Re: [GRASS-dev] CLI!=GUI

2010-12-01 Thread Hamish
 make build-core | build-xdriver | build-wxgui | build-tkgui

 make install-core | install-xdriver | install-wxgui | install-tkgui

as far as the debian grass package goes, perhaps we could have
a main grass package, which depended on grass-base and
grass-gui packages. The -base package wouldn't depend on the
-gui package, so folks who didn't want the GUI deps could avoid it.
also apt-get install grass would still install what people
expect: base+gui.

not sure if -base or -core is preferred, but whichever..


Hamish



  
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-30 Thread Maris Nartiss
MAKE system patches into studio! Be ready to deal also with any code
that migh assume GUI-related tools to be present by default.

Saving less than 10MB of uncompressed disk space is not worth a large effort.
GRASS has so many dependencies, that it's impossible to split-out all
tools that require some obscure lib/utility without ruining GRASS as
whole. I doubt that Debian now is providing ALL dependencies of ALL
GRASS modules - reports about module X requires Y will not go away
because of GUI/CLI split.

As GRASS lacks manpower - code or GTFO* ;)
Maris.

* A nice reference to Chatroulette

2010/11/30, Francesco P. Lovergine fran...@debian.org:

 Something like:

 make build-core | build-xdriver | build-wxgui | build-tkgui

 make install-core | install-xdriver | install-wxgui | install-tkgui

 would be more modular and clean. Of course, one can always now
 install the whole beast, use only what is of interest and avoid
 installing all dependencies. As said, it works too.
 But I - as packager - would prefer avoiding tons of silly reports
 about 'command X is not working because Y is missing' and installing
 most required dependencies for all commands provided.
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-30 Thread Markus Neteler
On Tue, Nov 30, 2010 at 4:12 AM, Glynn Clements
gl...@gclements.plus.com wrote:
 Francesco P. Lovergine wrote:

 Something like:

 make build-core | build-xdriver | build-wxgui | build-tkgui

 make install-core | install-xdriver | install-wxgui | install-tkgui

 would be more modular and clean. Of course, one can always now
 install the whole beast, use only what is of interest and avoid
 installing all dependencies. As said, it works too.
 But I - as packager - would prefer avoiding tons of silly reports
 about 'command X is not working because Y is missing' and installing
 most required dependencies for all commands provided.

 You can package it how you want, but I don't think that it's realistic
 to structure GRASS around individual packagers' policies (e.g. who
 says what is core and what isn't).

Well, I think that Francesco suggestion above is pretty clear.
It does not look like too strange to me and sufficiently generic.

Markus
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-30 Thread Glynn Clements

Michael Barton wrote:

 I'm not sure that GRASS will even run on Windows without the GUI. 
 Certainly xdrivers will not.

None of the standalone display drivers (d.mon ...) will run on
Windows. d.* commands always behave as if GRASS_RENDER_IMMEDIATE=TRUE,
i.e. they will save an image to $GRASS_PNGFILE. Other commands are
unaffected.

-- 
Glynn Clements gl...@gclements.plus.com
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-29 Thread Francesco P. Lovergine
On Sun, Nov 28, 2010 at 09:29:28PM +, Glynn Clements wrote:
 
 Hamish wrote:
 
If the only available packages (RPM, .deb, etc) insist
upon installing GUI libraries, complain to the people who
make the packages.
  
  Paolo:
   OK, so, now I'm complaining ;) : packagers, please have
   your saying here.
  
  disk space is very cheap. package maintaining time is not. IMHO
  unused libraries  a couple extra packages do no real harm
 
 It isn't a couple. Once you link against wx, you're looking at ~60
 extra libraries.
 
 I'd suggest that the core GRASS package shouldn't list X, wxWidgets or
 (in 6.x) Python as dependencies.
 
 It shouldn't be necessary to split the actual GRASS distribution into
 GUI and non-GUI components. Put everything in the non-GUI package, and
 have a separate GUI package which is empty (or contains a single dummy
 file if the packaging system doesn't allow empty packages) and exists
 solely to hold the GUI-specific dependencies.
 

As said in another mail of mine, it is not a problem brute-splitting GUI
dependencies (e.g. current squeeze release for 6.4 only *suggests*
wxpython dependencies). Moving GUI-related binaries in different
packages is a pain to maintain, but theoretically it can also be done. 

It remains a brute hack anyway, which is a symptom of
a fundamental design problem: the whole system is theoretically 
layered and modular, but in fact it cannot be really componentized
in a *clean* way. Something I find basically disturbing and not
elegant, sorry my purism. 

BTW, I tend to disagree about considering GUI maintainance not influent
in the releasing cycle. It *is* influent and caused many of the
past/present windows/macos delays. Having a sort of grass-toolbox 
and grass-gui sub-projects could help a lot (like CNES does 
with OTB and Monteverdi).

And like it or not, there are people that do not use the
default GUI or a GUI at all. Splitting would gain consensus
about the core, which is IMVHO the true value of GRASS.

Of course, my 2 cents, as always.

-- 
Francesco P. Lovergine
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-29 Thread Glynn Clements

Francesco P. Lovergine wrote:

 As said in another mail of mine, it is not a problem brute-splitting GUI
 dependencies (e.g. current squeeze release for 6.4 only *suggests*
 wxpython dependencies). Moving GUI-related binaries in different
 packages is a pain to maintain, but theoretically it can also be done. 
 
 It remains a brute hack anyway, which is a symptom of
 a fundamental design problem: the whole system is theoretically 
 layered and modular, but in fact it cannot be really componentized
 in a *clean* way. Something I find basically disturbing and not
 elegant, sorry my purism. 

You are really going to have to explain what you are talking about,
otherwise I (and possibly others) are going to conclude that you don't
have a clue and should just be ignored.

GRASS' structure is about as layered and modular as you can get. If
you install it on a system without GUI libraries, the only portions
which won't work are those which inherently cannot function without
those libraries (i.e. the GUI, digitiser, XDRIVER).

 And like it or not, there are people that do not use the
 default GUI or a GUI at all.

Like me. The only time I'll run the GUI is if I need to test some
aspect of the GUI. Which is why I have no idea where you're getting
these notions about integration from.

-- 
Glynn Clements gl...@gclements.plus.com
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-29 Thread Francesco P. Lovergine
On Mon, Nov 29, 2010 at 10:20:52PM +, Glynn Clements wrote:
 
 Francesco P. Lovergine wrote:
 
  As said in another mail of mine, it is not a problem brute-splitting GUI
  dependencies (e.g. current squeeze release for 6.4 only *suggests*
  wxpython dependencies). Moving GUI-related binaries in different
  packages is a pain to maintain, but theoretically it can also be done. 
  
  It remains a brute hack anyway, which is a symptom of
  a fundamental design problem: the whole system is theoretically 
  layered and modular, but in fact it cannot be really componentized
  in a *clean* way. Something I find basically disturbing and not
  elegant, sorry my purism. 
 
 You are really going to have to explain what you are talking about,
 otherwise I (and possibly others) are going to conclude that you don't
 have a clue and should just be ignored.
 
 GRASS' structure is about as layered and modular as you can get. If
 you install it on a system without GUI libraries, the only portions
 which won't work are those which inherently cannot function without
 those libraries (i.e. the GUI, digitiser, XDRIVER).
 

Something like:

make build-core | build-xdriver | build-wxgui | build-tkgui

make install-core | install-xdriver | install-wxgui | install-tkgui

would be more modular and clean. Of course, one can always now
install the whole beast, use only what is of interest and avoid
installing all dependencies. As said, it works too. 
But I - as packager - would prefer avoiding tons of silly reports 
about 'command X is not working because Y is missing' and installing 
most required dependencies for all commands provided. 

