlite

[Nick Alexander]

> > such that clients can query the online
> > database and avoid downloading 10GB worth of data?
>
> Definitely.

Having recently struggled with the database interface, it would be
_great_ if there was a networked db class that loaded data from
sage.math.  Then I could only download the (few) class polynomials I
need rather than the full 77meg db.

This might also be a good time to standardize the database
implementations; they are all subtly different internally.

Nick


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Timothy Clemans]

Jaap you've been added as owner. http://freshmeat.net/projects/sage

On 1/22/07, Jaap Spies <[EMAIL PROTECTED]> wrote:
>
> Iftikhar Burhanuddin wrote:
>
> >>>
> >> This medium is like talking via Mars
> >
> > The above is weird but if you create a login, activate account
> > and login things work smoothly.
>
> Sorry, I meant [EMAIL PROTECTED] and gmane.comp.mathematics.sage.devel
>
> I'm "member" of freshmeat since the late 90's
>
> Jaap
>
>
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE in the third world...

[Fernando Perez]

Hi all,

On 1/22/07, Jan Groenewald <[EMAIL PROTECTED]> wrote:

> What we need more than a mirror, is a lecturer. The rest of this
> email is a pitch towards that.

[...]

Let me add a bit of background info on this.  In April of 2006 I
traveled to Cape Town for an applied math conference at the University
of Stellenbosch (SANUM'06), and after the conference we organized a
workshop where I taught for 2 days an intensive python/numpy/scipy
mini-course with the assistance of several students from U.
Stellenbosch.  The original plan was to have this at AIMS with Jan's
help, but for various logistical reasons it ended up happening at U.
Stellenbosch instead.

The experience and results were great.  The feedback we received was
overall very positive, and the local academics who hosted me have told
me later that they've felt a significant impact in terms of python
uptake for numerics (largely replacing matlab).  Several of the
students who helped running and teaching the workshop subsequently
became core Numpy/Scipy developers with commit rights, and they've
been very productive on those projects.

I've been talking to them about returning for a repeat, since
apparently they've been hounded with requests.  I haven't committed to
anything yet simply because I'm sorting out travel time for this year,
but I would absolutely love to do it again.  I had a great time, the
people were both extremely kind and super enthusiastic about working
hard, and the country is stunningly beautiful.  I'm not a pro, but
these are some pics of the trip, in case you'd like to whet your
appetites:

- http://picasaweb.google.com/fdo.perez/SouthAfricaWesternCape
- http://picasaweb.google.com/fdo.perez/SouthAfricaKrugerPark

The second set is from a trip we took to the Kruger Park after the
workshop was over.  Highly recommended :)

I did promote SAGE as well at the workshop, though only lightly given
the time constraints.  But I mentioned it and wore my Sage Days
T-shirt while teaching, so hopefully that helped...

I hope this is useful to others in making up their minds.  I'll
probably decide on my own travel options for South Africa in February,
I'd be happy to coordinate things with others here if you want.  Let
me know if you have any questions, either on- or off-list.

Regards,


f

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: sqlite

[alex clemesha]

I have done some small tests with using
pysqlite from Twisted and it was very easy and convient,
so I vote +1 for sqlite.

Alex

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Jaap Spies]

Iftikhar Burhanuddin wrote:

>>>
>> This medium is like talking via Mars
> 
> The above is weird but if you create a login, activate account
> and login things work smoothly.

Sorry, I meant [EMAIL PROTECTED] and gmane.comp.mathematics.sage.devel

I'm "member" of freshmeat since the late 90's

Jaap


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Iftikhar Burhanuddin]

On Mon, 22 Jan 2007, Jaap Spies wrote:
> William Stein wrote:
> > Could somebody volunteer to add SAGE to freshmeat?:
> >
> > http://freshmeat.net/add-project/
> >
>
> This medium is like talking via Mars

The above is weird but if you create a login, activate account
and login things work smoothly.

Timothy please email the list once SAGE surfaces on freshmeat.

ciao

The Medium is the message




--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Jaap Spies]

William Stein wrote:
> Could somebody volunteer to add SAGE to freshmeat?:
> 
> http://freshmeat.net/add-project/
> 

This medium is like talking via Mars

Jaap


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Jaap Spies]

Timothy Clemans wrote:
> What is your username so I can add you to admin of SAGE project on
> Freshmeat.net?
> 
  jaapspies

Jaap


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Timothy Clemans]

What is your username so I can add you to admin of SAGE project on
Freshmeat.net?

On 1/22/07, Jaap Spies <[EMAIL PROTECTED]> wrote:
>
> William Stein wrote:
> > Could somebody volunteer to add SAGE to freshmeat?:
> >
> > http://freshmeat.net/add-project/
> >
>
> I'm logged in right now. What do you want me to include as
> a description, etc?
>
> Jaap
>
>
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Timothy Clemans]

I sumbitted.

On 1/22/07, Jaap Spies <[EMAIL PROTECTED]> wrote:
>
> William Stein wrote:
> > Could somebody volunteer to add SAGE to freshmeat?:
> >
> > http://freshmeat.net/add-project/
> >
>
> I'm logged in right now. What do you want me to include as
> a description, etc?
>
> Jaap
>
>
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Jaap Spies]

William Stein wrote:
> Could somebody volunteer to add SAGE to freshmeat?:
> 
> http://freshmeat.net/add-project/
> 

I'm logged in right now. What do you want me to include as
a description, etc?

Jaap


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: primality tests and pseudo primes?

[William Stein]

On Sun, 21 Jan 2007 23:20:03 -0800, Nick Alexander <[EMAIL PROTECTED]> wrote:

>
> Can someone tell me what primality tests are available in sage?  I'd
> like a fast probabilistic algorithm if possible, I don't need
> certificates.  Maybe what is pseudo-prime testing called in pari?

As far as I know PARI is the only component of SAGE that has respectable
primality testing.  In fact, in my experience it is *very* respectable,
beating everything else I know of for general integers.

Thus I refer you to the PARI documentation.  Feel free to report back
to this list with what you find.

  -- William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: [sage-devel] factor(n) reducing some polynomials over the rationals

[William Stein]

On Mon, 22 Jan 2007 09:36:55 -0800, Robert Bradshaw <[EMAIL PROTECTED]> wrote:
> Is there any reason x shouldn't be initially assigned to belong to ZZ
> [x], given that coercion happens naturally to QQ[x] if any of the
> coefficients are rational?

There is no reason to change to ZZ[x], since in the future (not sure exactly
when) x (and a,b,c,d,...,z) will all be some sort of symbolic indeterminates.
I.e., changing from QQ to ZZ is not worth doing, since a more dramatic
change is on the way (after sage-2.0).

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: freshmeat

[Timothy Clemans]

me

On 1/22/07, William Stein <[EMAIL PROTECTED]> wrote:
>
> Could somebody volunteer to add SAGE to freshmeat?:
>
> http://freshmeat.net/add-project/
>
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] freshmeat

[William Stein]

Could somebody volunteer to add SAGE to freshmeat?:

http://freshmeat.net/add-project/

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE and calculus

[Robert Bradshaw]

On Jan 22, 2007, at 7:34 AM, William Stein wrote:

> On Mon, 22 Jan 2007 00:01:43 -0800, alex clemesha  
> <[EMAIL PROTECTED]> wrote:
>> On 1/19/07, Timothy Clemans <[EMAIL PROTECTED]> wrote:
>>> On the issue of 3d plotting, why is mathplotLib's 3d plotting not  
>>> good
>>> enough?
>> You may have already seen examples of the 3D stuff that matplotlib  
>> can do
>> here:
>> http://www.scipy.org/Cookbook/Matplotlib/mplot3D
>>
>> and here is a link from the matplotlib TODO list (
>> http://matplotlib.sourceforge.net/goals.html)
>> http://sda.iu.edu/matplot.html
>>
>> Basically the code isn't nearly as good (implementation,  
>> functionality,
>> speed, etc)
>> as the 2D stuff ... but oh how I wish it were ;)
>>
>> It's really not clear what the best thing to do is ...
>> a) write 3D code from scratch for matplotlib (a lot of work)
>> b) improve what code already exists (possible fatal flaws in  
>> existing code)
>> c) search for another compatible  (for the notebook, etc) 3D plotting
>> library (still looking)
>>
>> Ideas/thoughts are welcomed.
>
> I think the best thing to do by far is to create or modify an existing
> Java applet for 3d plotting.  For 3d it is crucial to have realtime
> rotation to really be usable, and this requires hardware support,
> i.e., a java applet.
>
> William

We did some investigation on this over the break and right now are  
leaning towards using an applet to wrap http://www.xj3d.org/ , and  
output plots, etc as vrml/x3d files. This would allow someone to use  
their own preferred VRML viewer in the future if they want too. I  
already have some of the work done.

- Robert

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: [sage-devel] factor(n) reducing some polynomials over the rationals

[Robert Bradshaw]

Is there any reason x shouldn't be initially assigned to belong to ZZ 
[x], given that coercion happens naturally to QQ[x] if any of the  
coefficients are rational?

On Jan 21, 2007, at 4:37 PM, Timothy Clemans wrote:

> -- Forwarded message --
> From: Bobby Moretti <[EMAIL PROTECTED]>
> Date: Jan 21, 2007 3:59 PM
> Subject: Re: [sage-devel] factor(n) reducing some polynomials over  
> the rationals
> To: [EMAIL PROTECTED]
>
>
> The indeterminate 'x', predefined as a polynomial ring element, gets
> imported automatically:
>
>
> --
> | SAGE Version 1.7.1, Release Date: 2007-01-18   |
> | Type notebook() for the GUI, and license() for information.|
> --
> Loading SAGE library. Current Mercurial branch is: calc
>
> sage: type(x)
>  
>
> So if you define any polynomials in terms of arithmetic with x, it
> will assume that your polynomial is over the rationals.
>
>  If you want a polynomial over the integers, you have to specify this
> ring manually. If you do
>
> g. = PolynomialRing(ZZ)
>
> Then it overloads x, and x becomes an indeterminate for polynomials
> over the integers.
>
> sage: g. = PolynomialRing(ZZ)
> sage: type(x)
> 
> sage: factor(2*x^2+1)
> 2*x^2 + 1
>
>
>  On 1/21/07, Timothy Clemans <[EMAIL PROTECTED]> wrote:
>>
>> "sage: factor(n)" should only reduce input over the integers.
>>
>> sage: factor(2*x^2 + x + 1)
>> (2) * (x^2 + 1/2*x + 1/2)
>> sage: factor(2*x^2 + x + 2)
>> (2) * (x^2 + 1/2*x + 1)
>> sage: factor(2*x^2 + x + 3)
>> (2) * (x^2 + 1/2*x + 3/2)
>> sage: factor(2*x^2 + x + 4)
>> (2) * (x^2 + 1/2*x + 2)
>>
>> Is there a function in SAGE or one of the components that reduces
>> polynomials only over the integers?
>>
>>
>>>
>>
>
>
>
> -- 
> Bobby Moretti
> [EMAIL PROTECTED]
>
> 

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE in the third world...

[Iftikhar Burhanuddin]

On Mon, 22 Jan 2007, William Stein wrote:
>
> Based on everything I've seen, I think SAGE is totally wrong for OLPC, at 
> least
> at this point in time.   Just the hardware specs of the machine are such that
> SAGE wouldn't fit in it for starters.

In which case we should construct a low-cost machine to run SAGE :)

> But there could be an offshoot of SAGE, e.g., maybe a pure Python
> program that uses ideas from SAGE in some way and maybe from NZMATH,
> which could eventually be in OLPC.  This could be something that would
> come out of a student project.  It could be called "sagelite".

In this world a la the world of comic delivery, c'est tout le Timing!

ciao


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE in the third world...

[William Stein]

Hi,

Based on everything I've seen, I think SAGE is totally wrong for OLPC, at least
at this point in time.   Just the hardware specs of the machine are such that
SAGE wouldn't fit in it for starters.  But there could be an offshoot of SAGE,
e.g., maybe a pure Python program that uses ideas from SAGE in some way and
maybe from NZMATH, which could eventually be in OLPC.  This could be something
that would come out of a student project.  It could be called "sagelite".


  William

On Mon, 22 Jan 2007 07:04:39 -0800, Timothy Clemans <[EMAIL PROTECTED]> wrote:
> Development for OLPC is highly specialized. I think the notebook
> server could be a problem.,since the speed is much slower than the
> command line and at the end, SAGE would need to work with Sugar.
> http://wiki.laptop.org/go/Development_issues
>
> Something to thin about:
>
> ---QUOTE---
> Since the OLPC is rather different from the average PC, we need to
> keep in mind a set of guidelines while building educational
> applications.
>
>1. Remember that the end user is a kid.
>2. Keep the power consumption low. Avoid animated graphics or
> provide a way for the user to turn them off.
>3. Use simple datastores such as pickle or dbm. For relational
> databases use SQLite
>4. Keep your GUI simple. Remember that the final application will
> be running in a tabbed page inside Sugar
>5. Use UNICODE for all strings
>6. Use the gettext module for all string literals
>7. Get a friend to translate all string literals and test your
> program in their language
>8. Where possible, use international iconic symbols instead of text
>
> On 1/22/07, Iftikhar Burhanuddin <[EMAIL PROTECTED]> wrote:
>>
>> On Mon, 22 Jan 2007, Martin Albrecht wrote:
>>
>> > This might be an issue but can be addressed. When I start SAGE on my 
>> > machine
>> > it uses ~ 34MB of physical memory.
>> >
>> >Mass storage: 512 MiB SLC NAND flash, high speed flash controller
>> >
>> > This is definitely a problem. SAGE_LOCAL takes 450 MB on my machine. Maybe
>> > some compressed filesystem helps here.
>>
>> Or we fork a SAGE-OLPC branch, which is lightweight and targets the
>> highschool/undergrad student working on the OPLC laptop [1], which will
>> have the bare minimum SAGE functionality --- Calculus and plotting,
>> etc.(?) Thoughts?
>>
>> We should commit to the project depending on how Nicholas Negroponte's
>> excellent initiative takes off, as it is unlikely that it'll become a
>> global standard [2] as there are other projects in the works [3].
>>
>> Martin, thanks for the info.
>>
>> Regards,
>> Ifti
>>
>> [1] http://www.laptop.org/faq.en_US.html
>> What is the $100 Laptop, really?
>> The proposed $100 machine will be a Linux-based, with a dual-mode display
>> both a full-color, transmissive DVD mode, and a second display option that
>> is black and white reflective and sunlight-readable at 3 the resolution.
>> The laptop will have a 500MHz processor and 128MB of DRAM, with 500MB of
>> Flash memory; it will not have a hard disk, but it will have four USB
>> ports. The laptops will have wireless broadband that, among other things,
>> allows them to work as a mesh network; each laptop will be able to talk to
>> its nearest neighbors, creating an ad hoc, local area network. The laptops
>> will use innovative power (including wind-up) and will be able to do most
>> everything except store huge amounts of data.
>>
>> [2] http://wiki.laptop.org/go/OLPC_India
>> And, according to this story in the Times of India, India's HRD doesn't
>> like the idea, believing the money could be better spent in other ways.
>>
>> [3] http://www.simputer.org/
>>
>>
>> >
>>
>
> >
>


