I've been watching this list with great interest, coming from a few years background with OpenSolaris.<div><br></div><div>I wanted to make a very high-level suggestion, as a non-contributor (so take this with as many grains of salt as you like): I think Illumos should be scoped as small as is reasonable at first.</div>
<div><br></div><div>Or, put another way, let's not bite off more than we can chew.</div><div><br></div><div>Or, a third way, take the iPod approach.</div><div><br></div><div>What do I mean by that? I'm not arguing that the iPod is the greatest music player, but it's certainly the most successful. When it was first released, it had very few features (play, pause, shuffle, sync). Now it has the ability to play video, a built-in browser and has even become a gaming platform.</div>
<div><br></div><div>I think the lesson in the device's success is that it's best to start with just the things that you can do very well, and expand after you get an audience. In the case of the iPod, this meant simple-to-use software/hardware interaction. In the case of Illumos, this might mean stable, mature core tools (zfs, dtrace, network stack) and a really easy build process.</div>
<div><br></div><div>I guess I'm just encouraging focus on the core that makes Solaris great, rather than trying to compete with bells-and-whistles Linux distros or even Oracle Solaris. And, above all, lowering barriers to adoption of this core (drivers, build process, etc).</div>
<div><br></div><div>best,</div><div>Blake</div>