Search This Blog

Saturday, November 14, 2009

e-governance, FOSS and public administration transparency in Greece

Greece is facing a quite tough period at the moment. The economy is in ruins , unemployment is rising and public debt becomes a major issue. Political stability has been a major problem. The October national election result was a big blow to the "New Democracy" conservative party which saw its percentage falling to an all time low, since the restoration of democracy in Greece in 1974. That was a fair result, since ex Prime Minister Kostas Karamanlis failed to address major social and economic issues. His greatest failure was to secure fairness and transparency in public administration, a goal he vowed to achieve as part of his government campaign and a chronic problem of Greece.

The new tenants of the government buildings, the PASOK socialist party regained power and promise to address -amongst all the other major issues - the issue of fairness and transparency in public administration.  As part of their arsenal to tackle this problem, they propose an e-governance framework to reduce bureaucracy and assure fairness in the public sector hiring procedures, a major area where corruption has infiltrated over the years.  It is also clear that there is a clear motion from PASOK's inner circles to use Free Open Source Software (FOSS) and utilize already existing Greek frameworks/communities for that effort (ΕΛΛΑΚ).

There is no doubt that the FOSS adoption will meet opposition from well established Microsoft shops that use to dominate the public sector IT purchase market (and the taxpayer's pocket). In short, I have the following recommendation for the Greek government (if they ever bother to listen)

  • To follow/migrate to open source products in public administration not only on the desktop (Linux/Open Office) but also at the server end if possible. The cost benefits will be enormous, especially now that there is an active FOSS community.
  • To start teaching actively FOSS issues and software in secondary education. It is completely unacceptable to have an IT/Computing cirriculum that mentions almost nothing about FOSS.
  • Follow the paths of other countries (Netherlands, Norway) and making Open Document formats a standard for public services.
  • In response to the opengov.gr effort, this is a first and is a very positive move. However, the timing of the effort (the expectations of people) shows that the backend pipelines are still not able to work properly. The end result is that thousands of CVs are on standby and technologies to filter them out still do not exist. This could backfire badly and derail the effort.   

Happy 40th year Internet - A historical perspective on the evolution of computing


While I was at a conference in Mexico, I missed a very important date: The 29th of October should be a special day for all internet users and IT/Data networking professionals. The 29th of October 2009 marks the 40th anniversary of the Internet.

On October 29th 1969 , Leonard Kleinrock's team (Frank Heart, Willy Crowther, Dave Walden, Bernie Cosell, Bob Kahn, Ben Barker and others)  at UCLA's Samuel School of Engineering and Computer Science fired up his part of the Interface Message Processor (IMP) SDS Sigma-7 host machine and attempted to login/communicate with an SDS-940 computer located at the offices of the Douglas Engelbart's group (yes, the inventor of the computer mouse) of the Stanford Research Institute (now SRI International).


The very first transmission had to be episodic. The cleartext login banner made its way down the routes of the packet switch network, but a fault with the SDS-940 machine ensured that only just three characters of the login prompt made their way through. The fault was fixed about an hour later and the first successful remote login took place, writing history as the first successful ARPANET remote login, in fact the first ever successful login of the Internet era.

Every major historical event offers the opportunity for reflection. I entered the world of computing much later in 1987, during the era of the Zilog Z80 home computers (an Amstrad CPC 6128), and I remembered the excitement of connecting to my first BBS (as much as the screaming of my parents when they saw an elevated phone bill). 8 years later, a DECstation with the Mosaic browser got me into the era of the World Wide Web. Today, we talk about the semantic web and we have transformed our lives by means of using Internet technologies, to work, communicate, and even make social connections.

Despite the entanglement of rich content down the packet network, I can't help but notice that the very fundamental algorithms that govern the operation of networks (queue theory, routing) remain the same. Sure, speeds are up, encapsulated data complexity is up but the basics are still the same.

What's more fascinating is that this trend is true beyond data networking, and in my humble opinion is universal across the entire field of computing. In modern processors, for example, the miracle of wafer manufacturing and materials technology has made computing affordable for the masses and skyrocketing speeds possible. However, the very fundamental architectural framework behind the operation of all modern computers has been the same since the days of John Von Neumann. In computer programming, new high level languages have revolutionized the productivity of the programmer by letting him focus on the task at hand. Garbage collection, object orientation, interfacing and scripting are some of the techniques used today to shift the programmer's mind from the low level machine details to the details of the task to perform. However, at the compiler/interpreter level, you still have more or less C/C++ libraries and machine code to do the job. Many aspects of high level languages are still the same as Konrad Zuse's Plankalkül .


This raises important questions about the fundamental ground breaking research in the field of computer architecture. We sure have many computing inventions in the applications area, but we also have some big setbacks (connectionist architecture, AI), when it comes to true parallel operations. I am not pessimistic, I am just very aware that we "stand on the shoulders of giants".