Re: sage on windows
William Stein
Greetings Prof. Stein,
OK I don't even need to look to know that Sage on Cygwin is never going to work
(I hate Cygwin, that hermaphroditic monster!) - but tell me now, is Sage on MinGW an impossibility as well?
Are the problems intrinsic or Python-related? Thanks in advance!
-drl
----------------------------------------------
"I write a little. I erase a lot." - Chopin
William Stein
Professor of Mathematics
University of Washington
http://wstein.org
Emil Widmann
On Dec 23, 7:19 pm, William Stein <wst...@gmail.com> wrote:
> On Fri, Dec 23, 2011 at 11:02 AM, Danny Ross Lunsford <
>
> -- William
>
> > -drl
>
> > ----------------------------------------------
> > "I write a little. I erase a lot." - Chopin
>
> --
> William Stein
> Professor of Mathematics
> University of Washingtonhttp://wstein.org
technical sophistication < a tool like the combined
Virtual Box / Sage installer?
Download: http://boxen.math.washington.edu/home/emil/Win-Inst/Sage-Win-Installer.exe
with some screenshots http://boxen.math.washington.edu/home/emil/doc/html/en/Windows_Installer.html
I must say I am a bit puzzled - I thought that would be really a smart
way to present sage to the windows folks but I am mistaken,
interest seems to converge to zero ...
kcrisman
> > Moreover, with a virtual machine you can easily get full 64-bit support,
> > with likely better performance than any native port of Sage to windows
> > would ever give (because Linux is often faster than Windows at the sort of
> > things that Sage uses).
>
> > For Windows users with little technical sophistication, usinghttp://sagenb.orgisprobably their best bet.
>
>
> What do you think - is it an option to offer such > users with little
> technical sophistication < a tool like the combined
> Virtual Box / Sage installer?
> with some screenshotshttp://boxen.math.washington.edu/home/emil/doc/html/en/Windows_Instal...
> I must say I am a bit puzzled - I thought that would be really a smart
> way to present sage to the windows folks but I am mistaken,
> interest seems to converge to zero ...
test it to make sure it performs as advertised - for instance, I
haven't yet because I already have the VirtualBox on my (already
virtual) Windows. We need some good set of people who will test all
our Windows options on a regular basis! But I think this will
eventually be a good part of our offerings.
William Stein
> What do you think - is it an option to offer such > users with little
> technical sophistication < a tool like the combined
> Virtual Box / Sage installer?
> Download: http://boxen.math.washington.edu/home/emil/Win-Inst/Sage-Win-Installer.exe
> with some screenshots http://boxen.math.washington.edu/home/emil/doc/html/en/Windows_Installer.html
>
> I must say I am a bit puzzled - I thought that would be really a smart
> way to present sage to the windows folks but I am mistaken,
> interest seems to converge to zero ...
Here's the last email I got from a Windows user wanting to use Sage:
"William: I just wasted half an hour trying to figure out how to
download Sage. I give up. [...] Obviously everyone in your group knows
many programs I never heard of and how to use them to download. I went
around in circles, trying to follow links and directions starting at
the SAGE website. I don't know what Virtual Box is or how to use it. I
don't know what P2P is or how to use it. I do know I have Windows 7
and 64 bit operating system. That didn't help me figure out how to
interpret what to do. I'm just letting you know about this. I have to
bring my computer home now. So I won't be able to download anything
before the class, since I can't do it now. Maybe I'm the only person
in the class who wasn't born knowing all this. I don't know how to get
info about Windows 7 because our IT department just switched all
in-class systems and our laptops to it without training us. (I also
know some non-helpful IT person turned off my wireless connection
setting for internet connection and I have no clue how to turn it back
on. So I won't be able to connect to the online version of SAGE in the
class, either.) Same with everything else--no help and no training. I
can use Maple and Matlab, but it looks pretty hopeless to figure out
how to download Sage without making a mess of my office laptop. So I'm
giving up."
That's pretty bad. Even http://sagenb.org offers them no solace.
-- William
David Kirkby
> On Fri, Dec 23, 2011 at 10:09 PM, Emil Widmann <emil.w...@gmail.com> wrote:
>
>> What do you think - is it an option to offer such > users with little
>> technical sophistication < a tool like the combined
>> Virtual Box / Sage installer?
