Unit tests ... not just for the young

[snip]

Are you working alone on all your projects and no one ever has
to modify or re-factor your code? If no, do they have the
same defect rate as you are when working with your code? What
about yourself when you have to work on your code a year later?

Yura.

···

-----Original Message-----
From: Sean O'Dell [mailto:sean@celsoft.com]
Sent: Monday, June 07, 2004 12:35

It's hard to describe my technique. I just have a lot of
little habits that add up to a lot of smooth sailing.

Nowadays, I work alone mostly. When I work with others, we all have the same
type of habits, more of less. I haven't worked with a programmer younger
than 30+ in a long time, though. I have about the same defect rate as the
older programmers I've worked with. I've worked with some fresh guys right
out of college that had what I would a disastrous defect rate. In fact, I
don't recall a college-educated programmer at any place I've worked that
DIDN'T get fired for being highly unproductive...perhaps one, maybe two. I
return to ancient code all the time and it's fine. In fact, about a year
ago, I ported a good-sized C++ application (non-GUI) that was written for
VC++/MFC about 9 years ago to Linux and it took me a couple days to get
ported. I even got a pretty well-rounded C++ class library out of the port
that is almost completely portable between Linux and Windows; strings,
arrays, avl trees, file I/O, sockets, threads, mutexes...all sorts of stuff,
all portable.

  Sean O'Dell

···

On Monday 07 June 2004 16:13, Kloubakov, Yura wrote:

> -----Original Message-----
> From: Sean O'Dell [mailto:sean@celsoft.com]
> Sent: Monday, June 07, 2004 12:35
>
> It's hard to describe my technique. I just have a lot of
> little habits that add up to a lot of smooth sailing.

[snip]

Are you working alone on all your projects and no one ever has
to modify or re-factor your code? If no, do they have the
same defect rate as you are when working with your code? What
about yourself when you have to work on your code a year later?

I've worked with some fresh guys right
out of college that had what I would a disastrous defect rate.

How many have you worked with that had no college degree, but the same
amount of (in)experience as your grads?

Don't paint college education as the source of those problems.

Experience in what though? I've seen lots of college grads that couldn't even
perform the simplest of programming tasks very well, but NEVER a non-college
grad. The worse I've seen with the work-experienced hires are their skills
are sharp in something we didn't need, so they were let go. I've seen some
good college grad programmers, though. The worst coders I've seen, though,
the rock-bottom worst, were college grads. But that's usually because they
got the job when they were highly inexperienced, and placed into positions
where we expected a little something of them. They really should have gotten
a QA job or something somewhere and coded in their spare time to ramp up.
Non-college grads, though, never got hired unless they could prove their
experience.

  Sean O'Dell

···

On Tuesday 08 June 2004 07:55, Michael Campbell wrote:

> I've worked with some fresh guys right
> out of college that had what I would a disastrous defect rate.

How many have you worked with that had no college degree, but the same
amount of (in)experience as your grads?

Don't paint college education as the source of those problems.

Non-college grads, though, never got hired unless they could prove their
experience.

Sounds like the fault of those hiring.

I think my last truly bad experience regarding people with CS degrees goes
back to early/mid-90's. Also, I've been mostly solo since about 1999, so
things could be a lot different in real "companies" these days.

  Sean O'Dell

···

On Tuesday 08 June 2004 10:03, Michael Campbell wrote:

> Non-college grads, though, never got hired unless they could prove their
> experience.

Sounds like the fault of those hiring.

> > Non-college grads, though, never got hired unless they could prove their
> > experience.
>
> Sounds like the fault of those hiring.

I think my last truly bad experience regarding people with CS degrees goes
back to early/mid-90's. Also, I've been mostly solo since about 1999, so
things could be a lot different in real "companies" these days.

Maybe it depends on what your are doing, but I'm afraid for some
things it hasn't changed much. My belief (don't flame me) is that
those with inexperience do OK when they are working on things that
give visible feedback, like GUI's and such, but take number
crunching or data processing where errors don't readily
reveal themselves (unless you code defensively), then they tend to make
a lot of mistakes and/or leave behind a bunch of land mines.

···

On Wednesday, 9 June 2004 at 3:52:21 +0900, Sean O'Dell wrote:

On Tuesday 08 June 2004 10:03, Michael Campbell wrote:

  Sean O'Dell

--
Jim Freeze
Does the name Pavlov ring a bell?

Maybe it depends on what your are doing, but I'm afraid for some
things it hasn't changed much. My belief (don't flame me) is that
those with inexperience do OK when they are working on things that
give visible feedback, like GUI's and such, but take number
crunching or data processing where errors don't readily
reveal themselves (unless you code defensively), then they tend to make
a lot of mistakes and/or leave behind a bunch of land mines.

