Roadmap
|
|
Christoph Baudson (Wheregroup)
|
We don't really have a roadmap. We talked about creating one a lot, but
we never came around to actually doing it. I think it's time we should create a roadmap, follow it, and make it transparent to our users. We should make retrospectives at least every three months to see what we have accomplished. Please allow me to show you a roadmap that is guided by my impressions. (1) Improve visual appearance Mapbender looks just like it did five years ago. Users out there don't know that the technology under the hood has improved tremendously in these five years. It seems like Mapbender is a GIS antique shop from the good old Web 1.0 days. I propose we improve the appearance by these steps (a) rely more on the jQuery library it has a lot of interesting plugins, like tablesorters. It also offers a lot of unobtrusive effects. jQuery UI has a lot more of these plugins we should use, for example use their tabs module instead of our own. By this we remove code from Mapbender and make it leaner and easier to maintain. (b) adopt the jQuery UI style conventions I've written about this in a previous mail. We should use their class definitions and map them to our existing modules. People could create themes with the jQuery UI theme roller. Again, this will also make Mapbender easier to configure, and easier to maintain. (c) a new logo it's already in the make, but it should be finished soon. Could be used in a nice default splash screen. (d) possibly implement the pseudo slippy map. The Mapbender map is considered old-fashioned as well. It could be updated to a pseudo slippy map, a proof of concept exists. The OpenLayers viewer for untiled WMS is not spectacular either. (2) Allow other map clients we should be able to allow other mapping clients, in order to allow tiled maps as well. I see these options (a) map Mapbender API to Openlayers API. By this, the Mapbender and the OL client could be made interchangable. However, the OpenLayers API is huge and the question is where to stop. I assume this would be the hard way. (b) use Mapstraction. The Mapstraction library comes with a code template you have to fill with an adapter to your map client for each API function. The set of functions is not so big, and could be finished in reasonable amount of time. However, not all functionality is covered (like adding map services). But we would also be able to display other maps like Google or Yahoo without the OpenLayers overhead. Which could be nice for smaller apps. I've omitted other important things like update script or installer, basically because I think we should focus on marketing and show what Mapbender is capable of, without hurting the eyes of potential users. I would recommend we create a new folder for modules that have been refactored. By this, we would maintain the current status, and still be free to make huge changes. I think the Mapbender core is quite stable and will not be subject to severe changes. Storing the modules at another location will allow us to address the modules we think most important, and be able to create a stunning new application within the next three months. This new application should be made public and linked to intensively, in order to attract attention to the Mapbender project. I propose we create the roadmap before the next dev sprint, so we can focus on our top prioritiy and not get lost in discussion. We should not plan too much and only address the topics we deem most important in order to gain momentum and make good progress. Please contribute to this discussion. This is just my roadmap and maybe the project would benefit more from other improvements. Let's talk about it and vote before the dev sprint. Have a nice weekend Christoph -- ---------------------------------- Aufwind durch Wissen! Qualifizierte OpenSource-Schulungen bei der www.foss-academy.eu ---------------------------------- _______________________________________ W h e r e G r o u p GmbH & Co. KG Siemensstraße 8 53121 Bonn Germany Christoph Baudson Anwendungsentwickler Fon: +49 (0)228 / 90 90 38 - 15 Fax: +49 (0)228 / 90 90 38 - 11 [hidden email] www.wheregroup.com Amtsgericht Bonn, HRA 6788 _______________________________________ Komplementärin: WhereGroup Verwaltungs GmbH vertreten durch: Olaf Knopp, Peter Stamm _______________________________________ _______________________________________________ Mapbender_dev mailing list [hidden email] http://lists.osgeo.org/mailman/listinfo/mapbender_dev |
||||||||||||||||
Arnulf Christl (aka Seven)
|
Christoph,
thank you for taking this up, it is a very good suggestion. I cannot contribute as much as I would like to yet but I believe that all your points are good and probably also come iin the right priority. I would like to see some REST-oriented ideas come up, for example submitting WMS and WFS service capabilities Links or documents to Mapbender without having to log in and open a web administration interface. Mass-create accounts in the same way and enhance notifications, monitoring. I am pretty unsure what this would involve technical-wise and will unfortunately not be at the next sprint (at least I did not check yet). Good too see this happening. A side question: If something is broken on mapbender.telascience.org/current - should that also be reported using the general Mapbender ticket system? Regards, Arnulf. > We don't really have a roadmap. We talked about creating one a lot, but > we never came around to actually doing it. I think it's time we > should create a roadmap, follow it, and make it transparent to our > users. We should make retrospectives at least every three months to see > what we have accomplished. > > Please allow me to show you a roadmap that is guided by my impressions. > > (1) Improve visual appearance > > Mapbender looks just like it did five years ago. Users out there don't > know that the technology under the hood has improved tremendously in > these five years. It seems like Mapbender is a GIS antique shop from the > good old Web 1.0 days. I propose we improve the appearance by these steps > > (a) rely more on the jQuery library > > it has a lot of interesting plugins, like tablesorters. It also offers a > lot of unobtrusive effects. jQuery UI has a lot more of these plugins we > should use, for example use their tabs module instead of our own. By > this we remove code from Mapbender and make it leaner and easier to > maintain. > > (b) adopt the jQuery UI style conventions > > I've written about this in a previous mail. We should use their class > definitions and map them to our existing modules. People could create > themes with the jQuery UI theme roller. Again, this will also make > Mapbender easier to configure, and easier to maintain. > > (c) a new logo > > it's already in the make, but it should be finished soon. Could be used > in a nice default splash screen. > > (d) possibly implement the pseudo slippy map. The Mapbender map is > considered old-fashioned as well. It could be updated to a pseudo slippy > map, a proof of concept exists. The OpenLayers viewer for untiled WMS is > not spectacular either. > > > (2) Allow other map clients > > we should be able to allow other mapping clients, in order to allow > tiled maps as well. I see these options > > (a) map Mapbender API to Openlayers API. By this, the Mapbender and the > OL client could be made interchangable. However, the OpenLayers API is > huge and the question is where to stop. I assume this would be the hard > way. > > (b) use Mapstraction. The Mapstraction library comes with a code > template you have to fill with an adapter to your map client for each > API function. The set of functions is not so big, and could be finished > in reasonable amount of time. However, not all functionality is covered > (like adding map services). But we would also be able to display other > maps like Google or Yahoo without the OpenLayers overhead. Which could > be nice for smaller apps. > > > > I've omitted other important things like update script or installer, > basically because I think we should focus on marketing and show what > Mapbender is capable of, without hurting the eyes of potential users. > > I would recommend we create a new folder for modules that have been > refactored. By this, we would maintain the current status, and still be > free to make huge changes. I think the Mapbender core is quite stable > and will not be subject to severe changes. > > Storing the modules at another location will allow us to address the > modules we think most important, and be able to create a stunning new > application within the next three months. This new application should be > made public and linked to intensively, in order to attract attention to > the Mapbender project. > > > I propose we create the roadmap before the next dev sprint, so we can > focus on our top prioritiy and not get lost in discussion. We should not > plan too much and only address the topics we deem most important in > order to gain momentum and make good progress. > > Please contribute to this discussion. This is just my roadmap and maybe > the project would benefit more from other improvements. Let's talk about > it and vote before the dev sprint. > > Have a nice weekend > > Christoph > > -- > ---------------------------------- > > Aufwind durch Wissen! > > Qualifizierte OpenSource-Schulungen > bei der www.foss-academy.eu > > ---------------------------------- > > _______________________________________ > > W h e r e G r o u p GmbH & Co. KG > > Siemensstraße 8 > 53121 Bonn > Germany > > Christoph Baudson > Anwendungsentwickler > > Fon: +49 (0)228 / 90 90 38 - 15 > Fax: +49 (0)228 / 90 90 38 - 11 > [hidden email] > www.wheregroup.com > Amtsgericht Bonn, HRA 6788 > _______________________________________ > > Komplementärin: > WhereGroup Verwaltungs GmbH > vertreten durch: > Olaf Knopp, Peter Stamm > _______________________________________ > > _______________________________________________ > Mapbender_dev mailing list > [hidden email] > http://lists.osgeo.org/mailman/listinfo/mapbender_dev > -- Arnulf Christl Exploring Space, Time and Mind http://arnulf.us/ _______________________________________________ Mapbender_dev mailing list [hidden email] http://lists.osgeo.org/mailman/listinfo/mapbender_dev |
Astrid Emde (WhereGroup)
|
Hello devs,
we talked about the roadmap in the last IRC meetings. We agreed on the following points: * we want to create a roadmap, follow it, and make it transparent to our users. * we want to do retrospective every 3 month. We want to connect this meetings to the yearly events (FOSSGIS, Bolsena, INTERGEO, xmas) * we want to set up the roadmap before the next develper sprint (before 12.12.2009) * we want to build up the roadmap in the trac [4] Christoph made suggestions for the roadmap [3]. In the discussion on IRC we added a section on service security (owsproxy, http digest, basic authentication). We will further discuss the roadmap in the next IRC meetings [5]. If you want to join the discussion please send a mail or come to the next IRC meeting. Best regards Astrid [1] http://logs.qgis.org/mapbender/%23mapbender.2009-11-02.log [2] http://logs.qgis.org/mapbender/%23mapbender.2009-11-09.log [3] http://lists.osgeo.org/pipermail/mapbender_dev/2009-October/001868.html [4] http://trac.osgeo.org/mapbender/roadmap [5] http://www.mapbender.org/IRC_Meeting On Fri, October 30, 2009 10:40 pm, Arnulf Christl (Seven) wrote: > Christoph, > thank you for taking this up, it is a very good suggestion. I cannot > contribute as much as I would like to yet but I believe that all your > points are good and probably also come iin the right priority. > > I would like to see some REST-oriented ideas come up, for example > submitting WMS and WFS service capabilities Links or documents to Mapbender > without having to log in and open a web administration interface. > Mass-create accounts in the same way and enhance notifications, > monitoring. > > I am pretty unsure what this would involve technical-wise and will > unfortunately not be at the next sprint (at least I did not check yet). > > Good too see this happening. > > > A side question: If something is broken on > mapbender.telascience.org/current - should that also be reported using the > general Mapbender ticket system? > > > Regards, > Arnulf. > > >> We don't really have a roadmap. We talked about creating one a lot, but >> we never came around to actually doing it. I think it's time >> we should create a roadmap, follow it, and make it transparent to our >> users. We should make retrospectives at least every three months to see >> what we have accomplished. >> >> Please allow me to show you a roadmap that is guided by my impressions. >> >> >> (1) Improve visual appearance >> >> >> Mapbender looks just like it did five years ago. Users out there don't >> know that the technology under the hood has improved tremendously in >> these five years. It seems like Mapbender is a GIS antique shop from >> the good old Web 1.0 days. I propose we improve the appearance by these >> steps >> >> (a) rely more on the jQuery library >> >> >> it has a lot of interesting plugins, like tablesorters. It also offers >> a lot of unobtrusive effects. jQuery UI has a lot more of these plugins >> we should use, for example use their tabs module instead of our own. By >> this we remove code from Mapbender and make it leaner and easier to >> maintain. >> >> (b) adopt the jQuery UI style conventions >> >> >> I've written about this in a previous mail. We should use their class >> definitions and map them to our existing modules. People could create >> themes with the jQuery UI theme roller. Again, this will also make >> Mapbender easier to configure, and easier to maintain. >> >> >> (c) a new logo >> >> >> it's already in the make, but it should be finished soon. Could be used >> in a nice default splash screen. >> >> (d) possibly implement the pseudo slippy map. The Mapbender map is >> considered old-fashioned as well. It could be updated to a pseudo slippy >> map, a proof of concept exists. The OpenLayers viewer for untiled WMS >> is not spectacular either. >> >> >> (2) Allow other map clients >> >> >> we should be able to allow other mapping clients, in order to allow >> tiled maps as well. I see these options >> >> (a) map Mapbender API to Openlayers API. By this, the Mapbender and the >> OL client could be made interchangable. However, the OpenLayers API is >> huge and the question is where to stop. I assume this would be the >> hard way. >> >> (b) use Mapstraction. The Mapstraction library comes with a code >> template you have to fill with an adapter to your map client for each API >> function. The set of functions is not so big, and could be finished in >> reasonable amount of time. However, not all functionality is covered >> (like adding map services). But we would also be able to display other >> maps like Google or Yahoo without the OpenLayers overhead. Which could be >> nice for smaller apps. >> >> >> >> I've omitted other important things like update script or installer, >> basically because I think we should focus on marketing and show what >> Mapbender is capable of, without hurting the eyes of potential users. >> >> >> I would recommend we create a new folder for modules that have been >> refactored. By this, we would maintain the current status, and still be >> free to make huge changes. I think the Mapbender core is quite stable >> and will not be subject to severe changes. >> >> Storing the modules at another location will allow us to address the >> modules we think most important, and be able to create a stunning new >> application within the next three months. This new application should >> be made public and linked to intensively, in order to attract attention >> to the Mapbender project. >> >> >> I propose we create the roadmap before the next dev sprint, so we can >> focus on our top prioritiy and not get lost in discussion. We should not >> plan too much and only address the topics we deem most important in >> order to gain momentum and make good progress. >> >> Please contribute to this discussion. This is just my roadmap and maybe >> the project would benefit more from other improvements. Let's talk >> about it and vote before the dev sprint. >> >> Have a nice weekend >> >> >> Christoph >> >> >> -- >> ---------------------------------- >> >> >> Aufwind durch Wissen! >> >> >> Qualifizierte OpenSource-Schulungen >> bei der www.foss-academy.eu >> >> ---------------------------------- >> >> >> _______________________________________ >> >> >> W h e r e G r o u p GmbH & Co. KG >> >> >> SiemensstraÃe 8 >> 53121 Bonn >> Germany >> >> >> Christoph Baudson >> Anwendungsentwickler >> >> >> Fon: +49 (0)228 / 90 90 38 - 15 >> Fax: +49 (0)228 / 90 90 38 - 11 >> [hidden email] www.wheregroup.com Amtsgericht Bonn, HRA >> 6788 >> _______________________________________ >> >> >> Komplementärin: >> WhereGroup Verwaltungs GmbH >> vertreten durch: Olaf Knopp, Peter Stamm >> _______________________________________ >> >> >> _______________________________________________ >> Mapbender_dev mailing list >> [hidden email] >> http://lists.osgeo.org/mailman/listinfo/mapbender_dev >> >> > > > -- > Arnulf Christl > > > Exploring Space, Time and Mind > http://arnulf.us/ > > > _______________________________________________ > Mapbender_dev mailing list > [hidden email] > http://lists.osgeo.org/mailman/listinfo/mapbender_dev > > _______________________________________________ Mapbender_dev mailing list [hidden email] http://lists.osgeo.org/mailman/listinfo/mapbender_dev |
||||||||||||||||
| Free Embeddable Forum Powered by Nabble | Help |