>> Download: http://boxen.math.washington.edu/home/emil/Win-Inst/Sage-Win-Installer.exe
>> with some screenshots http://boxen.math.washington.edu/home/emil/doc/html/en/Windows_Installer.html
>>
>> I must say I am a bit puzzled - I thought that would be really a smart
>> way to present sage to the windows folks but I am mistaken,
>> interest seems to converge to zero ...
>
> Here's the last email I got from a Windows user wanting to use Sage:
> "William: I just wasted half an hour trying to figure out how to
> download Sage. I give up. [...] Obviously everyone in your group knows
> many programs I never heard of and how to use them to download. I went
> around in circles, trying to follow li
<snip>
> I don't know what Virtual Box is or how to use it. I
> don't know what P2P is or how to use it.
has he or she never herd of Google?
> I don't know how to get
> info about Windows 7 because our IT department just switched all
> in-class systems and our laptops to it without training us. (I also
> know some non-helpful IT person turned off my wireless connection
> setting for internet connection and I have no clue how to turn it back
> on. So I won't be able to connect to the online version of SAGE in the
> class, either.) Same with everything else--no help and no training. I
> can use Maple and Matlab, but it looks pretty hopeless to figure out
> how to download Sage without making a mess of my office laptop. So I'm
> giving up."
>
> That's pretty bad. Even http://sagenb.org offers them no solace.
>
> -- William
Whatever the issues may be with a Windows version of Sage (and of
course I'm aware there are issues), I get the feeling this individual
is not really putting the effort in to get Sage. He/she could easily
Google VirtualBox. I'm not sure where the P2P comes in, but if he/she
does not know what that is, then again its pretty simple to find out.
In my local amateur radio club there are several people in their 70's
who have used VirtualBox. They were not bought up on computers like
students are today, but they figure it out. A guy in my village, who
will be 80 this year, uses Linux. He was born before the first
transistor was produced, let alone an integrated circuit. If a young
person doing a degree can't figure these things out, I can't help feel
they are not really trying.
I see it as advantagous to make installing Sage as easy as possible,
but there is only so much one should reasonably do. I think you once
said there was an 8-year old who was contemplating contributing to
Sage developement. If an 8-year old has managed to do something useful
with Sage, I'm sure a college student should be able to work out how
to install a VirtualBox image.
The more time you devote to making the installation easier, the less
time you have to devote to do something else.
Dave
Dima Pasechnik
> can use Maple and Matlab, but it looks pretty hopeless to figure out
> how to download Sage without making a mess of my office laptop.
Emil Widmann
> giving up."
install
http://boxen.math.washington.edu/home/emil/Win-Inst/Sage-Win-Installer.exe
with the help of
http://boxen.math.washington.edu/home/emil/doc/html/en/Windows_Installer.html
Some people search excuses not reasons, but often wrapping sells
better than content.
Marco Streng
" To run Sage on Microsoft Windows you need the following:
This speaks about "VMware", while we switched to VirtualBox. And it tells them about a native port, but isn't that very far in the future? It also uses a lot of technical language that would scare the hell out of many people I know.
I randomly click the first link, and get
" Usage
- VirtualBox for Windows or VMWare Player
- Then run the downloaded VirtualBox or VMWare Distribution of Sage and follow these additional instructions.
Again the installation of Sage is split into 2 steps, but they do not correspond 1-1 to the first set of steps. I also need to make a choice between VirtualBox and VMWare, while not understanding the difference. But let's assume I am brave and smart enough to ignore all this and click one of the mirrors.
I get lots of readme files to choose from, and one with an ova extension that I've never heard of. So I click on "Please read the Sage Virtual Machine manual."
Then I get:
" This page describes how to use the virtual machine that you can download from http://www.sagemath.org/download-windows.html. If you want to create your own virtual machine, go to the VirtualBox wiki page instead. "
This line never even mentions Sage! And does our audience even know whether they want to create their own virtual machine?
I bravely ignore this and read on: " Download the file Sage-x.y.z.ova located in the directory where this README is." Which README? I'm on the wiki!
I don't think it is the Windows user's fault...
Marco
Emil Widmann
> I don't think it is the Windows user's fault...
marketing departments nightmare .
Various people complained about the confusing and misleading
documentation during the last months, including some "wake up" posts
of ckrisman on this lists. There is a ticket on trac about this
situation:
http://trac.sagemath.org/sage_trac/ticket/12042.
Some improvements were made, including the good wiki entry of Volker
Braun about the VM ova image, but as a whole the docs are still
misleading and confusing.
my comment about excuses and reasons was not coined only on the users,
but also on the shortcommings on the project side.
RegB
sophisticated" (-:
By simple analogy, they don't REALLY want to drive a car, they just
want/need
to get somewhere - they care even less about the design and
manufacture of cars,
about the history/evolution of the automobile, etc.
Railroads as an alternative... oh, never mind.
I don't think of it as laziness, apathy, ignorance etc., they just
"wanna get to the MATH part"
they may have NO interest in Virtual machines, Linux, etc., they are
Windoze users (.) Period
Is that SO unreasonable ?
> > -- William
>
> > > -drl
>
> > > ----------------------------------------------
> > > "I write a little. I erase a lot." - Chopin
>
> > --
> > William Stein
> > Professor of Mathematics
> > University of Washingtonhttp://wstein.org
>
> What do you think - is it an option to offer such > users with little
> technical sophistication < a tool like the combined
> Virtual Box / Sage installer?
> with some screenshotshttp://boxen.math.washington.edu/home/emil/doc/html/en/Windows_Instal...
>
William Stein
On Dec 30, 2011 7:29 AM, "RegB" <2regb...@earthlink.net> wrote:
>
> I would characterize such users as "differently technically
> sophisticated" (-:
>
> By simple analogy, they don't REALLY want to drive a car, they just
> want/need
> to get somewhere - they care even less about the design and
> manufacture of cars,
> about the history/evolution of the automobile, etc.
> Railroads as an alternative... oh, never mind.
>
> I don't think of it as laziness, apathy, ignorance etc., they just
> "wanna get to the MATH part"
> they may have NO interest in Virtual machines, Linux, etc., they are
> Windoze users (.) Period
> Is that SO unreasonable ?
It is not unreasonable. We should try harder to make Sage easy for them.
> --
> You received this message because you are subscribed to the Google Groups "sage-windows" group.
> To post to this group, send email to sage-w...@googlegroups.com.
> To unsubscribe from this group, send email to sage-windows...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/sage-windows?hl=en.
>
David Kirkby
>
> On Dec 30, 2011 7:29 AM, "RegB" <2regb...@earthlink.net> wrote:
>>
>> I would characterize such users as "differently technically
>> sophisticated" (-:
>>
>> By simple analogy, they don't REALLY want to drive a car, they just
>> want/need
>> to get somewhere - they care even less about the design and
>> manufacture of cars,
>> about the history/evolution of the automobile, etc.
>> Railroads as an alternative... oh, never mind.
>>
>> I don't think of it as laziness, apathy, ignorance etc., they just
>> "wanna get to the MATH part"
>> they may have NO interest in Virtual machines, Linux, etc., they are
>> Windoze users (.) Period
>> Is that SO unreasonable ?
>
> It is not unreasonable. We should try harder to make Sage easy for them.
I'd agree the documentation could be a bit clearer.
But is it unreasonable for someone who wants to use a complex bit of
software to spend time to evualte it?
I recently wanted to evaluate HFSS
http://www.ansoft.com/products/hf/hfss/
which is expensive and makes Mathematica's cost seem like a cheap kids game.
The process was like this:
1) Contact the vendor
2) Speak on the phone about getting an evaluation license. This was a
fairlly lengthly phone call.
3) Get sent a form by email, which I must sign and FAX. In that I must
agree to devote sufficient time to evaluate the software over a 14 day
trial period.
4) Get a link to download the software.
5) Send the MAC address and the host name to the vendor.
6) Install the software
7) Install the license
8) Install the license manager softare (the horid FlexLM)
9) Print off some of the documentation - unlike Sage, there are no
books written about HFSS.
10) Have a tutorial from an engineer. That lasted about 1.5 hours.
11) Exchange several emails with support engineer, exchanging files
for my model.
12) Asked, and was granted a free one-day course on HFSS (normally
several hundred pounds).
13) Will probably get the trial extended, as 14 days was not
sufficient to evaluate this expensive, complex software.
I must have put in 50 hours over a 2-week period of this software.
Had I given up after half an hour, like the person who tried to
install Sage, I'd be stuck at step #2.
