On Sat, 6 Aug 2005, Bryan Maynard wrote:
1. The business plan and revenue model
1.a I agree that this needs to be adressed. This issue could break this
whole effort if not handled properly.
Agreed. Both are being worked on and will be published as soon as they are
available.
I do not believe the world needs
another certification thrown into the pool known as "evidence of
knowledge". There are already many standards and certification bodies
that are both highly respected and very influential. If BSD
certification attempts to go head-to-head with MCSE or Linux
certification companies with it's business model it will lose.
My idea is to take an alternate approach and target a specific and
focused group - not only for certification, but also for costomers of
certified BSD professionals.
Agreed. Our audience is much smaller and the costs of the chosen testing
methodology will have to address that. And, unlike MCSE, our audience is
not primarily North American and the pricing scheme and testing
methodology will also have to address that.
* How will proctors be paid?*
(Are you willing to live with the risk of corruption of volunteer
proctors? We weren't. - LPI)
On the other hand, corruption risk is related more to the reputation and
perceived involvement in the project than it is to the amount of money
received. The BSDs already have a very well defined global infrastructure
of committers (both code and documentation), translators, user groups,
advocates and those who know each other from mailing lists, IRC and
conventions. Someone who feels involved and known within the community is much
less likely to risk their reputation and more likely to see a benefit in
giving back to the community. Having said that, part of the standards
process is creating disciplinary and appeals policies as well as a
well-defined set of expectations for both testing centers and proctors.
*Will there be official courseware that will augment the exam revenue?*
Official, no; recommended, yes. Much of this depends upon the audience. An
exam targetted at more experienced sysadmins should have very detailed
exam objectives, but probably won't need courseware. There should be
courseware available for those who are less experienced and this will take
time.
2. Actual demand
2.a This could seem like a dream-killer, but I don't think it is. The
BSDs popularity is growing (Mac OS X uses FreeBSD also). The BSDs
Represent a large chunk of the world's most secure OSs. Linux is the
most-hacked OS on the market. BSD certifications could very well
augment growth and provide the perception that the BSD's security is
not haphazard. Make the big-name clients of FreeBSD as widely known as
possible. Play up that quality IT professionals are recieving quality
training on the world's greatest operating systems. People have
enormous respect for UC Berkley - however, how many people even know
that the BSDs originated at Berkley? Play up the BSD's pedigree!
I believe the demand for the BSDs (and certified professionals to setup
and administer them) is there, it's just untapped. Many, many companies
are sick of Windows and looking for alternatives. I see many users,
developers, and sysadmins switching to FreeBSD on the mailing lists -
make this known!
This is happening. As an example, watch for the new and improved
www.freebsd.org coming to a web browser near you sometime in September. It
will include new whitepapers and case studies, many of which will also be
available for download from bsdcertification.org/resources.htm. At the
moment, I have a very nice queue of case studies for FreeBSD. I still have
an open call for anyone wishing to highlight their company, regardless of
the BSD or mix of BSDs in use. Anyone interested, email me direct and
we'll write one up and make sure a copy is available for download on the
relevant project's website.
I'm also interested in interviewing companies/educational institutions
etc. who want to go on record as saying "we can't wait to train/test for
BSD certification". Again, drop me an email if you're interested in
participating.
3. Partnerships
3.a This is potentially very sticky. Partnerships mean politics - and
politics can kill innocation and software. One of the greatest
strengths of the BSD's is their conftolled development. Random people
don't throw patches together and call it software. However, to get
noticed you have to have someone care about you. And to have someone
care, you have to give them what they want; this is where things can
get sticky. How to we give them what they want? Who's requests do we
answer first? with a limited development staff, what features get
implemented when? If we make partners unhappy, they won't be partners
anymore, so how do we please our supports without rushing software out
the door?
That is one way to do business, but not the only way. Another approach is
to know the needs of your audience and address those in a quality manner.
The best partners in the world won't help if your product is of low
quality. And partners have been known to drop you like a hot potato once
you're no longer the flavour of the week or you no longer serve their own
interests. This is just the other side of the coin and is not meant to
sound anti-partnership.
One thing to note is this: While all vendors provide solutions for small
to medium sized businesses none SPECIALIZE in small to medium sized
business solutions. This would be a good place for us to start. As I
mentioned in my previous post (Some ideas. . .), I think a good
approach would be to certify more than just sysadmins:
Produce certified software experts that know to build software
solutions for small to medium sized businesses. This certification
could provide assurance that the certified would not only know what
Open SOurce apps best fit specific needs, but also know how groups of
applications fit together, follow the development and community of
applications (maybe even provide active feedback to projects such as
OpenOffice as part of their job description - this could help build
healthy and influential partnerships).
Produce certified kernel experts that could shave the kernel to the
bare required minimum. These experts would also have intimate knowledge
of specific hardware support - what motherboards owrk best, what
rack-mounted servers are the easiest to setup and most stable, etc.
Produce performance engineers to squeeze every last drop of power from
a system. Given a combination of hardware and software, these experts
would be able to tune the system for maximum efficiency and speed.
These experts could also handle things like loading balancing and
banwidth shaping.
Produce watchdogs that provide optimal security environments. These
experts could take a system built by the other experts and secure it.
Finally, produce sysadmins. These experts could be brought in after a
system is designed, built, and installed. They would monitor the system
and modify it as necessary to meet changing business needs.
These are all very good, but not in our current mandate. Let's fulfill the
mission statement to provide quality sysadmin certification, see how that
goes and then move on from there. Perhaps some reincarnation of the
current certification Group will start to address these a few years down
the line. Or perhaps another group will form to address these in an
earlier time frame...
FYI: I am looking to start just such an open source consulting firm. One
of my friend's has done something similar and I am aware of several
other ventures that have started recently along the same lines. If we
could provide these consulting firms with IT professionls certified in
open source BSD solutions than we could simultaneously stimulate demand
for the BSDs, our certifications, and our client firm's solutions.
Sounds like you are also interested in advocacy. If so, email me off list
as there is much that can be done.
Dru
_______________________________________________
BSDCert mailing list
[email protected]
http://lists.nycbug.org/mailman/listinfo/bsdcert
