[sage-devel] Re: bug in show method for DiGraphs

[davidp]

>
> I haven't seen a definition of the laplacian matrix for looped digraphs,
> but sure, okay, I assume this is standard, then?  I don't think the
> laplacian_matrix function calculates this quantity for digraphs.  This
> might be a bug, then.

It is standard to take the Laplacian to be D - A where D is the
diagonal matrix of out-degrees (or sometimes in-degrees---I much
prefer the former) and A is the adjacency matrix.  Thus, the Laplacian
does not see loops.

>
> So is this laplacian matrix wrong?

Yes.  The 1,1 entry of the adjacency matrix is 1, the out-degree of
the first vertex is 2, and 2-1=1.  So the 1,1 entry of
G.laplacian_matrix() is wrong.

>
> sage: G = DiGraph({1:{1: 1, 2: 1}, 2:{1:1}})
> sage: G
> Looped digraph on 2 vertices
> sage: G.laplacian_matrix()
>
> [ 2 -1]
> [-1  1]
>

> Why do you expect that DiGraph(G.laplacian_matrix())
> should give you back the same graph?
>

Except for loops, I can reconstruct a graph from its Laplacian.  So I
expect DiGraph(G.laplacian_matrix()) to return the graph, removing
loops if present.

About the edge labels: the negative entries in the Laplacian come from
subtracting the adjacency matrix.  They are the negatives of the
number of edges connecting a vertex to adjacent vertices.  The edge
labels should tell me how many edges connect two vertices.  This is
pretty standard, and I have never seen the other convention (negative
labels) in mathematical papers.

> Jason

Thanks,
David
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: FInally the link to download SAGElwlcd ISO!

[Lucio Lastra]

Rob,

sure! when i started creating this metadistro 3.4 wasn't
available yet. The next release will be with it.

If you want to add it yourself before I do, check the instructions
I sent William to build the same .iso. He forwarded them to this
list.

You will have to download SAGE 3.4 from here:

http://sage.math.washington.edu/sage/linux/32bit/sage-3.4-linux-Ubuntu_8.10-i686-Linux.tar.gz

Then uncompress it in a folder and read the README of the instructions
to see exactly what to do.

Meanwhile it will be nice to listen to your opinion about it and if it's
not much to ask, tell some people you know about the SAGElwlcd, so
they try it and share their opinions too.

Greetings,

Lucio.

On Fri, Mar 20, 2009 at 7:11 PM, Rob Beezer  wrote:

>
> Lucio,
>
> That worked!  Then a right-click to get the fluxbox menu up for
> anybody else who is playing along.
>
> Any plans to catch-up to 3.4?
>
> Thanks,
> Rob
>
> On Mar 20, 1:34 pm, Lucio Lastra  wrote:
> > wait couple seconds and hit Enter and you are going to have
> > the bash prompt back again.
> >
> > ubu...@ubuntu~:
> >
> > After that type: sudo startx and voilá!
>
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: bug in show method for DiGraphs

[Jason Grout]

davidp wrote:
>>> 4. In the second call, there should not be loops at each vertex.
>> Yes, there should be.  It's a weighted graph with loops on the vertices,
>> since the laplacian has nonzero diagonal entries.
>>
>> Thanks,
>>
>> Jason
> 
> The diagonal entries in the laplacian give the out_degrees of the
> corresponding vertices minus the weight of any loops.

I haven't seen a definition of the laplacian matrix for looped digraphs, 
but sure, okay, I assume this is standard, then?  I don't think the 
laplacian_matrix function calculates this quantity for digraphs.  This 
might be a bug, then.

So is this laplacian matrix wrong?

sage: G = DiGraph({1:{1: 1, 2: 1}, 2:{1:1}})
sage: G
Looped digraph on 2 vertices
sage: G.laplacian_matrix()

[ 2 -1]
[-1  1]




   Here is a
> simpler example showing that something is wrong.
> 
> sage: G.laplacian_matrix()
> 
> [ 2 -1 -1]
> [-1  2 -1]
> [-1 -1  2]
> sage: G = DiGraph({1:{2:1, 3:1}, 2:{1:1, 3:1}, 3:{1:1, 2:1}})
> sage: G.laplacian_matrix()
> 
> [ 2 -1 -1]
> [-1  2 -1]
> [-1 -1  2]
> sage: G.show()
> sage: DiGraph(G.laplacian_matrix()).show()
> 
> In this case, G is an ordinary triangle---no loops at vertices.  Each
> vertex has out_degree 2.
> 
> 1.  G.show() is wrong since it misses some arrows.

Yes, I think this is a bug.


> 3.  DiGraph(G.laplacian_matrix()).show() is also wrong because is
> missing arrowheads.

Yep, I agree.



 > 2.  DiGraph(G.laplacian_matrix()).show() is wrong because it displays
 > loops at the vertices.
> 4.  To clarify an earlier question: why aren't G.show() and DiGraph
> (G.laplacian_matrix()).show() the same?
> 4.  To further complicate matters, I think that
> 
> DiGraph(G.laplacian_matrix()).show(edge_labels=True)




I think the answer to questions 1, 4(a) and 4(b) are to clarify what 
DiGraph(matrix) does.  Just to clarify, here's the matrix in question:


sage: G = DiGraph({1:{2: 1}, 2:{1:1}})
sage: G.laplacian_matrix()

[ 1 -1]
[-1  1]

This is how I understand the intent of the DiGraph initialization 
function:  If there are negative numbers in the matrix, then the entries 
of the matrix are treated as weights.  If they are all nonnegative, then 
the entries are treated as the numbers of edges.  Since the diagonal 
entries are nonzero, the digraph is given loops.  Since the off-diagonal 
entries are nonzero (and at least one is negative), the digraph is also 
given two edges, with weights corresponding to the values of the entries.

I think the functionality is working exactly as intended.  However, it 
may be that Sage is trying too hard to be intelligent about guessing 
what you want.  Why do you expect that DiGraph(G.laplacian_matrix()) 
should give you back the same graph?

