M
Matt Garrish
Lothar Scholz said:
I have to assume from your response that your English is not that strong,
hence your misunderstanding. I would suggest you look up the word emblematic
in a dictionary.
Matt
I have to assume from your response that your English is not that strong,
hence your misunderstanding. I would suggest you look up the word emblematic
in a dictionary.
Matt
C
Chris
Walter said:
Certainly a point worthy of consideration... I certainly didn't mean to
generalize managers in the short disoourse at the head of this
sub-thread. In fact, I see the very same points I made in the area of
programming equally applicable to management. 7-10 years of "management
experience" doesn't make one a great manager. It goes right back to raw
talent and the person again. I've found that a lot of the great
managers were the ones that knew good and well they weren't worth an
auto-increment in development; they get out of the way and let their
developers do their jobs. They are just as good at SEEING a coming
political or resource management problem as well as any great developer
guards against deadlocks...
So, the point, I believe, remains. The really good ones SEE past the
details (TQM, XP groups, TDM, RAD, etc) and are able to adapt and apply
using various tools be they management or development persons. Good
managers aren't good because they took a course on TQM and know TQM
buzzwords; they are good beecause they use the best tool necessary for
the job and aren't lost every time their company "reorgs" according to
the latest management fad or the next new thing the CEO or COO wants to
"try." Nor should they (nor ARE they) disqualified from a management
position at another company (usually not) because they haven't applied
TQM or XP (eg. extreme programming) at the last company they managed.
I feel this is the EXACT point being made. How much of a difference is
there between C, Perl, PHP, JavaScript, C++, C# and Java in and of
themselves? REALLY? Not much. And I can use an API reference as well
as anyone. And face it, PHP, .NET (C#), C++, and Java are HUGE. Very
few persons, even those using either of these "for years" can POSSIBLY
know all the classes, calls and interface possibilities and combinations
in ANY ONE of them. Again *some* merit has to be given to someone with
prolonged exposure, but how many good, comptent people are turned away
simply because their exposure is a bit less, but their potentiality,
adaptability, raw talent and true development ability (it's abstract
talent EXPRESSED in different so called "languages") would easily (in
very short order) outweigh someone that's just "done their time" but
doesn't really have a real clue about underlying technology?
I guess I am venting a little bit because I see this especially in the
..NET and J2EE arenas where I believe someone that is thoroughly intimate
with the "ins" and "outs" of raw web technology, IP, middleware, remote
proxies, and such infrastructure isn't fooled one bit by the smoke and
mirrors of .NET. And yet I see these ads where you're not going to get
a sniff if you've never fired up Visual Studio .NET and worked in that
IDE for at least 2 years. Please!! :-(
Chris
M
Marc Bissonnette
Indeed, I mirror Charleton's opinion/experience. Before starting my own
business, I was a project / middle manager for a couple of ISPs. I'd seen
a fair share of programmers with computer science degrees, as well as
those who were completely self-taught with *totally* unrelated college
experience (my own college experience, for example, was in heavy
equipment diesel mechanics) - the self-taught folks almost always
outshone the college folk for the simple fact (IMO) that they *loved*
what they did (I'm sure this is true of some college/university folk as
well, of course)
The self-taught folk tend to go out and learn what works and what works
best, whereas the college folks tended to do what the books told them to
- which isn't always what's best... (reminded of the thread involving
George Reese, Perl and MySQL...)
Apologies for blanket statements and recognition to the exceptions, of
course
J
Jochen Buennagel
Walter said:
I was thinking about the direct superior of the programmer, as in "team
leader" or "project manager". Also it depends on the organization: If
the ultimate goal of the company is to produce good software, then
managers are often interchangable while good programmers aren't. Even
the best manager can't succeed in this endeavor with bad programmers,
while good programmers can't be held back even by an averagly-abled manager.
Jochen
W
Walter Roberson
:I was thinking about the direct superior of the programmer, as in "team
:leader" or "project manager". Also it depends on the organization: If
:the ultimate goal of the company is to produce good software, then
:managers are often interchangable while good programmers aren't. Even
:the best manager can't succeed in this endeavor with bad programmers,
:while good programmers can't be held back even by an averagly-abled manager.
That's a faulty comparison. In the first branch, you have
"best" manager and "bad" programmers; in the second branch, you have
"good" programmers and "average" manager. "average" is a very different
category than "bad". Can the best manager succeed with average
programmers? Can the good programmers succeed with a bad manager?
My organization can always skim some bright young programmers off of
the local universities (and if not, we import 'em from other countries).
Finding pretty good programmers isn't hard. But good productive
programmers aren't necessarily all that good at analysis, only
at producing new code. A manager who -is- pretty good at analysis can
make a great difference in the overall results, even if the manager
isn't a programming whiz-kid. And finding technical people who are
pretty good at analysis is a lot harder than finding pretty good
programmers.
:leader" or "project manager". Also it depends on the organization: If
:the ultimate goal of the company is to produce good software, then
:managers are often interchangable while good programmers aren't. Even
:the best manager can't succeed in this endeavor with bad programmers,
:while good programmers can't be held back even by an averagly-abled manager.
