Update your readers! These are the new URLs:

• RSS 0.92
• RSS 2.0
• Atom

Posted by Adrian at 03:37 PM | Permalink

Hi everyone,

This is the last entry of this blog; I will come back soon with a new site, with a new look and feel and new contents. Thanks to all of you for your support, comments and suggestions! This blog will remain at this URL for reference purposes (particularly regarding the popular Ubuntu entries!).

Posted by Adrian at 02:36 PM | Permalink

After upgrading my good ol' PC to Dapper, I proceeded to download and burn a copy of it for PowerPC systems, and I have managed to install it in my old faithful G3 iBook in dual boot, with Mac OS X 10.2 "Jaguar" on the other partition. Here's how I did it.

Continue reading "How to Install Ubuntu 6.06 "Dapper" in an Apple G3 iBook" »

Posted by Adrian at 04:16 PM | Permalink | Comments (1)

"Dapper", the new version of Ubuntu has been released yesterday! So I downloaded the ISO file, burn the CD down and proceeded to install it over my old Breezy Kubuntu installation.

Of course, I wanted to use my good' old Linksys WPC54GS Wireless-G Network Adapter with Broadcom chipset, and I was lucky enough to find this page:

https://wiki.ubuntu.com/WifiDocs/Driver/bcm43xx?action=show&redirect=WifiDocs%2FDriver%2FBroadcom43xx#head-e70dd6b5c57894d32e3eddc4f3e21d7d6d02230f

It describes the whole Broadcom problem in Ubuntu, and gives the instructions needed to make the card work in Dapper:

1. Follow the instructions in http://www.kosmaczewski.net/blogs/tech/archives/2006/02/how_to_install_1.php from steps 1 to 7 (step 8 cannot be done, yet)
2. Type "echo 'blacklist bcm43xx' | sudo tee -a /etc/modprobe.d/blacklist"
3. Type "sudo rmmod bcm43xx",
   "sudo rmmod ndiswrapper" and
   "sudo modprobe ndiswrapper"
4. Do the step 8 of http://www.kosmaczewski.net/blogs/tech/archives/2006/02/how_to_install_1.php

And that's it! You've got the wireless card ready for use.

:) I'm happy to have found that page! It seems that the Broadcom drivers that come with Dapper do not work with all concerned wireless cards, so the thing is to "blacklist" them, and then to load ndiswrapper that works perfectly well.

Posted by Adrian at 12:18 AM | Permalink

Software Example

In the February 2006 issue of MSDN Magazine (http://msdn.microsoft.com/msdnmag/), Matt Neely describes a .NET implementation of mobile agents:

"The term agent originates in artificial intelligence and describes a logical entity that has some level of autonomy within its environment or host. A mobile agent has the added capability to move between hosts. In a computing context, a mobile agent is a combined unit of data and code that can move between different execution environments."

The idea described in the article is that of a small family of .NET classes that literally "jump" from a computer to another, performing tasks in the host computer, through a mechanism called Remoting:

"An example of a traveling agent app could perform operations control. An agent is sent out with a list of machines on the local network it should traverse to inventory hardware and software (...) built-in services that facilitate the componentization and mobility of code, namely object remoting and serialization.(...) Mobile agents have their uses and their pros and cons. The autonomous and mobile nature of mobile agents can lead to reduced network traffic, decentralization, increased robustness and fault-tolerance, and easy deployment."

(Source: Neely, 2006)

Continue reading "Intelligent Software Agents - A .NET Example" »

Posted by Adrian at 05:00 PM | Permalink

The other day I found this website, and created my personal dna report with it. Kind of interesting, actually, the results are not far from the truth :)

Continue reading "My personal dna report" »

Posted by Adrian at 10:55 PM | Permalink

This morning I said to myself; I should write something for my blog... but had no subject to write about; so I found a faster way: I went to http://pdos.csail.mit.edu/scigen/ and asked the SCIgen Automatic CS Paper Generator to create a document suitable to post here, to keep you entertained for a while. The results are amazing!