-- 
Francesco P. Lovergine
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-29 Thread Michael Barton
On Nov 29, 2010, at 5:28 PM, grass-dev-requ...@lists.osgeo.org wrote:

 Date: Tue, 30 Nov 2010 00:26:20 +0100
 From: Francesco P. Lovergine fran...@debian.org
 Subject: Re: [GRASS-dev] CLI!=GUI
 To: Glynn Clements gl...@gclements.plus.com
 Cc: Francesco P. Lovergine fran...@debian.org,
grass-dev@lists.osgeo.org
 Message-ID: 20101129232620.ga6...@frankie.is-a-geek.org
 Content-Type: text/plain; charset=us-ascii
 
 On Mon, Nov 29, 2010 at 10:20:52PM +, Glynn Clements wrote:
 
 Francesco P. Lovergine wrote:
 
 As said in another mail of mine, it is not a problem brute-splitting GUI
 dependencies (e.g. current squeeze release for 6.4 only *suggests*
 wxpython dependencies). Moving GUI-related binaries in different
 packages is a pain to maintain, but theoretically it can also be done.
 
 It remains a brute hack anyway, which is a symptom of
 a fundamental design problem: the whole system is theoretically
 layered and modular, but in fact it cannot be really componentized
 in a *clean* way. Something I find basically disturbing and not
 elegant, sorry my purism.
 
 You are really going to have to explain what you are talking about,
 otherwise I (and possibly others) are going to conclude that you don't
 have a clue and should just be ignored.
 
 GRASS' structure is about as layered and modular as you can get. If
 you install it on a system without GUI libraries, the only portions
 which won't work are those which inherently cannot function without
 those libraries (i.e. the GUI, digitiser, XDRIVER).
 
 
 Something like:
 
 make build-core | build-xdriver | build-wxgui | build-tkgui
 
 make install-core | install-xdriver | install-wxgui | install-tkgui
 
 would be more modular and clean. Of course, one can always now
 install the whole beast, use only what is of interest and avoid
 installing all dependencies. As said, it works too.
 But I - as packager - would prefer avoiding tons of silly reports
 about 'command X is not working because Y is missing' and installing
 most required dependencies for all commands provided.

This may work for folks on Linux but is incompatible with the way software is 
installed and used on Mac and Windows. I'm not sure that GRASS will even run on 
Windows without the GUI. Certainly xdrivers will not. 

Michael

 --
 Francesco P. Lovergine

___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-29 Thread Glynn Clements

Francesco P. Lovergine wrote:

 Something like:
 
 make build-core | build-xdriver | build-wxgui | build-tkgui
 
 make install-core | install-xdriver | install-wxgui | install-tkgui
 
 would be more modular and clean. Of course, one can always now
 install the whole beast, use only what is of interest and avoid
 installing all dependencies. As said, it works too. 
 But I - as packager - would prefer avoiding tons of silly reports 
 about 'command X is not working because Y is missing' and installing 
 most required dependencies for all commands provided. 

You can package it how you want, but I don't think that it's realistic
to structure GRASS around individual packagers' policies (e.g. who
says what is core and what isn't).

Every optional dependency can be disabled at configure time. At build
time, you can build individual directories. In 7.0, you can control
the target directory (ARCH_DISTDIR) separately from the directory
where previously built components are found (GISBASE).

Packaging involves work, and GRASS is no different to any other
package. E.g. most modern Linux distributions split libraries between
a run-time package containing the shared libraries and a -devel
package containing the headers, static libraries etc. The source tree
from which the package is built doesn't provide explicit support for
this; it's just something the packager has to do.

Moreover, integrating (and subsequently maintaining) the support for
various packaging systems would be significantly more work for the
GRASS developers than it would take for the package maintainers to
maintain their own packaging systems privately. A privately-maintained
system only needs to work to the level of works for me. Something
intended for inclusion into the source tree has to be more finished,
which can take more work than creating the original version.

-- 
Glynn Clements gl...@gclements.plus.com
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-28 Thread Hamish
Glynn ha scritto:
  If the only available packages (RPM, .deb, etc) insist
  upon installing GUI libraries, complain to the people who
  make the packages.

Paolo:
 OK, so, now I'm complaining ;) : packagers, please have
 your saying here.

disk space is very cheap. package maintaining time is not. IMHO
unused libraries  a couple extra packages do no real harm 
where it does matter (embedded systems) the user will probably
want to custom build from source anyway. I do not much like the
idea of making the packaging scripts any more complicated than
absolutely needed. I do not much like the idea of e.g. the grass
debian package merely recommending the GUI be installed as well,
even if it is technically possible not to require it.


