Re: what's up with GROW?
In article <199607101517.IAA19065@cygnus.com>, Michael Tiemann <tiemann@cygnus.com> writes:
> I didn't post there specifically, but John Gilmore may have done so a > while ago. What specific problem are the cyberpunks trying to solve > that GROW would facilitate? Sorry, I've just recieved so much complete incompetence with this issue that talking to someone who was on the ball took some adjusting. Hope I wasn't too rude. One of the main issue on cypherpunks right now (Black Unicorn did a sort of "white paper" on it recently, as in two days ago) is integrating cryptography with other apps, in particular WWW apps. It just seemed to me that if the kind of exstensibility which Emacs is famous for was going to be a part of GROW, which apparently was the attention, this would be a Good Thing for cryptography integration. I good point in this direction is that the most popular way of using PGP with email is emacs-based (i.e. more people sign their posts to cypherpunks using Mailcrypt, the emacs-PGP interface, then all other signers combined, so it must be pretty easy to use). I've used it, and it's incredibly slick. I had hopes the GROW to go as far as emacs but with more network awareness. Oh well. This has been crossposted to cypherpunks, BTW. -Robin PS: You may want to go through and terf ALL the grow pages, not just some of them.
To follow up further...we are very interested in deeply embedding network security in all software components we deliver as "enterprise solutions". This is the commercial analog (I believe) of what cypherpunks want in their personal space. Our approach right now is to view Kerberos as a solution to two problems (user authentication and key management), and to extend basic services such as web servers, web clients, etc., with Kerberos to build a coherent solution. We believe that Java has effectively solved one of the problems that GROW was intended to address: ubiquitous extensibility. While some may argue the finer points of just how powerful the Java model is compared to the scheme model, there is a market momentum that argues for a level of ubiquity that we could not have hoped to achieve through Scheme. That said, perhaps it would be worthwhile to compare notes on current Cygnus and Cyberpunks projects and approaches, to see if there are any good synergies to tap. Mark Eichin (our long-time V5 technical lead) will be visiting from our Boston office the week of 7/22 for the Usenix network security conference. Perhaps we could have a mini meeting of minds around that time. Mark is eichin@cygnus.com. Michael P.S. If the above sounds like more smoke than cyberpunks are accustomed to, let me know in private email, and I'll continue follow-ups among those who give me a positive response.
participants (2)
-
Michael Tiemann -
Robin Powell