Mailing-List: contact dev-help@couchdb.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@couchdb.apache.org
Content-Type: text/plain; charset=windows-1252
Mime-Version: 1.0 (Mac OS X Mail 7.1 \(1827\))
Subject: Re: Git Repository Creation
From: Robert Samuel Newson <rnewson@apache.org>
In-Reply-To: 
 <CAJNb-9o4U1onV2j-U9j99kZYTzL2rDCmJEpuiBN+u=Vewhn9wA@mail.gmail.com>
Date: Wed, 22 Jan 2014 13:15:32 +0000
Content-Transfer-Encoding: quoted-printable
Message-Id: <58924CEF-C77F-4A70-BE8A-96F64CFD8DE6@apache.org>
References: 
 <CAJ_m3YDruu=0P-dQZiQwDAp4ac=zXRrqwA9bMa+nJWW0ZRayWA@mail.gmail.com>
 <CAJNb-9oFctms3MVmsMV6jYdRSc3trWsp-b_Y22PAP0kQ+K_vkg@mail.gmail.com>
 <CAJ_m3YAGOSQ2Pcj4_Eqr_e-sPCQ0CO3RMEgqCG=OM51vimzBAQ@mail.gmail.com>
 <etPan.52de7d64.46e87ccd.15f6@akai.jsonified.com>
 <E1556D94-B196-4DE9-A2D6-30284F80F685@apache.org>
 <06B0A74C-D886-481B-92CC-548B7A05E327@apache.org>
 <CAJNb-9ru4pUwPZV3VjEq1zBAUeHEBrKJo1BjEH752RdpwpkDiA@mail.gmail.com>
 <CAHdjipKzPoEH=J9G0bS5S5-3T9Gh+NcsEB0z9TAWfuN7Fx5Ujg@mail.gmail.com>
 <CAJNb-9rC3URBkcC4yKY4cLE+QgZg_0xb3K7M4V1au9N+JM15XQ@mail.gmail.com>
 <CAHdjip+yZS+hW5OxN_7+AKEv5XfXZpA_x7rj9bmDS6m8X=G5=Q@mail.gmail.com>
 <A62DAF5E-517A-47BB-9679-1BFC25B51042@apache.org>
 <CAJNb-9r8kh_dX8xRi8468oUPQkcWCk8rPh1606gpcbsqW+Qddg@mail.gmail.com>
 <CAL+Y1nuLyoyuNV3Xgba1-g8xbxukTx3bCZ6AaXZVD=7jN8n0Lg@mail.gmail.com>
 <CAJNb-9o4U1onV2j-U9j99kZYTzL2rDCmJEpuiBN+u=Vewhn9wA@mail.gmail.com>
To: dev@couchdb.apache.org


Benoit,

I=92m aware of the changes in R15 and R16.

If we are discussing whether to raise the lowest supported version of =
erlang for CouchDB from its current value of R13B04, let=92s have a new =
thread. I will point out that R14 is stable and widely used still and =
we=92ll go from there.

B.

On 22 Jan 2014, at 13:10, Benoit Chesneau <bchesneau@gmail.com> wrote:

> On Wed, Jan 22, 2014 at 1:58 PM, Dave Cottlehuber <dch@jsonified.com> =
wrote:
>>=20
>>=20
>> The issues wrt to scheduler collapse are well known, and the concern
>> around stability under heavy load with NIFs is still valid. R16B02+
>> have not been out so long that anybody has made pronouncements about
>> considering it solved, unless I missed something.
>>=20
>=20
> A lot of improvements and fixes have been made in the scheduler, do =
you
> have any link to what you refer in mind? Nifs just appeared in R14.
>=20
> About r16b02, well I was not thinking to this version per see. The =
fact is
> that r14 isn't supported any more and R17 will introduce breaking =
changes
> (actually deprecations in latest R16) so it may be harder to maintain =
a
> workingf version on couch on every release. Also apart nifs there have =
been
> other improvements in Erlang.
>=20
> - benoit