[Ach] meta-question on algorithm agility

ianG iang at iang.org
Mon May 5 14:58:51 CEST 2014

On 5/05/2014 13:21 pm, René Pfeiffer wrote:
> On May 02, 2014 at 2320 +0100, ianG appeared and said:
>>>> Imagine that algorithm agility was banned.  No more choice!  How much
>> resource would this free up?
> What about scenarios like the SEED monoculture?
> http://kanai.net/weblog/archive/2007/01/26/00h53m55s

This history:

"The history goes back to 1998, when the 128 bit SSL protocol was still
not finalized (it was finalized by the IETF as RFC 2246 in Jan. ’99.)
South Korean legislation did not allow 40 bit encryption for online
transactions (and Bill Clinton did not allow for the export of 128 bit
encryption until December 1999) and the demand for 128 bit encryption
was so great that the South Korean government funded (via the Korean
Information Security Agency) a block cipher called SEED. SEED is, of
course, used nowhere else except South Korea, because every other nation
waited for the 128 bit SSL protocol to be finalized (and exported from
the US) and have standardized on that."

I never heard that rendition of history before.  I think it's wrong.  As
far as I know, everyone was happily using full strength crypto, except
those that followed rules of their government or some other government.
 The world was using SSL well before the IETF got involved, and if
people "waited" that's really their fault.

But anyway -- the cause of that story, and for South Korea -- is an
underlying problem of forcing differing standards;  discrimination if
you will.  By forcing different choices on different groups of people,
then every different culture went its own way, ending up with perverse
results like the local monoculture in South Korea [0].

If the SSL people had done the job properly and just specified 128 bit
ciphers and *that's it!* then none of this would have happened.

Discrimination is harmful on almost all grounds.  Just say no.

> I think this is a good argument in favour of algorithm agility.

Algorithm agility was used to paper over the original flaw which was
that the Americans pandered to their local government rules and wouldn't
export the crypto.  So it took a while for the rest of the world to
develop the non-USA crypto.  SSLeay and all that, until Eric was bought
out by RSA and it became OpenSSL.

Algorithm agility was the problem, and the solution was getting rid of
it entirely;  not using it to justify the political crap.


[0]  Also, recall that the browser wars were between M$ and Netscape.
Netscape's browser was $50 but typically free, so it had the problem of
low revenue, whereas, once M$ dropped Blackbird and decided to own the
web, it turned all its forces to it.  Cash rich ... it attacked Netscape
and beat them at their own game.  In that concext, SSL was originally a
neat trick by Netscape to try and develop some revenue from this thing
they briefly had control of, by inventing something called ecommerce.

Point being that the 40bit crypto thing was a skirmish in a bigger war.
 Old rant.


in which I probably disagree with what I wrote above...

More information about the Ach mailing list