I speak now as someone who spent years studying mathematics. I am very familiar with what it takes to really prove anything. You cannot prove that you exist. You cannot prove that arithmetic is consistent. You cannot prove that the Sun will come up tomorrow. You cannot prove that the world is round.

In short if you are unable to learn anything until it is proven to you, then there is very little that anyone will be able to teach you. If your ability to think about the process of learning ends with solid demonstrations, then you will be unable to achieve even the level of sophmoric philosophy, let alone appreciate or participate in something like the scientific process.

Therefore most of our knowledge about the world is provisional. That does not make our knowledge useless, and it does not mean that all knowledge is equally weak. For instance I believe, but cannot prove, that there is such a thing as gravity. I am very confident in that belief, to the point that if I was threatened by being dangled out of a tall window by my ankle, I would scream in terror. I also believe, but cannot prove, that AMD's 64-bit strategy is a better bet than Intel's. That is a rather weak belief, based on a leaning tower of opinions, suppositions, and third-hand information. I am not at all confident in that conclusion and would hasten to add that nobody should go out and short Intel stock or buy AMD stock based on my thinking that.

So I am faced with the need to operate based on provisional information and provisional beliefs. I have no proof and know (ie I am quite certain) that proof is not generally going to be available. My solution to that is to understand and analyze my beliefs so I know what assumptions they depend on, and why I believe them. Then when I talk to people I attempt to indicate not only the statement I believe to be true, but also the degree of certainty I have and (if it seems appropriate) something about my sources so that they can form their own opinions. Furthermore I attempt to test my beliefs. I state them, let people correct me. I try to learn more about the subject. I try to analyze the same situation from different points of view and compare conclusions. I develop my own understanding of what factors are likely to matter more than others.

This is all a good deal less certain and solid than anything you might call "proof". But certainty is cheap - every fanatic is certain. At the opposite extreme, solid facts are often unavailable, by the time you know for sure what the answer is, that information has become useless. Therefore there is real value at being willing to accept, evaluate, and work with ideas, even without having absolute proof of them.

But you like to see proof and examples. So I will give you a concrete example.

Your assertions to the contrary notwithstanding, absolute proof is by no means regarded as necessary in true academic debate. As proof I offer you two examples from math, often supposed to be the ultimate example of absolute proof. The first example in math, look up the debates on constructivism versus formalism early in this the last century. This was a key debate about how to do math, and yet the ultimate conclusions were achieved by rhetoric, not by proof. The second example is the classification of finite simple groups. This is one of the most important proofs of the last century. It was finished in 1983. However within a short time key players in the original proof, notably Daniel Gorenstein, became dubious about their own proof. The current status now, nearly two decades after the proof, is that it is not satisfactorily proven, and there is an active research effort (which involves key players in the original proof) to produce a satisfactory proof. This is despite the fact that there is absolutely no concrete evidence that the original proof is wrong, and nobody has any indication that the result is inaccurate!

Now I don't pretend to know where you aquired your misapprehensions about what makes up true academic debate. But most certainly real academics have a more sophisticated understanding of what makes up academic debate than your simplistic demand for proof. This is true even if you stick to mathematics, which is often cited as the paragon for absolute proof.

UPDATE
Brouwer, Hilbert, et al were debating constructivism in the early 1900's and not a few months ago. (Thanks ar0n for teasing me about it publically.)

OK. My use of the word "think" was inaccurate. I know that using Parse::FixedLength makes code more complex than using pack and unpack. Here's how I justify that statement:

pack and unpack are standard Perl built-in functions. This means that they are a) well documented, b) guaranteed to work on any Perl installation and c) very fast (as they are written in C. In addition to that their interface is based on concepts that I understand.

On the other hand, Parse::FixedWidth is not a standard part of Perl. It needs to be downloaded from CPAN and installed on every system where you want to use it. The documentation is patchy and, as pure Perl code, it isn't particularly fast. Furthermore the interface is based on concepts that seem foreign to me. I therefore find that it adds complexity to my script unnecessarily.

Obviously, there are times when adding complexity is a good idea (otherwise, I'd have no use for any CPAN modules) but the complexity needs to allow me to do things easier. I really can't see how Parse::FixedWidth makes my life any easier. What does it give me that I can't achieve with the standard Perl built-ins?