 The rationale is: decoupling CLI and GUI will make the release
 cycle smoother

I humbly suggest that in practice it will only complicate things
and add unneeded burden to the release manager(s).
From a code point of view they are decoupled already.


best,
Hamish



  
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-28 Thread Glynn Clements

Hamish wrote:

   If the only available packages (RPM, .deb, etc) insist
   upon installing GUI libraries, complain to the people who
   make the packages.
 
 Paolo:
  OK, so, now I'm complaining ;) : packagers, please have
  your saying here.
 
 disk space is very cheap. package maintaining time is not. IMHO
 unused libraries  a couple extra packages do no real harm

It isn't a couple. Once you link against wx, you're looking at ~60
extra libraries.

I'd suggest that the core GRASS package shouldn't list X, wxWidgets or
(in 6.x) Python as dependencies.

It shouldn't be necessary to split the actual GRASS distribution into
GUI and non-GUI components. Put everything in the non-GUI package, and
have a separate GUI package which is empty (or contains a single dummy
file if the packaging system doesn't allow empty packages) and exists
solely to hold the GUI-specific dependencies.

[FWIW: I once tried SuSE Linux. It kept wanting to install a good
chunk of KDE because what were essentially command-line tools had been
built with optional (and mostly trivial) GUI features enabled. I
lasted around three days before ditching it for Gentoo.]

-- 
Glynn Clements gl...@gclements.plus.com
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-27 Thread Martin Landa
Hi,

2010/11/27 Markus Neteler nete...@osgeo.org:
 On Sat, Nov 27, 2010 at 6:51 AM, Paolo Cavallini cavall...@faunalia.it 
 wrote:
 E.g., does anybody want to install libwxgtk2.8-0 on a server, where GRASS 
 can be used
 for WPS?

 Likely not, but --without-wxwidgets should help.

it will just disable wxGUI digitizer (wxNviz has been rewritten to python).

We could add new flag --without-gui which would disable compiling (and
installing selected components).

martin

-- 
Martin Landa landa.martin gmail.com * http://geo.fsv.cvut.cz/~landa
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-27 Thread Francesco P. Lovergine
On Sat, Nov 27, 2010 at 11:43:02AM +0100, Martin Landa wrote:
 Hi,
 
 2010/11/27 Markus Neteler nete...@osgeo.org:
  On Sat, Nov 27, 2010 at 6:51 AM, Paolo Cavallini cavall...@faunalia.it 
  wrote:
  E.g., does anybody want to install libwxgtk2.8-0 on a server, where GRASS 
  can be used
  for WPS?
 
  Likely not, but --without-wxwidgets should help.
 
 it will just disable wxGUI digitizer (wxNviz has been rewritten to python).
 
 We could add new flag --without-gui which would disable compiling (and
 installing selected components).
 
 martin
 

That would be the very first step, indeed. Of course IMHO a full solution would
imply also:

- having a componentized GUI which could be added to the core without rebuilding
  the whole beast (possibly componentizing 2d/3d would be also a good idea).
- decoupling gui/core releases: my impression is that the core could evolve much
  fastly than the gui. That's based on counting the current number of GUI 
related bugs 
  in comparison with core ones. 

-- 
Francesco P. Lovergine
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-27 Thread Hamish
for the record I regularly build grass 6.5svn on an old debian/etch
machine which has no wx2.8 avail. ie just for the CLI. It copies
python files into a gui/ dir at install but never uses them. C++
wx modules are not build (cleanly). no problems at all... zero.


as per dual packages, I'd say not necessary, extra work for very
little gain. it would just save a megabyte or two on the install.

the gui development is very good at exposing limitations in the
CLI versions of everything, so it's natural that they both grow
together. and it is already very well separated at the code
level. the only thing that aren't are interactive apps which
are not relevant to the CLI-only crowd.


