- From: Eastlake III Donald-LDE008 <Donald.Eastlake@motorola.com>
- Date: Fri, 18 May 2001 13:38:20 -0400
- To: "'xml-encryption@w3.org'" <xml-encryption@w3.org>
Forwarded with permission. Note: David confused which comments were mine and which were Joseph Ashwood's... Donald -----Original Message----- From: Wheeler, David M AZ [mailto:david.m.az.wheeler@intel.com] Sent: Thursday, May 17, 2001 11:53 AM To: 'Eastlake III Donald-LDE008'; 'Joseph Ashwood' Subject: RE: Early Draft Algorithms Section Donald and Joe, I'm not going to jump too deep into the fray here, primarily because I'm not sure of the appropriate 'rules' for contribution. I have been watching the list actively for about 4 months, though I have been rather busy on other things. My interest is that we will likely utilize the standard or an early draft within the next 6-12 months for inclusion in a product. Though interoperability will NOT be a requirement for this particular product, it is a concern going forward. So, that's some background on my interest. I'd like to respond to Donald's question: && Does anyone else in the WG have an opinion on including a stream cipher? Although I don't have any formal status in the WG, I'd like to voice an industry opinion. Stream ciphers are vital to streaming media (music & video), of which I'm sure you are both aware. However, I thought I might mention that, because I didn't see it in any of the correspondence. As we move forward with different technologies, the ability to stream media, and encrypt/decrypt quickly, are vital. The stream ciphers meet important real time requirements for streaming media, but also for other technologies, like voice (and voice over IP is becoming real important). IMHO, it would be a serious deficiency in the standard to ignore stream ciphers, or at least provide the ability to use a block cipher in streaming mode (OFB, CFB, etc), much like WAKE was proposed. HOWEVER, this specification of a stream cipher has got to be optional. I vehemently agree with Joe regarding code bloat and complexity to implement. > %% If the parties can't negotiate, then a profusion of algorithms > %% either breaks interoperability or puts an enormous burden, > %% especially on low end limited memory and processing devices, to > %% implement all these algorithms. I, for one, just do not believe > %% there is any good reason to provide for 39 symmetric algorithms. > %% 2 or 3 well chosen strong algorithms are a much > %% better idea from the viewpoint of the goals of interoperability and > %% widespread implementation, including lower end devices. Well said, Joe! Donald, I understand what you are saying, and the technologist within me wants to agree, however, business disagrees. Why? Time to market and risk management are the primary reasons. If I build a web server that exports information to a closed group, what is the likelihood that someone is going to put forward the effort to break the encryption of my data? Realistically? Let me share a little inside information. DES is still viable as an encryption algorithm for many solutions being built today because the financial gain from breaking the crypto and gaining access to the information provides no pay back for the effort to break the cipher in the first place. Besides, there are so MANY other ways to gain access to the information, and so many other weaknesses in the system to address! The algorithm is on one piece. To directly attack a good cipher like AES, 3DES, (and yes even DES), is highly unlikely. Yes, it's possible to break DES...but how many of the real threats on my system will actually have that capability? Under 1/10%. The same will be said of 3DES in 5 years. It is good enough for commercial use in 80% of the cases. I hope this input is useful to you both. I hope also that the WG will seriously consider adding an optional stream cipher, or at least streaming modes of operation to the standard. Thanks for you ear. Sincerely, David M. Wheeler Home Products Group Intel Corp. LEGAL DISCLAIMER: The views and opinions expressed are solely my own, and do not necessarily reflect the views and opinions of Intel Corporation.
Received on Friday, 18 May 2001 13:38:26 UTC