Enjoy! :)

A Case for Compilers

Adrian Kosmaczewski

Download a Postscript or PDF version of this paper.
Download all the files for this paper as a gzipped tar archive.

Abstract

Continue reading "A Case for Compilers - A Fake Paper" »

Posted by Adrian at 06:01 PM | Permalink | Comments (1)

Does OOP reflect a "natural" way of thinking? Is it a better choice than the procedural programming paradigm?

In computer science, to say that one approach is "better" than another is to miss a great detail: I do not think that there are "better" or "natural" paradigms per se, but just apppropriate answers to certain problems in a given context.

In his 1962 book "The Structure of Nature Revolutions", Thomas Kuhn introduces the idea of the "paradigm shift"; following this idea, human knowledge does not evolve gradually, but rather in discontinuous jumps, called "paradigm shifts" or "scientific revolutions":

"A scientific revolution occurs, according to Kuhn, when scientists encounter anomalies which cannot be explained by the universally accepted paradigm within which scientific progress has thereto been made. The paradigm, in Kuhn's view, is not simply the current theory, but the entire worldview in which it exists, and all of the implications which come with it. There are anomalies for all paradigms, Kuhn maintained, that are brushed away as acceptable levels of error, or simply ignored and not dealt with (a principal argument Kuhn uses to reject Karl Popper's model of falsifiability as the key force involved in scientific change)."

(Wikipedia, 2006)

In the case of the activity of software engineering, the paradigm shift from procedural to object-orientation is quite evident, both historically and technically speaking.

Continue reading "About OOP and other programming paradigms" »

Posted by Adrian at 12:55 PM | Permalink

Introduction

I have been working as a software developer since 1996, and as such I've used a variety of different languages, both compiled and interpreted. But the who languages that I know and use most today, are two somewhat different ones, C# and Ruby. I will begin my presentation with a short explanation of both, providing their major similarities and differences, and then providing some code samples of both.

Both languages are ranked #7 and #21 respectively in the TIOBE Programming Community Index, as of February 2006 (http://www.tiobe.com/tpci.htm).

Continue reading "Quick Comparison of C# and Ruby" »

Posted by Adrian at 12:25 PM | Permalink

Well the PowerPoint slides that I used during the TechDays 2006 conference have been published in the TechDays page. You can download them from this site as well:

• A204 - SharePoint Workflow (zipped, 6.5 MB)
• D308 - LINQ (zipped, 6.7 MB)

BTW, they are in French... and yes, the slide with Maradona on it is maybe the most important of them all :))

Posted by Adrian at 07:23 AM | Permalink

OK, so this time I've tried to make the same I've described before, but for Kubuntu 5.10. I have changed to Kubuntu since I like KDE more than gnome, and also, Kubuntu seems to run faster than Ubuntu. And so I said to myself, OK, these are the same guys who make Ubuntu and Kubuntu, the wireless stuff should work fairly easily. After all it's the same kernel...

Wrong.

Continue reading "Kubuntu 5.10 and the Linksys WPC54GS Wireless-G Network Adapter" »

Posted by Adrian at 12:01 AM | Permalink

Cuando AOL (America On Line) alla a fines de los 80, le puso a su sistema de correo electronico una voz de chabon que decia "You've got Mail!" cada vez que llegaba un nuevo mensaje, poco se imaginaba Ginebra de que diez años mas tarde hasta los gorriones tendrian direccion de e-mail.

Continue reading ""You've got Mail!"" »

Posted by Adrian at 01:41 AM | Permalink

You knew Microsoft Reporting Services, Microsoft Excel Services, Microsoft Services for UNIX, Microsoft Windows SharePoint Services, and of course, the Service Packs.

Well, let me introduce you Microsoft Catering Services (aka Service Pack for human beings, or the Microsoft Stomach Update Services):

Posted by Adrian at 05:15 PM | Permalink

Introduction

"The true operating system is the net itself"

This phrase, common marketing argument in the late nineties, made me remind that in the eighties, Sun Microsystems' founder, Scott McNealy, used the slogan "The Network is the Computer" to describe his vision:

Source: http://www.csg.is.titech.ac.jp/~mich/photos/visitsun/13/full/13004.jpg

But, can we safely mix both concepts?