2c,
Hamish



  
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-27 Thread Paolo Cavallini
Il 27/11/2010 23:40, Hamish ha scritto:
 for the record I regularly build grass 6.5svn on an old debian/etch
 machine which has no wx2.8 avail. ie just for the CLI. It copies
 python files into a gui/ dir at install but never uses them. C++
 wx modules are not build (cleanly). no problems at all... zero.

please think of normal users: believe it or not, people does not compile their
packages, and rely on executables.

 as per dual packages, I'd say not necessary, extra work for very
 little gain. it would just save a megabyte or two on the install.

this is not the point: it will save unnecessary dependencies. That's why in 
sane OSs
packages are split in several independent subpackages.

 the gui development is very good at exposing limitations in the
 CLI versions of everything, so it's natural that they both grow
 together. and it is already very well separated at the code
 level. the only thing that aren't are interactive apps which
 are not relevant to the CLI-only crowd.

if they are well separated, why not separating them, giving more freedom to 
users?
I agree that GUI and CLI will grow together, but why waiting in the release of 
one
part just because the other is still to be fixed?
The rationale is: decoupling CLI and GUI will make the release cycle smoother, 
and
allows a greater freedom, especially for 3rd party applications, either desktop 
or web.
I would like to see GRASS spreading further in the freeGIS arena, and I'm 
suggesting
this could be a way.

All the best.
-- 
Paolo Cavallini: http://www.faunalia.it/pc
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-27 Thread Glynn Clements

Paolo Cavallini wrote:

  Like Martin said, because GRASS is modular, anyone can compile it or use it 
  with
  or without the GUI. I use it heavily with the GUI for some things. For 
  others, I
  use it completely scripted, without any GUI, and called from a completely
  different environment. This kind of flexibility is a real value for this 
  piece of
  software.
 
 AFAIK this is not true: compiling (better: packaging - believe it or not, not
 *anyone* is capable of compiling) without the GUI requires changes
 in the source code.

Such as?

 Sorry to insist, I realize my position is not popular among GRASS devs, but 
 there is
 a number of situations where having a pure CLI package would bring great 
 advantages.
 E.g., does anybody want to install libwxgtk2.8-0 on a server, where GRASS can 
 be used
 for WPS?

If you don't want to install wxWidgets, don't install it. The only
part of GRASS which requires it is the GUI. The rest of GRASS will
work fine without it.

If the only available packages (RPM, .deb, etc) insist upon installing
GUI libraries, complain to the people who make the packages.

The only binaries (executables and libraries) which have a dependency
upon a GUI library are those which simply cannot function without one:
NVIZ, xganim, the OGSF library, etc.

[This isn't necessarily true in 7.0; if you build --with-cairo and
cairo has X support enabled, then the display library will have X as a
direct dependency.]

-- 
Glynn Clements gl...@gclements.plus.com
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-27 Thread Paolo Cavallini
Il 28/11/2010 02:51, Glynn Clements ha scritto:

 If the only available packages (RPM, .deb, etc) insist upon installing
 GUI libraries, complain to the people who make the packages.

OK, so, now I'm complaining ;) : packagers, please have your saying here.
All the best.

-- 
Paolo Cavallini: http://www.faunalia.it/pc
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-26 Thread Paolo Cavallini
Il 21/11/2010 21:36, Michael Barton ha scritto:

 Like Martin said, because GRASS is modular, anyone can compile it or use it 
 with
 or without the GUI. I use it heavily with the GUI for some things. For 
 others, I
 use it completely scripted, without any GUI, and called from a completely
 different environment. This kind of flexibility is a real value for this 
 piece of
 software.

AFAIK this is not true: compiling (better: packaging - believe it or not, not
*anyone* is capable of compiling) without the GUI requires changes in the 
source code.
Sorry to insist, I realize my position is not popular among GRASS devs, but 
there is
a number of situations where having a pure CLI package would bring great 
advantages.
E.g., does anybody want to install libwxgtk2.8-0 on a server, where GRASS can 
be used
for WPS?
Please, help us poor users: separate the two packages, and we'll all be 
happier. I do
not think it's a major work, and packagers can probably give a hand.
All the best.
-- 
Paolo Cavallini: http://www.faunalia.it/pc
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-26 Thread Markus Neteler
On Sat, Nov 27, 2010 at 6:51 AM, Paolo Cavallini cavall...@faunalia.it wrote:
 Il 21/11/2010 21:36, Michael Barton ha scritto:

 Like Martin said, because GRASS is modular, anyone can compile it or use it 
 with
 or without the GUI. I use it heavily with the GUI for some things. For 
 others, I
 use it completely scripted, without any GUI, and called from a completely
 different environment. This kind of flexibility is a real value for this 
 piece of
 software.

