On 11/8/2007 9:37 AM, Shawn Walker wrote: > On 08/11/2007, Marty Duey <[EMAIL PROTECTED]> wrote: >> On 11/7/2007 10:49 PM, Shawn Walker wrote: >>> On 07/11/2007, Moinak Ghosh <[EMAIL PROTECTED]> wrote: >>>> It is because the miniroot is 146M in size which on a 256M >>>> RAM system does not leave enough memory to run GNOME. >>>> A smaller miniroot combined with an alternative lighter >>>> weight WM like Xfce will be needed. >>> Actually, I think it would be possible to lower this requirement to as >>> little as 64 to 128mb for the text installer if we actually produced a >>> miniroot specifically designed for "headless systems". Mind you I >>> don't think a system with less than 512mb is going to be terribly >>> usable (at least w/ a graphical desktop of any sort). >>> >> The One Laptop per Child (OLPD) program championed by MIT is targeting a >> low cost laptop (less than $200 and moving toward $100) that will could >> potentially be put into the hands of millions of people in developing >> countries. The memory spec for the system is 256MB (see >> http://laptop.org/laptop/hardware/specs.shtml). >> >> Granted, the demographic OLPD is targeting is not necessarily a primary >> target of an OpenSolaris distribution, but it is a data point on that >> should be considered for the miniroot/WM combination. > > Yes, but it is important to remember that that is a very specialised > target. That's practically an embedded target in my mind. > > It is not logical (to me) to try to target "average desktops, laptops" > and at the same time an extremely specialised target like OLPC. > > Not only that, the OLPC has a significantly reduced functionality set > compared to an "average" system. > > I believe that such targets require special consideration that is > beyond the scope of our primary ones. > Shawn, I agree that the target audience is "special" (if not specialized). Perhaps you're more versed in the design parameters of OLPD than me, but from what I read one of the principles is not to limit the users ability to "...reshape, reinvent, and reapply their software, hardware, and content."
As for "significantly reduced functionalities" that depends on your perspective. No, it's not going to match up to a $1000 laptop in many ways, but in others (wifi, power) it will equal or surpass such. People can look at the list of source (http://dev.laptop.org/git) and the specs (above) and decide for themselves the level of functionality. Again, I'm not saying that school children in Ethiopia are anywhere on the list of target audiences, but the 256MB design point shouldn't be dismissed because of that. That's all. Marty _______________________________________________ indiana-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/indiana-discuss