I wouldn't necessarily disagree, but the absence or presence of a CS
(et. al.) doesn't necessarily skew that heavily in one way. I've been
in this industry a long time too, and I've seen this assertion made a
lot of times, almost exclusively by people w/o the degree they're
blaming poor workmanship on.

There's truth behind the assertion.

Degrees look GREAT in interviews. Given a young, sharply-dressed person with
a degree, compared with a middle-aged programmer with loads of experience but
no degree, people just get gooey about youth and pedigree. It's human
nature.

  Sean O'Dell

···

On Tuesday 08 June 2004 12:12, Michael Campbell wrote:

> Maybe it depends on what your are doing, but I'm afraid for some
> things it hasn't changed much. My belief (don't flame me) is that
> those with inexperience do OK when they are working on things that
> give visible feedback, like GUI's and such, but take number
> crunching or data processing where errors don't readily
> reveal themselves (unless you code defensively), then they tend to make
> a lot of mistakes and/or leave behind a bunch of land mines.

I wouldn't necessarily disagree, but the absence or presence of a CS
(et. al.) doesn't necessarily skew that heavily in one way. I've been
in this industry a long time too, and I've seen this assertion made a
lot of times, almost exclusively by people w/o the degree they're
blaming poor workmanship on.

Sean O'Dell wrote:

Degrees look GREAT in interviews. Given a young, sharply-dressed
person with
a degree, compared with a middle-aged programmer with loads of
experience but
no degree, people just get gooey about youth and pedigree. It's human
nature.

Some of the best programmer's I've had the pleasure of working with had
degrees in music.

Maybe there's a hidden relationship there! :slight_smile:

Curt

Sean O'Dell wrote:

···

On Tuesday 08 June 2004 12:12, Michael Campbell wrote:

Maybe it depends on what your are doing, but I'm afraid for some
things it hasn't changed much. My belief (don't flame me) is that
those with inexperience do OK when they are working on things that
give visible feedback, like GUI's and such, but take number
crunching or data processing where errors don't readily
reveal themselves (unless you code defensively), then they tend to make
a lot of mistakes and/or leave behind a bunch of land mines.

I wouldn't necessarily disagree, but the absence or presence of a CS
(et. al.) doesn't necessarily skew that heavily in one way. I've been
in this industry a long time too, and I've seen this assertion made a
lot of times, almost exclusively by people w/o the degree they're
blaming poor workmanship on.

There's truth behind the assertion.

Degrees look GREAT in interviews. Given a young, sharply-dressed person with a degree, compared with a middle-aged programmer with loads of experience but no degree, people just get gooey about youth and pedigree. It's human nature.

  Sean O'Dell

How about a middle-aged programmer with loads of experience AND a degree? :wink:

Gennady.

> I wouldn't necessarily disagree, but the absence or presence of a CS
> (et. al.) doesn't necessarily skew that heavily in one way. I've been
> in this industry a long time too, and I've seen this assertion made a
> lot of times, almost exclusively by people w/o the degree they're
> blaming poor workmanship on.

There's truth behind the assertion.

Degrees look GREAT in interviews. Given a young, sharply-dressed person with
a degree, compared with a middle-aged programmer with loads of experience but
no degree, people just get gooey about youth and pedigree. It's human
nature.

Again, that says nothing about the degree; it says something about the
interviewer and/or the interview process.

Gennady wrote:

>
How about a middle-aged programmer with loads of experience AND a
degree? :wink:

Hey, now you've got two counts against you! (welcome to my neighborhood! :slight_smile:

Curt

Some of the best programmer's I've had the pleasure of working with had
degrees in music.

Maybe there's a hidden relationship there! :slight_smile:

Not hidden at all; I don't have the link handy, but there was a study
done by Arthur Andersen years ago that showed a very high correlation
between success at "high logic" jobs (mathematician (et. al;
statistician, logician, etc.), computer science, engineering, and so
forth) and musical aptitude/training.

Too, and this might be urban legend, but I've read that the only
fields that have shown quantified "idiot savants" are mathematics and
music.

Good to go! My experiences haven't been bad with experienced programmers,
just the ones toting nothing more than a degree.

  Sean O'Dell

···

On Tuesday 08 June 2004 12:26, Gennady wrote:

Sean O'Dell wrote:
> On Tuesday 08 June 2004 12:12, Michael Campbell wrote:
>>>Maybe it depends on what your are doing, but I'm afraid for some
>>>things it hasn't changed much. My belief (don't flame me) is that
>>>those with inexperience do OK when they are working on things that
>>>give visible feedback, like GUI's and such, but take number
>>>crunching or data processing where errors don't readily
>>>reveal themselves (unless you code defensively), then they tend to make
>>>a lot of mistakes and/or leave behind a bunch of land mines.
>>
>>I wouldn't necessarily disagree, but the absence or presence of a CS
>>(et. al.) doesn't necessarily skew that heavily in one way. I've been
>>in this industry a long time too, and I've seen this assertion made a
>>lot of times, almost exclusively by people w/o the degree they're
>>blaming poor workmanship on.
>
> There's truth behind the assertion.
>
> Degrees look GREAT in interviews. Given a young, sharply-dressed person
> with a degree, compared with a middle-aged programmer with loads of
> experience but no degree, people just get gooey about youth and pedigree.
> It's human nature.

