[GRASSweb-list]bernhard: web related_projects.html,1.22,1.23

Bernhard Reiter bernhard at intevation.de
Thu Aug 7 07:17:39 EDT 2003


On Thu, Aug 07, 2003 at 12:49:13PM +0200, Markus Neteler wrote:
> On Thu, Aug 07, 2003 at 12:34:05PM +0200, Bernhard Reiter wrote:
> > On Thu, Aug 07, 2003 at 12:17:42PM +0200, Markus Neteler wrote:
> > > On Thu, Aug 07, 2003 at 11:34:05AM +0200, Bernhard Reiter wrote:
> > > > On Thu, Aug 07, 2003 at 11:14:14AM +0200, Markus Neteler wrote:
> > > > > On Wed, Aug 06, 2003 at 07:08:40PM +0200, grass at intevation.de wrote:
> > > > > > Author: bernhard
> > > > > > 
> > > > > > Update of /grassrepository/web
> > > > > > In directory doto:/tmp/cvs-serv5364
> > > > > > 
> > > > > > Modified Files:
> > > > > > 	related_projects.html 
> > > > > > Log Message:
> > > > > > Added hint that lx-viewer depends on a non-free library.
> > > > > 
> > > > > A comment:
> > > > > 
> > > > > In this case the licenses of all other products should be
> > > > > added. Or (preferred by me), we don't mention the licenses
> > > > > here as they are already mentioned on the individual
> > > > > project pages - to reduce our page maintenance efforts.
> > > > 
> > > > My proposal is to only warn users against licenses that are non-free.
> > > > GRASS is a Free Software project that is why most users will
> > > > expect us to be aware of non-free solutions.
> > > > This is really has value to our readers.
> > > > And we also make sure that if people decide to use that software
> > > > they are warned. 
> > > > Friends don't let friends use drugs^H^H^H^H^Hproprietary software.  ;)
> > > 
> > > Well, I am not sure if it is really our task to explain the
> > > external projects. For that 'www.freegis.org' exists.
> > 
> > Not it isn't, but in that case we should not link it.
> > In some situation it is very useful, because the user
> > is looking for solutions and is goal driven.
> > 
> > > We should not spend time on that, but invest more time
> > > in improving GRASS (write an 'installer', write an
> > > 'update' more like R-stats provides etc). There are many
> > > more important things to do. Or just make GRASS 5.1
> > > working on MacOSX and IRIX?
> > 
> > Yes I agree.
> > However it was a couple of minutes action
> > and I really consider it important.
> 
> But it is incomplete. Either rate all links or no link.
> I suggest not to rate as this is not the goal of the GRASS
> web site.

It is complete, because of the context as explained above.
Every link should lead to mainly Free Software 
unless an explicit warning is given.
This is the case for ltplus and lx-viewer.

I can add what is Free Software here to the others, if you insist.

> > On the strategy level Free Software looses a lot 
> > because potential users and developers are not properly informed.
> > (Just think of the damage the SCO smearcampaign did,
> > it would have been much more difficult if people had used GNU/Linux
> > instead of the missleading "Linux" for the operating system earlier.)
> > Anyway to leads to far.
> > 
> > > > Otherwise we also might need to link all the proprietary 
> > > > or semi-proprietary software that can do useful tasks
> > > > in combination with GRASS.
> > > 
> > > Yes, installation should be simplified.
> > 
> > I don't understand that remark.
> 
> Sorry: I wanted to point out that there are much more important
> things to do than spending time on rating the licenses of
> external links (remember the "polish the GRASS source code header
> file inclusion" issue we once talked about).
> First we should make GRASS easiy to use, then doing such extra
> work.

Now I understand it.

Naturally I disgree because of first hand experience
what bad information can cause. Missinforming the users is worse 
than having GRASS being harder to set up.

> > My argument was: Linking to usefuil applications which are proprietary 
> > from the GRASS pages should be an exception, 
> > because we don't want to link them all.
> 
> But "lx-viewer" is not proprietary. It is "GPL with exception".
> However, this discussion is not related to a GRASS web page (IMHO).

It _must_ have that proprietary library,
thus it is a clear path that is leadind to non-free software
and away from a long term benefit for GRASS.
Users need to be warned
I consider it irresponsible to not warn them.

> [this is off-topic: how to look at DXF/DWG data? Except "lx-viewer"
>  there is *nothing* working. There are not yet always Free Software
> solutions available, unfortunately.]

I won't go along with that all or nothing evaluation,
but as you said that is off topic. 

> > > Why not let the users decide? If someone want to use a non-free
> > > solution, that's not our decision. I would prefer to give the 
> > > users the freedom to decide themselves.
> > 
> > Yes, I completely agree,
> > that is why it makes sense to add a warning in the rare
> > case that we link a proprietary solution.
> 
> Again, "lx-viewer" is not proprietary. 

Correct, but still that warning is needed, see above.

> But thinking more generally,
> you would have to check all licenses of all linked software
> packages to make useful recommendations to the web users.

I basically did that, by in many cases having a best effort
relying on other credible work (like Debian, FSF or FreeGIS).
If I get aware of such a licensing trap leading to non-free software,
I'll add a warning for it.

> > Users have been warned and then can make an educated decision.
> 
> They can just read the related web pages were the licenses are
> also stated. Let us keep the GRASS site concentrating on GRASS.
> Educating the users about external projects is not our task
> (FSF can do that, FreeGIS, OSI, ...).

The problem with links to solutions that depend on proprietary software 
do not have an interest to explain it to users. We had to push strongly
with Lx-Viewer to remotely do this. Ltplus does not react to it
and keeps its readers in the dark.

If we link to them, we also implicitely give a recommendation
and they get a bit of our credit. So to protect our credit,
we have to warn our readers, if we put up the link.

> Well, enough said :-)

Yes, as long as some people in the GRASS project (like me)
find it important and do the work, it only benefits GRASS' users.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.osgeo.org/pipermail/grass-web/attachments/20030807/6d1f9188/attachment.bin


More information about the grass-web mailing list