| 1 |
Date: Thu, 8 Feb 2001 08:42:04 -0600 |
| 2 |
From: Vikram S. Adve <[email protected]> |
| 3 |
To: Chris Lattner <[email protected]> |
| 4 |
Subject: RE: Type notation debate... |
| 5 |
|
| 6 |
Chris, |
| 7 |
|
| 8 |
> Okay before you comment, please look at: |
| 9 |
> |
| 10 |
> http://www.research.att.com/~bs/devXinterview.html |
| 11 |
|
| 12 |
I read this argument. Even before that, I was already in agreement with you |
| 13 |
and him that the C declarator syntax is difficult and confusing. |
| 14 |
|
| 15 |
But in fact, if you read the entire answer carefully, he came to the same |
| 16 |
conclusion I do: that you have to go with familiar syntax over logical |
| 17 |
syntax because familiarity is such a strong force: |
| 18 |
|
| 19 |
"However, familiarity is a strong force. To compare, in English, we |
| 20 |
live |
| 21 |
more or less happily with the absurd rules for "to be" (am, are, is, been, |
| 22 |
was, were, ...) and all attempts to simplify are treated with contempt or |
| 23 |
(preferably) humor. It be a curious world and it always beed." |
| 24 |
|
| 25 |
> Basically, my argument for this type construction system is that it is |
| 26 |
> VERY simple to use and understand (although it IS different than C, it is |
| 27 |
> very simple and straightforward, which C is NOT). In fact, I would assert |
| 28 |
> that most programmers TODAY do not understand pointers to member |
| 29 |
> functions, and have to look up an example when they have to write them. |
| 30 |
|
| 31 |
Again, I don't disagree with this at all. But to some extent this |
| 32 |
particular problem is inherently difficult. Your syntax for the above |
| 33 |
example may be easier for you to read because this is the way you have been |
| 34 |
thinking about it. Honestly, I don't find it much easier than the C syntax. |
| 35 |
In either case, I would have to look up an example to write pointers to |
| 36 |
member functions. |
| 37 |
|
| 38 |
But pointers to member functions are nowhere near as common as arrays. And |
| 39 |
the old array syntax: |
| 40 |
type [ int, int, ...] |
| 41 |
is just much more familiar and clear to people than anything new you |
| 42 |
introduce, no matter how logical it is. Introducing a new syntax that may |
| 43 |
make function pointers easier but makes arrays much more difficult seems |
| 44 |
very risky to me. |
| 45 |
|
| 46 |
> In my opinion, it is critically important to have clear and concise type |
| 47 |
> specifications, because types are going to be all over the programs. |
| 48 |
|
| 49 |
I absolutely agree. But the question is, what is more clear and concise? |
| 50 |
The syntax programmers are used to out of years of experience or a new |
| 51 |
syntax that they have never seen that has a more logical structure. I think |
| 52 |
the answer is the former. Sometimes, you have to give up a better idea |
| 53 |
because you can't overcome sociological barriers to it. Qwerty keyboards |
| 54 |
and Windows are two classic examples of bad technology that are difficult to |
| 55 |
root out. |
| 56 |
|
| 57 |
P.S. Also, while I agree that most your syntax is more logical, there is |
| 58 |
one part that isn't: |
| 59 |
|
| 60 |
Arrays (without and with size): |
| 61 |
type ::= '[' type ']' | '[' INT ',' type ']'. |
| 62 |
|
| 63 |
The arrays with size lists the dimensions and the type in a single list. |
| 64 |
That is just too confusing: |
| 65 |
[10, 40, int] |
| 66 |
This seems to be a 3-D array where the third dimension is something strange. |
| 67 |
It is too confusing to have a list of 3 things, some of which are dimensions |
| 68 |
and one is a type. Either of the following would be better: |
| 69 |
|
| 70 |
array [10, 40] of int |
| 71 |
or |
| 72 |
int [10, 40] |
| 73 |
|
| 74 |
--Vikram |
| 75 |
|