[1194] in Kerberos_V5_Development
Re: key data version as array index
daemon@ATHENA.MIT.EDU (Theodore Y. Ts'o)
Thu May 16 12:18:47 1996
Date: Thu, 16 May 1996 12:18:33 -0400
From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Marc Horowitz <marc@MIT.EDU>
Cc: proven@cygnus.com, "Barry Jaspan" <bjaspan@MIT.EDU>, krbdev@MIT.EDU
In-Reply-To: Marc Horowitz's message of Wed, 15 May 1996 22:23:08 EDT,
<9605160223.AA14525@bill-the-cat.MIT.EDU>
Date: Wed, 15 May 1996 22:23:08 EDT
From: Marc Horowitz <marc@MIT.EDU>
>> This is an effiecency hack.
In the tradeoff between efficiency and readability, when only a few
instructions in a non-inner-loop block are concerned, I think it makes
more sense to opt for readability.
It's more than just readability; there's also future code
maintainability. If you have assumptions like that scattered all over
your code, it only takes someone else who is modifying the code to
forget one place where you've made such an assumption, and it's
all over.
- Ted
