todd@stiatl.UUCP (Todd Merriman) (01/05/90)
What actually measures the quality of software? "Quality" is an often-abused word used to describe software developed by software engineers, but the criteria by which to judge "quality" varies widely from place to place. Some of the ways I judge quality are: How well the pieces of a software system are integrated How few trouble reports there are How efficiently it performs How supportable is the code How modular is it? How easy is it to add new modules? How intuitive is the user interface? How well it satisfies the end-users Does it have acceptable functionality? What are some of the ways you judge the quality of software? ...!gatech!stiatl!todd Todd Merriman * 404-841-4000 * Atlanta, GA
hallett@pet16.uucp (Jeff Hallett x5163 ) (01/05/90)
In article <8439@stiatl.UUCP> todd@stiatl.UUCP (Todd Merriman) writes:
%What actually measures the quality of software? "Quality" is an
%often-abused word used to describe software developed by software
%engineers, but the criteria by which to judge "quality" varies
%widely from place to place. Some of the ways I judge quality
%are:
%
% How well the pieces of a software system are integrated
%
% How few trouble reports there are
%
% How efficiently it performs
%
% How supportable is the code
%
% How modular is it? How easy is it to add new modules?
%
% How intuitive is the user interface?
%
% How well it satisfies the end-users
%
% Does it have acceptable functionality?
%
%
%What are some of the ways you judge the quality of software?
%
% ...!gatech!stiatl!todd
% Todd Merriman * 404-841-4000 * Atlanta, GA
In general, I like the areas you target, but the problem is that most
of these are "unquanta" (to use a DeMarco-ism). I think we all have a
good idea of what kinds of things we want to measure: does it meet
requirements? can the user use it effectively? is it put together
fairly well? etc. However, the difficulty lies in how do we measure
these things? We can measure adherance to requirements and
satisfaction of the user's needs quantitatively. There still are no
impirical measures for the other things though (or, if there are, I'd
like to see them).
We need research into a lot of these things...we cannot control what
we cannot measure and unmeasurable criteria will probably leave us
worse off than having none at all.
--
Jeffrey A. Hallett, PET Software Engineering
GE Medical Systems, W641, PO Box 414, Milwaukee, WI 53201
(414) 548-5163 : EMAIL - hallett@gemed.ge.com
"Non Sequitor - Your facts are uncoordinated"; "I am Nomad: I am perfect."pgn@cbnewsc.ATT.COM (Novorolsky) (01/06/90)
From article <8439@stiatl.UUCP>, by todd@stiatl.UUCP (Todd Merriman): > What actually measures the quality of software? "Quality" is an > often-abused word used to describe software developed by software > engineers, but the criteria by which to judge "quality" varies > widely from place to place. Some of the ways I judge quality > are: > > How well the pieces of a software system are integrated > How few trouble reports there are > How efficiently it performs > How supportable is the code > How modular is it? How easy is it to add new modules? > How intuitive is the user interface? > How well it satisfies the end-users > Does it have acceptable functionality? > > > What are some of the ways you judge the quality of software? > Two more that I like to add to the list: Is the software available when needed (i.e., for newly developed stuff - as scheduled ) How much does it cost. (The "you get what you pay for" argument. Good software does not need to be expensive, expensive software isn't always "high quality", and I am willing to live with more faults if the product is really cheap.) These, BTW, are MY opinions, and not necessarily those of my employer. (Particularly, I don't want to imply that my employer is going to make or buy really cheap software that has lots of bugs. :-) ======================================================== **paul novorolsky ( !att!iwtpm!pgn, pgn@iwtpm.att.com, attmail!pnovorolsky) ========================================================