Dave
silvio.a
On Dec 23 2011, 8:19 pm, William Stein <wst...@gmail.com> wrote:
>
> By far the best way to use Sage under Windows is to run Linux in some way
> under Windows, and install Sage into that Linux instance. There are many
> options for doing this, including:
>
> . Microsoft's own Virtual machine stuff,
> . VirtualBox + any linux distro
> . Vmware + any linux distro
> . Premade virtual machines with Sage installed already:http://sagemath.org/download-windows.html
> . co-Linux (*not* a Virtual machine!):http://sourceforge.net/projects/colinux/files/
>
> Moreover, with a virtual machine you can easily get full 64-bit support,
> with likely better performance than any native port of Sage to windows
> would ever give (because Linux is often faster than Windows at the sort of
> things that Sage uses).
my comments about sage on windows:
I am working in industry, performing video quality analysis.
I use mainly numerical mathematics and statistics. I think
I am the type of customer matlab is targeted at.
I use Sage form time to time on windows using virtual box and ubuntu.
The notebook is *great* for the top-level analysis, including
plots and documentation.
For me, installation was very easy, (binary and from source)
- thanks a lot for your efforts !!
On the other hand, what I do *not* like about the virtual box
solution:
* I use many additional tools for my analysis, appart the ones
installed with sage. Some tools are windows only, others
exist on linux, e.g. ffmpeg to process video files.
I have all the tools installed on windows. But if I want
to use them from within Sage, I have to install them again
in Linux (and maintain the installation).
This is something the average windows user cannot do (or
is not willing to learn how to do it).
In addition, for larger software packages this is not very
practical, e.g. if you are a matlab user under windows,
and want to use matlab from within sage, you have to
install it again under linux.
* Some hardware is not supported by VirtualBox. E.g.
I hade once problems with data on external drives, which
are connected via fireWire. I could not access it from within
virtualBox.
I can partly understand all the difficulties of a windows port of
sage. On the other hand for many matlab users a windows port of
sage notebook + numpy + matplotlib + R (via rpy2)
would already be a *very,very useful application*.
Python, numpy, matplotlib, R all exist on windows (older versions
of rpy2, too).
How difficult would it be to make the
sage notebook run with these packages on windows?
Silvio.
Jason Grout
Have you looked at the html notebook provided with ipython 0.12? It
sounds like it might be a perfect fit for you, and it does run on windows.
I'm not saying that we can't or shouldn't do what you suggest, but you
might look at that as a possible solution for you.
Thanks,
Jason
Dima Pasechnik
Silvio.
William Stein
I've run the notebook under Windows not using cygwin before...
> sagenb does not depend on Sage (well, it might use few python decorators from there, but they can easily isolateled); it depends on several python packages that run on Windows (twisted and flack), and some javascript things.
> Help system uses ipython, as far as I know, which also runs on Windows...
That is a common misconception - I don't think the notebook uses ipython in any way, except as inspiration.
>
>
>>
>>
>>
>> Silvio.
>>
> --
> You received this message because you are subscribed to the Google Groups "sage-windows" group.
> To view this discussion on the web visit https://groups.google.com/d/msg/sage-windows/-/740pIJ9DZ1kJ.
Dmitry Semikin
Dima Pasechnik
kilucas
here:
- I didn't understand virtual machines or Linux
- it can be tricky enough to understand these well enough to do the
installation but:
- the infrastructure meant there were more places where problems
could arise (VM, Linux, Sage). For example, VMware disagreed violently
with my PC and called for a Windows reinstall, and disrupting all my
bother software installations in the process. This wasn't caused by
Sage but it was caused by the Sage installation process.
- I might've got by if everything worked on my PC but many things
didn't and, when something doesn't work and you have no
troubleshooting skills in these unfamilar environments, life gets very
complicated.
- So I googled a lot. I had to learn not only about VMs, Linux, and
Sage's infrastructure but how these integrate with each other too.
- After many weeks (after all, I have to go to work during the day), I
realised some important things:
1) I'd spent a great deal of time installing Sage but no time at
all using it!
2) I'd installed Microsoft Mathematics for some friends and,
although it's far less powerful than Sage obviously, it has one
characteristic in common with almost all the Windows software that I
use in that I just install it and use it. In fact I'm normally using
Windows software within minutes of beginning an installation, and
almost always in less than an hour. I've never installed Mathematica
but I'll guess that installing it is the same process. I can even
install the mammoth Microsoft Office and start using it in less than
an hour.