Jason


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: notebook share button

[Jason Grout]

Maurizio wrote:
> Let me throw this stone, it's just something that pops into my mind
> now: do you know what simulink is? That is a unique feature of MatLab
> (actually it's a toolbox), which has been pathetically replicated by
> somebody, but with no results in my opinion. To the best of my
> knowledge, I actually see SAGE as a viable platform to possibly work
> on something like that, even though we are MILES away to have the
> chance to do that. Before, we should use this brick to build all the
> necessary engineering walls, because we really need a comfortable
> building first, to think about such an additional package...
> My point is that we have a chance here, for something useful not only
> to mathematician, but also to a whole world of people, long awaiting
> for something to free them from the boring and expensive chain of
> Matlab and Mathcad and Mathematica...
> 
> My 2 cents
> 


We've had some threads on Simulink (and the scilab version of simulink) 
before.  You're right; we don't have sufficient framework for it right 
now.  It's certainly come up before, and I'm sure it will again.  I'm 
not sure anyone has very much experience with it, though.

There have been some attempts at some similar functionality in python. 
It would be interesting for someone that is familiar with simulink (and 
the audience simulink addresses) to comment on what is available in 
python or what is available, maybe in C, that could be wrapped by Sage.

You also might find it interesting to search the archives for Simulink.

Jason


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE possible improvement

[Jason Grout]

Maurizio wrote:
> Not yet... I think I was previously asking whether some of you guys
> are interested in trying to contact them, if you do think it does
> makes sense.
> 
> I mean, if this community is interested in having this feature, the
> Quantities developers are going to find some good feedback, otherwise,
> we could just probably end up losing a good occasion, because things
> are not ready yet (although I hope that's not the case)!
> 

I think there are a good number of people now that would like to have 
the feature, and as you well know, there are lots and *lots* of future 
users that would like to have the functionality.

I'd say it's time to contact them and see if they can help.  If there 
needs to be changes made to Quantities, then making them now, while it 
is still solidifying, would be advantageous.

Jason




> Regards
> 
> Maurizio
> 
> On 20 Mar, 01:01, Jason Grout  wrote:
>> Maurizio wrote:
>>> Hi
>>> can anyone give an advice on how to adapt Quantities to let it be
>>> compatible with SAGE? I don't know how SAGE modifies the way to
>>> represent numbers, and why it does not comply with Quantities.
>> Have you asked the Quantities developers to look at this thread?  My
>> guess is that they would be very interested in having Quantities
>> integrated in with Sage, as it immediately gives them a far greater user
>> base and exposure (and bug reports!) than otherwise.  Besides, they
>> would probably be better able to work on this than most other people.
>>
>> Thanks,
>>
>> Jason
> > 
> 


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: bug in show method for DiGraphs

[davidp]

>
> > 4. In the second call, there should not be loops at each vertex.
>
> Yes, there should be.  It's a weighted graph with loops on the vertices,
> since the laplacian has nonzero diagonal entries.
>
> Thanks,
>
> Jason

The diagonal entries in the laplacian give the out_degrees of the
corresponding vertices minus the weight of any loops.  Here is a
simpler example showing that something is wrong.

sage: G.laplacian_matrix()

[ 2 -1 -1]
[-1  2 -1]
[-1 -1  2]
sage: G = DiGraph({1:{2:1, 3:1}, 2:{1:1, 3:1}, 3:{1:1, 2:1}})
sage: G.laplacian_matrix()

[ 2 -1 -1]
[-1  2 -1]
[-1 -1  2]
sage: G.show()
sage: DiGraph(G.laplacian_matrix()).show()

In this case, G is an ordinary triangle---no loops at vertices.  Each
vertex has out_degree 2.

1.  G.show() is wrong since it misses some arrows.
2.  DiGraph(G.laplacian_matrix()).show() is wrong because it displays
loops at the vertices.
3.  DiGraph(G.laplacian_matrix()).show() is also wrong because is
missing arrowheads.
4.  To clarify an earlier question: why aren't G.show() and DiGraph
(G.laplacian_matrix()).show() the same?
4.  To further complicate matters, I think that

DiGraph(G.laplacian_matrix()).show(edge_labels=True)

should label edges by their weights, not by the negative of their
weights.

David




--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: notebook share button

[Maurizio]

thank you for adding fuel! :)

I honestly think that SAGE's user target should be slightly different
from that one. I mean, picking data from hardware drivers is just a
matter of companies agreement and capability to crunches large
quantities of data.

In this software, on the contrary, I see a powerful symbolic calculus
capability, which is not that easy to find! I see a big number of
design engineers out there, whose interest is system modeling, system
design, system analysis, and I don't see any capability to accomplish
this without any powerful symbolic or analytical system (unless you
just prefer to do thousands of simulations...).

Moreover, it's very important to deal with interfacing the analysis
results to data arrays of reasonable size. In this way, the engineer
is enabled to verify and match the analysis versus some simulation or
experimental results, without necessarily needing to physically
connect the measurement hardware to the pc. Most of the time, a couple
of waveforms is more than enough!

I have been able to take some advantage of MAXIMA in the past, but the
reality is that SAGE has some added value in this field:
1) nice and powerful data and analysis representation (I LOVE the
notebook)
2) easy sharing within a network (I LOVE the notebook)
3) easy interfacing with additional external packages
4) easy interfacing to numerical packages (like numpy)

Do you see why I do like SAGE? And I do see a LOT of potential for the
engineering world!! But I think we need some effort to further develop
that... hopefully I'll push for that :)

Regards

Maurizio

Let me throw this stone, it's just something that pops into my mind
now: do you know what simulink is? That is a unique feature of MatLab
(actually it's a toolbox), which has been pathetically replicated by
somebody, but with no results in my opinion. To the best of my
knowledge, I actually see SAGE as a viable platform to possibly work
on something like that, even though we are MILES away to have the
chance to do that. Before, we should use this brick to build all the
necessary engineering walls, because we really need a comfortable
building first, to think about such an additional package...
My point is that we have a chance here, for something useful not only
to mathematician, but also to a whole world of people, long awaiting
for something to free them from the boring and expensive chain of
Matlab and Mathcad and Mathematica...

