Julian Assange Dear Chap

Julian Assange a very Dear Chap
couldnt control his pecker
got caught in a honey trap
Should have kept that rubber on, Jules
Nordic Scandinavians may be easy but even they have rules

meanwhile Dear Chap’s Website the eponymous Wikileaks
is leaking revolution and democracy like  Vegas casino magic tricks
The Arabs read his website before Sentor Joe crashed it down
And now  Anglo Saxon allies in Egypy, Tunisia, Libya, Yemen, Bahrain are wearing a frown

Viva La Website Revolution Wikileaks
Merde to the Dear Chap\s pecker squeaks
Time up, time for all dictators to go and hide,
rulers Arabian, or Aussi hackers on a funny ride.

Google Realtime Live Updates on Egypt Yemen Tunisia Jordan..

Using Google RealTime, a small icon on the left margin, you can monitor the latest uprisings. Apparently you can still get shot in most of the world to ask for freedom. What a trillion dollars of industrial arms complex could not do in Iraq or Afghanistan, hackers at Wikileaks, Bloggers in Middle East and Media people at Al Jazzera are doing right now. I am probably too young in 1989 when communists fell, but watching dictators fall by people power than external arms is good, no.

Now if only a few more people could listen to some Chinese Democracy

Stuxnet DeMystified

Detail of a New York Times Advertisement - 1895
Image via Wikipedia

A fascinating article in New York Times details the fascinating details of the Stuxnet virus, apparently the most successful cyber weapon in recent times.

Given that Industrial Controllers are a part of a everything from factories to missile launch configurations, I believe this is a fascinating area of study for the world’s research scientists including creating variants and defenses for this.

https://www.nytimes.com/2011/01/16/world/middleeast/16stuxnet.html

Also a 2008 presentation by Siemens that the NYT was kind enough to link to- (whither Wikileaks ??)

Towards better quantitative marketing

Cycle of Research and Development, from "...
Image via Wikipedia

The term quantitative refers to a type of information based in quantities or else quantifiable data (objective properties) —as opposed to qualitative information which deals with apparent qualities (subjective properties)

http://en.wikipedia.org/wiki/Quantitative

Fear, uncertainty, and doubt (FUD) is a tactic of rhetoric and fallacy used in sales, marketing, public relations,[1][2] politics and propaganda. FUD is generally a strategic attempt to influence public perception by disseminating negative and dubious/false information designed to undermine the credibility of their beliefs.

Source-

http://en.wikipedia.org/wiki/Fear,_uncertainty_and_doubt

Top 5 FUD Tactics in Software and what you can say to end user to retain credibility

1) That software lacks reliable support- our support team has won top prizes in Customer Appreciation for past several years.

  • Our software release history-
  • graph of bugs filed-
  • turn around time box plot for customer service issues
  • quantitatively define reliability

2) We give the best value to customers. Customer Big A got huge huge % savings thanks to our software.

  • Pricing- Transparent – and fixed. For volume discounts mention slabs.
  • Cost to Customer- Include time and cost estimates for training and installation
  • Graphs of average ROIC (return on capital invested) on TCO (total cost of ownership)  not half a dozen outlier case studies. Mention Expected % return

3) We have invested a lot of money in our Research and Development. We continue to spend a lotto of money on R &D

  • Average Salary of R and D employee versus Average Tenure (Linkedin gives the second metric quite easily)
  • Mention Tax benefits and Accounting treatment of R&D expenses
  • Give a breakdown- how much went to research and how much went to legacy application support
  • Mention open source projects openly
  • Mention community source projects separately

4) Software B got sued. Intellectual property rights (sniff)

  • Mention pending cases with your legal team
  • Mention anti trust concerns for potential acquisitions
  • Mention links to your patent portfolio (or even to US PTO with query ?=your corporate name )

5) We have a 99.8% renewal rate.

  • Mention vendor lock in concerns and flexibility
  • Mention What-If scenarios if there are delays in software implementation
  • Mention methodology in calculating return on investment.

 

 

 

Also

http://blogs.computerworlduk.com/infrastructure-and-operations/2010/10/three-fud-statements-used-not-to-implement-standards-based-networking/index.htm

2011 Forecast-ying

Free twitter badge
Image via Wikipedia

I had recently asked some friends from my Twitter lists for their take on 2011, atleast 3 of them responded back with the answer, 1 said they were still on it, and 1 claimed a recent office event.

Anyways- I take note of the view of forecasting from

http://www.uiah.fi/projekti/metodi/190.htm

The most primitive method of forecasting is guessing. The result may be rated acceptable if the person making the guess is an expert in the matter.

Ajay- people will forecast in end 2010 and 2011. many of them will get forecasts wrong, some very wrong, but by Dec 2011 most of them would be writing forecasts on 2012. almost no one will get called on by irate users-readers- (hey you got 4 out of 7 wrong last years forecast!) just wont happen. people thrive on hope. so does marketing. in 2011- and before

and some forecasts from Tom Davenport’s The International Institute for Analytics (IIA) at

http://iianalytics.com/2010/12/2011-predictions-for-the-analytics-industry/

Regulatory and privacy constraints will continue to hamper growth of marketing analytics.

(I wonder how privacy and analytics can co exist in peace forever- one view is that model building can use anonymized data suppose your IP address was anonymized using a standard secret Coco-Cola formula- then whatever model does get built would not be of concern to you individually as your privacy is protected by the anonymization formula)

Anyway- back to the question I asked-

What are the top 5 events in your industry (events as in things that occured not conferences) and what are the top 3 trends in 2011.