How about a middle-aged programmer with loads of experience AND a
degree? :wink:

Do you feel I am criticizing people with degrees?

  Sean O'Dell

···

On Tuesday 08 June 2004 12:31, Michael Campbell wrote:

> > I wouldn't necessarily disagree, but the absence or presence of a CS
> > (et. al.) doesn't necessarily skew that heavily in one way. I've been
> > in this industry a long time too, and I've seen this assertion made a
> > lot of times, almost exclusively by people w/o the degree they're
> > blaming poor workmanship on.
>
> There's truth behind the assertion.
>
> Degrees look GREAT in interviews. Given a young, sharply-dressed person
> with a degree, compared with a middle-aged programmer with loads of
> experience but no degree, people just get gooey about youth and pedigree.
> It's human nature.

Again, that says nothing about the degree; it says something about the
interviewer and/or the interview process.

Curt Hibbs wrote:

Some of the best programmer's I've had the pleasure of working with had
degrees in music.

Maybe there's a hidden relationship there! :slight_smile:

Curt

Not all that hidden, actually :slight_smile: There's apparently a significant correlation industry-wide. Can't remember where I heard that, but I do know that, when I worked at Apple, a substantial number of the folks were musicians, ranging from weekend-jammer to seriously gifted.

- dan

···

--
...
... dhtapp is a cox dot net account
...

> Again, that says nothing about the degree; it says something about the
> interviewer and/or the interview process.

Do you feel I am criticizing people with degrees?

Do you feel you're not?

In fact, I don't recall a college-educated programmer at any place
I've worked that DIDN'T get fired for being highly unproductive...
perhaps one, maybe two.

You mention nothing of experience level here, placing the blame solely
on the fact that the person has a degree.

···

On Wed, 9 Jun 2004 04:35:32 +0900, Sean O'Dell <sean@celsoft.com> wrote:

Hmm, I'm no musician, but I consider myself an above average coder.
Also, the worst programmer I knew was a musician. Couldn't program
his way out of a paper bag.

···

On Wednesday, 9 June 2004 at 4:34:27 +0900, Michael Campbell wrote:

> Some of the best programmer's I've had the pleasure of working with had
> degrees in music.
>
> Maybe there's a hidden relationship there! :slight_smile:

Not hidden at all; I don't have the link handy, but there was a study
done by Arthur Andersen years ago that showed a very high correlation
between success at "high logic" jobs (mathematician (et. al;
statistician, logician, etc.), computer science, engineering, and so
forth) and musical aptitude/training.

--
Jim Freeze

I'm American, and I never finished college, but I've met Europeans who graduated with a CS degree in 3 years through more efficient programs, with what is essentially less education than I have - And they're fully qualified and working.

I had , if I remember right, the equivalent of 3 or 4 years worth of credit hours, and I was still a good two years away from getting my piece of paper.

The USA's school system is hobbled by lengthy bureaucracy and politics. The money is what dominates, and most schools will keep you as long as they can, just for the money. The private schools are better, but even they must do what the accreditation boards dictate...And therein lies the inescapable bureaucracy, politics, and technological inertia that makes US schools non-competitive with the best and most agile foreign schools.

As a broad generalization, I think I can safely say that most educational systems, anywhere in the world, are either hopelessly broken, or out of touch with reality. Not to say they have no value, they're just not as good as they could be, even the best of them. That said, I'd still give my left arm to be able to be a college student again, even in a crappy public USA school.

Sean O'Dell wrote:

···

On Tuesday 08 June 2004 12:31, Michael Campbell wrote:

I wouldn't necessarily disagree, but the absence or presence of a CS
(et. al.) doesn't necessarily skew that heavily in one way. I've been
in this industry a long time too, and I've seen this assertion made a
lot of times, almost exclusively by people w/o the degree they're
blaming poor workmanship on.

There's truth behind the assertion.

Degrees look GREAT in interviews. Given a young, sharply-dressed person
with a degree, compared with a middle-aged programmer with loads of
experience but no degree, people just get gooey about youth and pedigree.
It's human nature.

Again, that says nothing about the degree; it says something about the
interviewer and/or the interview process.

Do you feel I am criticizing people with degrees?

  Sean O'Dell