3) I wanted to make facilities available easily, but privately, to
pupils. (So using the sagenb site was not an option and hence my
desire to install my own copy). So I compared my Sage experience with
using GeoGebra. GeoGebra lets pupils use some mathematical
functionality for free and privately whether online, or locally for
personal use. And it installs easily and quickly. If pupils had
licences, I'm guessing they could install and use Mathematica fairly
straightforwardly too. Can you imagine what would be involved in
pupils installing their own Windows copies of Sage?
4) While one of my interests was single-user Sage, I expected to
have to put more effort into installing a facility that multiple users
could use. But how much more? Not the weeks I'd spent. After all, I
had multi-user Alfresco running inside an evening and that's not
trivial software either.
I am in awe of the effort that Sage developers put into projects like
this and I think it's extraordinary when people can obtain and use
such a thing for free. But the most important word in that sentence is
"use". If I don't get to use it, then I get no value from it - whether
it's free or not. Neither do the other folk who I'd hoped would use
the multi-user facility that I'd hoped to provide.
So, my ideal is an installation process that is no more complex for
the installer than common Windows installations and one which is no
more fragile. I've no idea whether this is possible and I completely
appreciate that developers may have other priorities. But I think it's
what's needed if you want to succeed with Windows users of which,
whatever we all think of Windows, there are very, very many.
Hoping the feedback is helpful.
Kevin
On Jan 13, 5:02 pm, Dima Pasechnik <dimp...@gmail.com> wrote:
> Here it's probably not the right place to ask...
> IPython has its mailing lists and an irc channel.
William Stein
Why do you think that sagenb site is not "private"? (OK, so somebody
could hack into, but it doesn't sound like you're talking about top
secret data here.)
> So I compared my Sage experience with
> using GeoGebra. GeoGebra lets pupils use some mathematical
> functionality for free and privately whether online, or locally for
> personal use. And it installs easily and quickly. If pupils had
> licences, I'm guessing they could install and use Mathematica fairly
> straightforwardly too. Can you imagine what would be involved in
> pupils installing their own Windows copies of Sage?
>
> 4) While one of my interests was single-user Sage, I expected to
> have to put more effort into installing a facility that multiple users
> could use. But how much more? Not the weeks I'd spent. After all, I
> had multi-user Alfresco running inside an evening and that's not
> trivial software either.
>
> I am in awe of the effort that Sage developers put into projects like
> this and I think it's extraordinary when people can obtain and use
> such a thing for free. But the most important word in that sentence is
> "use". If I don't get to use it, then I get no value from it - whether
> it's free or not. Neither do the other folk who I'd hoped would use
> the multi-user facility that I'd hoped to provide.
In the near future (almost now), someone like you should no more have
to install Sage to use it than you would have to install your own
Google Docs server to use Google docs.
-- William
ancienthart
Download and install virtualbox.
Download Volker's image and install into virtualbox.
Run the image and point your host browser at localhost:8000
I run this on my work machine, and the only hassle I have is that there's a piece of local-server technology already running on port 8000. To fix this problem, all I need to do is change the port forwarding so 8000 is mapped to 8080, and this is the biggest problem I've had with the new image.
Joal Heagney
David Kirkby
I run this on my work machine, and the only hassle I have is that there's a piece of local-server technology already running on port 8000. To fix this problem, all I need to do is change the port forwarding so 8000 is mapped to 8080, and this is the biggest problem I've had with the new image.
Joal Heagney
That will be fixed in Sage 5.0. It has been finally agreed to use port 8080 insted of 8000.
Dave
Emil Widmann
"normal users" have, while being immersed into the technical details
of a project very deeply.
I noticed:
> could arise (VM, Linux, Sage). For example, VMware disagreed violently
VirtualBox which should be standard by now. Unfortunately the
documentation and Information is still not consistent (this has been a
problem now for at least half a year). I don't know personally why,
because basically removing "VMware" from the frontpages would avoid
much confusion (see this "Ticket" http://trac.sagemath.org/sage_trac/ticket/12042).
So please, whoever can change this, please do it.
> - So I googled a lot. I had to learn not only about VMs, Linux, and
> Sage's infrastructure but how these integrate with each other too.
to bring sage to windows, you are also right that there is a lot to
learn the usage of the VM is not always straightforward and there is
still a great potential for improvement (I'd even say virtualisation
could be useful on other platforms, even some Linuxes fail a straight
compile from source). This is somewhat in contrast to other opinions
expressed here.
An example about the difficulties which can arise you can read in this
support thread:
http://www.mail-archive.com/sage-s...@googlegroups.com/msg25009.html
If you have to tell an average windows user that he has to recompile
the numerical libraries because his processor doesn't support the
instructions set of the virtual machine is everything else, but NOT
user-friendly.
> I realised some important things:
> 1) I'd spent a great deal of time installing Sage but no time at
> all using it!
> 2) I'd installed Microsoft Mathematics for some friends and,
> although it's far less powerful than Sage obviously, it has one
> characteristic in common with almost all the Windows software that I
> use in that I just install it and use it.
the Microsoft Program is probably a good choice for most people who
need
high school/early university level covered.
> 3) I wanted to make facilities available easily, but privately, to
> pupils. (So using the sagenb site was not an option and hence my
> desire to install my own copy). So I compared my Sage experience with
> using GeoGebra. GeoGebra lets pupils use some mathematical
> functionality for free and privately whether online, or locally for
> personal use. And it installs easily and quickly. If pupils had
> licences, I'm guessing they could install and use Mathematica fairly
> straightforwardly too.
would be
very hot in the edu field, but it has not happened until now.
> 4) While one of my interests was single-user Sage, I expected to
> have to put more effort into installing a facility that multiple users
> could use. But how much more? Not the weeks I'd spent. After all, I
> had multi-user Alfresco running inside an evening and that's not
> trivial software either.
which
is a pre-configured Sage server and combined with Virtualbox.
So the result is a "usual" Windows installer exe, which should install
the system with a view mouse clicks.
I promoted the concept, but was - until now - not successful to win
over other developers.
It pretty much conforms with the specifications you gave for your
intended usage.
Info & even Download link:
http://boxen.math.washington.edu/home/emil/doc/html/en/relase-notes-slvms-b.htm.htm
It includes a pre-configured server (29 simultaneous "worker
accounts") which should be fine for testing it in class.
So I hope you keep sage on your "watchlist" at least. The mission goal
is still
"Creating a viable free open source alternative to Magma, Maple,
Mathematica and Matlab"
and I trust this still includes the goal of installations to "Personal
Computers" and not only existence "in the cloud".
- best from Austrian Mountains
Emil Widmann
RegB
As a "regular Joe" I would like to add some hope to this.
We (regular and below average Joe and Jane s) are becoming more familiar with virtual
machines for other reasons.
One reason of course is that Oracle bought Sun, where VirtualBox originated (I am fairly sure of this).
On Oracle's technet there are a number of virtual machine images containing lab exercises for their
courses and developer days. These are very simple to download and import into VirtualBox.
I am reading into this that Oracle is USING what it (previously Sun) makes and generally that is good.
If Sage were to follow the style that Oracle is setting it MIGHT lead to easier acceptance.
If sage just publicizes more that it is using the same platform as Oracle - it would help.
OK, OK, so how much (how LITTLE) commonality of interest and career path there is between
database dudes and math types is open to speculation, I'm just suggesting that Oracle seems
to be setting a direction here - & BTW, they are LARGE and have influence on IT/IS trends.
My first experience with virtual machines was at dec with Virtual VMS, but that was in the late 80s,
same/similar concepts, but I digress even more.
William Stein
On Jan 29, 2012 3:26 PM, "RegB" <2regb...@earthlink.net> wrote:
>
>
> As a "regular Joe" I would like to add some hope to this.
> We (regular and below average Joe and Jane s) are becoming more familiar with virtual
> machines for other reasons.
> One reason of course is that Oracle bought Sun, where VirtualBox originated (I am fairly sure of this).
> On Oracle's technet there are a number of virtual machine images containing lab exercises for their
> courses and developer days. These are very simple to download and import into VirtualBox.
>
> I am reading into this that Oracle is USING what it (previously Sun) makes and generally that is good.
> If Sage were to follow the style that Oracle is setting it MIGHT lead to easier acceptance.
> If sage just publicizes more that it is using the same platform as Oracle - it would help.