Continue reading "About Operating Systems and Networks" »

Posted by Adrian at 09:39 PM | Permalink | Comments (1)

Since data and instructions are stored in RAM in pretty much the same way, a priori the CPU cannot distinguish each other, but by the cycle in which the binary chunk is fetched from memory. In the case of instructions, it then needs to decode the operation codes into instructions, with the added problem that if the operation is performed on data that is not implied by the operation code, the results are wrong or even catastrophic.

The question is: would it be useful if in hardware, each cell of data would carry its own type designation? I will discuss here the pros and cons of this approach, in respect to hardware and software architectures.

Continue reading "Hardware Polymorphism" »

Posted by Adrian at 12:43 PM | Permalink

Not only because of the fact that I love my Macs, but because now Apple supports booting in Windows XP from it as well!

That did make my jaw drop on the floor. Definitely.

More info here and here.

Posted by Adrian at 06:39 PM | Permalink | Comments (2)

While I was reading this blog post about "Easter eggs" in Microsoft products, I came accross these two (utterly brilliant) comments in the same page:

The first from Kevin Daly:

Let's be honest, we all know perfectly well that the real reason corporate customers don't like Easter eggs is that in the core of their shrivelled little souls they believe that having fun while you work is the same as stealing from the company.

This is the culture of the suit, the meaningless mission statement, and the ruthless elimination of all signs of joy and humanity.

The Open Source movement's got it all wrong: it's not software that needs to be made free, but the people who use it. And the people who write it.

The second from mark

The word you're missing when discussing Easter Eggs is "PRIDE". Easter eggs wind up in software because someone is proud of their work. The original easter egg in a game was put in there because the programmer wasn't getting any credit anywhere in the code, on the box, or in the manual... and he was proud of what he'd done.

You want software written by people who take pride in their work, the same way you want a car built by someone who isn't disgruntled at his employer or food served to you by someone who hates their customers.

Talking about pride...

Continue reading "A more boring world" »

Posted by Adrian at 11:25 AM | Permalink

Somewhere I read that it was a good thing to learn at least one new programming language every year; I think I have kept up that trend since 1992:

• 1992: QBasic
• 1993: Turbo Pascal
• 1994: C
• 1995: "Classic" Visual Basic
• 1996: Java
• 1997: JavaScript
• 1998: VBScript
• 1999: Transact-SQL
• 2000: C++
• 2001: C#
• 2002: Objective-C
• 2003: AppleScript
• 2004: Visual Basic.NET
• 2005: Ruby

And this year's winner is: LINQ. The main purpose of learning it is to prepare the LINQ conference in the TechDays next week... and this is huge indeed!

Posted by Adrian at 05:31 PM | Permalink

In his weblog, Michael Platt has posted an interesting article, comparing different definitions for the word "Architecture"; interesing read! http://blogs.technet.com/michael_platt/archive/2006/03/27/423300.aspx

Posted by Adrian at 10:02 PM | Permalink

Again from Redmond... A quick tip for all of you, ASP.NET developers in the field: stop everything you are doing, and learn all what you can about these two technologies:

• Windows Workflow Foundation
• Windows SharePoint Services and Office SharePoint Services (not the same thing...)

This applies particularly to those developers working in "pure" business environments with Microsoft technologies; do not reinvent the wheel; if your application involves any of the following:

• Tracking and reporting scenarios
• List or libraries with items or documents
• Forms or views of items of data (whatever the type)
• Applications with typical workflow logic
• Role-based security

Then stop looking and use SharePoint. There is not any other portal product in the market that is so tightly integrated with Windows, SQL Server, Active Directory or Office; I must say that what I've seen so far in the MOSDC 2006 is really impressive. The SharePoint platform provides a complete design, development, deployment, execution and management scenario for building business applications, and really, you should not reinvent the wheel.

Posted by Adrian at 01:17 AM | Permalink