Varnish Cache Project summer status

[Poul-Henning Kamp]

Hi Everybody,

This is a quick update from the Varnish Cache project, it wont take
long, but please spend 1 minute, there is important stuff here.


Where the heck is the 4.1 Release ?
-----------------------------------

It is now truly right around the door.  I have closed the gates for
any new features, only bugfixes and performance tunings allowed until
4.1-R is out the door.

We are busy getting tests run to make sure it lives up to expectations.

If you have any chance to try out the code in git/trunk on live
traffic on in test environments, it would *really* help the project
and speed up the release:  Give it a roll and send me an email with
any and all observation, even if it runs flawlessly.


Whats after 4.1-R ?
-------------------

We don't know yet if there will be a 4.2 release or just 4.1.X releases,
but we know for sure that there will be a 5.0 release.

I don't like to pre-announce code I have not written yet, so that is all
I want to say about 5.0 at this time.

However, it is quite obvious to everybody that our release process
for 4.x and to a lesser extent 3.x did not really work for anybody,
so I have decided a new release strategy starting from 5.0.

The new strategy is this:  We will make a head-of-tree release every
six months, come hell or high water, and whatever works at that
time is the feature-set of that release.

This will allow both developers and users to actually plan their Varnish
upgrades going forward.

More importantly it will eliminate the "You *have* to hold the
release so I can get this $PETFEATURE in, because I can't wait years
for the next release" perverse incentive to schedule slippage.

If you analyse this change deeper, it reflects the transition in the
project where we now have full time developers working on Varnish,
rather than Devops running websites doing part time development.
The downside to this, is that we do not have any people in the
developers group running bleeding edge Varnish all the time, and
consequently our release strategy cannot rely on that testing.


Community News
--------------

We have not had a Varnish User Group meeting for a very long time,
but the Varnish Developer Days have been happening on a semi-regular
basis.

We have talked about what the right frequency and format of these
events should be in the future, and more details about this will
be forthcoming once details fall into place.

The big picture is that we want to turn the VDDs into more of a
working gathering than a formal meeting, and we want to open
it up to other people than the core developers to come and hang
out and contribute to the Varnish Project code base and documentation.

We also want to get some kind of devops/user oriented annual event going,
but we're not quite sure how, we want to make it affordable and
easy for people to participate.  Any good ideas are most welcome
(email ruben at varnish-software.com and phk at varnish.org)


Varnish Moral License News
--------------------------

Brazil is going through a hard time, and therefore GLOBO.COM have
had to discontinue their economical support to the development and
maintenance of the Varnish Cache software.

I can fully sympathize with that. 

I want to thank GLOBO very much for the support they have provided
during the last four years, and I wish them speedy recovery from
their trouble.

Fortunately Bluehost.com, a major webhosting company out of Utah, USA
has decided to help support Varnish, so there is going to be no break
or disruption in the near term.

That means going forward the Varnish Cache development and maintenance
is funded by:

	Bluehost
	Fastly
	UPLEX
	Varnish Software

Many thanks to them all, it puts bread under my butter and quality
software on your computers.

As always full details at: http://phk.freebsd.dk/VML/index.html

And if you want to show your appreciation another way, my Amazon
Wishlist is here: https://www.amazon.com/gp/registry/ref=cm_reg_rd-upd?id=1XSZNDWXKIAS0


That's it, thanks for reading.

Poul-Henning


-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.