[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: going back to stone axes



In a galaxy far, far away, : 26 Feb 1996 16:01:21 PST
> Note that I have not included above the ability to compile a design 
> specification into an implementation.  Many of the implementation problems 

  In other words, ASN.1 is not good because it solves the
bit-stream encoding problem, but because it a nice descriptive language. Many
seem to feel that the bitstream encoding problem just isn't much a of a 
problem.
  I happen to agree. I haven't worked on "big" projects, just operating systems
kernels: one hopes they are *small*, not *big* (I'll let the guilty remain
anonymous).
  This suits me fine. It sounds to me that most of criticism of ASN.1 is 
really criticism of DER. It sounds that we can define an encoding for an email 
protocol to be ASCII based if we like.

  [BTW: I know ASN.1 enough to understand SNMPv2, but I couldn't write it. I
find trying to "get into" SNMPv2 to be very difficult for this reason. I had
no problems hacking around with HTTP/0.9 when that was current. I don't know
where our firewall product would be if it weren't for the three or four hundred
regression tests on our proxies written in Perl. Mostly ten lines or less.]




References: