Computer Languages and Knowledge

[Wyman]

I have been reading Quine's Word and Object, which is a tome of analytic, 20th century philosophy. In it, he laboriously 'analyses' language and shows pretty convincingly that, for intended purposes (most specifically mathematics and physics), we can analyse down to a basic system of logic, set theory and the objects of physics and express all we need to express. This is a huge oversimplification, but the import of the project is rigorous analysis of logical thought to a foundational or near-foundational level.

I wonder where that gets us, if anywhere.

A analogy that came to mind is that of machine language versus high level computer languages in the field of computer programming. Machine language deals with series of 1s and 0s and a lot of binary math. Higher level languages 'abstract' by joining elements of the machine language into larger and more complex (and more useful and easy to work with) groupings.

Of course, there is an analogy here with the way we think. In number theory, for instance, we can derive all the usual arithmetic of integers and even real numbers (and the calculus) from the axioms of set theory. But this is an extremely laborious process. Similarly, to code a complex program in machine language would be an exercise in masochism.

My question is, would a computer programmer who programs in and understands machine language (as wells as the high level language), be in some way more 'knowledgeable' than a programmer who has only learned high level languages? Both are experts of the same intelligence and proficiency within their realm of programming, both are creative and adept at solving the problems at hand. The only difference is that one is, in addition, proficient at the 'underlying' machine language.

[Impenitent]

the machine language programmer simply has an additional language at his disposal...

each language works on a separate compiler...

-Imp

[Wyman]

the machine language programmer simply has an additional language at his disposal...

each language works on a separate compiler...

-Imp

And so do I extrapolate that foundationalism is bunk?

[Impenitent]

the machine language programmer simply has an additional language at his disposal...

each language works on a separate compiler...

-Imp

And so do I extrapolate that foundationalism is bunk?

perhaps, if your belief is your justification that foundationalism is bunk...

Peter Pan has left the building...

-Imp

[attofishpi]

My question is, would a computer programmer who programs in and understands machine language (as wells as the high level language), be in some way more 'knowledgeable' than a programmer who has only learned high level languages? Both are experts of the same intelligence and proficiency within their realm of programming, both are creative and adept at solving the problems at hand. The only difference is that one is, in addition, proficient at the 'underlying' machine language.

The programmer who can also code directly in machine code should be considered more 'knowledgeable' with respect to computer programming.
Of course he\she would rarely or never bother to code in machine language, unless the higher language could not achieve a desired result.

When programming in C (3rd level) you have the ability to insert 'inline' assembly (2nd level) code should you need to. Many moons ago it was useful for syncing the screen write of a graphics memory dump with the screen refresh rate permitting non-flicker rendering.

'Knowing' how to achieve something that is not available with a higher level language -unless you recode the compiler- is extra knowledge...no?

[Wyman]

My question is, would a computer programmer who programs in and understands machine language (as wells as the high level language), be in some way more 'knowledgeable' than a programmer who has only learned high level languages? Both are experts of the same intelligence and proficiency within their realm of programming, both are creative and adept at solving the problems at hand. The only difference is that one is, in addition, proficient at the 'underlying' machine language.

The programmer who can also code directly in machine code should be considered more 'knowledgeable' with respect to computer programming.
Of course he\she would rarely or never bother to code in machine language, unless the higher language could not achieve a desired result.

When programming in C (3rd level) you have the ability to insert 'inline' assembly (2nd level) code should you need to. Many moons ago it was useful for syncing the screen write of a graphics memory dump with the screen refresh rate permitting non-flicker rendering.

'Knowing' how to achieve something that is not available with a higher level language -unless you recode the compiler- is extra knowledge...no?

Absolutely. I was being an annoying 'philosopher' (albeit amateur; and perhaps Hex's designation of 'cozy chatterer' fits better) and trying a kind of thought experiment.

Suppose that knowing the lower language had no useful purpose over and above the higher language any longer.

An analogy, somewhat close, would be whether the Trigonometry or Calculus student 'knows' less about those subjects (after having mastered them) before he takes analytic geometry or set theory. The computer thought experiment is a little cleaner because in the case of the math student, one would have to cordon off and strictly define what is meant by 'mastery' and 'subject matter.'

[attofishpi]

Knowing how to make bricks and not just houses is always going to be more 'knowledge.able'

[hammock]

My question is, would a computer programmer who programs in and understands machine language (as wells as the high level language), be in some way more 'knowledgeable' than a programmer who has only learned high level languages? Both are experts of the same intelligence and proficiency within their realm of programming, both are creative and adept at solving the problems at hand. The only difference is that one is, in addition, proficient at the 'underlying' machine language.

The primal-programmer would just have the advantage of knowing in-depth what the other levels rest and depend upon (is the "god" who provides the "substrate" that other beings manipulate for additional inventions / creations). The hardware expert or engineer in turn knows the "ones and zeros" are illusionary interpretative crap, or useful labels. A language and the particular "expressions" falling out of it ultimately being structural arrangements of physical components organized to hold / not hold patterns of electrical charges, magnetized areas, etc. The patterns enabled or becoming causally potent within the "machine" via the latter's complex, mechanistic-like relationships designed with functional purposes.

Even the idea of the microscopic, connectively-constituted but sometimes widely distributed forms being "data" is beholding to the interpreters who know the underlying template which the specific brands of patterns conform to. To the causal alien visitor who never sees end-products displayed on a monitor or the resulting activity of other devices, there are only technological guts inside the computer manipulating electricity to no immediately discerned practical purpose. Though, like the linguists once trying to decipher Mayan hieroglyphs, the template schemes which "this and that" conform to might eventually be figured-out.