[xml-dev] A question about REST and transaction isolation

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

Mitigation is always the first smart step, the 
network effect of that is positive.  The next step 
should be to cure the disease.  Isolation will be 
the first step because the risks of having some 
systems blown off the air by easy and stupid means 
are too great.

What does REST contribute to the concept of transport 
agnostic formats?  Can we do that if we are stuffing 
HTTP namespace names everywhere?  Does REST care?

1.  Not much.  REST lives above the transport format.
2.  Yes, but it will be confusing.
3.  HTTP shouldn't care about TCP/IP vulerabilities, 
    but tunneling through systems can exploit this, yes?

I'm not a network wonk.  But I may have to become one 
just to answer these rose colored glassed RFPs.

len

From: David Megginson [mailto:dmeggin@attglobal.net]

Bullard, Claude L (Len) wrote:

> one should look at the middle ages during the plagues when those that 
> could left the cities and formed closed enclaves to save what they could 
> of their own resources.  It didn't always work, but when it worked, the 
> local effect was positive.

 From what I recall of my limited reading, the areas with the lowest
mortality rates from the Black Death were the big northern Italian cities 
like Milan, which had effective governments who introduced good sanitation 
and public health measures.  Retreating from the cities was not a very 
useful way of dealing with the plague, especially since the bubonic version 
was not transmitted directly from person to person.

To the late-medieval mind, simply cleaning up the streets and providing 
clean water and a clean living environment must have seemed like a stupid 
and easy approach compared to mass public pennance, pilgrimages to holy 
sites, retreat into the country, pogroms against Jewish residents, etc., but

it often cut the death toll in half all the same.