>
> OK, OK, so how much (how LITTLE) commonality of interest and career path there is between
> database dudes and math types is open to speculation, I'm just suggesting that Oracle seems
> to be setting a direction here - & BTW, they are LARGE and have influence on IT/IS trends.
>
Cool. I definitely wantto reiterate that I think pushing the VM option as hard as we can is the best strategy for supporting MS Windows.
Volker - how hard woud it be to generate a 64-bit vm as well?
William
> My first experience with virtual machines was at dec with Virtual VMS, but that was in the late 80s,
> same/similar concepts, but I digress even more.
>
>
>
> -- the
> You received this message because you are subscribed to the Google Groups "sage-windows" group.
> To view this discussion on the web visit https://groups.google.com/d/msg/sage-windows/-/rSagRxStqkQJ.
RegB
http://www.oracle.com/technetwork/community/developer-vm/index.html
for some sample appliances.
I am fairly sure there is a How-to on building appliances there too, ,,,somewhere.
I am reasonably sure that one would just build the model O/S, build Sage, decide on a suitable name
and click export. There are such questions as running it with minimum size of memory
and whether one core would be wise, users can adjust those upwards at import if they feel they can afford it.
One would probably delete sources, temp files, etc. before the export.
I have an Ubuntu 11.10 64 bit VM running Sage 4.8 now, though it built with some doctest errors.
It is time consuming, but in no way technically difficult.
kilucas
Thanks for the feedback and I may indeed look at the VirtualBox
option. It would be great to have this running if I can just do so
without having to become a technical expert.
Kevin
Dave Crowe
(snip...)
> I am in awe of the effort that Sage developers put into projects like
> this and I think it's extraordinary when people can obtain and use
> such a thing for free. But the most important word in that sentence is
> "use". If I don't get to use it, then I get no value from it - whether
> it's free or not. Neither do the other folk who I'd hoped would use>
> the multi-user facility that I'd hoped to provide.
>
> So, my ideal is an installation process that is no more complex for
> the installer than common Windows installations and one which is no
> more fragile. I've no idea whether this is possible and I completely
> appreciate that developers may have other priorities. But I think it's
> what's needed if you want to succeed with Windows users of which,
> whatever we all think of Windows, there are very, very many.
>
> Hoping the feedback is helpful.
>
> Kevin
Amen! I am astounded at the amount of time and effort Sage represents, and truely thank the community in toto. Likewise, the resource at http://boxen.math.washington.edu/home/emil/sagelive/ is singularly astounding in its usefulness by non-*nix users.
I briefly used Sage as a grad student, and have since taken a (staff) position with the Comp Sci dept. I now am responsible for supporting Sage in our labs - the same labs I was using just last year! For what it's worth, I agree with the general position that guiding windows users to select the Live CD
makes a great deal of sense. Virtual machines are useful, and certainly have their place. Just as certainly, there are no 'blanket answers' that will be the best fit for every circumstance.
The then-current version I ran as a student have a fairly straight-forward ability to do a local Pup\Sage install from the Live CD. Indeed, the Sage live *.exe (based on 511-47) at the site above has this functionality [From the ReadMe: 'Use the wizards in the Menu / Setup (Puppy Universal Installer)"]. it really seemed to be an elegant way to bring Sage to us mere windoze types. The process allowed testing Pup + Sage on the end-user's hardware platform directly. If the user liked it, they could install it with a minimum of fussing with techy, non-windoze stuff. It put Pup on it's own little partition, stuck sage on it, and Bob's you uncle. If the user didn't like it, reboot and they're back in the familiar windoze world, safe from all thing *nix.
This completely side-steps the need for porting, recompiling, fixing, or dragging Sage into the non-native M$ world, yet allows the M$ captives a glimpse at Linux-promised freedom. Further, it eliminates any performance hit from any type of full OS + v-box, and lets Pup's performance shine through on nearly any platform.
Am I missing something, or has that been omitted from 511-53? If it's there and I am not seeing it due to user error, I certainly apologize and hope not too many of you will point and laugh. However, I surely cannot find anything that offers to install Pup - frugal, full, or otherwise.
As a stop-gap, I can use the sagelive.exe from above to create local installations, but believe newcomers to Sage that are unaware of the (seemingly) ommitted functionality may not know that this route is a possibility: and this may negatively affect their impression/evaluation of Sage.
I hope I'm not coming across as whining, but rather as offering positive suggestions based on personal experience.