login
Header Space

 
 

Re: HAMMER update 06-Feb-2008

Previous thread: Re: HAMMER update 06-Feb-2008 by Matthew Dillon on Thursday, February 7, 2008 - 7:57 pm. (1 message)

Next thread: vkernel manager by Aggelos Economopoulos on Saturday, February 9, 2008 - 3:04 am. (7 messages)
To: <kernel@...>
Date: Friday, February 8, 2008 - 8:22 pm

Yes, and I've agonized over this very possibility.  Maybe the thing
    to do is to poll the people on kernel@.  HAMMER won't be ready for
    sure (things take however long they take), but the hardest part of it
    is working and stable and I'm just down to garbage collection and
    crash recovery.  Crazily enough, that is what all the major surgery
    yesterday, and the continuing work, is about.

    So what do people think?  Should this month's release be 1.12 or 2.0 ?

					-Matt
					Matthew Dillon 
					&lt;dillon@backplane.com&gt;
To: <kernel@...>
Date: Saturday, February 9, 2008 - 11:08 pm

I'd go with 1.2 for now.  A nice, known-stable release with an
experimental feature which isn't quite ready yet.  And then
release 2.0 with HAMMER known to be good, solid, and well-tested.
If you're going to make a big deal about HAMMER, then it would be
doubly-plus-good to be confident that all the pieces of it have
seen some stress-testing.

-- 
Garance Alistair Drosehn            =   gad@gilead.netel.rpi.edu
Senior Systems Programmer           or  gad@freebsd.org
Rensselaer Polytechnic Institute    or  drosih@rpi.edu
To: <kernel@...>
Date: Sunday, February 10, 2008 - 6:01 am

Too bad we can't reuse version numbers ;).
-- 
         Thomas E. Spanjaard
         tgen@netphreax.net
To: <kernel@...>
Date: Saturday, February 9, 2008 - 10:29 pm

I'd vote 1.12

Save up 2.0 for a really big milestone.
To: <kernel@...>
Date: Saturday, February 9, 2008 - 11:17 pm

Make this a 1.12, because HAMMER is only in an alpha state. Then when HAMMER 
is production ready release, change the numbering scheme altogether. The 
roman numerals for example are nice, and would attract attention. DragonFly 
XIII sure looks good, and the june release DragonFly XIV. Maybe dump the 
odd-even numbering altogether.

Petr
To: <kernel@...>
Date: Sunday, February 10, 2008 - 4:42 am

Whatever else is decided, as one saddled with a roman numeral surname 
suffix these past 63 years, I'm dead-set against those!

Give a thought to my numerous Irish friends, who casually drop 
'aitches', and what happens to 'William B Hacker III'.

;-)

- and fast forward with 'X' = [unknown | mystery], and find

XIII == 'DragonFlyBSD's [unknown | mysterious] turd release'

There's gotta be a better way?


Bill
To: <kernel@...>
Date: Saturday, February 9, 2008 - 7:51 pm

1.12

-- 
Robert Luciani
Chalmers University of Technology, SWE
Department of Computer Science and Engineering
http://death.olf.sgsnet.se:8080/public.key
To: <kernel@...>
Date: Saturday, February 9, 2008 - 9:34 pm

1.12




-- 
System Programmer
--
I'm Searching For Perfection,
So Even If U Need Portability U've To Use Assembly ;-)
--
http://libosdk.berlios.de
To: <kernel@...>
Date: Saturday, February 9, 2008 - 3:41 pm

I'm for 1.12.

Joe
To: <kernel@...>
Date: Saturday, February 9, 2008 - 9:14 am

I prefer 1.12

Best Regards,
sephe

-- 
Live Free or Die
To: <kernel@...>
Date: Saturday, February 9, 2008 - 5:17 am

If we want to release 2.0 as the big feature[tm] release, we should go
with 1.12, although I don't like the 1.x thing.  If 2.0 should be a
release with HAMMER as pre-alpha version, we could go with 2.0.
I would vote for 2.0

	Matthias
To: <kernel@...>
Date: Saturday, February 9, 2008 - 2:19 am

With that choice, I'd go for 1.12, but then our __Version is totally 
broken.  Hasso's 2.2 isn't bad either, fooling people nicely :)

I'd actually drop the "1." completely, and make it 12-RELEASE, followed by 
13-DEVEL.  I mean, what constitutes "enough" change to go from 1.* to 2.*, 
then to 3.*, etc.?  That's quite arbitrary.  Dropping that digit would 
avoid that alltogether.