My 2 cents

On 20 Mar, 02:02, kcrisman  wrote:
> > So, up to now, my wishlist is:
> > - better Laplace, Fourier, Zeta, any other transform management
> > (especially in symbolic)
> > - unit of measurement integration
> > - extensible comparison between different implementations of all those
> > features in the different packages (better to do the limit/integral/
> > anything else with maxima, or with sympy, or with sympycore, or with
> > pynac???)
> > - NEW: labels in the notebook :) colors would be VERY nice as well!
>
> Just to add fuel to the fire, I was told point-blank by one of my
> colleagues that unless Sage (or Octave or any other OSS) haddrivers
> to connect to measurementhardwarein his labs, he didn't see the
> point of trying to use Sage, because vendors supply/selldrivers
> hooking Matlab to the type of equipment he uses, and he wasn't about
> to write a driver.
>
> I don't really know whether this is true or not (that is, that it's
> too annoying to get data into Sage directly from things) and I don't
> even know exactly what sort of measurement device he was talking
> about, but this seems very plausible, as well as something the current
> developer network is not going to be able to solve any time soon.
> Food for thought, at least, esp. since a large piece of the current
> college calculus market is pre-engineers (see David Bressoud's recent
> articles in the Notices and FOCUS, e.g.) and those are the future
> users of Sage or any other similar program.
>
> - kcrisman
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: SAGE possible improvement

[Maurizio]

Not yet... I think I was previously asking whether some of you guys
are interested in trying to contact them, if you do think it does
makes sense.

I mean, if this community is interested in having this feature, the
Quantities developers are going to find some good feedback, otherwise,
we could just probably end up losing a good occasion, because things
are not ready yet (although I hope that's not the case)!

Regards

Maurizio

On 20 Mar, 01:01, Jason Grout  wrote:
> Maurizio wrote:
> > Hi
>
> > can anyone give an advice on how to adapt Quantities to let it be
> > compatible with SAGE? I don't know how SAGE modifies the way to
> > represent numbers, and why it does not comply with Quantities.
>
> Have you asked the Quantities developers to look at this thread?  My
> guess is that they would be very interested in having Quantities
> integrated in with Sage, as it immediately gives them a far greater user
> base and exposure (and bug reports!) than otherwise.  Besides, they
> would probably be better able to work on this than most other people.
>
> Thanks,
>
> Jason
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: notebook share button

[Maurizio]

Hi Minh,

you got the Bode plot I was talking about. Please, try to compare it
to our demo (we talked about that in this thread:
http://groups.google.com/group/sage-devel/browse_frm/thread/221ebcb20940d8dc/94e1551399325c5c?lnk=gst&q=bode+plot#94e1551399325c5c):

The demo is at this notebook: http://www.sagenb.org/pub/285/
Even though this has just a minor percentage of the whole lot of
possibility there, I'm still quite happy about that, I really hope you
guys have a look at that.

I hope your patience stays always this high, as you can see, I'm not
giving up so easily ;)

Maurizio

On 20 Mar, 00:48, Minh Nguyen  wrote:
> Hi Maurizio,
>
> On Thu, Mar 19, 2009 at 11:35 PM, Maurizio  wrote:
>
> > I hope I didn't get misunderstood :)
>
> > From my point of view, I was talking about enhancing the engineer
> > oriented SAGE functions.
>
> OK, I thought you meant the *software* engineering part of Sage
> itself, as opposed to say the features that are useful for an
> engineer. Thank you very much for clarifying this issue, and your
> growing wish list is certainly very welcome. Keep them coming!
>
> > My personal list is continuously growing as long as I use SAGE (it's
> > just 4 months now...not that much).
>
> > For example, my first concern was how to produce nice and useful Bode
> > plots. I could figure out how to do that with matplotlib (even though
> > I would have appreciated something like a semilogx or loglog plot
> > function, please forgive me but I have used Matlab for so long!).
> > Then I didn't like the inability to zoom and pan within those plots.
> > And so, mostly thanks to Kenny's help, I was able to have Bode plots
> > in FLOT, which allows me some beautiful online interaction (I mostly
> > use the notebook).
>
> With regards to Bode plot, do you mean something like the following?
>
> http://en.wikipedia.org/wiki/Bode_plot
>
>
>
> > Then, I went through the integration issues, and Laplace and Fourier
> > transform. In SAGE there is no support for delta functions, step
> > functions, and all those important features for powerful symbolic
> > representation in these transform domains.
>
> > Then, I recalled how useful was unit of measurements management with
> > Mathcad, and now I'm trying to deal with it.
>
> > So, up to now, my wishlist is:
> > - better Laplace, Fourier, Zeta, any other transform management
> > (especially in symbolic)
> > - unit of measurement integration
> > - extensible comparison between different implementations of all those
> > features in the different packages (better to do the limit/integral/
> > anything else with maxima, or with sympy, or with sympycore, or with
> > pynac???)
> > - NEW: labels in the notebook :) colors would be VERY nice as well!
>
> > I hope you don't ban me,
>
> No offense taken.
>
> Keep the feature requests coming. It's one way of knowing what's
> missing in Sage. I personally think that developers of maths software
> should be attuned to their users.
>
> > but I can tell you I will come with many more
> > requests soon... I'm afraid to be to annoying, but I really enjoy this
> > stuff! Many thanks!
>
> --
> Regards
> Minh Van Nguyen
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: FInally the link to download SAGElwlcd ISO!

[Rob Beezer]

Lucio,

That worked!  Then a right-click to get the fluxbox menu up for
anybody else who is playing along.

Any plans to catch-up to 3.4?

