All of lore.kernel.org
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: "Philip Oakley" <philipoakley@iee.org>
Cc: "Git List" <git@vger.kernel.org>
Subject: Re: [RFC/PATCH] howto/maintain-git.txt: new version numbering scheme
Date: Mon, 03 Feb 2014 15:58:12 -0800	[thread overview]
Message-ID: <xmqqa9e7bvfv.fsf@gitster.dls.corp.google.com> (raw)
In-Reply-To: <9E6F99D96D124571897121E4227508EF@PhilipOakley> (Philip Oakley's message of "Mon, 3 Feb 2014 22:03:36 -0000")

"Philip Oakley" <philipoakley@iee.org> writes:

> If we are progressing from V1.9 to V2.0 quickly (one cycle?), which I
> understand is the plan, then mixing the minor development items (patch
> series which progress to master) with the maintenance fixes over the
> next few months, thus only having 1.9.x releases, sounds reasonable.
>
> If there is going to be separate maintenance fixes from the patch series
> developments then keeping to the previous 1.9.x.y for maintenance would
> be better.
>
> Will the new rapid counting continue after V2.0, such that we get to
> V2.9 -> V3.0 rather more quickly than V1.0 -> V2.0 ?
>
> The key discriminator would be to say when V2.0 will be out for deciding
> the V1.9 sequence.

I do not quite follow.  The time distance between v1.9 and v2.0
should not affect anything.  If it is a long road, there may be
v1.10, v1.11, v1.12, ... before we have v2.0.  If not, v2.0 may
immediately follow v1.9 as a new feature release.  There may be
maintenance releases based on v1.9 that does not add any new
features.

Right now, if you count the maintenance releases, there are
potentially four kinds of version gaps:

 - Between v1.8.5 and v1.8.5.1, there are fixes but no new features;

 - Between v1.8.5 and v1.8.6, there are new features but no
   compatibility worries;

 - Between v1.8.6 and v1.9.0, there are new features, no
   compatibility worries, but somehow the jump is larger than the
   one between v1.8.5 and v1.8.6; and

 - Between v1.9.0 and v2.0.0, there are new features and also
   compatibility concerns.

Switching to 2-digit scheme and calling the upcoming one v1.9 (and
the next major one v2.0) was meant to make the naming more flat, as
the third item in the above list "somehow the jump is larger" does
not seem to add much value to the end users.  So the logical
numbering becomes more like this:

 - Between v1.9 and v1.9.1, there are fixes but no new features;

 - Between v1.9.x and v1.10, there are new features but no
   compatibility worries;

 - Between v1.9.x and v2.0, there are new features and also
   compatibility concerns.

With a twist, though.  There seem to be many places where at least
three digits are assumed to exist in our version numbers, so in
order to make life easier, the updated document says vX.Y (a feature
release) will identify itself as vX.Y.0

  reply	other threads:[~2014-02-03 23:58 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-31 23:14 [RFC/PATCH] howto/maintain-git.txt: new version numbering scheme Junio C Hamano
2014-02-03 22:03 ` Philip Oakley
2014-02-03 23:58   ` Junio C Hamano [this message]
2014-02-04 21:03     ` Philip Oakley

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=xmqqa9e7bvfv.fsf@gitster.dls.corp.google.com \
    --to=gitster@pobox.com \
    --cc=git@vger.kernel.org \
    --cc=philipoakley@iee.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.