re:
http://www.garlic.com/~lynn/2014e.tml#23 Is there any MF shop using AWS service?

TCP/IP Might Have Been Secure From the Start If Not For the NSA
http://beta.slashdot.org/story/200323

Note NSFNET backbone was precursor to modern internet (and cloud
computing).
http://www.technologyreview.com/featuredstory/401444/grid-computing/

we had been working with various players and were suppose to get $20M to
tie together the various NSF supercomputer sites. Then congress cuts the
budget and some other things happen. Finally they come out with an RFP
but internal politics prevent us from bidding. The director of NSF tries
to help by writing the company a letter, but that just makes the
internal politics worse (as does comments about what we already have
running is at least 5yrs ahead of bid responses). Some old nsfnet
related email
http://www.garlic.com/~lynn/lhwemail.html#nsfnet
posts mentioning NSFNET backbone
http://www.garlic.com/~lynn/subnetwork.html#nsfnet

We had project with T1 and faster speed links on the internal network
.. some of the past posts
http://www.garlic.com/~lynn/subnetwork.html#internalnet

... one of the differences was that all internal links had to be
encrypted ... which effectively required link encryptors. some old crypt
related email 
http://www.garlic.com/~lynn/lhwemail.html#crypto

includes some proposal for a PGP-like implementation in 1981. One of the
issues was that software DES for sustained full-duplex T1 would have
required dedicating 100% of both processors of large mainframe 3081K.

Now I didn't like what I had payed for T1 link encryptors and finding
link encryptors faster than T1 was really hard ... so I got involved in
doing our own; the design was to be able to handle several megabytes
(not megabits) per second sustained and could be built for under
$100. At first the corporate crypto product group claimed that it
significantly reduced DES crypto strength. It took me 3months to figure
out how to explain to them what was going on and convince them it was
significantly stronger than DES rather than significantly
weaker. However it was hollow victory, and I realized that there were
three kinds of crypto 1) the kind they don't care about, 2) the kind you
can't do, 3) the kind you can only do for them (I was told I could build
as many as I wanted, but they would have to all be sent to an address in
Maryland; and I couldn't use any of them).

Later (after we left), we were brought in as consultants to a small
client/server startup that wanted to do payments on their server; they
had developed this technology they called "SSL" they wanted to use, the
result is now frequently called "electronic commerce". We had to map the
technology to payment business process, audit/walk-thru these new
businesses selling SSL digital certificates, and establish deployment
requirements. Almost immediately webservers found that "SSL" cut their
throughput 80-90% and they dropped back to just using "SSL" for
checkout/payment.

Note, basic SSL assumption was that users understood the relationship
between the webserver they wanted to talk to and the URL they typed
in. The browser would then use SSL to validate that the webserver being
talked to corresponded with the URL typed in. Both were needed for the
webserver being talked to was the webserver the user thought they were
talking to. Webservers dropped back to only using SSL for
checkout/payment. Now the URL the user typed in was no longer
validated. Then payment URL was provided by clicked on button from the
unvalidated webserver. The result was that now SSL established that the
webserver being talked to was the webserver it claimed to be (but not
necessarily the webserver the user thought it was).

-- 
virtualization experience starting Jan1968, online at home since Mar1970

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to