Thanks,
Rob

On Mar 20, 1:34 pm, Lucio Lastra  wrote:
> wait couple seconds and hit Enter and you are going to have
> the bash prompt back again.
>
> ubu...@ubuntu~:
>
> After that type: sudo startx and voilá!

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: p-adic precision loss in charpoly

[William Stein]

On Fri, Mar 20, 2009 at 2:22 PM, David Kohel  wrote:
>
> William,
>
> Thanks for pointing out that I was wrong about the lifting problem
> and original of this slow behavior.
>
> Rather than just putting in place the solution by lifting to ZZ,
> which
> you show to be much faster for reasonable size, I hope someone
> can profile arithmetic or linear algebra for ZZ/nZZ and figure out
> why this determinant is so "dog slow".   Are there too many GCD's
> or what is going on?

It's slow becase it's using the O(n!) expansion by minors algorithm,
since one isn't over a domain and hence Sage doesn't use Hessenberg
form.

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: p-adic precision loss in charpoly

[David Kohel]

William,

Thanks for pointing out that I was wrong about the lifting problem
and original of this slow behavior.

Rather than just putting in place the solution by lifting to ZZ,
which
you show to be much faster for reasonable size, I hope someone
can profile arithmetic or linear algebra for ZZ/nZZ and figure out
why this determinant is so "dog slow".   Are there too many GCD's
or what is going on?

I hope the original p-adic problem also gets a track rather than
being
lost by my side comment, since there the problem was that a wrong
answer was being returned -- a much more serious issue!

Cheers,

David

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: p-adic precision loss in charpoly

[Bill Hart]

Tuning is always difficult, especially when the search space is
multidimensional.

FLINT 2.0 will come with a tuning facility which will tune all linear
cutoffs at build time.

I'm afraid I don't have anything to offer for multidimensional
cutoffs. Usually the search space is just too large to have an
efficient tuning implementation.

Something I do notice is that multiprecision arithmetic tends to need
less adjustments to its tuning than stuff over Z/pZ for word sized p
for example, or for single precision stuff.

Trying to find a single algorithm that doesn't require some tuning
cutoffs is almost certainly impossible, if you want maximum
performance.

Bill.

2009/3/20 William Stein :
> On Fri, Mar 20, 2009 at 10:18 AM, Nick Alexander  
> wrote:
>>
 Instead I worked around this by computing the determinants
 mod 2 and mod 13 and using CRT (if the determinants were
 both units).  The time was then almost trivial.  Suppose I
 replace this problem over ZZ/25ZZ or ZZ/256ZZ. I would
 still hope that the problem would NOT be lifted to ZZ for
 computation, since this would certainly not terminate in
 reasonable time for a dense matrix.
>>
>> In general doing this via the CRT requires factoring, so some cutoff
>> must be made, which will become outdated...
>
> Writing fast code that works for a range of inputs almost always
> involves making one more explicit cutoffs, and every single one that
> is ever made will eventually be non-optimal in the future, or even on
> differently configured computers now.
>
> I just spent some time making the linear algebra over ZZ and QQ 5-250
> times faster in lots of cases by special casing small dimensions to
> much more quickly use PARI (etc.), then switching to some
> asymptotically fast native sage algorithm for big input.   This
> literally speeds up a lot of code by a factor of 5-250 that uses small
> linear algebra.  I just hardcoded some defaults for now -- e.g., use
> PARI for dets of matrices with less than 50 rows.   Of course, it will
> be much better to factor out all this "tuning info" in a global object
> that can be optimized for a range of hardware and be updated over time
> (sort of like a Sage version of ATLAS).   That said, even with
> hardcoded values, doing this special casing is *dramatically* -- I
> mean *dramatically* -- better than not doing anything.  It also makes
> it much easier to factor out the "tuning" object, when we do that
> later.
>
> Bill Hart -- you might want to make some remarks at this point, since
> you are constantly dealing with such "tuning" issues in MPIR and
> FLINT.
>
>  -- William
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: FInally the link to download SAGElwlcd ISO!

[Lucio Lastra]

Rob,

after this appears:

ubu...@ubuntu~: nm-system-settings:  Adding default connection 'Auto
eth0' for /org/freedesktop/Hal/devices/net

wait couple seconds and hit Enter and you are going to have
the bash prompt back again.

ubu...@ubuntu~:

After that type: sudo startx and voilá!

Greetings,

Lucio.
On Fri, Mar 20, 2009 at 4:49 PM, Rob Beezer  wrote:

>
> Lucio,
>
> Thanks for the reply.
>
> Downloaded the iso and burned a new CD.  Tested it on same 2 machines,
> plus one more:
>
> 6-month old Toshiba laptop, with admittedly screwy wireless hardware
> (twice)
>
> Older Dell desktop, wired networking (twice)
>
> 2-month old home-built desktop, wired networking (once)
>  RTL8111/8168B PCI Express Gigabit Ethernet controller
>
> All three normally run some variant of Ubuntu.  Xubuntu seems to run
> and then all five attempts hang at a command prompt saying
>
> ubu...@ubuntu~: nm-system-settings:  Adding default connection 'Auto
> eth0' for /org/freedesktop/Hal/devices/net with_underscores?>
>
> (That's typed-in rather than cut/paste so might not be 100% correct.)
>
> Rob
> On Mar 20, 8:33 am, Lucio Lastra  wrote:
> > Update:
> >
> > I just downloaded the ISO form Diego's Blog, burn it into a CD and
> > no problems, no data corruption, the same as the original at home.
> >
> > Booted it and did everything as described above.
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: p-adic precision loss in charpoly

[William Stein]

On Fri, Mar 20, 2009 at 10:18 AM, Nick Alexander  wrote:
>
>>> Instead I worked around this by computing the determinants
>>> mod 2 and mod 13 and using CRT (if the determinants were
>>> both units).  The time was then almost trivial.  Suppose I
>>> replace this problem over ZZ/25ZZ or ZZ/256ZZ. I would
>>> still hope that the problem would NOT be lifted to ZZ for
>>> computation, since this would certainly not terminate in
>>> reasonable time for a dense matrix.
>
> In general doing this via the CRT requires factoring, so some cutoff
> must be made, which will become outdated...

