Re: [TLS] Heartbleed / protocol complexity

[Watson Ladd <[e-mail]>]

On Wed, Apr 9, 2014 at 4:06 PM, Nico Williams <[e-mail]> wrote:
> On Wed, Apr 9, 2014 at 5:42 PM, Salz, Rich <[e-mail]> wrote:
>>> TLS has an ad-hoc IDL and encoding, and it IIUC doesn't adhere to its own conventions tightly enough that we could now standardize a compatible IDL+encoding and develop tooling for it.
>>
>> I wrote a parser for "TLS IDL" and posted it to the list.  There are a handful of corrections that need to be made in order for the definitions to match the defined syntax.  The biggest one being "ASN.1Cert" looks like a field name, not a type.  The posts are in the archives, if anyone cares.  I'll mail the code to anyone who cares.
>
> That's great news.  Can you say anything about my "IIUC" above?

Parsing an IDL isn't the problem. The problem is developing a
validated parser generator for the language which the IDL expresses.
The further problem is that the semantics of the Hello messages are
extremely ugly. Peter Gutmann compared it to ordering from a Chinese
menu at one point when discussing ECC ciphersuites, and I don't think
that's the only extension with that property.

Even if you succeed in parsing, the extensibility of TLS is (largely)
ridiculous. Security protocols and their implementations are part of
the TCB. They need to be simple.

Sincerely,
Watson Ladd



-- 
"Those who would give up Essential Liberty to purchase a little
Temporary Safety deserve neither  Liberty nor Safety."
-- Benjamin Franklin