Board index » delphi » Re: Cargo Cult Programming (return of the revenge)

Re: Cargo Cult Programming (return of the revenge)


2006-11-05 03:47:01 AM
delphi70
Jim Rowell writes:
...
Quote
That's one reason I was asking for examples. Your's all have to do
with simple speed.
Speed is just the result, the thing you can measure. More important is what
you see when a good developer is at work. The stuff simply flows, the brain
is not doing any grunt work as the main task. You are faster, but the point
is that you can thing in larger pictures without fighting with the basic
steps. But yes, for the project speed is exactly what matters. If someone is
10 times faster to deliver the same high quality code, then take him!
Quote
There are many other things one can bring to a
team and I have seen many examples in my life where a person was
phenomonal if just allowed to start slow and ramp up. That was just
the way their brain worked. They ended being way (I mean /way/)
better than typical.
In my experience, it is _very_ rare that someone learns full imersion high
speed mode in a very specialized area when he becomes 25 or older. It is a
bet I'd not take, because statistics (ok, gut feeling) says that I will
probably lose. In my whole life, I have not seen one single person who
called himself a developer, had only some basic knowledge, and was not able
to get a problem at once and solve it, and then transformed to a star
developer. Someone I can give a problem, knowing that he will come back with
a solution in 15 minutes, and working code in a day - max. Someone who asks
only the questions that have do be asked. And no, I don't want to work with
average developers. The factor 10 is common in the industry, and I have to
pay maybe 30% more for someone who is at least 3 times faster than the
average. Funny enough, some of the best developers I know cost less than
many of the mediocre type. And I don't buy anybody who is less good than 3x
average, no matter how cheap. that is like begging for problems.
Ralf
 
 

Re: Cargo Cult Programming (return of the revenge)

Jim Rowell writes:
Quote
I used to think I would make a great interviewer but not anymore. I've
seen too many really good people that couldn't interview and too many
{*word*99}py that could and too many poor managers where it didn't matter
anyway. <g>
That didn't come out right. I meant good people that couldn't handle /being/
interviewed.
 

Re: Cargo Cult Programming (return of the revenge)

Quote
In my experience, it is _very_ rare that someone learns full imersion high
speed mode in a very specialized area when he becomes 25 or older.
I don't think age is the problem. I usually take longer to understand things
(has always been the case) because I want to know the basics, and simply can't
use information that is somehow contradictory, incomplete/inaccurate.
I have to see it written to play with it, think about some fundamental stuff,
even sometimes I look for the etymology of words/terms that are used. Often I
look for more descriptive names as it is so common that terms are too short or
synonyms artificially made different.
Once you know that, you can slip into this idea/theory/problem like in a body
and move its arms legs, bend, jump and dance like you want.
The first step takes time, sometimes a lot. After that you are incredebly fast,
and can often guess only by the symptoms of a bug what you have to correct.
Without this time you become less flexible, because you are merely applying
techniques all the time instead of thinking about the nature of things. With 25
and above you are more likely to do routine work, which is perhaps why you have
observed such "immersion problems".
 

Re: Cargo Cult Programming (return of the revenge)

Mal Hrz writes:
...
Quote
The first step takes time, sometimes a lot. After that you are
incredebly fast, and can often guess only by the symptoms of a bug
what you have to correct.
That works if you are familiar with the structures, the scaffold. It's
easier to learn developing when you studies math or physics. It is way
harder when you learned something where pure logic is not a common tool. And
it is harder when you are not familiar with that immersion state. If you
never worked with abstract ideas and only pen and paper or a computer
available in way where you just flow through the structures, then, after you
reach an age where flexibility becomes harder, you will most probably never
learn it. There are always exceptions, of course. But they are _very_ rare,
and I'd never bet on them. Same with development: if you don't know how
to solve all the small problems (the flow, describe it as you want) while
you code in eg. 3 years, then you will most probably never get it. And that
is the average developer.
Ralf
 

Re: Cargo Cult Programming (return of the revenge)

"Dan Downs" <XXXX@XXXXX.COM>writes
Quote
How about a course on debugging, the prof hands out broken programs (say
from other students in lower classes) and they have to make it work. It'll
probably be a good thing to have their noses rubbed in some strangers code
for awhile, then it might click as to why there are certain ways you want
and should do things.

Wow, now there is a really creative idea. I will have to
think about how to take advantage of that.
--
Thanks,
Brad.
 

Re: Cargo Cult Programming (return of the revenge)

On Fri, 3 Nov 2006 16:20:12 -0500, "Dan Downs"
<XXXX@XXXXX.COM>writes:
Quote
How about a course on debugging, the prof hands out broken programs (say
from other students in lower classes) and they have to make it work.
Sounds similar to a class I heard about where the prof divided the
class into two, each one should implement a program (the same one for
each group, that is). Then after they were done, they had to extend
the program to include new features. Only they had to extend the other
groups program :)
Such things should be mandatory, imho.
- Asbjrn
 

Re: Cargo Cult Programming (return of the revenge)

Wayne Niddery [TeamB] writes:
Quote
John Jacobson <jake[nospam]@jsnewsreader.com writes:
>Look at what we do to the young mind.

Jake, while I have similar views on this point, it is really off-topic here.
Issues of faith, pro or con, need to be confined to off-topic or appropriate
usenet groups.

I don't get this?
It is on topic. John is right. The basics necessary for math and reading
are not at the levels they were 30 years ago when I went through
University. This has nothing to do with issues of specific beliefs, but
all with the ability to reason. The SATs and other tests have been
skewed over the years to cover up the loss of skills. Managers are not
any better as I am a firm believer in the Peter Principle.
See this for some facts:
mwhodges.home.att.net/new_96_report.htm
James Gibbons
www.folding-hyperspace.com/
 

Re: Cargo Cult Programming (return of the revenge)

James Gibbons" <"james at ventek-inc.com writes:
Quote

It is on topic. John is right.
Discussing reasoning skills, education standards, etc, as applied to
programming, is fine. Advocating or denigrating religion is not.
--
Wayne Niddery - Winwright, Inc (www.winwright.ca)
"True peace is not the absence of tension, but the presence of
justice." - Martin Luther King, Jr.
 

