FW: Network Analysis

[Robert Howard-2-3]

 

 

Robert Howard
Phoenix, Arizona

 

  _____  

From: Robert Howard [mailto:[hidden email]]
Sent: Sunday, May 27, 2007 10:56 PM
To: 'The Friday Morning Applied Complexity Coffee Group'
Subject: RE: [FRIAM] Network Analysis

 

Phil,

 

I?d like to see a diagram of a network of self-referential ?cells? with no
inputs or outputs. I?m visualizing a bunch of polka dots on a white board.
:-)

 

To answer your questions, the SHA algorithm is a one-way trapdoor hashing
function. It takes any string input (like a password) and returns a
constant-length output string that statistically maps bijectively to the
input string. No two inputs should have the same output (if they do, it?s
called a collision). Any input always (statistically) produces the same
output. But any given output cannot (easily) generate the input, which is
why they call it a ?one way? bijective function, and smugly brag about its
security. People use SHA to create fingerprints for documents, and then sign
these fingerprints with asymmetric cryptographic algorithms, like elliptic
curves.

For a network analogy, SHA is like a finite looping network. It loops 80
times. Each loop has a single input and a single output. In the diagram
<http://en.wikipedia.org/wiki/SHA-1> , during any loop, the input (the ABCDE
at the top) splits into one of nine branches where some are multi-input
transformations (like shifts and lookup tables). These branches ultimately
make their way down to the output (the ABCDE at the bottom).

 

Each loop (or iteration) produces an equivalent set of nodes (or cells) in a
feed-forward network mesh. Sort of like if you did this (but 80 blocks
long):



So there?s exactly one way (i.e. deterministic) to flow from left-to-right,
but there were originally estimated to be 2^80 ways to go from
right-to-left, which is why it?s considered secure. Put your password
?FLUFFY? in the left, and you get something like
?CHGDTSHJDGFBCGAFSRWFDGFBCHXGAY? out. Computers store that output in files
to authenticate users. If the file is stolen, well, no one can know your
password UNLESS they put the long output string in the right side of the
network and trace it back to the left side for something that?s
recognizable, like ?FLUFFY?. But, they?ll never guess the correct
right-to-left path because there are too many to consider, and so they?re
forced to try them all (brute force) looking for a match, which takes
forever.

However, recently, thanks to some Chinese smarty pants who did their own
network analysis (probably much like what your conference is talking about),
the number has dropped to 2^69.

When that number degrades down to about 2^40, a laptop computer can start
forging signatures.

So that?s why I?m interested in the latest techniques for network analysis.
It doesn?t matter whether the network is software generated, or physically
?out there?. The problem domain is the same.

 

Robert Howard
Phoenix, Arizona

 

  _____  

From: [hidden email] [mailto:[hidden email]] On Behalf
Of Phil Henshaw
Sent: Sunday, May 27, 2007 3:41 PM
To: 'The Friday Morning Applied Complexity Coffee Group'
Subject: Re: [FRIAM] Network Analysis

 

I think the thing I find different about the NetSci models is that they
approach networks as self-referential 'cells' of interconnection, with no
inputs or outputs....and with almost no property of interest but
'connectedness'.   That has a similarity to the feedback networks of growth
systems.  How to embed self-referential networks into environments is the
trick, and I'd have a proposal for how to do that.  

 

'Getting it' is 2/3 of the puzzle with most modeling concepts I think.
Like, I don't understand at all what the function of a SHA function is,
though I can read the lettered blocks and connecting lines of the diagram.
The form of diagram is nothing like the 'hairballs' of NetSci diagrams.  It
looks more like a software decision-making diagram of some sort.  What's the
SHA hash function diagram represent?

