Fri Nov 18 22:52:17 PST 2005
Jeff Waugh <firstname.lastname@example.org> writes:
> See, this is the usual rationalisation in action... Feature, feature, blah,
> blah, *handwave*, kinda-like-this-stuff-here, *handwave*. Nothing to say of
> the benefits driven by the features. Sure we can kinda-sorta do this stuff
> already, but not with the clarity, manageability or shake-out benefits that
> ZFS provides. It's bloody good stuff, and we shouldn't be to snotty to say
> it! :-)
See the thing is, while the outside may be "pretty" and "user
friendly" it mostly looks like a massive layering hack to me.
Not sure how to define a good interface to LVM for dynamically sizing
partitions ... no problem, just hack the entire LVM into the FS. Not
sure how to put checksumming in the block layer to give reliable
storage ... no problem, just hack parts of the block layer into the FS.
Of course some of this is probably an instinctive reaction to the
copious marketing of ZFS as the second coming.
>> > It is wonderful to have Sun kicking Linux up the arse with some
>> > innovative, useful and *OPEN SOURCE* technologies. :-)
>> I see it more as being Sun playing catchup on the open source bandwagon;
>> the rest feels a little bit like too little, too late.
> Catching up? They're making great technology. Cooler stuff than we have in
> some cases... and there's nothing offensive or pathetic about that at all,
> it just shows we're seeing competition (which we pushed). Rock on!
Yes, as in ... they're the 7th "open" Unix like OS, 5 months after
opening they still don't even have a buildable distro., there is no
open technical discussion that I can find, AIUI you still need half
of a Linux box userland to make the binary only blob bearable and
they'll have a Linux comparable network stack "coming real soon now".
 And they can't share code with anyone else, although they should
be able to take code from the BSD projects it can't easily go the
 James's 3 tiered approach to running an "Open" project:
1. Release binaries and source to build said binaries under a useful
2. Have public technical discussions about direction of source.
3. Have more multiple competing contributors.
...Sun almost have 1 done.
James Antill -- email@example.com
More information about the linux-elitists