How does code metrics affect your motivation? Academic survey

[RLC]

I am conducting research on the effects source code quality
measurements have on developers’ general motivation and well-being.

If you write code for a living, please give us just a few minutes of
your time and respond to this questionnaire.

http://www.smart-survey.co.uk/v.asp?i=46061csldl


Many thanks,

Roi

 

[Nick Keighley]

I am conducting research on the effects source code quality
measurements have on developers’ general motivation and well-being.

If you write code for a living, please give us just a few minutes of
your time and respond to this questionnaire.

http://www.smart-survey.co.uk/v.asp?i=46061csldl

the examples they give arefrom a particular tool. It smells like
advertising for the particular tool.

 

[Victor Bazarov]

the examples they give arefrom a particular tool. It smells like
advertising for the particular tool.

The advertising is probably very immaturely made, then. I've gone
there, answered questions, and I can't recall either the name of the
tool or their publisher... Nor do I care for those, really.

V

 

[Jorgen Grahn]

The advertising is probably very immaturely made, then. I've gone
there, answered questions, and I can't recall either the name of the
tool or their publisher... Nor do I care for those, really.

I went there, answered questions ... and wondered how well *percieved*
quality and metrics from some tool correlate. Sure, I want more focus
on code quality, but I think I'd prefer to have myself and my coworkers
say what's acceptable and what's not.

/Jorgen

 

[Nick Keighley]

I went there, answered questions ... and wondered how well *percieved*
quality and metrics from some tool correlate.  Sure, I want more focus
on code quality, but I think I'd prefer to have myself and my coworkers
say what's acceptable and what's not.

I've got similar opinions. Such metrics might be handy for spotting
potential problems; "list all the methods that are >100 LOC" or exceed
some value for "complexity". But when I've had such things readily
available I didn't make great use of them. I can also imagine the
reaction to making such things compulsary (moaning and whining,
followed by malicious compliance)

 

[Miles Bader]

I've got similar opinions. Such metrics might be handy for spotting
potential problems; "list all the methods that are >100 LOC" or exceed
some value for "complexity". But when I've had such things readily
available I didn't make great use of them.

Hmm, that's a nice-sounding metric, as it's really simple to calculate
for typical programming languages, and probably is useful for the
purpose you state.

You're right that most people aren't going to be tracking such info
constantly, but still, every once in a while one tends to go into
"spring-cleaning mode", and it's nice to have such tools available then.

-Miles

 

[Bill Davy]

Hmm, that's a nice-sounding metric, as it's really simple to calculate
for typical programming languages, and probably is useful for the
purpose you state.

You're right that most people aren't going to be tracking such info
constantly, but still, every once in a while one tends to go into
"spring-cleaning mode", and it's nice to have such tools available then.

-Miles

I use http://www.campwoodsw.com/sourcemonitor.html (donations welcome) if I
have time. It is easy to set up and use. Then just check the functions
with the worst stats. I wish there was some way of making it happier about
large switch statemenst (as used to implement FSM quite cleanly and simply).

And although it does not produce "metrics" as such (imho), PC-Lint is a very
worthwhile tool as well.

 

[Joe keane]

I've got similar opinions. Such metrics might be handy for spotting
potential problems; "list all the methods that are >100 LOC" or exceed
some value for "complexity".

As well; "nm -s" works fine for me. I'm not a big fan of 'all gt <NoIS>
are wrong'; some functions should be, but it is nice to point out when
asomething has escaped your notice.

 

[Nick Keighley]

As well; "nm -s" works fine for me.  I'm not a big fan of 'all gt <NoIS>
are wrong'; some functions should be, but it is nice to point out when
asomething has escaped your notice.

what I meant to be my point. A 100LOC function isn't neccesarily
wrong. But its a warning notice and should be looked at hard.

 

[ILostMyKeys]

what I meant to be my point. A 100LOC function isn't neccesarily
wrong. But its a warning notice and should be looked at hard.

Why are you here?

 

[Nick Keighley]

Why are you here?

to learn a bit of C, to share my knowledge of C, to discuss
programming issues based around C. There are some very smart
programmers on this ng.

Why are you here?

 

[Henrik Faber]

to learn a bit of C, to share my knowledge of C, to discuss
programming issues based around C. There are some very smart
programmers on this ng.

Why are you here?

That's kind of obvious. He lost his keys and is looking in this NG.

Henrik