Package categories

[Chris Walker-2]

It would be nice to see science packages better categorised. I guess
this is ultimately best done using debtags.

For example http://cdd.alioth.debian.org/science/tasks/physics.php lists
packages to do Finite element analysis, optical simulation, xray
absorption spectroscopy, ab inito quantum mechanics and computer algebra.

I propose to post to this list or the wiki a list of potential
subcategories along with packages that would go in them, and see where we go from there.

Comments?


Chris

PS I think that http://cdd.alioth.debian.org/science/tasks/physics.php
is nice, but a denser format would also be useful - more like the
https://help.ubuntu.com/community/UbuntuScience (but again with better
subcategories).



--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Andreas Tille]

On Sun, 29 Jun 2008, Chris Walker wrote:

> For example http://cdd.alioth.debian.org/science/tasks/physics.php lists
> packages to do Finite element analysis, optical simulation, xray
> absorption spectroscopy, ab inito quantum mechanics and computer algebra.
>
> I propose to post to this list or the wiki a list of potential
> subcategories along with packages that would go in them, and see wheres
> we go from there.

The current technique to build these pages is not able to create subcategories
and there is no plan to add this feature for reasons given below.  I would
not recommend using a wiki for the categorisation - other CDDs had the
experience that these wikis become unmaintained and outdated very quickly.

> PS I think that http://cdd.alioth.debian.org/science/tasks/physics.php
> is nice, but a denser format would also be useful - more like the
> https://help.ubuntu.com/community/UbuntuScience (but again with better
> subcategories).

I perfectly agree that physical packages deserve a better categorisation.
DebTags is great and should definitely be used - just go for a tagging
of these packages.

The reason why we do not have a better categorisation is that we started
Debian Science CDD for those sciences that did not yet have enough man power
to maintain an own CDD.  I would really love to see a Debian Physics CDD.
Would you volunteer to care for this?

My suggestion would be to start with creating tasks files builded according
to the known sheme and move them to the cdd/projects space.  Then we could
build physics specific pages easily and will see whether this idea attracts
enough people to work on a self-contained CDD.  Just ask if you need any
help.

Kind regards

         Andreas.
--
http://fam-tille.de


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Chris Walker-2]

On Sun, Jun 29, 2008 at 04:30:35PM +0200, Andreas Tille wrote:

> On Sun, 29 Jun 2008, Chris Walker wrote:
>
> >For example http://cdd.alioth.debian.org/science/tasks/physics.php lists
> >packages to do Finite element analysis, optical simulation, xray
> >absorption spectroscopy, ab inito quantum mechanics and computer algebra.
> >
> >I propose to post to this list or the wiki a list of potential
> >subcategories along with packages that would go in them, and see wheres
> >we go from there.
>
> The current technique to build these pages is not able to create
> subcategories
> and there is no plan to add this feature for reasons given below.  I would
> not recommend using a wiki for the categorisation - other CDDs had the
> experience that these wikis become unmaintained and outdated very quickly.
>

It was the outdatedness of the science wiki pages that prompted me to
think something needs doing, and that there was a contribution I could
make.

The other thing was seeing some discussion about FEA packages going in
- but it not being clear whether there was a complete fea system I
would be able to use.


> >PS I think that http://cdd.alioth.debian.org/science/tasks/physics.php
> >is nice, but a denser format would also be useful - more like the
> >https://help.ubuntu.com/community/UbuntuScience (but again with better
> >subcategories).
>
> I perfectly agree that physical packages deserve a better categorisation.
> DebTags is great and should definitely be used - just go for a tagging
> of these packages.
>

I'm not sure there are enough categories - physics is too coarse a
granularity - perhaps I should post some examples.


> The reason why we do not have a better categorisation is that we started
> Debian Science CDD for those sciences that did not yet have enough man power
> to maintain an own CDD.  I would really love to see a Debian Physics CDD.
> Would you volunteer to care for this?

I don't think I will have the time unfortunately.

In addition, although I'm a physicist, much of what I've done has been
crystallography, so packages useful to me would pull in a significant
part of chemistry.


>
> My suggestion would be to start with creating tasks files builded according
> to the known sheme and move them to the cdd/projects space. Then we could
> build physics specific pages easily and will see whether this idea attracts
> enough people to work on a self-contained CDD.  Just ask if you need any
> help.
>

I'll list some tasks and packages, and see where we go from there.

