Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.c > #111628

Re: And they say C is obscure

Show all headers | View raw

On 06/06/17 14:26, Malcolm McLean wrote:
> On Tuesday, June 6, 2017 at 12:09:37 PM UTC+1, David Brown wrote:
>> On 06/06/17 12:47, Malcolm McLean wrote:
>>
>>> "As in E. coli, so in elephant" is a famous aphorism in biochemistry.
>>> They're both organisms. What we see in one tells us quite a bit about the 
>>> other, and therefore about all organisms, or maybe all organisms except
>>> a narrow subset. But it's not a mathematical rule, it doesn't even always
>>> hold.
>>
>> It is not science either - because in science, to be a rule it should
>> always hold as far as we know.  It's just a saying that can be a useful
>> guideline - like in mathematics you might say "this looks like a pattern".
>>
>> You are making a distinction between mathematics and other sciences that
>> does not exist - pretending that this lets you off the hook for writing
>> something that is at best a vague idea of yours, formulated as though it
>> were a clearly defined absolute rule.
>>
> It depends on your science.
> 
> A lot of science, like psychology, is mainly descriptive.

Sorry, I was talking about /science/.  You know, theories, evidence,
data, experiments, predictions, repeatable results - that sort of thing?

Psychology is not a science.  It is a "craft" - there is a certain basis
in logical and scientific theory, but much of it is about how the
practitioner prefers to do things.  The "descriptive" bits are not
science - they are a collection of ideas and experiences.  (This is not
a condemnation of psychology.)

> Skinner tried to turn
> psychology into a hard science like physics by ruthlessly stripping behaviour
> down to its bare essentials. The "Skinner box" was closed box in which a rat 
> lived, with only levers to interact with. He did succeed in making his results
> reproducible. But the conclusion on his work was that it wasn't applicable,
> it didn't tell us much about anything except rats in boxes.
> 
> A psychological observation - that three is the limit on parentheses that can
> be easily parsed

/Your/ personal observation, unrelated to anyone else or other people's
experience.

> (it should go without saying that we mean similar symbols in
> short expressions, and without devices to increase symmetry)

Why should that "go without saying"?  Do you just mean some extra
conditions that you forgot to mention when first declaring your absolute
fixed rule, and that you have added in afterwards?

> - leads to a
> theory - since we can express array dimensions as nesting, and we know why
> three is the limit for visualising arrays, doesn't the same limit apply for the same
> underlying reason to nesting.

That is not a "theory" - certainly not in the scientific sense.  I would
not even count it as a "theory" in a pseudo-scientific sense.  It is a
vague idea, with no evidence or basis.  That is all.

> 
> But that remains just a theory. It's hard to test. The first test is to make sure that
> the observations are correct, that three is in fact a limit. That seems to be pretty
> solid, the exceptions that have been given are either cheats or failure to understand
> the self similarity requirement.

How often are you going to deny reality?  Even when you change the rules
as you go along, such as now introducing a "self similarity
requirement", it is still nonsense.

> 
>>
>>  /Why/ are you so obsessed with the idea that 3 is a magic number in all this?
>>
>> It is pointless to speculate why "three seems to be the limit" when it
>> has been so clearly demonstrated that there is no particular limit at all.
>>
> Because it's potentially a useful result.

No, it is not.

Even if your "rule of three" were approximately correct, and even if
your explanation for it were approximately correct, it still would not
be useful.  We already know that we should write code that is clear and
easy to understand - what benefit could it possibly be to anyone to know
that the reason we can't understand more than 3 levels of nesting is
that we live in a 3D world?

Now, if you had said "because it would be an /interesting/ result", I
would agree.  It is still nonsense, of course, but if it had been true,
it would have been interesting.

> The observation that the more complex
> the expression the harder it is to read doesn't tell us anything. 

It tells us that we should keep our expressions simple.  And that is
about as concrete a rule as we can get here - the details all vary
according to circumstance.

> If we can put
> a number on it then we can being to understand. That's how science works, you
> start with description (these caves are full of bats), then try to assign numbers
> (2,300 bats live in this cave), then to formulae (we have 50 bats per metre squared,
> will that apply to other caves, are populations limited by roosting space or by
> other factors?). If there's "no particular limit" you're just giving up. 

It is not "giving up" - it is accepting that programming is not a
science (it is better described as a craft), and that some things do not
have a single number answer.  Your idea is to count the 2300 bats living
in one cave, and declare that all caves contain 2300 bats, perhaps
because there are 2300 trees in the woods near your house.

> Bats live in
> the cave, something must limit their numbers, even if it is only cave volume
> divided by bat volume. Humans can parse expressions, something must limit the
> complexity of those expressions, even if it's just the size of the visual field.

Great.  With something about 100 million retina cells, and say a 6x8
minimal readable font, that leaves us room for about 1,000,000 pairs of
brackets.  If you want to claim that humans cannot parse more a million
sets of nested brackets at a glance, I will not argue with you (unless
it is because you think that is the average total length of a person's
hair, in cm).

