On Thu, Jul 15, 2010 at 10:53 AM, Garrett D'Amore <garr...@nexenta.com> wrote: > The *code* is probably not going away (even updates to the kernel). > Even if the community dies, is killed, or commits OGB induced suicide.
1. You used correct word: "probably". 2. No community = stale outdated code. > There is another piece I'll add: even if Oracle were to stop releasing > ZFS or OpenSolaris source code, there are enough of us with a vested > interest (commercial!) in its future that we would continue to develop > it outside of Oracle. It won't just go stagnant and die. So you're saying "let's fork it". Let's imagine some red-eyed zealots decided to do so and did that. They have a shiny new Mercurial repo. Now what? Yet another very dead GNU/Hurd? Let's think through: to fork is to hope for the new product will take off and will be popular in the hackerdom, so the geeks can make new stuff for that, use it, fix it, build knowledge base how to fix foobar when it happens, some best practices etc. The hackerdom IS the place where new real specialists are made. So if Solaris will be not any free, then nobody gives a shell about this OS and it will be as "popular" as AIX from IBM, where you simply can not find any specialists that could support it. Why? Because nobody knows AIX and does not want to know that. There is no much enhancements to AIX other than done by IBM in their Frankenstein's way. But hey, why to fork ZFS and mess with a stale Solaris code, if the entire future of Solaris is a closed proprietary payware anyway? And opposite to ZFS, we have totally free BTRFS that has been moved to the kernel.org and is *free* and is for Linux that is *already* popular AND *free*? Yes, Linux is not the best OS, if you compare to Solaris in some technical parts that would make things just more sophisticated. But on the other hand Linux is totally free, cheap and you can live with these inconveniences perfectly (just drink more water and breath more deeply). You can curse these inconveniences, but at the end it still works cheap and reliable and is just OK to get things done. Well, BTRFS sucks at some points (software RAID at kernel level comes to mind), but it is still better FS for Linux in many places than extN, but it is still free and more popular. Maybe today BTRFS is not the right answer as ZFS is to the market, but tomorrow it probably will be just as opposite, I think: geeks will use BTRFS and Linux and soon Oracle will deeply regret they're killed Solaris, but no one will throw their energy to make Solaris at least as strong as Linux is now. > I believe I can safely say that Nexenta is committed to the continued >development and enhancement of this code base -- and to doing so in the open. Yeah, and Nexenta is also committed to backport newest updates from 140 and younger builds just back to snv_134. So I can imagine that soon new OS from Nexenta will be called "Super Nexenta Version 134". :-) Currently from what I see, I think Nexenta will also die eventually. Because of BTRFS for Linux, Linux's popularity itself and also thanks to the Oracle's help. Sorry telling this to you, working @nexenta.com though... You, guys, are doing a very good job, but in fact, your days are also doomed, I think. -- Kind regards, BM Things, that are stupid at the beginning, rarely ends up wisely. _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