Re: Cargo Cult Programming (return of the revenge)

"Wayne Niddery [TeamB]" <XXXX@XXXXX.COM>writes
<XXXX@XXXXX.COM>
Quote
James Gibbons" <"james at ventek-inc.com writes:
>
>It is on topic. John is right.

Discussing reasoning skills, education standards, etc, as applied to
programming, is fine. Advocating or denigrating religion is not.
Unfortunately, that makes any real discussion of the problem impossible, at
least in relation to a country like the United States. There ARE reasons why
logic is not easy for most people, reasons that have nothing to do with innate
abilities. When you can not talk about the biggest reason, we might as well just
give up and become potato farmers.
--
***Free Your Mind***
Posted with JSNewsreader Preview 0.9.4.2921
 

Re: Cargo Cult Programming (return of the revenge)

John Jacobson writes:
Quote
When you can not talk about the biggest reason, we might as well just
give up and become potato farmers.
You are free to talk about it all you like, Jake. Just not here.
--
Nick Hodges
Delphi/C# Product Manager - Borland DTG
blogs.borland.com/nickhodges
 

Re: Cargo Cult Programming (return of the revenge)

John Jacobson <jake[nospam]@jsnewsreader.com writes:
Quote

Unfortunately, that makes any real discussion of the problem
impossible,
No it doesn't. You can talk about irrationality and illogic, again regarding
programming, without specifically bringing religion into it. Many people are
irrational and illogical for all sorts of different reasons that have
nothing to do with religion. The *why* of those conditions is properly
off-topic here (whether you wish to blame religion, government, the
education system, or Martians). You can discuss the what, and you can
discuss the *consequences* of those conditions.
--
Wayne Niddery - Winwright, Inc (www.winwright.ca)
"The only reason some people get lost in thought is because it's
unfamiliar territory." - Paul Fix
 

Re: Cargo Cult Programming (return of the revenge)

The human 'belief system' is the bigger set, 'religion' is only a
subset, so you could discuss in the higher terms, keeping away from the
subset.
Programmers have their 'beliefs' about how
computers/compilers/languages (should) work. No connection to other
belief subsets necessary.
--
Anders Isaksson, Sweden
BlockCAD: web.telia.com/~u16122508/proglego.htm
Gallery: web.telia.com/~u16122508/gallery/index.htm
 

Re: Cargo Cult Programming (return of the revenge)

"Wayne Niddery [TeamB]" <XXXX@XXXXX.COM>writes
<45514eef$XXXX@XXXXX.COM>
Quote
No it doesn't. You can talk about irrationality and illogic, again regarding
programming, without specifically bringing religion into it. Many people are
irrational and illogical for all sorts of different reasons that have
nothing to do with religion. The *why* of those conditions is properly
off-topic here (whether you wish to blame religion, government, the
education system, or Martians). You can discuss the what, and you can
discuss the *consequences* of those conditions.
Religion is the largest source of irrationality, according to my religious
beliefs. Being unable to discuss religion as a possible source is restricting
my speech because of my religion. That is religious discrimination, which is
perfectly legal in a private forum such as this as far as I know. After all,
contrary to popular misconception, free speech is not a fundamental universal
right here in the United States.
But I am not fooled into thinking this is a full and proper discussion of why it
is so hard to find good programmers that can think logically.
--
***Free Your Mind***
Posted with JSNewsreader Preview 0.9.4.2939
 

Re: Cargo Cult Programming (return of the revenge)

John Jacobson writes:
Quote

Religion [snip]
Post cancelled, entirely off-topic.
I won't argue this with you any further Jake.
--
Wayne Niddery - Winwright, Inc (www.winwright.ca)
"True peace is not the absence of tension, but the presence of
justice." - Martin Luther King, Jr.
 

Re: Cargo Cult Programming (return of the revenge)

Anders Isaksson <i.rather@not>writes
<4551a2bc$XXXX@XXXXX.COM>
Quote
The human 'belief system' is the bigger set, 'religion' is only a
subset, so you could discuss in the higher terms, keeping away from the
subset.

Programmers have their 'beliefs' about how
computers/compilers/languages (should) work. No connection to other
belief subsets necessary.
But in the real world no set of beliefs occur in the vacuum. For example, if
you look at the non-science views of Nobel Prize scientists you find that they
differ from that of the public at large, significantly, on specific issues that
they themselves feel science can't or does not address at all. It seems
reasonable to think that software developers are somewhat similar in the way
their ability to think logically (the top skill required for programming) more
than likely also has a similar correlation with those same specific issues that
scientists feel science does not or can't address.
To be more blunt about it, I don't think it is just a coincidence that Nobel
Prize winners in science also are more than likely to share the same set of
negative beliefs about X. I think it occurs because the way of thinking that
causes one to excel in science also leads one to reject X. I bet if you polled
top programmers, you would also find them more likely to reject X, because the
way of thinking that causes one to excel in programming also causes one to tend
to reject X.
Now X can be any number of a set of things. I know what I think is in X, but it
is obvious that some others in this newsgroup disagree. The point though is
that there does appear to be a set X of non-programming beliefs that
nevertheless are highly negatively correlated with programming ability. We may
not be allowed to say what we think is in X, but some of have observed it, or
so we believe anyway.
--
***Free Your Mind***
Posted with JSNewsreader Preview 0.9.4.2921