Writing fast code that works for a range of inputs almost always
involves making one more explicit cutoffs, and every single one that
is ever made will eventually be non-optimal in the future, or even on
differently configured computers now.

I just spent some time making the linear algebra over ZZ and QQ 5-250
times faster in lots of cases by special casing small dimensions to
much more quickly use PARI (etc.), then switching to some
asymptotically fast native sage algorithm for big input.   This
literally speeds up a lot of code by a factor of 5-250 that uses small
linear algebra.  I just hardcoded some defaults for now -- e.g., use
PARI for dets of matrices with less than 50 rows.   Of course, it will
be much better to factor out all this "tuning info" in a global object
that can be optimized for a range of hardware and be updated over time
(sort of like a Sage version of ATLAS).   That said, even with
hardcoded values, doing this special casing is *dramatically* -- I
mean *dramatically* -- better than not doing anything.  It also makes
it much easier to factor out the "tuning" object, when we do that
later.

Bill Hart -- you might want to make some remarks at this point, since
you are constantly dealing with such "tuning" issues in MPIR and
FLINT.

 -- William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: bug in show method for DiGraphs

[Jason Grout]

davidp wrote:
> Hi,
> 
> It seems to me that the show method for DiGraphs is not working
> correctly in Sage 3.4.
> 
> EXAMPLE:
> sage: G = DiGraph({1:{2: 1}, 2:{1:1}})
> sage: G.show()
> sage: DiGraph(G.laplacian_matrix()).show()
> 
> 1. The first call to show draws the graph with an edge missing.

This seems like a bug.


> 2. The second call produces edges with no arrows.

This is a bug, but probably not for the reason that you think it is. 
The bug is that weighted graphs do not show arrows, but should.


> 3. Why should these be different?

Your second graph is a weighted graph with loops.  Look at what 
G.laplacian_matrix() gives to see why.

What should be the same is DiGraph(G.adjacency_matrix()).show().  Is it?


> 4. In the second call, there should not be loops at each vertex.

Yes, there should be.  It's a weighted graph with loops on the vertices, 
since the laplacian has nonzero diagonal entries.

Thanks,

Jason


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: FInally the link to download SAGElwlcd ISO!

[Rob Beezer]

Lucio,

Thanks for the reply.

Downloaded the iso and burned a new CD.  Tested it on same 2 machines,
plus one more:

6-month old Toshiba laptop, with admittedly screwy wireless hardware
(twice)

Older Dell desktop, wired networking (twice)

2-month old home-built desktop, wired networking (once)
  RTL8111/8168B PCI Express Gigabit Ethernet controller

All three normally run some variant of Ubuntu.  Xubuntu seems to run
and then all five attempts hang at a command prompt saying

ubu...@ubuntu~: nm-system-settings:  Adding default connection 'Auto
eth0' for /org/freedesktop/Hal/devices/net

(That's typed-in rather than cut/paste so might not be 100% correct.)

Rob
On Mar 20, 8:33 am, Lucio Lastra  wrote:
> Update:
>
> I just downloaded the ISO form Diego's Blog, burn it into a CD and
> no problems, no data corruption, the same as the original at home.
>
> Booted it and did everything as described above.
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] bug in show method for DiGraphs

[davidp]

Hi,

It seems to me that the show method for DiGraphs is not working
correctly in Sage 3.4.

EXAMPLE:
sage: G = DiGraph({1:{2: 1}, 2:{1:1}})
sage: G.show()
sage: DiGraph(G.laplacian_matrix()).show()

1. The first call to show draws the graph with an edge missing.
2. The second call produces edges with no arrows.
3. Why should these be different?
4. In the second call, there should not be loops at each vertex.

David
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Coloring in Sage (was notebook share button)

[Florent Hivert]

  Dear Nick 

> > This coloring feature is very cool. I activated it right after  
> > reading your
> > e-mail... Until I launch emacs... Then I realize that emacs will not  
> > recognize
> > anything - and in particular the prompt - anymore. Is it possible to
> > deactivate to feature dynamically from sage itself, for example when  
> > we issue
> > the import sage_emacs as emacs ?

> I have partway done this in my tree, ie it will wait for a prompt,  
> detect colors, and issue a %colors NoColor command.  Whenever I post a  
> new spkg it will be there, but it's not a priority.

Thanks for your quick answer. I didn't imagine that ipython/sage accept during
the session a "%colors NoColor" so that I didn't even try it. Otherwise I
would not have sent my email. This was just what I was asking. I'm deeply
sorry for my stupidity... I should think twice before sending an e-mail.  

> But in general this is not the way to go: parsing ANSI and associated  
> color commands in emacs is supported but slow, according to a few  
> blogs I have read.  Why not customize font-lock-mode declarations for  
> inferior-sage-mode?  Much better.

Sure. I've already done this. 

Cheers,

Florent

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Coloring in Sage (was notebook share button)

[Nick Alexander]

On 20-Mar-09, at 1:07 AM, Florent Hivert wrote:

>
>  Dear All,
>
>> Have you tried editing
>>
>> $HOME/.sage/ipython/ipythonrc
>>
>> ?  It has numerous color options, which are off by default since it  
>> is
>> not possible to tell if the user's terminal has a white or black
>> background, and any choice of colors looks like crap if you guess
>> wrong about the background color.
>
> This coloring feature is very cool. I activated it right after  
> reading your
> e-mail... Until I launch emacs... Then I realize that emacs will not  
> recognize
> anything - and in particular the prompt - anymore. Is it possible to
> deactivate to feature dynamically from sage itself, for example when  
> we issue
> the import sage_emacs as emacs ?

I have partway done this in my tree, ie it will wait for a prompt,  
detect colors, and issue a %colors NoColor command.  Whenever I post a  
new spkg it will be there, but it's not a priority.

But in general this is not the way to go: parsing ANSI and associated  
color commands in emacs is supported but slow, according to a few  
blogs I have read.  Why not customize font-lock-mode declarations for  
inferior-sage-mode?  Much better.

Nick

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: notebook share button

[Nick Alexander]

>> I recently had a meeting with a maths educator who is very interested
>> in using Sage for undergraduate maths and cryptography. The above
>> coloured output issue is one of his feature requests.
>
> Have you tried editing
>
> $HOME/.sage/ipython/ipythonrc

Or customizing the font-lock-mode of inferior-sage-mode?  I don't care  
for this at all, but if you know what you want and can describe it  
with regular expressions it is easy.

Nick

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: p-adic precision loss in charpoly

[Nick Alexander]

>> Instead I worked around this by computing the determinants
>> mod 2 and mod 13 and using CRT (if the determinants were
>> both units).  The time was then almost trivial.  Suppose I
>> replace this problem over ZZ/25ZZ or ZZ/256ZZ. I would
>> still hope that the problem would NOT be lifted to ZZ for
>> computation, since this would certainly not terminate in
>> reasonable time for a dense matrix.

In general doing this via the CRT requires factoring, so some cutoff  
must be made, which will become outdated...

Nick

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: FInally the link to download SAGElwlcd ISO!

[Lucio Lastra]

Update:

I just downloaded the ISO form Diego's Blog, burn it into a CD and
no problems, no data corruption, the same as the original at home.

Booted it and did everything as described above.

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: Jasymca 2

[Burcin Erocal]

On Thu, 19 Mar 2009 16:17:58 -0700
William Stein  wrote:

> 
> Helmut,
> 
> Thanks for posting about your GPL'd math software project.  I found
> the description of the precise strategy you use for symbolic
> integration interesting:
> 
> http://webuser.hs-furtwangen.de/~dersch/jasymca2/Jasymca2en/node51.html


There's also a good summary of the heuristics used for integration in
Maple in the introduction to Chapter 11 of Algoritms for Computer
Algebra by Geddes, Czapor and Labahn. It's available here:

http://books.google.at/books?id=9fOUwkkRxT4C&pg=PR2&hl=en#PPA473,M1


Cheers,
Burcin

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: Jasymca 2

[Hazem]

Actually, William, I posted the message. It was forwarded from another
google discussion group (symbolic math group), where Helmut posted the
original.

If you wish to thank him, you should post something in that forum.

Best regards,

Hazem

On Mar 19, 7:17 pm, William Stein  wrote:
> Helmut,
>
> Thanks for posting about your GPL'd math software project.  I found
> the description of the precise strategy you use for symbolic
> integration interesting:
>
> http://webuser.hs-furtwangen.de/~dersch/jasymca2/Jasymca2en/node51.html
>
>
>
>
>
> On Thu, Mar 19, 2009 at 4:09 PM, Hazem  wrote:
>
> > -- Forwarded message --
> > From: Helmut Dersch 
> > Date: Mar 16, 3:45 am
> > Subject: Jasymca 2
> > To: sci.math.symbolic
>
> > Jasymca is an interactive system for solving math problems. It
> > supports arbitrary precision numbers and symbolic variables. Scalars,
> > vectors, and matrices can be built from all datatypes and used in
> > calculations. From the pseudoinverse of symbolic matrices over
> > trigonometric simplifications to symbolic solutions of integrals and
> > systems of equations, the main functionalities of CAS-programs are
> > provided. Additionaly, high performance numerical routines from LAPACK
> > and a plotmodule are implemented. The user interface can be selected
> > from either an Octave/Matlab/SciLab-like language or a GNU-Maxima
> > style. Three versions of Jasymca are provided which cover almost any
> > computer platform: A Midlet version for portable devices like
> > cellphones or PDAs, a java application for desktop PCs, laptops and
> > workstation, and an applet which can be integrated in webpages.
> > Jasymca is free software covered by the GNU public license.
>
> > Homepage with live applet:
> > 
>
> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://wstein.org- Hide quoted text -
>
> - Show quoted text -
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: FInally the link to download SAGElwlcd ISO!

[Lucio Lastra]

Rob,

I wrote you about it just about a minute ago, but think I sent you
the message to you exclusively, sorry.

When you write back may you post the answer to the list?

I'll download the .iso from Diego's blog, burn it and try it, maybe
it
got corrupted while he uploaded it. I got the original ISO at home
and as I told you, no problems.

Greetings,

Lucio.

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: FInally the link to download SAGElwlcd ISO!

[Lucio Lastra]

Rob,

I just burnt the ISO on one CD (28x) ran it and no troubles.  While
testing it I
used a 2 Gb pen drive and had any troubles either. The ISO has been
tested
on 4 different computers already with no problems at all.

I just put the cd on the reader, reboot, when the menu comes out, I
select
English, then the first option and everything loads normally. When the
Xubuntu
progress bar goes off you get a prompt, then type sudo startx and
that's it.

The idea is that Network Manager sets up everything automatically
(hardware,
connections, etc), and when you call it from the fluxbox menu you can
select
the wifi network it does detect, or disable it and go wired, etc.

Can you be more specific about when does it hung? (i.e.: while
loading, when
calling the Network Manager on the fluxbox menu, etc).  Can you
provide the
models of the network cards ?

Greetings,

Lucio.