Chris


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Chris Walker-2]

On Sun, Jun 29, 2008 at 05:46:23PM +0100, chrisw wrote:

> On Sun, Jun 29, 2008 at 04:30:35PM +0200, Andreas Tille wrote:
> > On Sun, 29 Jun 2008, Chris Walker wrote:
> >
> > >For example http://cdd.alioth.debian.org/science/tasks/physics.php lists
> > >packages to do Finite element analysis, optical simulation, xray
> > >absorption spectroscopy, ab inito quantum mechanics and computer algebra.
> > >
> > >I propose to post to this list or the wiki a list of potential
> > >subcategories along with packages that would go in them, and see wheres
> > >we go from there.

And here it is.
== Overview ==

I find that categorising packages into just physics/chemistry/maths is
not quite granular enough. It would be nice to have better granularity
- and I think that debtags is probably the way to go here. What do
other people think?

The debtags available don't seem to have quite enough granularity -
but perhaps I've missed something - so I've knocked up a very
incomplete list of sections that packages might drop into.

Nb all the packages here have been packaged for debian.

I guess the main thing I'm aiming for is some discussion of whether
there is need for addition package tags, or perhaps I've missed something.

Comments?

== Physics ==
field::physics


=== Finite Element Analysis ===
proposed -- field::physics:fea
 (it isn't clear to me that this should be in physics rather than engineering, so maybe field::fea would be better)


Exotk
Code_Aster
OpenFoam
OpenFLOWer
Salome

=== Diffraction/crystal structure solution ===

(all the below are listed as having unofficial packages by Carlo Segre).
mpr peak fitting and x-ray diffraction data viewer
expgui GUI for GSAS (OK for Debian but GSAS isn't)
fox crystal structure solution
mcmaille x-ray structure solution
tpf x-ray profile refinement
xgen macromolecular crystallography


=== Particle Physics ===
CERNLIB


== Chemistry ===
field::chemistry

=== Structure calculation ===

==== Abinitio  ====
(should DFT be listed separately here?

Abinit (currently under physics)
MPQC
psi


==== Semi Empirical ====
mopac


=== Molecular graphics ===
Bkchem
rasmol
pymol
Gabedit (possibly also referenced in abinitio)
Chemtool
garlic
gchempaint
grcystal
Gdis
jmol


== Maths ==

===Symbolic calculation ===
maxima

===Matlab like ===
octave
pdl
matplotlib
scilab


== Data Acquisition ==
libcomedi
libgpib










--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Andreas Tille]

On Sun, 29 Jun 2008, Chris Walker wrote:

> I'm not sure there are enough categories - physics is too coarse a
> granularity - perhaps I should post some examples.

Sure.  We should try with some pragmatism here.

> In addition, although I'm a physicist,

Me too - but I'm not working as a physicist any more ...

> I'll list some tasks and packages, and see where we go from there.

Sounds like a reasonable start - lat's see where it will go to ...

Kind regards

            Andreas.

--
http://fam-tille.de


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Adam C Powell IV]

On Sun, 2008-06-29 at 13:55 +0100, Chris Walker wrote:

> It would be nice to see science packages better categorised. I guess
> this is ultimately best done using debtags.
>
> For example http://cdd.alioth.debian.org/science/tasks/physics.php lists
> packages to do Finite element analysis, optical simulation, xray
> absorption spectroscopy, ab inito quantum mechanics and computer algebra.
>
> I propose to post to this list or the wiki a list of potential
> subcategories along with packages that would go in them, and see where we go from there.
>
> Comments?

Rather than "subcategories", I would prefer "keywords", because many
packages cover multiple subcategories.  But then, debtags are more
keyword than (sub)category in this sense.

-Adam
--
GPG fingerprint: D54D 1AEE B11C CE9B A02B  C5DD 526F 01E8 564E E4B6

Engineering consulting with open source tools
http://www.opennovation.com/

[George Serbanut]

I agree with you, Adam. There are packages touching more than one subcategory. I am physicist in the elementary particle physics and high energies, working on software for that field, and I can say that, in my field at least, there cannot be physics without math and computer science. So, I support Adam in his comment.

