On Sat, Jan 17, 2015 at 7:28 PM, Reza Naghibi < [email protected]> wrote:
> Do you have any feedback for 2.0 [0]? So we are doing a few things: > > -partitioning detection into device, browser, and os. No explanation here, > should be clear why this is needed. > > This has always been the case in existing data, BrowserDataSource and OperatingSystemDataSource were not fully used by even the W3C compliant API, but all of its definitions there already. So we should make proper use of such separation, can't "invent" it, since that was pretty well done by W3C DDR already;-) > -creating a specification for data and apis to interact [1]. Similar to > the above partitioning, this will allow for data and code to evolve > independent of each other. This is already happening with the potential > introduction of new classifier clients, like your ruby client...! > > 2.0 is still in its very early stages, so feedback is important. > > The reference client will be a clear guide to 2.0 implementation. This > will be coming in the near future. > > [0] http://wiki.apache.org/devicemap/DeviceData2 > > [1] http://wiki.apache.org/devicemap/DataSpec2 > > > Still have to look into this in more detail, ideally examples could be closer to the domain like "Series 6" or similar rather than "I am 47 years old". Switching to Git for the 2.x artifacts would be worth a thought btw. We may leave existing SVN structure but start using Git for some of the new stuff. WDYT? Werner
