Hi,

rauldmil...@gmail.com wrote:
> On Sat, Aug 26, 2017 at 4:36 AM,  <leo_...@volny.cz> wrote:
>> The greater the body of code is, the smaller our understanding, or at
>> least our ability to grok the code.
>>
>> Even in the UNIX world, 'duckspeak' code -- just doing what seems right
>> without realizing the longer-term implications -- is unfortunately very
>> common.
>>
>> I don't think that we can really afford that in the modern world.
>
> Could you be more specific?

Lack of foresight. (I hope that indeed qualifies as 'more' specific and 
not 'less'...)

> What problem are you trying to solve?

Not a single, discrete one. More like a range of potential problems that
are, from my pov, just sitting there, waiting to manifest themselves.

As fun and useful as it must've been (it predates me! :), we're no
longer just playing research on PDPs. The margin of error has decreased
enormously since then. How many people use OpenBSD in their firewall? I
suppose many. Apart from the inconvenience of applying patches, do we
really want to take the risk of sites getting pwned just 'cause we fixed
a problem in one place and not in another? And that's just security.

Any bit of code is only as decent as the assumptions of the programmer.

Isn't it a *little* too easy to assume, that we know what we're doing?

I suppose that what I propose, is that we be more sceptical of our own
abilities, and act accordingly. By taking into account the flaws of
wetware: we're relatively good at invention, but relatively lousy at
verbatim repetition. We're good at quick-and-dirty workarounds, but
long-term solutions take us a lot more effort (not that they're not
worth it! :).

A well-designed computer can complement our capabilities. It cannot
replace our capabilities, nor can we replace the computer's. To try to
do either would be foolish.

But then again, we're all fools in the end, anyway =)

And then there's the issue of us limiting ourselves to operate within
the designs we ourselves create...

Sorry if this all sounds trivial, but given the state of a typical piece
of (even UNIX) code, I'm reminded of this stuff everyday.

I thought it wouldn't hurt to share it.

        --schaafuit.

Reply via email to