That sort of models the modern amateur radio
transceiver manufacturers.  You may get an
upgrade infrequently, after it has been
tested many times.  We are familiar with
that process.


One group of hams are attracted to the
Flex-Radio products because of the
performance and features.

Another group enjoys the dynamics of a
Software-Defined Radio and thrives on the
improvements and new features.

Do we limit the second group's access
to placate the first group's fears of a
software problem?  Had the development
been closed, we would have missed the
excellent contributions from those not
in the original developer group.

What if instead of talking about SVN or
alpha versions, we called it experimental
or prototype versions?  It implies the
possibility of problems and the unsupported
nature of the software.  A kind word off-
reflector to those who talk about issues
with the experimental versions here should
be adequate.


Mike - AA8K


Dave Gomberg wrote:


IBM developed this fork in the 1970's when an operating system might support thousands of concurrent users on one CPU and cost millions per day to be down (like the principal Merril-Lynch machine that did about 40% of the NYSEs volume).

Development started out experimentally, when the developer was sure it was right, he ran it on his own work machine in production. When he survived, he ran it on a machine shared with a few development buddies. When it passed that test, it ran on a shared machine for the whole development lab, in a version that might contain several changes all slotted for the same release schedule.

When it was believed stable, it was shipped to the research labs for use in their environments (this was called alpha testing, since it was the first non-development test). When it passed alpha test, it was released to volunteer real users in the real world (this was called beta test, nobody risked big money on a beta test). Beta testers got direct to level 3 support so that problems could be quickly described and resolved.

Finally a "general availability" release was announced, prepared, and delivered (in that order). Usually 60% of installations were on the latest release, 30% one release back, and sprinklings of others (who got tired of hearing that the fix to their problem was to upgrade).


_______________________________________________
FlexRadio Systems Mailing List
FlexRadio@flex-radio.biz
http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz
Archives: http://www.mail-archive.com/flexradio%40flex-radio.biz/
Knowledge Base: http://kc.flex-radio.com/  Homepage: http://www.flex-radio.com/

Reply via email to