What follows is a short interview with Dr Rich Murphy. We will be doing a more in depth interview in October.

Question 1: Could you tell us a little about your company?

We do computer security software, and in particular, host-based intrusion prevention software (http://www.whiteoaklabs.com).

We have been contributing to Asterisk development, and have assigned copyright to Digium for all work we contribute so that it can be integrated and leveraged fully by Digium and the asterisk community.

Question 2: What are your thoughts on the commoditization of the IP-
PBX?

I believe that Asterisk is positioned quite well to lead the market in commoditization of the IP-PBX. Such commoditization occurs naturally as the market matures. The asterisk licensing has fostered a community that drives rapid development and thus puts asterisk in a position to maintain a leadership role.

Question 3: Are there any benefits running Asterisk on FreeBSD vs Linux? i.e. Benefits specific to the OS (e.g. TCP/IP stack) and not personal preference factors (e.g. administration).

As a founding FreeBSD core team member, anything I say should probably be labeled as personal preference.

That said, I would point to obvious differences between binary package distribution of RedHat/Suse/Mandrake/Debian and source based package systems of Gentoo and FreeBSD. Source based package systems tend to have fewer binary compatibility issues between applications and the rest of the system. Binary incompatibility with libraries or the kernel can exhibit subtle, seemingly unrelated symptoms. Source based package distribution also tends to have varying performance benefits due to compile-time optimizations.

There are a number of other differences in the core OS technology such as file systems, IP stack and VM. But that requires some concrete details to be fair to each.

Question 4: Where do you see the Asterisk on FreeBSD project going in the future?

Maxim Sobalev ported Asterisk to FreeBSD, and maintained it as a 'port' in the FreeBSD distribution. I have focused on integrating support into the current asterisk distribution, extending the support to all of the BSD distributions, and extending support to all of the current asterisk features. In the future, asterisk support should extend beyond FreeBSD, OpenBSD, NetBSD and Darwin/OSX to all the BSDs, including derivatives such as Firefly. The drivers will eventually be extended to support these, but at a slower rate due to greater effort required to integrate and maintain differences in driver APIs. The effort is more than just making the code work on those platforms. The code must also avoid any negative impact on stability or performance on the Linux platforms, and it must be consistent with the style and design methods currently used in the asterisk distribution.

Question 5: Could you give us a little personal background?

I collaborated with Bill Jolitz on his first public release of 386BSD (called 'no name'), prior to the unofficial 386BSD 'patch kit' that later became FreeBSD, that later spawned NetBSD and OpenBSD. Bill Jolitz is truly an unsung hero. Berkley contracted Bill Jolitz to port BSD to the x86 platform, and he negotiated terms of his contract that required his source code to be released publicly. That was the genesis of all these things, including my participation as founding member of the FreeBSD core team and later a founding member of the XFree86 project.

This illustrates the value of encouraging others to leverage one's work in successive efforts. I believe that collaborating with Mark Spencer will foster the same sort of string of successes, and I believe that the current asterisk community represents a successful effort to do so.
More info at

http://www.whiteoaklabs.com/about-us.php

Comments (Click to post)

Name:
Subject:
Website:
Message: