On the Democratic Operation of Bugzilla

[Rob Cozens]

Garrett:

> This is not about influencing the direction of the product.  This is 
> about how bug reports should be directly given to the company, the 
> company should track it internally and insure that it's taken care of. 
>  Users should not have to do anything else, that's why they pay 
> Runtime for the product.
>

RR bug reports are directly given to the company by posting a report in 
the Bugzilla database... preferably  after checking to see if the 
problem has been previously reported.  The information recorded [except 
votes] is the same information that would be asked of you if you were 
reporting the problem via telephone.

RRLtd and any interested users can track bugs internally by querying 
BZ.  Once a problem is reported, one does not _have_ to do anything 
else.

I believe the crux of your issue is the "insure that it's taken care 
of", and I suggest that has little to do with the way it's reported and 
tracked.  How do you suppose RRLtd would process bug reports submitted 
by telephone?  Don't you imagine the Tech Support person taking the 
call would enter the information in a bug database like BZ?

So unless you take issue with going online and reporting a bug to BZ 
instead of taking up the time of someone who could be fixing bugs but 
instead must sit on the phone and ask you to relate the information, or 
take issue with the fact that users as well as RRLtd staff can track 
the information, I don't see BZ as the culprit.

I get the feeling that you are taken back by the number of items on the 
BZ bug list and the amount of time some items remain unresolved.

Based on my thirty year's experience in the field, I suggest that there 
is NO bug-free application of any scope or complexity on the market 
today.  When I ran DG Minis for Oakland Police Department, I would 
receive monthly a 350+ page book listing all known bugs in Data General 
software.  Those bugs did not prevent us from performing our daily dp 
tasks.

Most companies keep their bug lists internal, but virtually all 
companies have them.  The philosophy of the original owners of FlexWare 
was "we won't make our bug list public because people will think our 
product is no good" (and perhaps Dan and others' panning of BZ may 
prove their point).  What I saw was people responding "what's the 
matter with the people at Flexware that they don't know about this 
problem" when they crashed the system doing something they wouldn't 
have done if they had been warned on a bug list.

Counting bugs gives one little indication of the overall quality of the 
product without taking into account their nature and severity.  How 
many "the rectangle graphic is rendered one pixel short on XP systems 
when the width is odd" bugs equate to one "Rev 2.7.1 crashes in Win XP 
when I copy to the Clipboard" bug?  How many bugs in BZ are of the 
former type?  How many are the latter?  I think you need to know this 
before you can make quality judgments of Revolution based on the number 
of entries in Bugzilla.

Rob Cozens
CCW, Serendipity Software Company

"And I, which was two fooles, do so grow three;
Who are a little wise, the best fooles bee."

from "The Triple Foole" by John Donne (1572-1631)