Hi Andrea,
WMS allows extensions and GetPrint is one of them. BTW: GetPrint is what
many organizations find quite attractive. If we improve GetPrint to also
allow to generate data-sheets and reports by using Atlas functionality
it will be even more powerful. So I'd rather see GetPrint continuously
improved than removed.
It is true that in some cases we need to improve to be more standards
compliant. This is a group effort. Everybody can contribute. Removing
GetPrint from GetCapabilities, or implementing in a standard compliant
way, would help. We could still have it listed in GetProjectSettings.
Your other claims, that QGIS server is not ready for real deployments is
wrong. There are several cities and provinces using QGIS server in
production mode, including my own organization for 3-4 years now. Speed
is comparable to other WMS - some times slower, some times faster. You
can try other Server at http://gis.uster.ch/ and will see that it isn't
slower than other WMS. On the contrary - people often mention that it is
quite fast.
QGIS server is lacking some features you listed (including proper SLD
support, multiple styles, name groups, etc.), but some of the features
you listed are already there (e.g. the hidden layers (you can exclude
layers in the OWS configuration tab)). In addition it supports stuff
that other WMS servers don't support.
QGIS server is and will be what the users want it to be - so everybody
can shape and influence where it should go. With development work or by
hiring developers to implement missing features. Having said that, and
citing Vincent, you cannot just remove stuff that others rely on.
Just ranting about things does not help.
Andreas
Am 2014-06-10 22:38, schrieb aperi2007:
Il 10/06/2014 18:09, Vincent Picavet ha scritto:
We rely on this command for some applications, just removing it is not
an
option for us right now.
People should stop thinking their use case is the only one and if they
shout
louder their desiderata will be executed.
Talking about norms compliancy and passing OGC CITE tests is
important, but
should be done rationnaly.
In this case, backward compatibility should be taken into account as
well.
Vincent
Il 10/06/2014 18:09, Vincent Picavet ha scritto:
We rely on this command for some applications, just removing it is not
an
option for us right now.
People should stop thinking their use case is the only one and if they
shout
louder their desiderata will be executed.
Talking about norms compliancy and passing OGC CITE tests is
important, but
should be done rationnaly.
In this case, backward compatibility should be taken into account as
well.
Vincent
Hi Vincent
I guess your point of view is respectable.
You like to protect your work.
This is really a good point of view.
But it is exactly why I'm really unfair for this story of wms
complaints.
Infact the problem is exactly when someone put a bad and incompatible
solution inside the qgis-server and other start to use it.
Invest on it.
After this is hard to remove it because
they say:
"hey! my work is important for me. The backward should be taken in
account".
This is why I don't like this kind of situation.
And is why I fork my qgis-server to have qgis-server compliant with
the need of ou public-administration.
If this need to have a different product in the future rather than
qgis original. I don't know actually.
I hope no of-course. But I really don 't know now.
To response to you point of view:
I can only say that we
are fund some plugin.at 1.8 level.
With the break api of the 2.0 we must to refund the repair of plugins.
With the 2.2 the plugin stop again to work, and with the 2.4 they
still don't work.
And we are planning to refund again the repair when the 2.4 will go
out.
And more probably with the 2.6 we will are again to fund for them.
I guess we could say as you:
"hey our work is imortant for us. The backward was taken in account".
So stop to change again and again.
Don't ?
So what we should do .
Hibernate all ?
QGIS Desktop is a good software .
QGIS-Server is no so good software. Is too early software.
It need really many investment to reach the level of other software
Gfoss solutions in WMS world.
Of course if an user need only a map-dispenser where there is only 1
user every minute or more surely could use the qgis-server
is more simple and fast to developer (perhaps)
But My experience say that the user need growth rapidly and
qgis-server is not capable to offer more than to usual solutions.
Is surely too few for the actual modern need of a WMS.
No hidden layer, No named groups, No multistyles, No html templating
for response, No GML response in wms, etc....
All need that a modern GIS system ask to offer a full experience to the
user.
You guess the wysiwyg capability of qgis is a "must have" in wms
systems ?
I guess yes but only if this mean Have also all the functionality in a
WMS and only if compatible.
Otherwise I prefer other solution.
Because the wms point of view is
one do the work and all the other use it
So if the one must write a mapfile or an sld file is no so bad.
The most important is what see and use the remote user.
The functionality I list over and that qgis as not, are really more
important of the wysiwyg capabiity of qgis-server.
QGIS has no one of all them ... but has "GetPrint".
I say "ok, sorry I forgto to say qgis-server is also incopatible with
all the others".
:))
The shortcut solution is not always the best.
In this case the shortcut is also the worst.
Because it became a blocking for new evolutions.
:)
Your point of view is the confirm of my disagree.
Regards,
Andrea.
_______________________________________________
Qgis-user mailing list
Qgis-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-user
_______________________________________________
Qgis-user mailing list
Qgis-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-user