 


Phil Henshaw                       ????.?? ? `?.????
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
680 Ft. Washington Ave
NY NY 10040                      
tel: [phone]                
e-mail: pfh at synapse9.com          
explorations: www.synapse9.com <http://www.synapse9.com/>    

-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf
Of Robert Howard
Sent: Saturday, May 26, 2007 12:44 PM
To: 'The Friday Morning Applied Complexity Coffee Group'
Subject: [FRIAM] Network Analysis

Phil wrote: Hey, did you know that network maps are real categories of real
systems that are amenable to mathematical analysis??  That's a first!   I
luckily had a great conference on the subject come to town and am enjoying
it thorough ally.   The pure theory guys are still sort of at sea I think,
but I would anyway whether they were or not, right ??   :))

 

I always thought network maps would probably cause slow progress in the
formal mathematical analysis arena. If the nodes are symmetric (or fixed in
a similar manner) then Graph Theory and Stochastic Markov Chains can be
applied. But when a network contains dissimilar node types, the test cases
break the symmetry of node traversals. Each traversal becomes a convoluted
composite recursive function.

It should be no surprise that the complexity that arises in these types of
functions is deliberately exploited in cryptographic message digests.

Just look at one hop in the SHA-1 algorithm
http://www.md5security.com/sha-hash-functions/cat.php. It uses 80 hops just
generate the final hash, which is like a network (list) of 80 chained nodes.
A real network is neither a chain nor a tree. It has ugly stuff, like
cycles, state, reconfiguring properties, and errors.

 

I would be interesting in where the analysis has evolved to.

 

Robert Howard
Phoenix, Arizona

 


  _____  


From: [hidden email] [mailto:[hidden email]] On Behalf
Of Phil Henshaw
Sent: Tuesday, May 22, 2007 5:46 PM
To: 'The Friday Morning Applied Complexity Coffee Group'
Subject: Re: [FRIAM] Global warming - the real story + TeslaMotors and
SpaceX

 

well, a 100mpg car that performs like a 15mpg car does indeed sound so sweet
you wonder if it's true, and if it comes with open roads where you can still
do that sort of thing..., especially when you realize that what we actually
need to keep from multiplying resource consumption with continued growth is
1000 mpg cars by the end of the century and 32000mpg cars by the end of the
next.   I find those efficiencies from a 2nd law obeying systems of any kind
worse than dubious...    

 

Hey, did you know that network maps  are real categories of real systems
that are amenable to mathematical analysis??  That's a first!   I luckily
had a great conference on the subject come to town and am enjoying it
thorough ally.   The pure theory guys are still sort of at sea I think,...
but I would anyway whether they were or not, right ??   :))

 

 


Phil Henshaw                       ????.?? ? `?.????
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
680 Ft. Washington Ave
NY NY 10040                      
tel: [phone]                
e-mail: pfh at synapse9.com          
explorations: www.synapse9.com <http://www.synapse9.com/>    

-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf
Of Robert Howard
Sent: Tuesday, May 22, 2007 4:27 PM
To: 'The Friday Morning Applied Complexity Coffee Group'
Subject: Re: [FRIAM] Global warming - the real story + TeslaMotors and
SpaceX

Technologies like this give me hope: http://www.teslamotors.com
<http://www.teslamotors.com/>  

Projects like this give me hope:
http://www.wired.com/science/space/magazine/15-06/ff_space_musk 

The company is owned by Elon <http://en.wikipedia.org/wiki/Elon_Musk>  Musk,
founder of PayPal, and now SpaceX <http://en.wikipedia.org/wiki/SpaceX>

 

Robert Howard

Phoenix, Arizona

 

 

-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf
Of Stephen Guerin
Sent: Monday, May 21, 2007 11:10 AM
To: 'The Friday Morning Applied Complexity Coffee Group'
Subject: Re: [FRIAM] Global warming - the real story

 

Googling "Connie Meskimen" shows her to have a sarcastic wit:

http://www.snopes.com/humor/letters/daylight.asp

 

-Steve

 

> -----Original Message-----

> From: Robert Holmes [mailto:robert at holmesacosta.com]

> Sent: Monday, May 21, 2007 10:04 AM

> To: FRIAM

> Subject: [FRIAM] Global warming - the real story

>

> At last, someone has had the courage to tell the real story

> behind global warning.