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: Patch for sloane.A001694

[William Stein]

On Mon, 22 Jan 2007 06:59:18 -0800, Jaap Spies <[EMAIL PROTECTED]> wrote:

> William Stein wrote:
>
>>> It is long compared with Mathematica that takes only a few seconds.
>>
>> On my laptop it takes 15 seconds in Mathematica.
>> In SAGE this one line makes the list (it's more matheamtica like, I guess):
>>
>>   sage: time v=[n for n in range(2,21) if min(e for _,e in factor(n)) > 
>> 1]
>>   CPU times: user 47.23 s, sys: 8.47 s, total: 55.70 s
>>
>> I think the runtime is likely completely dominated by overhead associated
>> with the factor command (which currently uses the PARI C library) and
>> the structure in which it stores factorizations.
>> Of course that should get optimized.  However, as a shortcut, one could just
>> compute the list of poweful numbers directly using PARI, then eval the
>> result to get something in SAGE.  I did this, and now:
>>
>> sage: a = sloane.A001694
>> sage: time a(1000)
>>
>> sage: a = sloane.A001694
>> sage: time a(1000)
>> CPU times: user 0.93 s, sys: 0.01 s, total: 0.93 s
>> Wall time: 0.94
>> 253472
>>
>
> Great!
>
>>> I think it is extremely important to have a good SAGE program for this
>>> sequence! It will possibly be criticized.
>>
>> No worries -- this is now about 20 times faster than Mathematica on my 
>> system.
>>
>
> Indeed. One thing, the Mathematica program is not correct, because 1 is
> also a powerfull number by definition: p|n -> p^2|n, ex falso sequitur 
> quodlibet.
>
> What about the plans for one file one sequence? Shall I wait, or can I
> still submit new sequences?

Just keep submitting sequences.  And I liked your one file per related group
of sequences idea better than one file per sequence.

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: sqlite

[William Stein]

On Mon, 22 Jan 2007 08:55:18 -0800, Martin Albrecht <[EMAIL PROTECTED]> wrote:
>> Relational databases are much different because they offer a very
>> sophisticated query language and automated creation of indexes.
>
> I had my fair share of SQL queries, so I agree on that :-) I just didn't know
> that ZODB was so limited with respect to querying.

It's very limited.  It's a totally different thing.

>> The datasets used in
>> math computations are mostly large read-only tables, so a heavier database
>> (mysql or postgresql) is total overkill.
>
> I agree. Is there any data yet that would be put in a sqlite database?

Yes.  E.g., there is a large table of graphs that one person made, and another
very large table of reflexive polytopes in dimension <= 4.  Both are best
queried using SQL.   Gonzalo Tornario once made a very nice SQLite database
of Cremona's tables of elliptic curves.  I have a lot of data that might
best be stored as a SQL table.

> I would
> say the inclusion of a database is worth the compile time & space if there is
> any interesting data in the database. I assume most math packages don't put
> their data in SQL tables. Also, we should probably set up a SAGE SQL master
> server (this could be postgresql)

Having used postgresql a lot in the past, I hate postgresql.  The thought of 
using
it churns my stomache.  I would prefer trying to use SQLite only if possible,
for simplicity.

> such that clients can query the online
> database and avoid downloading 10GB worth of data?

Definitely.

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: sqlite

[Martin Albrecht]

> Relational databases are much different because they offer a very
> sophisticated query language and automated creation of indexes.  

I had my fair share of SQL queries, so I agree on that :-) I just didn't know 
that ZODB was so limited with respect to querying. 

> The datasets used in
> math computations are mostly large read-only tables, so a heavier database
> (mysql or postgresql) is total overkill.

I agree. Is there any data yet that would be put in a sqlite database? I would 
say the inclusion of a database is worth the compile time & space if there is 
any interesting data in the database. I assume most math packages don't put 
their data in SQL tables. Also, we should probably set up a SAGE SQL master 
server (this could be postgresql) such that clients can query the online 
database and avoid downloading 10GB worth of data?