Perl Training in the UK <http://www.iterative-software.com>

The reason I might use something like Parse::FixedLength (not the module as it stands, though) is that currently I define my fixed width data in one place with something like this (update: and imagine this is 20,50, or 100 fields, which for me makes using just plain unpack unwieldy and unmaintainable without using something like the below):

# fieldname:length
my @format = qw(
 field1:10
 field2:20
 field3:8
);
[download]

Then I have to jump through a couple of hoops (though I admit its not hard) to convert that to an array of just '@fieldnames', and a format string for unpack, and so now my one @format array turns into two arrays and a scalar floating around my code. I think it would save a bit of time and maybe be a bit cleaner to encapsulate that logic in a module (in a way I've suggested earlier), and I just may get around to doing it.

Update: Ok, for good or ill, I've taken princepawn up on his offer to let me take over the Parse::FixedLength module (look for v5.01 in a CPAN near you). Comments/criticisms welcome :-)

How about

my @format = ( #qw deleted
field1=>10,    #other delimiter
field2=>20,
field3=>8
);
#and use as
@fields{keys %format}=unpack"A".join("A",values %format),$_
[download]

?
Update: I know, the Order is not preserved, but I hope you get the Idea.

Runrig's example seems sufficient for the task at hand. Indeed, by the time tilly posted, bitwise had already acknowledged that the example was sufficient to solve the problem being discussed. What more was required in this thread?

If you mean examples of using pack and unpack to parse fixed-length files, well, there are several related nodes already in the archives as well as perlfaq5 (which describes these as the most efficient approach). Are these insufficient? Or is there another issue at work? One unrelated to the current thread?

If the former, what's wrong with posting an SOPW of your own? One outlining the sticky bits of confusion? If the latter, why not carefully compose a Meditation and, as diplomatically as possible, start a discussion on the lack of examples? If neither, then why not discuss the problem with the monks in question via private /msg's?

If you have constructive ideas on how to improve local posting practices, why not consider leading by example? Demonstrate your ideas in your own nodes. Or more simply: treat your fellow monks the way you want them to treat you.

I think many people have gone to great lengths to acknowledge your skills, contributions, and strengths. Most suggestions regarding your work have been, I believe, offered constructively. While some people are less diplomatic than others, it might help to start by thinking that we're all (generally) trying to help.

Granted, there have been some flames, too. The History colors many responses, including mine, yours, and others. However, I've noticed that we tend to find what we're looking for, which may not be what we need to find.

--f

And I can find post after post after post where both of you say something but provide no examples or support as to why what you say is reasonable. But of course since you both are the darlings of Perlmonks, you still get copious up-voting on something which leaves no instruction or enlightenment for posterity and which would have you downvoted in a true academic debate.

Now if it were me, I'd listen close to that man, not debate him.
Of course, there are times when such debate does occur, and at different levels. I probably won't be discussing source hacking anytime soon, but will discuss CGI and some database stuff. Sometimes, ain't nobody wanna talk shop.

Sometimes people just aren't interested in your subject.

Sometimes there's no good interface to the audience

Please excuse this frustrated post, but you people neeed to wake up and get out of your bad habits.

I am posting this publicly because I feel that this time you have gone too far. Last time I sent you such a message, you responded with "yes, but calling chromatic inane was immature and worthy of --".

In this case I agree with everyone else. This case is worthy of -- because you refer to "you people" and you mock the original poster. You ask us to "excuse" "this frustrated post." There is no excusing here. With voting, you are subject to the opinions of everyone else here. Don't ask us to excuse you, and don't be an asshole.

tilly, one of the people you call by name here, has told me many times that it is often best to wait a few minutes before posting, and revise what you were going to say. I think you could learn from this process and not be so inflammatory. Not gain so many --'s. Not be the 'village ass' (which you have been referred to).

Consider this a notice. I have normally supported you, but this time I cannot. I hope you take notice of this and change your behaviour accordingly.

brother (yes, brother) dep.

--
Laziness, Impatience, Hubris, and Generosity.