On 13 April 2013 15:37, <<a href="mailto:forster@ozonline.com.au" target="_blank">forster@ozonline.com.au</a>> wrote: <div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div class="im">> 1 Android kernel + Ported linux libraries + Sugar > 2 Android kernel + Datastore/Collaboration replacement + Sugar rewritten in > HTML > 3 Full Android + Datastore/Collaboration replacement + Sugar activities > rewritten in HTML > 4 Full Android + Datastore/Collaboration replacement + Sugar activities > rewritten with native Android API </div>Thanks Daniel That explains things for me. I was not fully understanding the technical discussions. With options 1 &2 a lot of existing functionality could be lost: The phone (unless a Sugar dialer was written) The alarm clock (unless a Sugar one written) Skype Some power management controls Airplane mode/wifi/phone modem controls Facial recognition screensaver delay Multiple file selection Excel spreadsheet viewer Anything we need Gnome to do on an XO Lots more A lot of Android devices are going to come into the posession of kids in developing countries, cheap second hand devices, old phones etc. Millions of them. Options 1&2 are not likely to be installed because they will result in a significant loss of functionality. Purchases of new tablets by government education departments with options 1&2 is viable. My guess is that cheap privately owned devices will outnumber education department devices by orders of magnitude. The privately held devices will also be used in a way that is more consistent with Sugar principles, experimentally and playfully, the education department devices may well be locked down. It would be good if Sugar's affordances for playful learning could exist alongside the full Android. I understand that we may not have the resources to do this. </blockquote><div> Hi Tony, I certainly understand this point of view... Most of the features you are mentioning could be implemented also with the 1/2 approach too but certainly we would get there a lot more slowly. And you are right that having to install a custom OS is a barrier than *lots* of people will never cross. I want it to be clear that the preference for 1/2 is just my personal inclination. I know of people that would like to research 3 when they find the time. Also most of the work towards 2 is also useful for 3, one doesn't exclude the other. Actually, in an ideal world, I think we would do both 2 and 3, it's "just" a matter of resources. A part of the work which has been discussed for the next release is useful to both 2 and 3, the other part is integration with the current OS, to not leave it behind. So I don't think we are going to make a call between 2 and 3 soon. </div></div>