Not exactly true Marc, it's just much less developed for mobile support and desktops than you think. Since 1992 it's been specifically for servers, making it robust but not viable to non- enterprises. It's just how it is, but given it's only been 2 1/2 years they've done an unimaginable task. To put this in another way, Linux wasn't even viable except for small tasks until 1998, 7 years, and even then it didn't catch on for non-specialists until around 2003-2005, so I say OpenSolaris is still viable, but it's impossible to expect all the features everyone needs which are consumer and at the time niche. Radius w/ TTLS WPA2 is a niche, believe me or not, outside of universities, only one paranoid person I know actually uses it, so maybe you should be aware of the core guidelines OpenSolaris has to follow and the tasks at hand to make it even semi-usable for non-enthusiasts. Personally I believe the areas are as follows:
Wireless drivers and WPA/WPA2 support for every possible card (No Radius, this is an advanced feature) Desktop experience (Updating everything to use Hal, Dbus, integrate with Solaris device infrastructure, improve performance on X.org and GNOME (JDS), and finally be in sync with the mainstream X.org tree) Documentation for most used and most important API's in every subsystem, as well as documentation for every option available to every single command-line application used by developers and users The rest is really just icing, that alone is impossible to do in less than 2 1/2 years without all Sun employees plus the community working on it. My estimate is that GNU and associated developers, Linux kernel developers, equal roughly 1/2 of the Linux userbase, having multiple installations, at minimal 18000 people working on software for the community for at least the last 5 years. Solaris is lacking in advanced wireless features, bluetooth, power management, and in FOSS software (It really isn't good to depend on sfw, and blastwave was originally made for Solaris 10) For common uses, the desktop experience has much improved, enough for developers to like it, and even some power users. Ethernet drivers and documentation are also top-notch, but again, mobile features, especially advanced ones, are outside the goals the project has at hand, which are to deliver a modern computing experience by making Solaris viable to general power users and to the developers which are the key target to get OpenSolaris viable. James On Nov 26, 2007, at 12:09 AM, Marc Lobelle wrote: > James Cornell wrote: >> It's dependent on which card and driver you have. I don't work >> with the >> driver team, but iwl and atheros will be the first most likely, but >> my >> eta is at least another 3 months. > Hi James, > > I replaced the intel wifi card in my notebook last year with an > atheros > in order to get better solaris wifi support and wpa with pkip and tls > are working fine. However EDUROAM (allowing wifi roaming between > universities) is apparently using ttls, so universities select it > and as > long as solaris does not support it, it makes solaris notebooks not > usable in universities: hard to make it popular in this case. It is a > pity when solaris becomes usable by non-specialists. > > Best regards > > Marc >> >> James >> On Nov 25, 2007, at 2:38 PM, Marc Lobelle wrote: >> >>> I'm using wifi with wpa with PKIP since several months at home. >>> (using an atheros mini-pci card) >>> In my university, there are several wifi networks available but they >>> are using ttls authentication with the radius server. Apparently >>> this >>> ttls is much more common than tls authentication that is currently >>> supported by solaris. >>> >>> Is there any hope to get ttls support with wpa soon ? >>> >>> Thanks >>> >>> Marc >>> >>> >>> This message posted from opensolaris.org >>> _______________________________________________ >>> laptop-discuss mailing list >>> laptop-discuss at opensolaris.org >> >
