Mailinglist Archive: opensuse-factory (1134 mails)

< Previous Next >
Re: Post-12.2 Market Research / Planning (was Re: [opensuse-factory] Calling for a new openSUSE development model)
  • From: Jos Poortvliet <jos@xxxxxxxxxxxx>
  • Date: Tue, 19 Jun 2012 13:45:16 +0200
  • Message-id: <38291389.MDneApaXRU@linux-6upc>
On Sunday 17 June 2012 13:02:26 M. Edward Borasky wrote:
On Sun, Jun 17, 2012 at 3:57 AM, Robert Schweikert <rjschwei@xxxxxxxx>
This, as you state requires planing, which is what we have
avoided/ignored so far. To this point our releases have mostly grown
organically and this has put us into hot water w.r.t. the time based
release schedule every now and then, more so now than previously.

Following the "planning" argument would lead to another discussion

"Do we as a community want a feature planned release?"

I'm going to state some opinions here. They're biased and I haven't
done much data gathering, so my assumptions about the "state of the
world" are probably off, but here goes:

1. There are two classes of Linux distros - commercially-supported
(commercial) and community-supported (community). RHEL, SLES/SLED,
Ubuntu and Oracle are the four best-known commercial ones. I
personally can't name another, but I'm sure there are some.

2. Of the community distros, the best known are Mint, Fedora, Debian
and openSUSE. There are hundreds of others, as a visit to Distrowatch
will show. Fedora is supported in part by Red Hat and openSUSE is
supported in part by SUSE/Attachmate.

3. Here's where my knowledge is hazy, but I'm going to take a stab at
it. I think the most commonly used *community* distros in server
deployment are CentOS, a respin from source RPMs of RHEL, and Debian.
There are a lot of servers running Ubuntu but I'm calling that a
commercial distro, not a community one. My impression is that there
are few servers running openSUSE; most Linux servers are either
running a supported commercial distro, CentOS or Debian. If someone
can point me to recent and accurate survey data on Linux server usage,
I'd be eternally grateful. ;-)

4. Linux on desktops / laptops is essentially irrelevant. Again, I
haven't seen recent figures, but my impression is that at *most* Linux
represents three percent of the market, with the rest split somewhere
in the 85:15 ratio between Windows and MacOS X. Linux netbooks were
not commercially viable.

The general principles of strategic marketing are fairly well known
and I don't want to go into a lecture here. Briefly, you want to
invest in *growing* markets and capture market share of *paying
customers*, not *users*. The markets that I see growing are cloud /
IaaS / PaaS, "big data / data science / business intelligence / text
analytics", and "social / local / mobile". That's where Red Hat /
Fedora are investing, and that's where Canonical / Ubuntu is
investing. And that's where I think SUSE/openSUSE needs to invest.

1. I think we need an openSUSE PaaS. Red Hat/Fedora has OpenShift, and
Ubuntu is the core of VMware's Cloud Foundry. Both Fedora and Ubuntu
ship with OpenStack Essex. I think 12.2 should ship with OpenStack
Essex, not Diablo, but that's not going to happen. I can live with the
B1 Systems repositories for now.

2. I think we need to package the major "big data" infrastructure
tools. Hadoop for sure, but probably damn near everything the Apache
Foundation maintains - Mahout, Cassandra, the UIMA natural language
processing infrastructure, etc. There's a fair amount of yak shaving
there; in particular making all the Apache magic run with OpenJDK is a
bit of effort, but somebody needs to do it. OBS is the logical place
to do that.

NoSQL: we've got MongoDB and CouchDB and Node.js and Redis already,
but Riak is in and out because of build failures, and I don't think
Neo4J has ever been packaged.

3. Ubuntu is absolutely crushing the rest of the distros in social /
local / mobile. They've got docking with Androids, an app store, music
services, ... Fedora doesn't seem to care about this at all, and
perhaps Red Hat made a conscious business decision based on hard data.
We need to get some hard data and make the conscious business
decisions - can SUSE/openSUSE *win business* in social / local /
mobile, and how do we do it? Partner with Intel / Samsung / Tizen? In
case you missed it, Nokia dropped Meego, HP flushed WebOS and RIM is
on the ropes. I think there's an opportunity here.

As Karl said - this really is the wrong list for this. If you want to
discuss strategy, come to a meeting (oSC or Summit, I'd say). This is hard
to do over the web. And not something for a development list at all.
< Previous Next >