That's a faulty comparison. In the first branch, you have
"best" manager and "bad" programmers; in the second branch, you have
"good" programmers and "average" manager. "average" is a very different
category than "bad". Can the best manager succeed with average
programmers? Can the good programmers succeed with a bad manager?
My organization can always skim some bright young programmers off of
the local universities (and if not, we import 'em from other countries).
Finding pretty good programmers isn't hard. But good productive
programmers aren't necessarily all that good at analysis, only
at producing new code. A manager who -is- pretty good at analysis can
make a great difference in the overall results, even if the manager
isn't a programming whiz-kid. And finding technical people who are
pretty good at analysis is a lot harder than finding pretty good
programmers.
J
Jochen Buennagel
Eric said:
If the above points where all there is about "Software Crafsmanship",
then it would indeed be totally superfluous (sp?), and I would certainly
not recommend reading it.
The book adresses the still common misconception that writing software
is an engineering discipline, embodied in the term (and process) of
"software engineering". McBreen's main point is that programming is a
craft in the old fashioned sense of the word. (E.g. it should not be
tought at a university but by a master to an apprentice.)
I recommend this book especially if you've read and agree with "Mythical
Man-Month" (or "Pragmatic Programmer", or DeMarco's works for that matter).
Jochen
G
G Klinedinst
Marc Bissonnette said:
I have had vastly different experience personally. In my experience
self-taught people tend to be more one-dimensional whether their
specialty is networking, programming, etc. Also they tend to not
follow basic principles of programming(modularizing code, etc), which
trained programmers do because it was hammered into them for years at
school. On the other hand I work/live in a computer tech center of the
country where a lot of great programmers flock, so it may not be
representative of the country as a whole. Where I went to school there
were certainly CS majors who were just there to get a piece of
paper(one senior in CS couldn't install his own CD-ROM drive).
I am going to try to say this without generalizing so stick with me:
This statement is what worries me. People out there programming and
doing what "they" think is best. Remember that for the most part the
books are written by people who have dedicated their lives to the
understanding of information processing. Having someone doing what
"they" think works rather than what is taught in the books doesn't
strike me as necessarily a good thing, unless they can cite a specific
reason where their solution is clearly better.
On the other hand I respect both your(Marc) and Charlton's
experiences. I haven't worked at enough companies to be able to make a
generalization about skill level across the board. If you feel more
comfortable with non-CS people then do what you feel is the right
thing for your business.
-Greg
G
G Klinedinst
Charlton Wilbur said:
You are right. I was speaking for people with degrees. OTOH I am a
proponent of AS or BS of CS degree for programmers(I don't think it's
ever a BA but I could be wrong), so the bias was intentional. I was
trying to imply that aspiring programmers _should_ get formal training
in CS. I go into that a little in a reply post I made to Marc who
replyed to your post also. If you have a minute to read it take a
look, because I talk a little bit about why I am biased that way.
-Greg
C
Charlton Wilbur
JB> - It is possible for an ungifted programmer to actually
JB> decrease the overall productivity of a development team. (As
JB> the old saying goes: "Hiring him is like losing two good
JB> men.")
I've worked with people like that. I've also burned out and *been*
someone like that.
JB> Once the above sinks in, hopefully some companies will
JB> realize:
JB> - It is logical that a good programmer can (and should) earn
JB> more than the manager he's working under.
JB> This last point will be the hardest for managers to
JB> accept. After all they live in a world where the hierarchy
JB> must always reflect the compensation.
Well, one of the other lessons that companies need to learn is that
you get what you reward. If you reward politicking and ass-kissing
more than technical competence -- and in places where the only way to
advance is to be promoted to management, politicking and ass-kissing
often get rewarded substantially more than technical competence --
then you get a company full of politicians and ass-kissers.
Charlton
JB> decrease the overall productivity of a development team. (As
JB> the old saying goes: "Hiring him is like losing two good
JB> men.")
I've worked with people like that. I've also burned out and *been*
someone like that.
JB> Once the above sinks in, hopefully some companies will
JB> realize:
JB> - It is logical that a good programmer can (and should) earn
JB> more than the manager he's working under.
JB> This last point will be the hardest for managers to
JB> accept. After all they live in a world where the hierarchy
JB> must always reflect the compensation.
Well, one of the other lessons that companies need to learn is that
you get what you reward. If you reward politicking and ass-kissing
more than technical competence -- and in places where the only way to
advance is to be promoted to management, politicking and ass-kissing
often get rewarded substantially more than technical competence --
then you get a company full of politicians and ass-kissers.
Charlton
J
Jeffrey Silverman
Exactly. My wife is from PGH. The only people I have met that omit
the verb "to be" from statements like "That needs to be washed" or "The
book needs to be returned" ar people from Pittsburgh.
"That needs washed"
"That needs returned"
to be! to be!
I'm used to it, at this point, at the very least.
later...
