[lug] Freedom formatting (was: ubuntu wireless discovery tool)

Rob Nagler nagler at bivio.biz
Thu Nov 26 08:07:24 MST 2009


Tom password protectedly writes:
> do hope they appreciate my creative work-around for *their* gmail bug. :)

I have to say, Tom, this was pretty creative.  It's interesting the
differences between PDF readers.  Preview tells me the doc is password
protected.  pdftotext tells me this, too.  Acrobat "Professional"
let's me select the text, copy it, and drops it on the floor.
Fortunately, the OCR works (use captchas next time :-), which is why
some people say that the Internet routes around censorship.  On this
subject, you might want to read: http://bitworking.org/news/2009/11/authority

This brings me to your other point:

> If that. is so, then we do not have a failure of technology here; we
> have a failure of courtesy.

Courtesy it is, but I think it goes much deeper.  Courtesy is a word
we hide behind when we want to be polite to one another.  It's when we
don't want to look deeper into why people are being discourteous.  The
reasons can get ugly, but I believe in cutting to the chase, one level
deeper...

I think *the* prime motivator is fear.  I think people are terrified
of knowing that their code isn't perfect.  Their fear is compounded by
the fact that laypeople blame themselves for crappy software: anxiety
is a multiplier.  This is quite similar to "if it's in print, it must
be true".

Programmers have a love/hate relationship with authority, and their
fear of testing (computable authority) is the reason crappy software
abounds.  No, it's not about "not having enough time to test".  It's
about not having the guts to test and to put as much energy into tests
as code.

While I only program in Perl now, I'm not a big fan of the language.
It's an accident of history bOP is written in Perl.  If I had time, I
would convert all our code to Lisp (another accident of history).
However, I do agree with some of the cultural tenets of Perl.  Testing
is one of them.  It's something that has been in Perl since day one, I
believe. 

Perl is one of the first programming languages built from the ground
up with a comprehensive validation suite besides Ada.  Testing has
become common, but Perl was a decade ahead.  Other languages added
validation suites ex facto, but that's not the same.  The reason
Pascal and Modula-2 were ill-defined (and some say ill-conceived ;-)
is that Wirth didn't write automated tests.  He just *knew*
everything, just like Djikstra (search on "djikstra testing").

I wish I was so smart.  I know one thing: I'm human.  I was just lucky
enough to have gotten the testing bug early in my career, like Larry,
and I assume you, too, Tom. :)

If there is a historical legacy to Perl, it's that Perl was the first
language to make testing first class.  Thanks for the courtesy, Tom &
Larry! 

Happy TG, 
Rob

PS. More comments on fear of testing here:
http://www.viarob.com/my/page/The_Psychology_of_Software_Testing



More information about the LUG mailing list