[lug] can't make this stuff up, folks... My 2 lines of code, errr, I mean, my 2 cents....

[Nate Duehr]

I took a Java programming class from a former ADA programmer who
had a pet peeve...

"If you can't see syntax bugs from a print-out without leaning on
the compiler to find them for you, you don't understand the
language."

She had a good point. She could find all our bugs with nothing
but a red pen and a printout of our code.  She never had to run a
single lick of it to see if it worked or not.

She was also the instructor who did an hour long session
off-topic one night in class on the possible evils of putting
things that really didn't need to be in an RDBMS, into an RDBMS.


This was long before "privacy" issues were even interesting to
programmers/computer folks, let alone the general public
demanding to know what people were going to do with their
personal information.

She knew where it was headed... and that any database can be
merged with another to find "interesting" or even "private"
information, with a simple OUTER JOIN...

:-)

She spent a while talking about "shopper loyalty" cards as a
possible example, and what someone could figure out from your
grocery list, if the DB data were supoenaed away from the grocery
store... and this was in the mid 90's, if even that late... (I
forget.)
--
  Nate Duehr
  nate at natetech.com


On Tue, 20 Oct 2009 12:08 -0600, "Bear Giles"
<bgiles at coyotesong.com> wrote:

  On Tue, Oct 20, 2009 at 11:15 AM, Gordon Golding
  <[1]gordongoldin at aim.com> wrote:



  Re: can't make this stuff up, folks...  My 2 lines of code,
  errr, I mean, my 2 cents....





  My Dad set up what was at the time the largest data center in
  the world.





  Standard productivity of a big development center like that in
  lines of code written per programmer per day?





  2 lines of code per programmer per day.....   ;-)





Was that back in the day when programmers wrote on gridded paper
that was handed over to typists who punched the actual punch
cards and somebody eventually ran the program and the programmer
got the results in their inbox the next morning?
(In college I used punch cards in my Fortran class but we got the
results back in 15 minutes or so.  We even had accounts and every
run cost a few cents.  It wasn't uncommon for people to run out
of "money" before they got their homework to run.)
Our current productivity shouldn't be that much higher.  Well, a
couple orders of magnitude since we have dedicated workstations
and can usually get nearly-instant results, but closer to a
hundred lines/day than a thousand lines/day.  Remember that
"productivity" includes the time to design, test, document, etc.
Here's today fun question: does "lines of code per day" include
test code?  I know some people who claim that you should spend at
least an hour writing tests for every hour you spend writing
code.  Ideally closer to a 2:1 ratio.

References

1. mailto:gordongoldin at aim.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lug.boulder.co.us/pipermail/lug/attachments/20091020/6f72c310/attachment.html>