[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Source code statistics ...

From: Andy Levy <andy.levy_at_gmail.com>
Date: Thu, 23 Jul 2009 13:58:59 -0400

On Thu, Jul 23, 2009 at 13:55, Parrish, Ken<KParrish_at_gomez.com> wrote:
> Thank you all for your responses.  Of course, these kinds of statistics are all but useless in any real management or statistical context.  In our case, it is even more complicated because the surface area of our application set is HUGE and involves the creation and deletion of multiple source code branches during development, integration of runtime code developed overseas, all manner of machine generated code, etc., etc.
>
> Fortunately, I believe the purpose for gathering this information is strictly PR related, internal or external--not in any sense a metric for measuring from a business or management perspective.

Using LOC changed or whatever they're measuring here strictly for PR
purposes sounds suspiciously like a "mine's bigger than yours" contest
:)

There's a Dilbert cartoon with a sign hanging outside the Marketing
department reading "2 drink minimum."

> -----Original Message-----
> From: Dallman, John [mailto:john.dallman_at_siemens.com]
> Sent: Thursday, July 23, 2009 1:26 PM
> To: users_at_subversion.tigris.org
> Subject: RE: Source code statistics ...
>
> Andy Levy [mailto:andy.levy_at_gmail.com] writes:
>
>> That's pretty much what I told the PM in my example. Several times
>> over. I told him I judge progress on the number of features/requirements
>> which have been completed successfully, not how many times I hit the
>> Enter key. Even explained that sometimes I'll remove lines along the
>> way because I found ways to re-use stuff I'd written before. Didn't
>> matter. He liked raw LOC numbers. My guess is he was comparing notes
>> with some of his friends about their programmers.
>
> It can be worth pointing out analogies in fields that pointy-haired
> managers might understand a little better. Measuring programmer
> productivity by lines of code is like:
>
> * Measuring the performance of a stockbroker by the number of trades
>  he makes, rather than the amount of profit that comes out of them.
>
> * Measuring the performance of a commedian by the number of jokes in
>  his script, rather than by how many people come to his shows.
>
> * Measuring the performance of a shole salesman by how many people he
>  sees, rather than how many shoes he sells.
>
> Hey, this is fun...
>
> Lines of code are not *completely* useless as a measure: they can give
> you a rough idea of how much disk space you need, how long builds will
> take, and so on.
>
> --
> John Dallman
> Parasolid Porting Engineer
>
> ------------------------------------------------------
> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=2374914
>
> To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
>
> ------------------------------------------------------
> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=2374919
>
> To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
>

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=2374920

To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
Received on 2009-07-23 19:59:50 CEST

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.