Cheers,
George

On Mon, Jun 30, 2008 at 5:39 AM, Adam C Powell IV <hazelsct@...> wrote:

On Sun, 2008-06-29 at 13:55 +0100, Chris Walker wrote:
> It would be nice to see science packages better categorised. I guess
> this is ultimately best done using debtags.
>
> For example http://cdd.alioth.debian.org/science/tasks/physics.php lists
> packages to do Finite element analysis, optical simulation, xray
> absorption spectroscopy, ab inito quantum mechanics and computer algebra.
>
> I propose to post to this list or the wiki a list of potential
> subcategories along with packages that would go in them, and see where we go from there.
>
> Comments?

Rather than "subcategories", I would prefer "keywords", because many
packages cover multiple subcategories.  But then, debtags are more
keyword than (sub)category in this sense.

-Adam
--
GPG fingerprint: D54D 1AEE B11C CE9B A02B  C5DD 526F 01E8 564E E4B6

Engineering consulting with open source tools
http://www.opennovation.com/

[Charles Plessy-12]

Le Sun, Jun 29, 2008 at 07:03:58PM +0100, Chris Walker a écrit :
>
> The debtags available don't seem to have quite enough granularity -
> but perhaps I've missed something - so I've knocked up a very
> incomplete list of sections that packages might drop into.

Hi Chris,

If I understand correctly, the Debtags are not granular on puropse.
Better categorisation is supposed to be acheived by combining them. For
instance Field::Chemistry, Use::Viewer instead of
Field::Chemistry:Molecular grahpics.

This said, there are cases where creating subfacets are definitely the
best solution. For instance for biology, we obtained the creation of the
Field::Biology:molecular, :bioinformatics and :structural subfacets.

Have a nice day,

--
Charles


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Christophe Prud'homme]

> === Finite Element Analysis ===
> proposed -- field::physics:fea
>  (it isn't clear to me that this should be in physics rather than engineering, so maybe field::fea would be better)
agreed or field::engineering::fea if field::engineering exists
note that it could also fit in math / numerical methods for those who
are interested in developing  numerical methods in the codes below

>
> Exotk
> Code_Aster
> OpenFoam
> OpenFLOWer


> Salome
to my knowledge Salome does not provide a fe code !


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Chris Walker-2]

"Christophe Prud'homme" <prudhomm@...> writes:

> > === Finite Element Analysis ===
> > proposed -- field::physics:fea
> >  (it isn't clear to me that this should be in physics rather than engineering, so maybe field::fea would be better)
> agreed or field::engineering::fea

That would be field::engineering:fea by anology with the biology subfields.

> if field::engineering exists

It doesn't - but a science-engineering metapackage has been created,
and therefore perhaps it should.

> note that it could also fit in math / numerical methods for those who
> are interested in developing  numerical methods in the codes below
>
> >
> > Exotk
> > Code_Aster
> > OpenFoam
> > OpenFLOWer
>
>
> > Salome
> to my knowledge Salome does not provide a fe code !
>

AFAICT from http://www.salome-platform.org/home/presentation/overview/
while salome doesn't perform FEA calculations, it can be used to
create meshes and display results from FEA - which is why I suggested
it in that category. It wouldn't however fit in a numerical methods
category.

Chris


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Chris Walker-2]

Charles Plessy <plessy@...> writes:

> Le Sun, Jun 29, 2008 at 07:03:58PM +0100, Chris Walker a écrit :
> >
> > The debtags available don't seem to have quite enough granularity -
> > but perhaps I've missed something - so I've knocked up a very
> > incomplete list of sections that packages might drop into.
>
> Hi Chris,
>
> If I understand correctly, the Debtags are not granular on puropse.
> Better categorisation is supposed to be acheived by combining them.

Yes. Furthermore, there are some programs that cross categories. For
example, if one were to propose a tag Field::Crystallography [1], then
that tag would be appropriate to molecular visualisation programs that
had support for crystals - eg by allowing multiple unit cells etc.


> For
> instance Field::Chemistry, Use::Viewer instead of
> Field::Chemistry:Molecular grahpics.

How does that distinguish it from the plotting program kst [2] - which
is currently tagged Field::Chemistry and use::viewing

Is kst incorrectly tagged? If not, should all the other scientific
plotting programs be tagged chemistry too - or does kst have some
chemistry specific features?


