Hi! I sent the attached email last week to the mapcss mailinglist. Unfortunately I didn't get any answers, which is sad as I think that MapCSS could need some development. Therefore I send this email to the osm-dev mailinglist too, maybe there are some interested persons.
In the meantime I copied the MapCSS specification to a repository on Github (which was fairly easy, as Github can render Mediawiki files): https://github.com/plepe/mapcss What do you think, anyone interested in improving MapCSS? greetings, Stephan ----- Forwarded message from Stephan Bösch-Plepelits <[email protected]> ----- Date: Fri, 23 Aug 2013 09:31:14 +0200 From: Stephan Bösch-Plepelits <[email protected]> To: [email protected] Subject: [Mapcss] Decision procedure for MapCSS Hi! As you might have read, I'm developer of a new MapCSS implementation, PGMapCSS: https://github.com/plepe/pgmapcss/ I think that MapCSS has huge potential, as it is more general than other definition languages like Cascadenik or CartoCSS. I've used Cascadenik long enough to see its downsides (like complex database queries; no calculations). My dream is that I can use the same map style on different devices: A web site with a TMS/WMS based map; mobile devices with interactivity ; printed maps; editors for OpenStreetMap data. Why not, if there are MapCSS implementations for all these different devices. On the other hand, MapCSS 0.2 is still very callow. Some definitions are vague and important features are missing. The Wiki history shows, that development has been very much in the hands of individuals. The conclusion is the observable incompatibility between the different MapCSS implementations. I haven't found a procedure how new features are proposed, discussed and finally decided. I also read the mailinglist thread from March 2013 with no real outcome: http://lists.openstreetmap.org/pipermail/mapcss/2013-March/thread.html I thought about proposing stuff on the Wiki Talk page, but as it is mostly deserted since 2011 I don't have much hope for feedback. I also don't want to change the specification myself - there should be some kind of decision process, don't you think? I don't want to whine. I think we need a strategy how to overcome this solution. My proposal: * We move the MapCSS specification to a GitHub repository, preferably "openstreetmap/mapcss". * If someone wants to propose a new feature, he/she creates a pull request, which will be merged by a maintainer after (say) a fortnight if no objections are raised. * We can use the issue tracker to raise ideas and discuss details. * We create a couple of reference styles for testing purposes, which are part of the repository. * We finalize MapCSS 0.2 and allow only clarifications. * We create a branch MapCSS 0.3 which will hopefully harmonize the incompatibilities. * We document this decision process somewhere[tm]. There a several examples where specifications are developed on Github, e.g.: * https://github.com/mojombo/semver * https://github.com/mustache/spec * https://github.com/perl6/specs Maybe we can take an example on them. What do you think? I'd love to see MapCSS develop ... greetings, Stephan PS: If we decide to move to Github, I offer to take care of the transition process. _______________________________________________ Mapcss mailing list [email protected] http://lists.openstreetmap.org/listinfo/mapcss ----- End forwarded message ----- -- Seid unbequem, seid Sand, nicht Öl im Getriebe der Welt! - Günther Eich ,---------------------------------------------------------------------. | Stephan Bösch-Plepelits, | | Technische Universität Wien - Studien Informatik & Raumplanung | | Projects: | | > openstreetbrowser.org > couchsurfing.org > tubasis.at > bl.mud.at | | Contact: | | > Mail: [email protected] > Blog: plepe.at | | > Twitter: twitter.com/plepe > Jabber: [email protected] | `---------------------------------------------------------------------' _______________________________________________ dev mailing list [email protected] http://lists.openstreetmap.org/listinfo/dev
