Shirt a Phobia

Sample Page

Requested move

Discussion for the swap is taking place at Talk:Array data structure#Requested_move. —Cybercobra (talk) 05:41, 16 May 2009 (UTC)[reply]

The move has been rejected.–Aervanath (talk) 18:11, 25 May 2009 (UTC)[reply]

Removed illogical note about elevator buttons in France

This note was (sorry) stupid for the following reason: in french there is no equivalent to floor, instead the word used (and thought) étage means so-to-say “over-floor”. Thus, étage 1 is english floor 2. And floor 1/ground floor logically is étage 0, that is in fact a non-étage; this also fits pretty well with underground levels numbered from -1 down. (I guess in english-speaking countries there probably is no 0 button, just like there is no year 0?)

denis ”spir” (talk) 17:31, 9 December 2012 (UTC)[reply]

@Denispir:: Well, I hear that in Britain, the first floor is a level above the ground floor… —SamB (talk) 22:43, 4 July 2015 (UTC)[reply]

Dubious

Regarding this bit: Thus, an array of numbers with 5 rows and 4 columns, hence 20 elements, is said to have dimension 2 in computing contexts, but represents a matrix with dimension 4-by-5 or 20 in mathematics. My informal impression is that it’s common to speak of a matrix as -dimensional, but much less so to call the product the dimension. Something like the latter only happens when one considers a set of matrices as a vector space; for example, the set of all self-adjoint matrices forms a -dimensional vector space, in which the natural counterpart of the dot product is the Hilbert–Schmidt inner product. But that notion of dimension depends upon which space the matrix is regarded as belonging to. XOR’easter (talk) 14:55, 7 January 2023 (UTC)[reply]

I agree. The language here is subtle, and there is a real danger of editors coining their own idioms. It is correct, both technically and culturally, to say that the set of 4 x 5 matrices forms a 20-dimensional vector space. But it is incorrect, at least culturally, to say that a 4 x 5 matrix is 20-dimensional. A big reason is that 4 x 5 matrices and 2 x 10 matrices (for example) are really dissimilar, despite having the same overall “dimension”. Mgnbar (talk) 19:44, 7 January 2023 (UTC)[reply]

This article is beyond biased

by presenting 

for (i = 0; i < 5; i += 1)

as if it’s a unique construct of zero-based indexing paradigms.

One could use those EXACT same parameters with one-based indexing,

with EVEN LESS verbosity, by doing 

for (i = 0; i++ < 5; )

thus eliminating the need for the 3rd portion of the for() loop construct. In either construct, the final value of i would be 1 more than the highest index of the array. So it’s beyond ridiculous to claim how zero-based indexing is any better, or how it “prevents one-off error”. The risk of one-off errors are identical. It’s two sides of the same coin.

Because if it were, all the math oriented software and platforms would have adopted it.

You know what ACTUALLY introduces plenty of one-off errors ? Having to substring or string slice at indices off by one from the human speech position of the letter(s) by having languages that tell people “the index position 2 letter of the word ONE isn’t the 2nd letter”, or suggesting people do lookup of July 24th by accessing 

 calendar[6][23]

How intuitive of zero-based languages indeed. 2603:7000:3C3D:4840:0:0:0:3C3 (talk) 18:46, 8 August 2024 (UTC)[reply]

I have remove the text in question, because it was too detailed for this article (but still not detailed enough to convince the reader of its point). I do think that the criticism that the article (rather than a couple of sentences in a deep paragraph) was beyond biased (rather than merely biased, or mistaken, or incomplete, etc.) was overblown. Mgnbar (talk) 01:39, 11 August 2024 (UTC)[reply]

Language neutrality

While I believe that #This article is beyond biased was a bit too harsh, there are NPOV issues. In particular,

  1. Some hardware, e.g., Sigma 7,[1] automatically scales index registers; at the time I thought that it was well meant but misguided.
  2. Representing arrays as arrays of pointers may be common in the C world, but there are languages that don’t do that.
  3. Possibly TMI, but some languages that define direct computation of addresses from subscripts specify row major while others specify column major.
  4. There is some variation in the details when enumerated types are allowed for subscripts.

I believe that the article would be better if it gave examples on each side of every choice that it mentions.

To clarify, FORTRAN used parentheses for 1-based subscripts, ALGOL 60 used brackets for defined-range subscripts, PL/I used FORTRAN (foo,bar,baz) syntax but ALGOL defined high:low range subscripts while C used ALGOL [foo,bar,baz] for 0-based subscripts. — Shmuel (Seymour J.) Metz Username:Chatul (talk) 13:39, 24 March 2026 (UTC)[reply]

References

  1. ^ “General Purpose Features” (PDF). Xerox SIGMA 7 – Computer Reference Manual (PDF). October 1973. p. 5. 90 O9 50J. Retrieved March 24, 2026. Displacement Indexing. The technique of indexing by means of a “floating” displacement permits the user to access the desired unit of data without the need to consider its size. The index registers automatically align themselves appropriately; thus, the same index register can be used on arrays with different data sizes.