Closing the gap
Getting Solaris 10 installed onto a large user base will require more development. It's a bit of a chicken/egg situation: you need a large user base to develop stuff, but you need stuff to attract users in the first place. If the company only wants to work with OEMs and see pre-loaded X-86 boxes shipped, then it's already in pretty good shape. But if it wants to see a community develop and flourish, it will have to take the user experience and desktop functions much more seriously.
First in line is, of course, device drivers. I would imagine that this is Sun's top priority.
Second would be multimedia: let's face it, even in a work environment, people play with their computers. They've got to be able to burn a disk, play a song, watch a video clip. There is a ton of open source stuff available, and we hope to see some of it turning up in Solaris 10 very soon.
Support for other file systems, even if it's read-only in some cases, would be nice.
Online updates should be free for non-commercial users. If you want people to stick with a product, especially early in its development, you can't have them worrying that some security hole or bug they haven't heard of has left them open to remote exploitation, or susceptible to some fatal error that might wipe out months of work. (You want me to trust your stuff? Then don't leave me guessing.)
The Solaris Management Console needs to be extended. With YaST, you can configure devices, add and remove software packages, perform an online update, track dependencies, edit your boot config, start and stop services and daemons and assign their runlevels, edit the /etc config files, configure networks, partition disks, plus just about everything the SMC provides, except for the real-time process interface, for which you can substitute the command ps.
KDE is certainly more popular than Gnome among Linux users, and most would agree that it's by far the better of the two desktops. It has more tools, it has better tools, and it's almost infinitely customizable. Adding a full version of KDE 3.4.x and including it on the boot menu as a desktop option would make the lives of many Linux users easier, and Solaris a good deal more attractive to them.
People develop software for many reasons, chief among them the fact that they're paid to. But a certain amount of development comes from hobbyists who have a system that they like, and want to make better, or want to make do something it doesn't already do. To attract the user base and developer interest that will really propel Solaris 10 forward, Sun would do well to think about it as a PC as well as a workstation. Generating enthusiasm and attracting a broad base of developers does involve giving people some fun in return, after all. Making SuSE Pro a fun distro and an excellent PC doesn't make it any less of a workstation, server platform, or development environment, a fact apparently lost on Red Hat. Sun should take a careful look at the cultural differences evident in Novell's and Red Hat's distros, and the differences in user and developer enthusiasm, and draw the obvious conclusion.
We've had fun with Solaris 10. It's got virtues that we definitely admire. What it needs to compete with Linux will be easier to bring about than what it's already got. It could become a Linux killer, or at least a serious competitor on Linux's turf. The only question is whether Sun has the will to see it through. ®
Note: The licensing scheme is another issue with implications for the future of Solaris 10, but that's a topic for a separate, forthcoming article.
Sun coughs up an OpenSolaris kernel
Sun plays hide and seek with key Solaris 10 goodies
Sun suffers self-imposed Linux lobotomy
VMware embraces Sun and even Solaris x86
Sun gets liquored up on own code
Sun slams Red Hat