I define my industry as being online technology writing- research (with a heavy skew on stat computing)

My top 5 events for 2010 were-

1) Consolidation- Big 5 software providers in BI and Analytics bought more, sued more, and consolidated more.  The valuations rose. and rose. leading to even more smaller players entering. Thus consolidation proved an oxy moron as total number of influential AND disruptive players grew.

 

2) Cloudy Computing- Computing shifted from the desktop but to the mobile and more to the tablet than to the cloud. Ipad front end with Amazon Ec2 backend- yup it happened.

3) Open Source grew louder- yes it got more clients. and more revenue. did it get more market share. depends on if you define market share by revenues or by users.

Both Open Source and Closed Source had a good year- the pie grew faster and bigger so no one minded as long their slices grew bigger.

4) We didnt see that coming –

Technology continued to surprise with events (thats what we love! the surprises)

Revolution Analytics broke through R’s Big Data Barrier, Tableau Software created a big Buzz,  Wikileaks and Chinese FireWalls gave technology an entire new dimension (though not universally popular one).

people fought wars on emails and servers and social media- unfortunately the ones fighting real wars in 2009 continued to fight them in 2010 too

5) Money-

SAP,SAS,IBM,Oracle,Google,Microsoft made more money than ever before. Only Facebook got a movie named on itself. Venture Capitalists pumped in money in promising startups- really as if in a hurry to park money before tax cuts expired in some countries.

 

2011 Top Three Forecasts

1) Surprises- Expect to get surprised atleast 10 % of the time in business events. As internet grows the communication cycle shortens, the hype cycle amplifies buzz-

more unstructured data  is created (esp for marketing analytics) leading to enhanced volatility

2) Growth- Yes we predict technology will grow faster than the automobile industry. Game changers may happen in the form of Chrome OS- really its Linux guys-and customer adaptability to new USER INTERFACES. Design will matter much more in technology on your phone, on your desktop and on your internet. Packaging sells.

False Top Trend 3) I will write a book on business analytics in 2011. yes it is true and I am working with A publisher. No it is not really going to be a top 3 event for anyone except me,publisher and lucky guys who read it.

3) Creating technology and technically enabling creativity will converge at an accelerated rate. use of widgets, guis, snippets, ide will ensure creative left brains can code easier. and right brains can design faster and better due to a global supply chain of techie and artsy professionals.

 

 

How to Analyze Wikileaks Data – R SPARQL

Logo for R
Image via Wikipedia

Drew Conway- one of the very very few Project R voices I used to respect until recently. declared on his blog http://www.drewconway.com/zia/

Why I Will Not Analyze The New WikiLeaks Data

and followed it up with how HE analyzed the post announcing the non-analysis.

“If you have not visited the site in a week or so you will have missed my previous post on analyzing WikiLeaks data, which from the traffic and 35 Comments and 255 Reactions was at least somewhat controversial. Given this rare spotlight I thought it would be fun to use the infochimps API to map out the geo-location of everyone that visited the blog post over the last few days. Unfortunately, after nearly two years with the same web hosting service, only today did I realize that I was not capturing daily log files for my domain”

Anyways – non American users of R Project can analyze the Wikileaks data using the R SPARQL package I would advise American friends not to use this approach or attempt to analyze any data because technically the data is still classified and it’s possession is illegal (which is the reason Federal employees and organizations receiving federal funds have advised not to use this or any WikiLeaks dataset)

https://code.google.com/p/r-sparql/

Overview

R is a programming language designed for statistics.

R Sparql allows you to run SPARQL Queries inside R and store it as a R data frame.

The main objective is to allow the integration of Ontologies with Statistics.

It requires Java and rJava installed.

Example (in R console):

> library(sparql)> data <- query("SPARQL query>","RDF file or remote SPARQL Endpoint")

and the data in a remote SPARQL  http://www.ckan.net/package/cablegate

SPARQL is an easy language to pick  up, but dammit I am not supposed to blog on my vacations.

http://code.google.com/p/r-sparql/wiki/GettingStarted

Getting Started

1. Installation

1.1 Make sure Java is installed and is the default JVM:

$ sudo apt-get install sun-java6-bin sun-java6-jre sun-java6-jdk$ sudo update-java-alternatives -s java-6-sun

1.2 Configure R to use the correct version of Java

$ sudo R CMD javareconf

1.3 Install the rJava library

$ R> install.packages("rJava")> q()

1.4 Download and install the sparql library

Download: http://code.google.com/p/r-sparql/downloads/list

$ R CMD INSTALL sparql-0.1-X.tar.gz

2. Executing a SPARQL query

2.1 Start R

#Load the librarylibrary(sparql)#Run the queryresult <- query("SELECT ... ", "http://...")#Print the resultprint(result)

3. Examples

3.1 The Query can be a string or a local file:

query("SELECT ?date ?number ?season WHERE {  ... }", "local-file.rdf")
query("my-query.rq", "local-file.rdf")

The package will detect if my-query.rq exists and will load it from the file.

3.3 The uri can be a file or an url (for remote queries):

query("SELECT ... ","local-file.db")
query("SELECT ... ","http://dbpedia.org/sparql")

3.4 Get some examples here: http://code.google.com/p/r-sparql/downloads/list

SPARQL Tutorial-

http://openjena.org/ARQ/Tutorial/index.html

Also read-

http://webr3.org/blog/linked-data/virtuoso-6-sparqlgeo-and-linked-data/

and from the favorite blog of Project R- Also known as NY Times

http://bits.blogs.nytimes.com/2010/11/15/sorting-through-the-government-data-explosion/?twt=nytimesbits

In May 2009, the Obama administration started putting raw 
government data on the Web. 
It started with 47 data sets. Today, there are more than
 270,000 government data sets, spanning every imaginable 
category from public health to foreign aid.