> Once upon a time Jake Messinger shaped the electrons to say...
> >I would think they need to release something immediately. Didnt anyone at
> >Livingston even TEST the compression of the OR-HS before they put up the
> >COM OS 3.7 for the OR? >
>
> You can't test every feature in every build - period. No one does that.
Many would disagree with statement. If you are adding a NEW feature to an
operating system, you should TEST the FINAL build before it is released,
if only for the new enhancements. According to your release notes on COM
OS 3.7, compression is the third thing on the list, indicating to me that
it must have been quite an important addition to the OS. The FIRST
sentence in the "Compression" paragraph states that software compression
was added to the Office Router.
> You'd never have a release. Everything was ok in the test builds. Things
> were rolled up, and the build was made. A typo got in. Shit happens.
This is why you have beta testers. You have them test it BEFORE it goes to
official release. Then when it DOES get out on your ftp site, you should
try downloading it and installing it to make sure THAT step was done
correctly. I have seen cases where software was made available on web
sites/ftp sites, but it was incomplete or damaged during its placement on
the ftp server.
I know and understand that Livingston is exerting most of its efforts on
the PM 3 and people would agree that this is where the primary efforts
should be. But you should have someone out there that uses the OR-HS's
that could Beta the new enhancements for you. I dont have HS models but I
will certainly offer to beta test for the models that we do use.
> Every time you make a build you can't go back and test every last feature,
> let alone every combination. Even with automation it just can't be done.
I didnt suggest that you test every last feature. You are hyperbolizing my
arguement.
> Well, it CAN be - but not if you want to stay in business and release
> anything on a realistic time scale. If a feature works and you don't mess
> with it, you have to presume it still works in the next. Nothing was
> messed with in the code, it was just a human error doing the build.
Again, the compression was a NEW feature added, it was a MAJOR addition.
It looks like there were only a very few features that were added/enhanced
that apply to the OR line and there appears to be only a few OR models.
Those NEW features should have been tested in the FINAL form.
> It would be as ridiculous as those who want us to test the PM-3 with
> every modem out there, or even just every Sportster. Not realistic in the
> least.
Hardly. This is a totally unrealistic comparison. There are
hundreds of modem types out there and you have no control over them.
Livingston manufactures the OR router and Livingston wrote the COM-OS for
that router. It is Livingston's responsibility as a manufacturerer, to
ensure that their products work as they advertise, not to make excuses as
to why they don't.
BTW, I think the OR and its com os is coming along quite
nicely and blows away the equivalent Ascend products at this point.
> Anyone with development experience understands this.
I have been writing software, for the Medical industry, primarily for the
past 18 years. I teach information sciences at the University of Houston.
I can say I have some knowledge in development at least on the software
side. One of the 8 steps in the Development Life Cycle is "Testing". This
step should take 50% of the total life to develop. Testing and then
"Documenting" are the 2 steps that seem to receive the least amount of
attention in practice.
I'm not trying to come down hard on Livingston, and I am not trying to
point fingers. I just want to help Livingston make a better product and be
perceived in a positive light by their users and potential customers and
investors. I made the statement because I thought it was a substantive
issue that an official release was made and the 3rd item on the
enhancements list does not function according to one of your customers.
(this may all be academic, of course, if it actually DOES work).
More importantly, your unprofessional response suggests that Livingston
makes a policy of releasing COM OS's with new, yet untested, enhancements.
I'm certain this is NOT the case, and I understand how a simple typo can
cause major problems. I have had my share of typos released software and
we always handle it quickly and professionally as soon as it is discovered
or reported. But I take issue now with your response on this matter.
You are the most vocal Livingston employee on this list and you are
perceived as the voice of Livingston, however, I do not believe you are
voicing what would be Livingston's response as a corporation to this
issue.
I would like to see a post from others at Livingston regarding this issue,
perhaps from the production management or the quality control department.
~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~
Jake Messinger 713-772-6690 jake@ams.com
Advanced Medical Systems, Inc. jake@uh.edu
8300 Bissonnet #400
Houston, Texas 77074 http://www.ams.com/~jake
~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~