>>
>> If you are going to make a claim as an absolute rule, as you have been
>> doing, then it is perfectly reasonable to insist that you give a
>> reasonable level of proof.  
>>
> You yourself provided a reasonable level of proof. Three is your limit.
> As In David Brown, so in Malcolm McLean.
> 

Look it up - I made a mistake with /three/ levels of nesting.  If it is
proof of anything, it is that /two/ levels is the limit.

Back to comp.lang.c | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Re: And they say C is obscure "J. Clarke" <j.clarke.873638@gmail.com> - 2017-05-28 16:03 -0400
  Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-05-30 02:21 -0700
    Re: And they say C is obscure "J. Clarke" <j.clarke.873638@gmail.com> - 2017-06-04 20:55 -0400
      Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-05 01:44 -0700
        Re: And they say C is obscure raltbos@xs4all.nl (Richard Bos) - 2017-06-05 09:47 +0000
          Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-05 03:51 -0700
            Re: And they say C is obscure Gareth Owen <gwowen@gmail.com> - 2017-06-05 15:43 +0100
              Re: And they say C is obscure Tim Rentsch <txr@alumni.caltech.edu> - 2017-06-05 13:51 -0700
              Re: And they say C is obscure raltbos@xs4all.nl (Richard Bos) - 2017-06-06 17:26 +0000
      Re: And they say C is obscure Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-06-05 11:59 +0100
        Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-05 04:12 -0700
          Re: And they say C is obscure asetofsymbols@gmail.com - 2017-06-05 05:54 -0700
          Re: And they say C is obscure Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-06-05 15:03 +0100
            Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-05 07:22 -0700
              Re: And they say C is obscure GOTHIER Nathan <nathan.gothier@gmail.com> - 2017-06-05 16:22 +0200
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-05 07:40 -0700
                Re: And they say C is obscure GOTHIER Nathan <nathan.gothier@gmail.com> - 2017-06-05 16:41 +0200
              Re: And they say C is obscure Gareth Owen <gwowen@gmail.com> - 2017-06-05 15:44 +0100
              Re: And they say C is obscure Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-06-05 16:12 +0100
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-05 09:08 -0700
                Re: And they say C is obscure David Kleinecke <dkleinecke@gmail.com> - 2017-06-05 12:08 -0700
                Re: And they say C is obscure asetofsymbols@gmail.com - 2017-06-05 13:10 -0700
                Re: And they say C is obscure Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-06-05 23:54 +0100
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-06 01:15 -0700
                Re: And they say C is obscure David Brown <david.brown@hesbynett.no> - 2017-06-06 11:10 +0200
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-06 02:39 -0700
                Re: And they say C is obscure David Brown <david.brown@hesbynett.no> - 2017-06-06 12:15 +0200
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-06 03:47 -0700
                Re: And they say C is obscure David Brown <david.brown@hesbynett.no> - 2017-06-06 13:09 +0200
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-06 05:26 -0700
                Re: And they say C is obscure GOTHIER Nathan <nathan.gothier@gmail.com> - 2017-06-06 14:25 +0200
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-06 05:34 -0700
                Re: And they say C is obscure GOTHIER Nathan <nathan.gothier@gmail.com> - 2017-06-06 15:41 +0200
                Re: And they say C is obscure David Brown <david.brown@hesbynett.no> - 2017-06-06 15:04 +0200
                Re: And they say C is obscure raltbos@xs4all.nl (Richard Bos) - 2017-06-06 17:37 +0000
                Re: And they say C is obscure David Brown <david.brown@hesbynett.no> - 2017-06-07 10:17 +0200
                Re: And they say C is obscure Keith Thompson <kst-u@mib.org> - 2017-06-06 08:49 -0700
                Re: And they say C is obscure Gareth Owen <gwowen@gmail.com> - 2017-06-06 19:17 +0100
                Re: And they say C is obscure Gareth Owen <gwowen@gmail.com> - 2017-06-06 19:16 +0100
                Re: And they say C is obscure Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-06-06 12:44 +0100
                Re: And they say C is obscure raltbos@xs4all.nl (Richard Bos) - 2017-06-06 17:34 +0000
                Re: And they say C is obscure David Brown <david.brown@hesbynett.no> - 2017-06-06 21:14 +0200
                Re: And they say C is obscure Tim Rentsch <txr@alumni.caltech.edu> - 2017-06-06 13:10 -0700
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-06 14:34 -0700
                Re: And they say C is obscure Keith Thompson <kst-u@mib.org> - 2017-06-06 15:27 -0700
                Re: And they say C is obscure Tim Rentsch <txr@alumni.caltech.edu> - 2017-06-11 16:06 -0700
                Re: And they say C is obscure Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-06-06 12:10 +0100
                Re: And they say C is obscure Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2017-06-06 04:36 -0700
                Re: And they say C is obscure raltbos@xs4all.nl (Richard Bos) - 2017-06-06 17:32 +0000
              Re: And they say C is obscure David Brown <david.brown@hesbynett.no> - 2017-06-05 17:58 +0200

csiph-web