On 20 mar, 01:17, Rob Beezer  wrote:
> This looks like a great idea, but has anybody had any luck with this?
>
> I've burned one CD, and on two very different machines, it has hung
> trying to setup networking at the same place - an "nm-system-settings"
> command, where I'm reasonable sure from the rest of the command that
> the "nm" refers to "Network Manager."
>
> Before I go burn some more coasters and reboot more machines, I
> thought I'd check to see if anybody else had either success or the
> same problem.  Thanks,
>
> Rob
>
> On Mar 19, 1:31 pm, William Stein  wrote:
>
> > -- Forwarded message --
> > From: Lucio Lastra 
> > Date: 2009/3/19
> > Subject: FInally the link to download SAGElwlcd ISO!
> > To: William Stein 
>
> > William,
>
> > there it goes:
>
> >http://diego.qualisyss.com/joomla/index.php
>
> > I hope you enjoy it. Please feel free to comment,
> > question and challenge it :)
>
> > You may forward it to the SAGE-devel list.
>
> > As always I'll be glad to read your opinion about it.
>
> > Greetings,
>
> > Lucio.
>
> > P.S.:  Any downloads should be done in between 16 and 7 hs.
> > (afternoon to next day morning) Ours is taken from GMT - 3
> > (Montevideo, Uruguay time).
>
> > --
> > William Stein
> > Associate Professor of Mathematics
> > University of Washingtonhttp://wstein.org

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: NetworkX v0.99

[Gustav Delius]

Mike, thanks a lot for having looked into this.

Gustav

On Mar 20, 12:30 pm, Mike Hansen  wrote:
> Hello,
>
> On Mar 20, 12:09 am, Gustav Delius  wrote:
>
> > I read that Sage was upgraded to using NetworkX v0.99 as of Sage
> > version 3.3. However onwww.sagenb.org, one still has to use
> > networkx.XGraph() class even though that has been removed from
> > NetworkX v0.99. Why is that? I believewww.sagenb.orgisrunning Sage
> > version 3.4. Was the upgrade to NetworkX v0.99 reversed when going to
> > Sage v 3.4? Why?
>
> Thanks for letting us know about this.  It was supposed to have been
> upgraded athttp://trac.sagemath.org/sage_trac/ticket/4860, but
> the .spkg is messed up so that the old version is installed.  (The
> networkx-0.99 directory was under src/.)  We'll fix this in the next
> release.
>
> --Mike
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: NetworkX v0.99

[Mike Hansen]

Hello,

On Mar 20, 12:09 am, Gustav Delius  wrote:
> I read that Sage was upgraded to using NetworkX v0.99 as of Sage
> version 3.3. However onwww.sagenb.org, one still has to use
> networkx.XGraph() class even though that has been removed from
> NetworkX v0.99. Why is that? I believewww.sagenb.orgis running Sage
> version 3.4. Was the upgrade to NetworkX v0.99 reversed when going to
> Sage v 3.4? Why?

Thanks for letting us know about this.  It was supposed to have been
upgraded at http://trac.sagemath.org/sage_trac/ticket/4860 , but
the .spkg is messed up so that the old version is installed.  (The
networkx-0.99 directory was under src/.)  We'll fix this in the next
release.

--Mike
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Error in Debian with Sage from sources(binary doesn't not work either)

[William Stein]

On Fri, Mar 20, 2009 at 2:13 AM, Francisco Cordobés  wrote:
>
> Hello, I have compiled Sage from source(version 3.4) and although the
> process is long it goes without any error. The problem is when trying
> to do things in the notebook. I cannot even to plot(sin,1,2) because
> maxima is not able to start. I also tried maxima alone and got the
> following error:
>
> $sage-3.4/local/bin$ ./maxima
>
> *** - invalid byte sequence #xCD #x53 in CHARSET:UTF-8 conversion
> Break 1 [4]>
>
> Any hint on what is going wrong?
>

You have a file -- maybe in the same directory where you start Sage --
that has a non-ascii character in it.  Rename that file and the probem
will go away.There's also some language environment variable that
will fix this, which I vaguely recall being mentioned on one of the
sage-devel or sage-support mailing lists in the last few weeks.

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Error in Debian with Sage from sources(binary doesn't not work either)

[Francisco Cordobés]

Hello, I have compiled Sage from source(version 3.4) and although the
process is long it goes without any error. The problem is when trying
to do things in the notebook. I cannot even to plot(sin,1,2) because
maxima is not able to start. I also tried maxima alone and got the
following error:

$sage-3.4/local/bin$ ./maxima

*** - invalid byte sequence #xCD #x53 in CHARSET:UTF-8 conversion
Break 1 [4]>

Any hint on what is going wrong?

Thanks,

Paco Cordobés

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] NetworkX v0.99

[Gustav Delius]

I read that Sage was upgraded to using NetworkX v0.99 as of Sage
version 3.3. However on www.sagenb.org, one still has to use
networkx.XGraph() class even though that has been removed from
NetworkX v0.99. Why is that? I believe www.sagenb.org is running Sage
version 3.4. Was the upgrade to NetworkX v0.99 reversed when going to
Sage v 3.4? Why?

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Fwd: FInally the link to download SAGElwlcd ISO!

[Alfredo Portes]

If you dont want to burn the iso and have vmware player or workstation
you can download:

http://sage.math.washington.edu/home/alfredo/sagelivecd/sage-vmware-livecd.zip

Edit the file SageCD.vmx where it says sage-3.4.iso to be the name of
your iso. Then
just run it inside vmware.

You can also try:

http://sage.math.washington.edu/home/alfredo/index.html

Regards,

Alfredo

On Thu, Mar 19, 2009 at 11:17 PM, Rob Beezer  wrote:
>
> This looks like a great idea, but has anybody had any luck with this?
>
> I've burned one CD, and on two very different machines, it has hung
> trying to setup networking at the same place - an "nm-system-settings"
> command, where I'm reasonable sure from the rest of the command that
> the "nm" refers to "Network Manager."
>
> Before I go burn some more coasters and reboot more machines, I
> thought I'd check to see if anybody else had either success or the
> same problem.  Thanks,
>
> Rob
>
> On Mar 19, 1:31 pm, William Stein  wrote:
>> -- Forwarded message --
>> From: Lucio Lastra 
>> Date: 2009/3/19
>> Subject: FInally the link to download SAGElwlcd ISO!
>> To: William Stein 
>>
>> William,
>>
>> there it goes:
>>
>> http://diego.qualisyss.com/joomla/index.php
>>
>> I hope you enjoy it. Please feel free to comment,
>> question and challenge it :)
>>
>> You may forward it to the SAGE-devel list.
>>
>> As always I'll be glad to read your opinion about it.
>>
>> Greetings,
>>
>> Lucio.
>>
>> P.S.:  Any downloads should be done in between 16 and 7 hs.
>> (afternoon to next day morning) Ours is taken from GMT - 3
>> (Montevideo, Uruguay time).
>>
>> --
>> William Stein
>> Associate Professor of Mathematics
>> University of Washingtonhttp://wstein.org
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Sun Article

