Re: Thou shalt have no other gods before the ANSI C standard

From: Brian Inglis (Brian.Inglis_at_SystematicSW.Invalid)
Date: 02/18/05

  • Next message: kumarh_at_gmail.com: "Initialization vector usage" 
    Date: Fri, 18 Feb 2005 17:28:45 GMT

    On Fri, 18 Feb 2005 06:28:05 +0000 (UTC) in alt.folklore.computers,
    daw@taverner.cs.berkeley.edu (David Wagner) wrote:

    >Brian Inglis wrote:
    >>There were also a *lot* of hardwired line length restrictions in AT&T
    >>derived code: either .5KB or 2KB; but they did check, report, and fail
    >>immediately when any line exceeded the bounds.
    >>That's when I found out about the GNU project, their "no arbitrary
    >>limits" guideline, and started using their code instead of AT&T
    >>derived versions to get work done.
    >>That's also a reason I'm against ABC: brittle code with arbitrary
    >>failures does not get work done and alienates the user, because
    >>there's no way they can work around it, except by using a different
    >>piece of software.
    >
    >Sounds like a good reason to be against programs that use fixed-size
    >buffers. Not sure how that translates to a reason to be against (or
    >for) ABC. ABC doesn't on its own make code any more or less brittle;
    >it just reduces the impact of buffer overrun bugs. If your code is
    >brittle with ABC enabled, it was brittle with ABC disabled, too.
    >
    >Ways for working around ABC are dangerous. If there are ways to work
    >around the bounds checking, then that is the first thing an attacker
    >is going to try.

    I prefer programs without any arbitrary bounds that could be checked
    automatically, and code that just "does the right thing". 

    -- 
Thanks. Take care, Brian Inglis 	Calgary, Alberta, Canada
Brian.Inglis@CSi.com 	(Brian[dot]Inglis{at}SystematicSW[dot]ab[dot]ca)
    fake address		use address above to reply
  • Next message: kumarh_at_gmail.com: "Initialization vector usage"

    Relevant Pages

    • Re: Thou shalt have no other gods before the ANSI C standard
      ... Sounds like a good reason to be against programs that use fixed-size ... for) ABC. ... ABC doesn't on its own make code any more or less brittle; ... it just reduces the impact of buffer overrun bugs. ...
      (sci.crypt)
    • Re: Thou shalt have no other gods before the ANSI C standard
      ... >>it just reduces the impact of buffer overrun bugs. ... >>brittle with ABC enabled, it was brittle with ABC disabled, too. ... > I prefer programs without any arbitrary bounds that could be checked ...
      (sci.crypt)
    • Re: equation* vs. [ ... ]
      ... \noindent abc ... environments use a different method than the other environments, ... avoid the align environment, not only for this reason, but also for the ... Morten once showed a hacked version of an amsmath env that still have the 'move-up' feature that equation have but the amsmath env's doesn't. ...
      (comp.text.tex)
    • Re: New level of paranoia on ABC
      ... Whoopi and Joy used the term "F-word". ... I can't possibly think of a reason why. ... ABC is a Mickey Mouse company. ...
      (rec.arts.tv)
    • Re: Jay Leno v. 20/20 v. Numbers
      ... transition I can no longer get ABC. ... I lost all three of these ... CBS NUMB3RS - For some reason I find this show boring. ...
      (rec.arts.tv)
    • Quantcast