>

> Robert

>

>

> From the Arkansas Democrat-Gazette, 16 April 2007, "Letters"

> section, page 6B:

>

> "You may have noticed that March of this year was

> particularly hot. As a matter of fact, I understand that it

> was the hottest March since the beginning of the last

> century. All of the trees were fully leafed out and legions

> of bugs and snakes were crawling around during a time in

> Arkansas when, on a normal year, we might see a snowflake or two.

>

> This should come as no surprise to any reasonable person. As

> you know, Daylight Savings Time started almost a month early

> this year. You would think that members of Congress would

> have considered the warming effect that an extra hour of

> daylight would have on our climate. Or did they?

>

> Perhaps this is another plot by a liberal Congress to make us

> believe that global warming is a real threat. Perhaps next

> time there should be serious studies performed before

> Congress passes laws with such far-reaching effects.

>

> Connie M. Meskimen

> Hot Springs"

>

>

>

 

 

============================================================

FRIAM Applied Complexity Group listserv

Meets Fridays 9a-11:30 at cafe at St. John's College

lectures, archives, unsubscribe, maps at http://www.friam.org

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070527/9757f977/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 19833 bytes
Desc: not available
Url : http://redfish.com/pipermail/friam_redfish.com/attachments/20070527/9757f977/attachment-0001.jpe 

[Phil Henshaw-2]

that's what all those super-duper 'hairballs' are...  Here are the links
I assembled for others.   I don't think cryptography algorithms
(throughput channels) define a network in the sense of forming 'a'
network.   Until anyone listens to my suggestion ( a couple notes ago )
I think they're bound to be 'windowless monads' of a sort, but still
highly useful.
 
NetSci conference:    <http://www.nd.edu/~netsci/conference.html>
http://www.nd.edu/~netsci/conference.html 

All the videos and displays of the 6 days of talks are expected to be
made available.  Roughly half were purely theoretical, concerning the
abstract properties of networks of links, and half were about the
networks identified in various kinds of biological, social or economic
systems.  If there's one that stuck in my mind, both for clarity and for
potential value, it was Ricardo Hausmann's model of 'product space'
(Wednesday) identifying specific learning pathways for national economic
development based on the empirical networks of knowledge communities
found in their mix of different products and services.

Much of the content of the conference was highly technical, concerning
the use of fancy new analytical tools, developed from scratch by the
presenter or with using the new open source software built with national
research grants, such as:

?Network Workbench? -  <http://nwb.slis.indiana.edu/>
http://nwb.slis.indiana.edu/

?Social Action? -  <http://www.cs.umd.edu/hcil/socialaction/>
http://www.cs.umd.edu/hcil/socialaction/

Links to the main centers are

Notre Dame  <http://www.nd.edu/~networks/> http://www.nd.edu/~networks/
and Laszlo's page  <http://www.nd.edu/~alb/> http://www.nd.edu/~alb/

MIT  <http://web.media.mit.edu/~barahona/main/links/SNA.htm>
http://web.media.mit.edu/~barahona/main/links/SNA.htm 

Harvard  <http://www.iq.harvard.edu/blog/netgov/>
http://www.iq.harvard.edu/blog/netgov/

There are several other research communities exploring other aspects of
complex information analysis and display, visualization of data is the
general term, helping to push the limits of what we can understand...
and hopefully also discover why we constantly have to.

 
 
 

Phil Henshaw                       ????.?? ? `?.????
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
680 Ft. Washington Ave
NY NY 10040                      
tel: [phone]                
e-mail: pfh at synapse9.com          
explorations: www.synapse9.com <http://www.synapse9.com/>    

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070528/b1ce42f4/attachment.html