Comments on: Why computer scientists count from zero

By: Chris

Chris — Fri, 23 Apr 2010 13:15:23 +0000

Dijkstra has a nice and very short little paper on this subject:
http://userweb.cs.utexas.edu/users/EWD/transcriptions/EWD08xx/EWD831.html

By: Aaron Evans

Aaron Evans — Thu, 22 Apr 2010 20:34:10 +0000

if you treated item 1 as index 1 (and not index 0) you could do lots of cool things like refer to an array with index 0. Then you could do

for (array++)
… check array type, whether it’s empty, etc.
… then do something with each iteration

By: Hello, world! « Ramblings of a Mind

Hello, world! « Ramblings of a Mind — Tue, 27 Jan 2009 22:10:06 +0000

[...] The definition starts out from zero instead of one because computer scientists always start counting from zero. [...]

By: John

John — Mon, 30 Jun 2008 22:59:27 +0000

Mathematicians are fairly consistent with using half-open intervals too. It’s most common to start indexing from 1 in math, but it varies by context. Matrix element subscripts, for example, start from 1, but power series coefficients start from 0.

By: Bob Carpenter

Bob Carpenter — Mon, 30 Jun 2008 20:41:14 +0000

We computer scientists also describe intervals using half-open notation, starting at the start, but ending one before the end. This makes loops simple, as in this loop over [0,k), that is, from 0 inclusive to k exclusive: for (int i = 0; i < k; i++) { } We tend to index strings and compute substrings the same way, with "abcde".substring(0,3) being the same as "abc". But to keep you on your toes, we sometimes use start+length rather than start+end. Set theorists do the same thing, implicitly starting from zero (coded as the empty set in logical theories of arithmetic) and indexing summations with

Σi < k.  Because zero is the smallest natural number they code, they get away without even writing the zero!



By: Codewiz51
Codewiz51 — Fri, 27 Jun 2008 00:54:56 +0000
We don’t count from zero, we index from zero.  The first item, has an index of zero in our little corner of the world. So item one is referred to as item[0]. It was a long time before humans had a concept of zero as a number (around the 4th century BC if my history serves me. I don’t think it was really codified until around 600 AD.)



By: John
John — Thu, 26 Jun 2008 16:01:42 +0000
By vascillating I meant inconsistent approaches in the same code base. Imagine this conversation. 
“I wonder whether this index is one-based or zero based.”
“Who wrote that part of the code?”
“Looks like it was Sam.”
“Oh, then it’s probably one-based.”



By: John Venier
John Venier — Thu, 26 Jun 2008 15:42:20 +0000
I think a preprocessor / kludge is really bad, probably worse than vascillating.  Unless that’s what you mean by vascillating.  I really hated the code in Numerical Recipies for C because of the kludges they used in converting their Fortran to C.  Double-plus ungood.
In a kindergarten somewhere a child of a programmer is counting to ten by saying, “Zero, one, two, three, four, five, six, seven, eight, nine.”