C
Charlton Wilbur
GK> On the other hand I respect both your(Marc) and Charlton's
GK> experiences. I haven't worked at enough companies to be able
GK> to make a generalization about skill level across the
GK> board. If you feel more comfortable with non-CS people then do
GK> what you feel is the right thing for your business.
I think the final determination is made by considering the person
himself or herself. I think that to be a good programmer, you need a
mix of good theory, good practice, and good team habits; you can learn
all three in college, and you can learn all three by being
self-taught.
Since you raise (in the part you snipped) a point about breadth of
knowledge: one of the major deficiencies I've seen in college-educated
programmers is an inability to learn effectively outside the
classroom, without a professor providing a syllabus and regular tests.
Anyone who is self-taught obviously does not have this problem, since
they learned what they know outside that context. In the long run,
this is a major advantage, and leads to breadth of knowledge.
A related point, though, is perspective: one of the things that you
ought to come away from a college education in programming with is the
notion that different problems have different solutions. The notion
that PHP or Perl is the cure to all ills, for instance, is something
that self-taught programmers have, usually because PHP or Perl is the
only language they have any serious experience with. A college-
educated programmer (if the college is worth its salt, at least) ought
to have played with at least a dozen languages among various
paradigms, and ought to have a sense that there's more out there than
Perl and PHP and C and Java, even if he or she doesn't actually use
them on a regular basis.
And I think the core problem is that a good CS education doesn't
always come with a nice parchment certificate, and a nice parchment
certificate doesn't always indicate the presence of a good CS
education.
Charlton
GK> experiences. I haven't worked at enough companies to be able
GK> to make a generalization about skill level across the
GK> board. If you feel more comfortable with non-CS people then do
GK> what you feel is the right thing for your business.
I think the final determination is made by considering the person
himself or herself. I think that to be a good programmer, you need a
mix of good theory, good practice, and good team habits; you can learn
all three in college, and you can learn all three by being
self-taught.
Since you raise (in the part you snipped) a point about breadth of
knowledge: one of the major deficiencies I've seen in college-educated
programmers is an inability to learn effectively outside the
classroom, without a professor providing a syllabus and regular tests.
Anyone who is self-taught obviously does not have this problem, since
they learned what they know outside that context. In the long run,
this is a major advantage, and leads to breadth of knowledge.
A related point, though, is perspective: one of the things that you
ought to come away from a college education in programming with is the
notion that different problems have different solutions. The notion
that PHP or Perl is the cure to all ills, for instance, is something
that self-taught programmers have, usually because PHP or Perl is the
only language they have any serious experience with. A college-
educated programmer (if the college is worth its salt, at least) ought
to have played with at least a dozen languages among various
paradigms, and ought to have a sense that there's more out there than
Perl and PHP and C and Java, even if he or she doesn't actually use
them on a regular basis.
And I think the core problem is that a good CS education doesn't
always come with a nice parchment certificate, and a nice parchment
certificate doesn't always indicate the presence of a good CS
education.
Charlton
C
Charlton Wilbur
WR> That's a faulty comparison. In the first branch, you have
WR> "best" manager and "bad" programmers; in the second branch,
WR> you have "good" programmers and "average" manager. "average"
WR> is a very different category than "bad". Can the best manager
WR> succeed with average programmers? Can the good programmers
WR> succeed with a bad manager?
The quality of the manager makes an incredible amount of difference,
in my experience; much more so than the quality of the individual
programmers. (Of course, this may be because I've been among the
individual programmers on a small team, and so the quality of the
programmers is somewhat consistent.)
So I'd answer your questions thus: the best manager can do a lot with
averageprogrammers; the good programmers with a bad manager might as
well seek employment elsewhere, because group success and job
satisfaction are likely to evade them completely.
Charlton
WR> "best" manager and "bad" programmers; in the second branch,
WR> you have "good" programmers and "average" manager. "average"
WR> is a very different category than "bad". Can the best manager
WR> succeed with average programmers? Can the good programmers
WR> succeed with a bad manager?
The quality of the manager makes an incredible amount of difference,
in my experience; much more so than the quality of the individual
programmers. (Of course, this may be because I've been among the
individual programmers on a small team, and so the quality of the
programmers is somewhat consistent.)
So I'd answer your questions thus: the best manager can do a lot with
averageprogrammers; the good programmers with a bad manager might as
well seek employment elsewhere, because group success and job
satisfaction are likely to evade them completely.
Charlton
C
Charlton Wilbur
GK> You are right. I was speaking for people with degrees. OTOH I
GK> am a proponent of AS or BS of CS degree for programmers(I
GK> don't think it's ever a BA but I could be wrong), so the bias
GK> was intentional.
I am the holder of a BA in computer science and music, so such things
exist. Of course, my college was a liberal arts college, and the
computer science major was in the process of spinning off from the
mathematics major when I studied there.
Charlton
GK> am a proponent of AS or BS of CS degree for programmers(I
GK> don't think it's ever a BA but I could be wrong), so the bias
GK> was intentional.
I am the holder of a BA in computer science and music, so such things
exist. Of course, my college was a liberal arts college, and the
computer science major was in the process of spinning off from the
mathematics major when I studied there.
Charlton