Martin

-- 
name: Martin Albrecht
_pgp: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
_www: http://www.informatik.uni-bremen.de/~malb
_jab: [EMAIL PROTECTED]


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: sqlite

[William Stein]

On Mon, 22 Jan 2007 08:28:31 -0800, Martin Albrecht <[EMAIL PROTECTED]> wrote:
> Am I right that ZODB (shipped with SAGE) is an object oriented database?

Yes.

> I am
> not very into databases thus I ask: What advantage does sqlite (being a
> relational database) over ZODB?

It's a completely different thing, and solves different problems.  Basically,
ZODB is very good at making large data structures (basically B-tree 
dictionaries)
persistent on disk.  You can make a 10GB dictionary, and work with it like
it is in memory, but in fact maybe only 1GB of RAM is used instead of 10GB.
But all modifications get appended to the end of the file (so ZODB sucks
if one does a lot of updating to the data, but is good for reading).  One 
problem
is the data is only readable by Python via ZODB.  A nice feature that doesn't 
get
used in SAGE yet is that ZODB databases can be served over the internet 
directly.

Relational databases are much different because they offer a very sophisticated
query language and automated creation of indexes.   ZODB doesn't give you nearly
the same level of querying functionality, except what you might add yourself.
Moreover data stored in a sqlite database can be queried without using Python.
Based on past experience, I think sqlite will scale much better to large
datasets than ZODB.  The datasets used in math computations are mostly large
read-only tables, so a heavier database (mysql or postgresql) is total overkill.

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: sqlite

[Jaap Spies]

William Stein wrote:

>(1) Try
>  sage -i sqlite-3.3.11
>to install it into SAGE.  Any problems?
>sage: import sqlite3
>sage: sqlite3.version
> 

Builds and installs fine on FC 5

sage: from pysqlite2 import dbapi2 as sqlite

sage:

>(2) Assuming sqlite installs OK for everyone, the constraint to adding
>it standard to SAGE is that it increases the build time by 1-2 minutes
>and the tarball size by 1.7MB (with the sqlite tests included).
>Is it worth it?
> 
>

???

Jaap


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: sqlite

[Martin Albrecht]

On Monday 22 January 2007 17:14, William Stein wrote:
> Hello,
>
> For a while now I've been considering including a relational database
> in SAGE, since in the long run this will make it much much easier to
> create certain types of databases that are easy to query.   Probably
> the obvious best choice would be:
>   http://www.sqlite.org/
>
> QUESTIONS:
>(1) Try
>  sage -i sqlite-3.3.11
>to install it into SAGE.  Any problems?
>sage: import sqlite3
>sage: sqlite3.version
>

  works fine (Debian Etch/AMD64 on Core 2 Duo)

>(2) Assuming sqlite installs OK for everyone, the constraint to adding
>it standard to SAGE is that it increases the build time by 1-2
> minutes and the tarball size by 1.7MB (with the sqlite tests included). Is
> it worth it?

Am I right that ZODB (shipped with SAGE) is an object oriented database? I am 
not very into databases thus I ask: What advantage does sqlite (being a 
relational database) over ZODB? 

Martin

-- 
name: Martin Albrecht
_pgp: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
_www: http://www.informatik.uni-bremen.de/~malb
_jab: [EMAIL PROTECTED]


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] sqlite

[William Stein]

Hello,

Errata -- in the previous email the test of sqlite should be this:

   sage: from pysqlite2 import dbapi2 as sqlite
  (doesn't go boom)

Ignore the test from the previous email.

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] sqlite

[William Stein]

Hello,

For a while now I've been considering including a relational database
in SAGE, since in the long run this will make it much much easier to
create certain types of databases that are easy to query.   Probably
the obvious best choice would be:
  http://www.sqlite.org/

QUESTIONS:
   (1) Try
 sage -i sqlite-3.3.11
   to install it into SAGE.  Any problems?
   sage: import sqlite3
   sage: sqlite3.version

   (2) Assuming sqlite installs OK for everyone, the constraint to adding
   it standard to SAGE is that it increases the build time by 1-2 minutes
   and the tarball size by 1.7MB (with the sqlite tests included).
   Is it worth it?

If the answer to (2) were obvious I wouldn't be asking, so I'd like
feedback.

William


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE and calculus

[William Stein]

On Mon, 22 Jan 2007 00:01:43 -0800, alex clemesha <[EMAIL PROTECTED]> wrote:
> On 1/19/07, Timothy Clemans <[EMAIL PROTECTED]> wrote:
>> On the issue of 3d plotting, why is mathplotLib's 3d plotting not good
>> enough?
> You may have already seen examples of the 3D stuff that matplotlib can do
> here:
> http://www.scipy.org/Cookbook/Matplotlib/mplot3D
>
> and here is a link from the matplotlib TODO list (
> http://matplotlib.sourceforge.net/goals.html)
> http://sda.iu.edu/matplot.html
>
> Basically the code isn't nearly as good (implementation, functionality,
> speed, etc)
> as the 2D stuff ... but oh how I wish it were ;)
>
> It's really not clear what the best thing to do is ...
> a) write 3D code from scratch for matplotlib (a lot of work)
> b) improve what code already exists (possible fatal flaws in existing code)
> c) search for another compatible  (for the notebook, etc) 3D plotting
> library (still looking)
>
> Ideas/thoughts are welcomed.

I think the best thing to do by far is to create or modify an existing
Java applet for 3d plotting.  For 3d it is crucial to have realtime
rotation to really be usable, and this requires hardware support,
i.e., a java applet.

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE in the third world...

[Timothy Clemans]

Development for OLPC is highly specialized. I think the notebook
server could be a problem.,since the speed is much slower than the
command line and at the end, SAGE would need to work with Sugar.
http://wiki.laptop.org/go/Development_issues

Something to thin about:

---QUOTE---
Since the OLPC is rather different from the average PC, we need to
keep in mind a set of guidelines while building educational
applications.

   1. Remember that the end user is a kid.
   2. Keep the power consumption low. Avoid animated graphics or
provide a way for the user to turn them off.
   3. Use simple datastores such as pickle or dbm. For relational
databases use SQLite
   4. Keep your GUI simple. Remember that the final application will
be running in a tabbed page inside Sugar
   5. Use UNICODE for all strings
   6. Use the gettext module for all string literals
   7. Get a friend to translate all string literals and test your
program in their language
   8. Where possible, use international iconic symbols instead of text

On 1/22/07, Iftikhar Burhanuddin <[EMAIL PROTECTED]> wrote:
>
> On Mon, 22 Jan 2007, Martin Albrecht wrote:
>
> > This might be an issue but can be addressed. When I start SAGE on my machine
> > it uses ~ 34MB of physical memory.
> >
> >Mass storage: 512 MiB SLC NAND flash, high speed flash controller
> >
> > This is definitely a problem. SAGE_LOCAL takes 450 MB on my machine. Maybe
> > some compressed filesystem helps here.
>
> Or we fork a SAGE-OLPC branch, which is lightweight and targets the
> highschool/undergrad student working on the OPLC laptop [1], which will
> have the bare minimum SAGE functionality --- Calculus and plotting,
> etc.(?) Thoughts?
>
> We should commit to the project depending on how Nicholas Negroponte's
> excellent initiative takes off, as it is unlikely that it'll become a
> global standard [2] as there are other projects in the works [3].
>
> Martin, thanks for the info.
>
> Regards,
> Ifti
>
> [1] http://www.laptop.org/faq.en_US.html
> What is the $100 Laptop, really?
> The proposed $100 machine will be a Linux-based, with a dual-mode display
> both a full-color, transmissive DVD mode, and a second display option that
> is black and white reflective and sunlight-readable at 3 the resolution.
> The laptop will have a 500MHz processor and 128MB of DRAM, with 500MB of
> Flash memory; it will not have a hard disk, but it will have four USB
> ports. The laptops will have wireless broadband that, among other things,
> allows them to work as a mesh network; each laptop will be able to talk to
> its nearest neighbors, creating an ad hoc, local area network. The laptops
> will use innovative power (including wind-up) and will be able to do most
> everything except store huge amounts of data.
>
> [2] http://wiki.laptop.org/go/OLPC_India
> And, according to this story in the Times of India, India's HRD doesn't
> like the idea, believing the money could be better spent in other ways.
>
> [3] http://www.simputer.org/
>
>
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE in the third world...

[Iftikhar Burhanuddin]

On Mon, 22 Jan 2007, Martin Albrecht wrote:

> This might be an issue but can be addressed. When I start SAGE on my machine
> it uses ~ 34MB of physical memory.
>
>Mass storage: 512 MiB SLC NAND flash, high speed flash controller
>
> This is definitely a problem. SAGE_LOCAL takes 450 MB on my machine. Maybe
> some compressed filesystem helps here.

Or we fork a SAGE-OLPC branch, which is lightweight and targets the
highschool/undergrad student working on the OPLC laptop [1], which will
have the bare minimum SAGE functionality --- Calculus and plotting,
etc.(?) Thoughts?

We should commit to the project depending on how Nicholas Negroponte's
excellent initiative takes off, as it is unlikely that it'll become a
global standard [2] as there are other projects in the works [3].

Martin, thanks for the info.

Regards,
Ifti

[1] http://www.laptop.org/faq.en_US.html
What is the $100 Laptop, really?
The proposed $100 machine will be a Linux-based, with a dual-mode display
both a full-color, transmissive DVD mode, and a second display option that
is black and white reflective and sunlight-readable at 3 the resolution.
The laptop will have a 500MHz processor and 128MB of DRAM, with 500MB of
Flash memory; it will not have a hard disk, but it will have four USB
ports. The laptops will have wireless broadband that, among other things,
allows them to work as a mesh network; each laptop will be able to talk to
its nearest neighbors, creating an ad hoc, local area network. The laptops
will use innovative power (including wind-up) and will be able to do most
everything except store huge amounts of data.

[2] http://wiki.laptop.org/go/OLPC_India
And, according to this story in the Times of India, India's HRD doesn't
like the idea, believing the money could be better spent in other ways.

[3] http://www.simputer.org/


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE in the third world...

[Martin Albrecht]

> Third world discussion spawns the following question: The One Laptop One
> Child project has been in the news lately,
>
> http://www.laptop.org/
>
> I haven't done the spadework and was wondering if SAGE could be made to
> run on this machine. Yes?

I think we could make it happen, but it looks like if some work is needed:
 
 CPU: AMD Geode [EMAIL PROTECTED](datasheet) 
 CPU clock speed: 366 Mhz 
 Compatibility: X86/X87-compatible 

No problem here, thought, it is a bit slow, we might want to tune SAGE a bit 
to start faster maybe.

 DRAM memory: 128 MiB dynamic RAM 

This might be an issue but can be addressed. When I start SAGE on my machine 
it uses ~ 34MB of physical memory.

   Mass storage: 512 MiB SLC NAND flash, high speed flash controller 

This is definitely a problem. SAGE_LOCAL takes 450 MB on my machine. Maybe 
some compressed filesystem helps here. 

Harware Specs were taken from 
http://wiki.laptop.org/go/Hardware_specification . You may find information 
about the software platform at http://wiki.laptop.org/go/Software . E.g. 
Python is shipped with the OLPC and we should use that instead of our own.

Martin

-- 
name: Martin Albrecht
_pgp: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
_www: http://www.informatik.uni-bremen.de/~malb
_jab: [EMAIL PROTECTED]


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE in the third world...

[Iftikhar Burhanuddin]

* http://sage.math.washington.edu/sage
* http://modular.fas.harvard.edu/sage
* http://cocoa.mathematik.uni-dortmund.de/sage/
* http://echidna.maths.usyd.edu.au/sage
* http://sage.scipy.org/sage

There has been discussion on this list about hosting SAGE out of South
America and India, that leaves us imo with China and Russia (unless the
Euro/German mirror would suffice for the latter.)  Folks, please work your
contacts.

Third world discussion spawns the following question: The One Laptop One
Child project has been in the news lately,

http://www.laptop.org/

I haven't done the spadework and was wondering if SAGE could be made to
run on this machine. Yes?

Regards,
Ifti

ps: Jan, thanks for that illuminating email, I (and I'm sure others) will
keep it in mind when making future plans. Please enlighten me about the
what's happening with the OLPC in Africa, from your perspective. If you
deem it off-topic, direct the email to moi.

On Mon, 22 Jan 2007, Jan Groenewald wrote:

>
> Hi
>
> On Sun, Jan 21, 2007 at 05:21:07PM -0800, William Stein wrote:
> > --- Forwarded message ---
> > From: "David R. Kohel" <[EMAIL PROTECTED]>
> > To: "William Stein" <[EMAIL PROTECTED]>
> > Subject: Re: Fwd: Crypto in Sage
> > Date: Sun, 21 Jan 2007 16:28:14 -0800
> >
> > Another important role for SAGE is mathematics in the third world.  When
> > visiting there, I found the African Institute for Mathematical Sciences in
> > Cape Town (for obvious reasons) was very pro-open source.
> > http://www.aims.ac.za/english/
>
> Pro Free Software, even ;)
>
> > At the time, I gave a talk on Magma, but they would not install Magma even
> > if given for free (since students should not learn or be dependent on 
> > software
> > not available when returning to their home universities across Africa).
>
> Hehe, that was me. We still follow this policy. Plus we've been travelling
> around Africa where we find mounds of illegal softwarez. It just doesn't
> make sense. Never mind poor universities, students can use the software
> on their personal machines, or those who become entrepreneurs might even
> use it in a small business without the formidable barrier-to-entry of
> costly proprietary software.
>
> > I think it would be a powerful statement to have a SAGE mirror there.
>
> Well, the deb package is in our apt-cacher cache already. I know this is
> probably a contentious issue, monolithic vs modular, but being packaged
> for debian and derivatives (and other distros) would go a long way
> towards being available already -- there are various linuxy mirrors
> around (mirror.ac.za for all universities, ftp.is.co.za countrywide).
>
> Allow me to test the limits of off-topic for a moment:


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---

[sage-devel] SAGE in the third world...

[Jan Groenewald]

Hi

On Sun, Jan 21, 2007 at 05:21:07PM -0800, William Stein wrote:
> --- Forwarded message ---
> From: "David R. Kohel" <[EMAIL PROTECTED]>
> To: "William Stein" <[EMAIL PROTECTED]>
> Subject: Re: Fwd: Crypto in Sage
> Date: Sun, 21 Jan 2007 16:28:14 -0800
> 
> Another important role for SAGE is mathematics in the third world.  When
> visiting there, I found the African Institute for Mathematical Sciences in
> Cape Town (for obvious reasons) was very pro-open source.
> http://www.aims.ac.za/english/

Pro Free Software, even ;)

> At the time, I gave a talk on Magma, but they would not install Magma even
> if given for free (since students should not learn or be dependent on software
> not available when returning to their home universities across Africa).
 
Hehe, that was me. We still follow this policy. Plus we've been travelling
around Africa where we find mounds of illegal softwarez. It just doesn't
make sense. Never mind poor universities, students can use the software
on their personal machines, or those who become entrepreneurs might even
use it in a small business without the formidable barrier-to-entry of
costly proprietary software.

> I think it would be a powerful statement to have a SAGE mirror there.

Well, the deb package is in our apt-cacher cache already. I know this is
probably a contentious issue, monolithic vs modular, but being packaged
for debian and derivatives (and other distros) would go a long way
towards being available already -- there are various linuxy mirrors
around (mirror.ac.za for all universities, ftp.is.co.za countrywide).

Allow me to test the limits of off-topic for a moment:

What we need more than a mirror, is a lecturer. The rest of this
email is a pitch towards that.

This institute (AIMS) has as its main programme a postgraduate 
diploma for the best math students from around Africa, who often
come from poor conditions (such as strikes, no electricity, few 
or no computers, few or no books, little local expertise and 
this little choice in science). Here in Cape Town they live in
an academic hothouse environment; institute, computer lab, and 
accomodation in one. Often their computing experience is somewhat
lacking at first, having learnt programming on paper. Their current
abilities belie this history. There are three phases to their year: 

1) Skills courses, bring everyone to the same level:
http://www.aims.ac.za/docs/2006-7_Skills.pdf
2) Review courses, students choose 9 of these:
http://www.aims.ac.za/docs/2006-7_Review.pdf
3) Essay phase:
Student writes an essay, supervised by some external lecturer (we have
no permanent academic staff, African and international lecturers come
here for three weeks at a time, to live in the institute). For example
last year students chose from these topics:
http://www.aims.ac.za/resources/essays/
Some examples from Algebra:
http://www.aims.ac.za/resources/essays/abstract.php?id=171
http://www.aims.ac.za/resources/essays/abstract.php?id=108
http://www.aims.ac.za/resources/essays/abstract.php?id=115

After AIMS many of the 50 annual students continue to an MSc in South
Africa or elsewhere, and are now, after three years of our existence, 
in very high demand around the country. We pride ourselves in preparing
Africa's brightest for research.  (Apart from the popstgraduate
diploma, we might host research groups here in future.)

Due to the fast pace of the course, and many other obvious reasons, 
we teach python and scipy now (no longer C or Octave -- C was way to 
confusing to new programmers, and Octave didn't scale the same way, 
and just ends up being unfavourably compared to Matlab) in between 
the math and physics -- each course has a strong component of modelling
on computer, for many courses it is 50%. Most work, but not all,
is numerical. Symbolic packages we use are maxima (ok, soon a slightly 
friendlier wxmaxima) and Singular (btw, also not officially packaged
for debian/ubuntu), and gap. More and more lecturers are joining us 
Now that's three different interfaces and syntaxes while the students 
have become proficient in Python, in finding and using libraries, and
in glueing together code, and SAGE would provide a familiar 
point-of-entry to the various Free symbolic packages, adding Pari-GP 
to that list. Disclaimer: I'm the sysadm here, but part time studying
algebra, so I might personally benefit from inviting sages here.

The AIMS year runs from September to June. We teach python programming
the first 6 weeks of the academic year. I suspect SAGE would be too
advanced to bring in at this point. We usually teach a computational
algebra course around January; currently Barry Green is wrapping up
a three week course: http://academic.sun.ac.za/maths/BWG/
We encourage lecturers from within Africa to work with international
experts in various fields, so someone with a good experience in teaching
and SAGE could join that. We are currently accepting proposals for
courses for the 2007/8 academic year: http://www.

[sage-devel] Re: SAGE and calculus

[alex clemesha]

On 1/19/07, Timothy Clemans <[EMAIL PROTECTED]> wrote:
>
>
> On the issue of 3d plotting, why is mathplotLib's 3d plotting not good
> enough?


You may have already seen examples of the 3D stuff that matplotlib can do
here:
http://www.scipy.org/Cookbook/Matplotlib/mplot3D

and here is a link from the matplotlib TODO list (
http://matplotlib.sourceforge.net/goals.html)
http://sda.iu.edu/matplot.html

Basically the code isn't nearly as good (implementation, functionality,
speed, etc)
as the 2D stuff ... but oh how I wish it were ;)

It's really not clear what the best thing to do is ...
a) write 3D code from scratch for matplotlib (a lot of work)
b) improve what code already exists (possible fatal flaws in existing code)
c) search for another compatible  (for the notebook, etc) 3D plotting
library (still looking)

Ideas/thoughts are welcomed.

-Alex

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~--~~~~--~~--~--~---