> This said, there are cases where creating subfacets are definitely the
> best solution. For instance for biology, we obtained the creation of the
> Field::Biology:molecular, :bioinformatics and :structural subfacets.

As well as

    * biology::emboss: EMBOSS
      Packages related to the European Molecular Biology Open Software Suite.

    * biology::format:aln: Clustal/ALN
      Used in multiple alignment of biological sequences.

    * biology::format:fasta: Nexus
      Popular format for phylogenetic trees.

    * biology::nuceleic-acids: Nucleic Acids Software that works with
      sequences of nucleic acids: DNA, RNA but also non-natural
      nucleic acids such as PNA or LNA.

    * biology::peptidic: Proteins
      Software that works with sequences of aminoacids: peptides and proteins.


Can someone explain the rationale for having both field::biology and
biology::*.


Chris


[1] I do wonder about a field::cystallography - to encompas packages
for real and reciprocal space study (X-ray, neutron and electron
diffraction) of crystalline materials.

[2] kst: "A KDE application used for displaying scientific data

    This is a metapackage for kst which installs all of the relevant packages.

    kst is a program for examining data streams which can plot x-y
    plots, power spectra, histograms and equations (including
    equations of data streams). It can also be used to examine data in
    files which are being updated as data is being logged, in which
    case it can act as a plotter for a chart recorder.


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Charles Plessy-12]

Le Thu, Jul 24, 2008 at 02:30:04PM +0100, Chris Walker a écrit :
>
> Is kst [A KDE application used for displaying scientific data] incorrectly tagged?

In my opinion, yes. The drawback with Debtags is that tagging is
anonymous, so we can't discuss with the person who tagged kst
Field::Chemistry. But I would suggest asking the maintainer and CCing
Debichem to confirm that this tag can be removed.


> Can someone explain the rationale for having both field::biology and
> biology::*.

Unfortunately, they stem from a disagreement between me and the Debtags
team. I wanted a Suite::EMBOSS and a few works-with and
works-with-format tags related to biology, and they objected that it was
too specialised. They created the biology::facet instead. I do not think
that I want to use it.

Here is the complete discussion.

http://lists.alioth.debian.org/pipermail/debtags-devel/2007-September/001680.html

I will go through the packages relevant to Debian Med and reopen the
discussion with solid numbers about how many packages would use this or
that proposed tag.

Have a nice day,

--
Charles


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Adam C Powell IV]

On Thu, 2008-07-24 at 13:44 +0100, Chris Walker wrote:

> "Christophe Prud'homme" <prudhomm@...> writes:
> > > Salome
> > to my knowledge Salome does not provide a fe code !
> >
>
> AFAICT from http://www.salome-platform.org/home/presentation/overview/
> while salome doesn't perform FEA calculations, it can be used to
> create meshes and display results from FEA - which is why I suggested
> it in that category. It wouldn't however fit in a numerical methods
> category.

Indeed: Salomé proper doesn't include a solver, but it does just about
everything else (meshing, MED file editing, post-processing).  And
Salomé-MECA adds modules to set up and monitor/control a complex Aster
run, so in a sense it is a complete FEA front end.

Unfortunately, Salomé recent source code is not available, and most of
-MECA source has never been released. :-(  However, Sylvestre is in
close contact with developers, and it looks like there will be progress
by the end of the year.

-Adam
--
GPG fingerprint: D54D 1AEE B11C CE9B A02B  C5DD 526F 01E8 564E E4B6

Engineering consulting with open source tools
http://www.opennovation.com/

[Felipe Figueiredo-2]

On Thursday 24 July 2008 03:29:00 Christophe Prud'homme wrote:
> > === Finite Element Analysis ===
> > proposed -- field::physics:fea
> >  (it isn't clear to me that this should be in physics rather than
> > engineering, so maybe field::fea would be better)

In the particular example of FE, it's "made" in a Mathematics/Numerical
Analysis framework, but most popularly used in Physics and Engineering. IMO,
this is the core of this discussion.

Does the tagging take into account the Field that uses the technique, or the
one that makes/improves it? Maybe there should be some discussion in regard
to policy to this matter, if there hasn't already been one.

regards
FF


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Michael Banck]

On Sun, Jun 29, 2008 at 07:03:58PM +0100, Chris Walker wrote:
> == Chemistry ===
> field::chemistry
>
> === Structure calculation ===
>
> ==== Abinitio  ====
> (should DFT be listed separately here?