 AFAIK this is not true: compiling (better: packaging - believe it or not, not
 *anyone* is capable of compiling) without the GUI requires changes in the 
 source code.
 Sorry to insist, I realize my position is not popular among GRASS devs,

It is moreover a question how many work is involved...

 but there is
 a number of situations where having a pure CLI package would bring great 
 advantages.

I compile GRASS myself quite often as CLI package.

 E.g., does anybody want to install libwxgtk2.8-0 on a server, where GRASS can 
 be used
 for WPS?

Likely not, but --without-wxwidgets should help.

 Please, help us poor users: separate the two packages, and we'll all be
 happier. I do not think it's a major work, and packagers can probably
 give a hand.

A concrete work estimate would be desired.

Markus
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-22 Thread Paolo Cavallini

Il 21/11/2010 21:36, Michael Barton ha scritto:


There is nothing inherently wrong with releasing different parts of
GRASS at different times.But trying to manage a single release cycle
for GRASS has been pretty complicated and my hat is off to Markus.
Trying to manage multiple release cycles would be more complicated.


My suggestion is that decoupling CLI and GUI will make the release cycle 
simpler, not more complicated. We do not need additional complexity.

All the best.
--
http://www.faunalia.it/pc
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-21 Thread Francesco P. Lovergine
On Sat, Nov 20, 2010 at 10:49:05PM +0100, Martin Landa wrote:
 Hi,
 
 2010/11/20 Paolo Cavallini cavall...@faunalia.it:
 
  I know it's an old thread, and that not everybody agrees, but I still 
  think, after
  talking with people more knowledgeable than me, that separating the core of 
  GRASS
  (libraries+CLI) from the GUI(s) will make the release and packaging process 
  faster
  and smoother, and the integration with other software, both desktop and 
  web, easier
  and cleaner.
  Can we revive the discussion about this?
 
 well, my option is very subjective - wxGUI is going to be a solid GUI
 for GRASS. Anyone can build it's own GRASS distribution without any
 GUI (libraries and subset of the GRASS modules). But it's not task for
 the core GRASS developers. They are creating solid environment ---
 GRASS libraries + modules (CLI) and also the GUI. Feel free to take
 what you what from this composition.
 
 Martin
 

What Paolo is probably trying (also?) to say is that GUIs and core could
have completely different release cycles. The GUI should be a component
like any other, to be released when ready, but that should not become
a permanent blocker of the release process, like did happen in the 
6.4 release. Many people (like me and many other) use Grass as a 
scripting language and are completely uninterested in having a stable 
GUI at all, but are interested in having a working stable core in 
reasonable times. Of course, IMHO applies.

-- 
Francesco P. Lovergine
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev


Re: [GRASS-dev] CLI!=GUI

2010-11-20 Thread Martin Landa
Hi,

2010/11/20 Paolo Cavallini cavall...@faunalia.it:

 I know it's an old thread, and that not everybody agrees, but I still think, 
 after
 talking with people more knowledgeable than me, that separating the core of 
 GRASS
 (libraries+CLI) from the GUI(s) will make the release and packaging process 
 faster
 and smoother, and the integration with other software, both desktop and web, 
 easier
 and cleaner.
 Can we revive the discussion about this?

well, my option is very subjective - wxGUI is going to be a solid GUI
for GRASS. Anyone can build it's own GRASS distribution without any
GUI (libraries and subset of the GRASS modules). But it's not task for
the core GRASS developers. They are creating solid environment ---
GRASS libraries + modules (CLI) and also the GUI. Feel free to take
what you what from this composition.

Martin

-- 
Martin Landa landa.martin gmail.com * http://geo.fsv.cvut.cz/~landa
___
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev