[Mapbender-dev] road map Q2

Arnulf Christl (Seven) seven at arnulf.us
Mon Mar 8 09:33:16 EST 2010


Christoph,
this roadmap looks great. Thanks for taking the lead yet again and
preparing this document.

Just as an explanation from my perspective: We really need the
slimming-process and it will *not* mean that Mapbender will be less
powerful after we are done with it. Instead, we will be able to
concentrate better on the core capabilities of the Mapbender-project, as
he have specified it in the Typification [1] At least this is what the
"slimming" means from my perspective. But this is an open project, so if
you have objections or want to add another emphasis to the project, please
speak up.

When it comes to doing things you will not see any code from me. But I
commit myself to help sort out some issues that we have with the backend
administration, potentially find some funding, and hopefully will produce
some human-digestible documentation (the prose that goes beyond the
automated lyrics).

Best regards,
Arnulf.


> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi friends of Mapbender,
>
> it's time to agree on a road map for Q2. We have already decided some
> key points at FOSSGIS 2010
>
> * release date (June, end of Q2)
> * version number (3.0)
>
> now we should decide what will be part of 3.0 and what not. I think by
> choosing 3.0 we have already decided that only the 3.0 contents will be
> guaranteed to be stable, older application elements from 2.x might no
> longer work. Please make yourself heard if you have another opinion!
>
> As I have said at the code sprint I feel we need more momentum in the
> project. I believe we can only get there if we slim Mapbender
> ("gesundschrumpfen") to a size that the existing team can manage, and
> then use the momentum of the OSGeo stack (like OpenLayers, GeoNetwork,
> etc) to grow again, sustainably. I think Q2 will deal almost exclusively
> with the slimming process.
>
> I had already proposed a road map on Feb 22nd, and I still think it's a
> good starting point. Having said that, I strongly feel that we need to
> focus even more on individual aspects in order to get the release done.
>
> Now, I would propose this road map, ordered by priority
>
> (a) automated testing
> - - create concept
> - - prototypical implementation
> - - build script which automatically runs all JS and PHP tests
> - - write a tutorial how to write tests
> - - write tests for code you touch or create (and make it part of
> Mapbender "Done")
>
> (b) administration applications
> - - re-evaluate Mapbender's administration concept (users, services,
> applications, authorization), document it
> - - create a coding concept, document it
> - - refactor old administration modules (or more likely, rewrite from
> scratch)
> - - create a new administration application (wrt the Mapbender "Done") and
> remove the old ones
> - - write tutorials
>
> (c) map applications
> - -  finishing touches on Mapbender core (remove obsolete code, API fine
> tuning, nothing spectacular but should be done)
> - - decide which template applications will be part of 3.0
> - - decide which application elements will be included, which will be
> deprecated
> - - create concept for deprecated application elements
> - - refactor or rewrite application elements (wrt the Mapbender "Done")
> and build template applications
>
>
> Our road to 3.0 should be very transparent. We should use the
> telascience installation as a demo platform where the community can
> monitor our progress and make suggestions as we move along. We should
> have a small review and retrospective every Monday at the IRC meeting to
> see what we have achieved and what we want to do next. We should send
> updates to the mailing lists regularly to get community input early and
> to keep each other motivated.
>
> We should keep in mind that we have no official sponsors yet for our
> endeavour. We should ask FOSSGIS e.V. if they want to sponsor our
> community events. I will also ask WhereGroup if they are willing to
> sponsor us, maybe by allowing their employees to spend some hours on
> Mapbender 3.0. Google could be a sponsor, in the course of GSoC 2010.
> Arnulf (who generously wants to sponsor the logo creation) also
> mentioned he will be looking for sponsors. Please try to find other
> sponsors, or if you are willing to sponsor, let us know.
>
> Let's discuss this proposal or suggest a different road map. We should
> not lose time, please reply this week so we can vote on the road map
> next Monday (March 15th) and then commit to working on it.
>
> I think FOSSGIS proved that Mapbender can rely on a great team, and I'm
> optimistic that we can pull this off, if we keep each other motivated
> and collaborate intensely. Thanks for your support!
>
> Christoph
>
>
> Christoph Baudson (Wheregroup) wrote:
>> With FOSSGIS being just around the corner, our first milestone of the
>> year is almost reached. It's time to think about what we want to do in
>> the next quarter.
>>
>> In my opinion, the next quarter should be used for consolidation of
>> existing functionality. We should focus on this
>>
>> (1) refactor existing modules (admin and map apps), and document them
>>
>> * All modules should be compliant to the Mapbender JavaScript API and
>> implement the jQuery UI CSS framework. Modules that do not match these
>> criteria will not be included in the next version.
>>
>> * Some modules should be merged with another, and obsolete modules
>> should be removed (especially WFS modules). It should be transparent to
>> an admin what each module does, and how it relates to other modules. We
>> should agree upon a minimal set of modules, and mark others as untested.
>>
>> * the admin modules should be implemented following kmq's approach from
>> GSoC 2009. The work flow of some components should be re-evaluated and
>> maybe refactored (again, especially WFS modules)
>>
>>
>> (2) release new version and make it stable
>>
>> * integrate Geoportal-RLP components first
>>
>> * release stable version in summer, like mid-june (and maybe call it
>> 3.0, and skip 2.7)
>>
>> * the release should be done with a group in the Linux hotel
>>
>>
>> (3) create an infrastructure for automated testing
>>
>> * concept and prototypical implementation of tests
>>
>> * build script which automatically runs all JS and PHP tests
>>
>>
>>
>> I have thought about it for quite some time, and maybe it's best to
>> really focus on a release, and have a consistent and coherent product.
>> There is no conference or fair in summer so maybe this is the only time
>> of the year where there's no pressure to come up with new stuff.
>>
>> (If we were to focus on consolidation, we would even more need someone
>> to sponsor us. I will invite WhereGroup, our primary sponsor so far, to
>> the IRC meeting and maybe we can discuss how our road map is consistent
>> with theirs. Maybe the people employed at WhereGroup could do some of
>> these tasks during their regular hours, which would be a great way to
>> sponsor the project. Maybe we could in return list WhereGroup as our
>> sponsor at the Mapbender web site. Another sponsor could be Google, if
>> we were to be given a slot in GSoC 2010)
>>
>> With a solid new version, we would again be open to new additions. For
>> Q3, I would like to address some interesting stuff for FOSS4G like
>>
>> * GeoNetwork interface
>> * GeoExt support
>> * Mapstraction support or full OpenLayers support
>>
>> Q3 could be kicked off in Bolsena and last until FOSS4G.
>>
>> Please comment or discuss later at the IRC meeting,
>>
>> Christoph
>>
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.9 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
>
> iEYEARECAAYFAkuTyDIACgkQGtMIfbycMX6h7gCaAqkOjqltwOkCYNjhaZrLLU1G
> AtMAn2UKoOrzWfEmbfJ11p84g6iTfBSS
> =O+KW
> -----END PGP SIGNATURE-----
> _______________________________________________
> Mapbender_dev mailing list
> Mapbender_dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/mapbender_dev
>


-- 
Exploring Space, Time and Mind
http://arnulf.us/



More information about the Mapbender_dev mailing list