DFT is ab initio, despite all rumours.
 

> Abinit (currently under physics)
> MPQC
> psi
>
>
> ==== Semi Empirical ====
> mopac
>
>
> === Molecular graphics ===
> Bkchem
> rasmol
> pymol
> Gabedit (possibly also referenced in abinitio)
> Chemtool
> garlic
> gchempaint
> grcystal
> Gdis
> jmol

Certainly the chemistry stuff should get better sorted, and this is on
my TODO list for debichem.  In particular, the last group needs more
grouping (molecular modelling, 2D structure editing etc.)

At the same time, I am wondering why abinit got uploaded to pkg-scicomp
without consulting the debichem team first?  If abinit is really only
interesting to physicists I guess that's fine, but the description seems
to indicate otherwise.  It seems pointless to have competing packaging
groups around, I thought pkg-scicomp was mainly packaging software which
is of general interest to scientific computing (like lapack, blas,
etc.), not specific applications for which other packaging groups (and
debian-science more generally) are already available.


cheers,

Michael


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Andreas Tille]

On Sun, 27 Jul 2008, Michael Banck wrote:

> On Sun, Jun 29, 2008 at 07:03:58PM +0100, Chris Walker wrote:
>> === Structure calculation ===
>>
>> ==== Abinitio  ====
>>
>> ==== Semi Empirical ====
>>
>> === Molecular graphics ===

So we would have four categories for DebiChem?

> Certainly the chemistry stuff should get better sorted, and this is on
> my TODO list for debichem.  In particular, the last group needs more
> grouping (molecular modelling, 2D structure editing etc.)

... or even more.

> At the same time, I am wondering why abinit got uploaded to pkg-scicomp
> without consulting the debichem team first?

I'm a real fan of communication - so please be nice to each other and
at least drop some notes to potentially interested groups ...

> If abinit is really only
> interesting to physicists I guess that's fine, but the description seems
> to indicate otherwise.

Well, there is nothing that speaks against listing a package in more than
one category (meta package).  We know that we are listing DebiChem maintained
packages in med-bio and there is much room for listing packages in more
than one task.

> It seems pointless to have competing packaging groups around,

Yes.  As long as a package is solidly maintained there is no reason
to blame anybody.

> I thought pkg-scicomp was mainly packaging software which
> is of general interest to scientific computing (like lapack, blas,
> etc.), not specific applications for which other packaging groups (and
> debian-science more generally) are already available.

I admit that for me the role of pkg-scicomp is somehow unclear and the
communication is not really the best.  That's why we did not waited for
the pkg-scicomp team to push the Debian Science effort but did this on
our own and hope for a reasonable cooperation.

Kind regards

         Andreas.

--
http://fam-tille.de


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[Muammar El Khatib-2]

Hi,

I'm sorry if maybe this would be a little bit off topic. However, I cannot talk
to many people about Quantum mechanics :) So here I go.

Michael Banck wrote:

> On Sun, Jun 29, 2008 at 07:03:58PM +0100, Chris Walker wrote:
>> == Chemistry ===
>> field::chemistry
>>
>> === Structure calculation ===
>>
>> ==== Abinitio  ====
>> (should DFT be listed separately here?
>
> DFT is ab initio, despite all rumours.
>  

Are you sure about this? I mean as far as I have studied, DFT is not considered
as Ab Initio (some people say yes, others no). Maybe calculation quality is as
good as Ab Initio (DFT takes care about electronic correlation) but DFT starts
from the premise that the electronic density is taken as the basic quantity. So
the electronic density is the function taken for the functional (in this case
the energy) while Ab Initio (like Hartree - Fock for instance) are based on the
many-electron wavefunctions. On the other hand DFT uses parameters derived from
_empirical data_, or from more complex calculations and in that moment is when
DFT does not follow the "calculations from first principles" (where no empirical
data is used).

IMHO DFT should be listed separately.


Regards,
--
Muammar El Khatib.
Linux user: 403107.
Key fingerprint = 90B8 BFC4 4A75 B881 39A3  1440 30EB 403B 1270 29F1
http://teorex.org | http://taciturna.com
  ,''`.
 : :' :
 `. `'
   `-


--
To UNSUBSCRIBE, email to debian-science-request@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...