[John Cremona]

2009/3/20 William Stein :
>
> On Fri, Mar 20, 2009 at 12:40 AM, Florent Hivert
>  wrote:
>>
>>> 
>>> The quote:
>>>
>>> particularly with respect to numbers theory
>>>
>>> should really be
>>>
>>> particularly with respect to *number* theory
>>>
>>> (my emphasis)
>>> 
>>
>> Fun !!! I didn't realize until now that English people where doing "number
>> theory" whereas us French do "théorie des nombres" (plural). So which one do
>> you pick for your study ? 0, 1 maybe pi ? :-) It sounds very weird to me.
>
> When I tell random non-mathematicians that I study 'number theory',
> they often look at me like I'm crazy for a while, then ask "What's so
> interesting about the number three?"   The point is that they
> misunderstandard "number theory" as "number three".  This has
> literally happened to me probably 10 times or more in my life.

It must be your funny accent, it has never happened to me!

Alternatively, you could say that you worked in Arithmetic. :)

John

>
> William
>
> >
>

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Sun Article

[Florent Hivert]

> A mathematician I know told me once she only had two tricks.  She
> could multiply by one and add zero.  But always very creatively.
> There's a grain of truth in that...

Well ! This remembers me Alain Lascoux which is a quite famous
combinatorialist, published some papers entitled:

 * Addition of +-1, application to arithmetic. 
 * About division by 1.
 * Couper les alphabets en 4 (which he translated to "Alphabet splitting") 

Cheers,

Florent

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Sun Article

[Rob Beezer]

A mathematician I know told me once she only had two tricks.  She
could multiply by one and add zero.  But always very creatively.
There's a grain of truth in that...

On Mar 20, 12:59 am, Jason Grout  wrote:
> I wonder what people would think if I started saying that my life's work
> was to study the number zero...

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Sun Article

[William Stein]

On Fri, Mar 20, 2009 at 12:40 AM, Florent Hivert
 wrote:
>
>> 
>> The quote:
>>
>> particularly with respect to numbers theory
>>
>> should really be
>>
>> particularly with respect to *number* theory
>>
>> (my emphasis)
>> 
>
> Fun !!! I didn't realize until now that English people where doing "number
> theory" whereas us French do "théorie des nombres" (plural). So which one do
> you pick for your study ? 0, 1 maybe pi ? :-) It sounds very weird to me.

When I tell random non-mathematicians that I study 'number theory',
they often look at me like I'm crazy for a while, then ask "What's so
interesting about the number three?"   The point is that they
misunderstandard "number theory" as "number three".  This has
literally happened to me probably 10 times or more in my life.

William

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Coloring in Sage (was notebook share button)

[Florent Hivert]

  Dear All,

> Have you tried editing
> 
> $HOME/.sage/ipython/ipythonrc
> 
> ?  It has numerous color options, which are off by default since it is
> not possible to tell if the user's terminal has a white or black
> background, and any choice of colors looks like crap if you guess
> wrong about the background color.

This coloring feature is very cool. I activated it right after reading your
e-mail... Until I launch emacs... Then I realize that emacs will not recognize
anything - and in particular the prompt - anymore. Is it possible to
deactivate to feature dynamically from sage itself, for example when we issue
the import sage_emacs as emacs ?   

Cheers,

Florent

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Sun Article

[Jason Grout]

Florent Hivert wrote:
>> 
>> The quote:
>>
>> particularly with respect to numbers theory
>>
>> should really be
>>
>> particularly with respect to *number* theory
>>
>> (my emphasis)
>> 
> 
> Fun !!! I didn't realize until now that English people where doing "number
> theory" whereas us French do "théorie des nombres" (plural). So which one do
> you pick for your study ? 0, 1 maybe pi ? :-) It sounds very weird to me.
> 

A computational number theorist once told me that all (his field's) 
proofs could be boiled down, in the end, to whether something was zero 
or one.  I suppose studying the number zero (and all things equal or not 
equal to it) is be pretty deep!

I wonder what people would think if I started saying that my life's work 
was to study the number zero...

Jason


--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Sun Article

[Minh Nguyen]

Hi Florent,

On Fri, Mar 20, 2009 at 7:40 AM, Florent Hivert
 wrote:
>
>> 
>> The quote:
>>
>> particularly with respect to numbers theory
>>
>> should really be
>>
>> particularly with respect to *number* theory
>>
>> (my emphasis)
>> 
>
> Fun !!! I didn't realize until now that English people where doing "number
> theory" whereas us French do "théorie des nombres" (plural). So which one do
> you pick for your study ? 0, 1 maybe pi ? :-) It sounds very weird to me.

Instead of saying number theory, in English one can also refer to that
discipline as the theory of numbers, not *numbers theory* :-)

>
> It's good to have those kink of papers about sage anyway !

-- 
Regards
Minh Van Nguyen

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Sun Article

[Florent Hivert]

> 
> The quote:
> 
> particularly with respect to numbers theory
> 
> should really be
> 
> particularly with respect to *number* theory
> 
> (my emphasis)
> 

Fun !!! I didn't realize until now that English people where doing "number
theory" whereas us French do "théorie des nombres" (plural). So which one do
you pick for your study ? 0, 1 maybe pi ? :-) It sounds very weird to me.

It's good to have those kink of papers about sage anyway !

Cheers,

Florent

--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---