Not an easy decision...

cheers
   simon

-- 
Serve - BSD     +++  RENT this banner advert  +++    ASCII Ribbon   /"\
Work - Mac      +++  space for low €€€ NOW!1  +++      Campaign     \ /
Party Enjoy Relax   |   http://dragonflybsd.org      Against  HTML   \
Dude 2c 2 the max   !   http://golden-apple.biz       Mail + News   / \
To: <kernel@...>
Date: Saturday, February 9, 2008 - 12:34 pm

Simon 'corecode' Schubert wrote:

This way we'd obviously have the highest version number of ALL open 
source OS'es, clearly demonstrating superiority and advancednessitude!
To: <kernel@...>
Date: Saturday, February 9, 2008 - 6:34 pm

DragonFly XIII-DEVEL
-- 
         Thomas E. Spanjaard
         tgen@netphreax.net
To: <kernel@...>
Date: Saturday, February 9, 2008 - 7:16 pm

Even better :)

-- 
Serve - BSD     +++  RENT this banner advert  +++    ASCII Ribbon   /"\
Work - Mac      +++  space for low €€€ NOW!1  +++      Campaign     \ /
Party Enjoy Relax   |   http://dragonflybsd.org      Against  HTML   \
Dude 2c 2 the max   !   http://golden-apple.biz       Mail + News   / \
To: <kernel@...>
Date: Saturday, February 9, 2008 - 1:13 am

Neither.

1.2X wherein HAMMER fs is 'experimental'

- With an allowance for 1.3X and HAMMER fs 'stable' (as in does not 
break, crash the OS, or lose data) if not yet fast. Let us not have a 
parallel to Reiserfs troubles.

I'd like to see 2.0 reserved for 'ready for prime time'. i.e. industrial 
strength stability if not (yet/ever) a speed champ on fs or SMP issues.

JM2CW - but we are having to convert servers from &lt;another&gt;BSD to &lt;a 
different&gt;BSD 'coz - for servers - performance can be addressed with 
hardware. Stability only with hours of our lives. And we grow weary of 
chasing fragile patches.

Bill
To: <kernel@...>
Date: Friday, February 8, 2008 - 11:22 pm

I want 2.0 to have a significant user-visible change to match the major
revision, and HAMMER seems to be that thing.  Do we need another 6 months
to reach that point?

I say "2.0", if people can use HAMMER's features, even if it has bugs or
stability issues; people (ahem) hammering on it is the best way to test
when it's something so basic as a file system.

Have we ever established what the criterion are for declaring HAMMER ready
for release?
To: <kernel@...>
Date: Friday, February 8, 2008 - 10:44 pm

For the most part, I don't really care about version numbers and 
people's alleged expectations. So I wouldn't object to 1.12 if you don't 
feel that HAMMER's current status meets the goals set for 2.0.

Sascha

-- 
http://yoyodyne.ath.cx
To: <kernel@...>
Date: Friday, February 8, 2008 - 9:42 pm

I vote for 2.0rc1.  I've watched both Linus and BillG+UncleFester go through
many release cycles over the years, and I'd hate to see DragonFly duplicate
the Windows Vista debacle just to meet an arbitrary/imaginary deadline.

Bill has made billions by promising products he can't (and doesn't) deliver,
but those good-ole-days seem to be over now, thankfully!
To: <kernel@...>
Date: Friday, February 8, 2008 - 8:46 pm

I don't really care about version number marketing and I don't think that 
it matters at all. Therefore I think that upcoming stable release should 
be 2.2 already :).

But whatever decision will be, I wish that __DragonFly_version would make 
logical sense again.


-- 
Hasso
To: <kernel@...>
Date: Friday, February 8, 2008 - 8:32 pm

1.12.
-- 
         Thomas E. Spanjaard
         tgen@netphreax.net
Previous thread: Re: HAMMER update 06-Feb-2008 by Matthew Dillon on Thursday, February 7, 2008 - 7:57 pm. (1 message)

Next thread: vkernel manager by Aggelos Economopoulos on Saturday, February 9, 2008 - 3:04 am. (7 messages)
speck-geostationary