Results of my first poll ever

A week ago, I decided to publish the first ever poll on the Lightweight Linux. I asked my readers if the can write code and if they actually do it.

Interestingly, most of my readers either write code at least sometimes or would like to learn to write program code.

This should not come as a surprise: the Linux users are, after all, geekier than the general audience. Most of us probably have read a few computing books and can at least write an occasional script.

During the last week I've been busy working on some client search engine optimizing (SEO) and search engine marketing (SEM) cases. At the moment, my main software tool for keeping the spreadsheet data in order is Excel, even if I very often decide to export the data into CSV files that can be searched, sorted and analysed very easily with tools like grep and sort.

The main reason for my interest in polishing my coding skills is that I want to be more efficient in analysing the business data. Some scripting is always necessary for being able to transform the data into actionable knowledge. Only then the data can be used to inform me in making better use of client money in marketing.

Fortunately, the OS X command line provides me with all the tools I need for this. Basically all the experience I have gained with using Linux is transferable to OS X shell. Thus there is no real need for building a dual boot system with Linux.

Moreover, it would be far more useful to install Windows on the Macbook as the "free" Excellent Analytics tool can only be used with Windows version of Excel.

Of course, I would be more than happy if there is a really free alternative framework that can be used for connecting Libre Office with Google Analytics. In that case I would have something to tell my colleagues at work -- and a reason for pushing free software forward in a business environment.

Random thoughts about Linux and my job

Knowledge of Linux probably helped me indirectly to get my job -- even if I don't actually need to do any hacking as part of my job. People geekier than me can do the heavyweight php scripting much more efficiently than I can. In addition, I decided to use OS X as main main desktop system at work.

Even as OS X user my Linux background (if using Linux on desktop for more than 10 years qualifies as a background) helps me with some of my daily computing needs.

For example, I'm not very efficient Excel user. Because of this, I very often do some keyword analysis using the OS X command line. For me it is simply easier to export huge Excel files into CSV and to use grep and its friends for finding the spreadsheet lines that contain interesting strings or substrings.

A few oneliners can get me very far when analysing spreadsheets with thousands of lines and I think it is easier than learning to use Excel really well. After all, it was never designed to be a tool for natural language processing.

Unfortunately, I have never had time to learn Perl that certainly would be the best tool for this kind of tasks. Its syntax has always seemed a bit too cryptic for my taste and I have never really needed to do any quantitative analysis of textual data.

Maybe, just maybe, later this year I would have time for learning Perl.

Or should I rather learn to code in Python?

In the late 1980s, I used to code quite a lot with my CBM-64. So algorithmic thinking is nothing new for me, even if the modern languages are completely different from 6510 assembly code.

So any modern language is pretty much completely new to me. I just have to decide which direction I want to take in my career.

I don't have any plans to become a website coder, so I doubt I would never really need Php + Html more than I understand now.

I code my MSc theses in R, but I don't think it is very useful for a marketing consultant.

Perl would be fine for analysing textual data, but I'm not so sure if it would be an overkill for my needs.

Python, on the other hand, would probably be fine for most of my tasks - at least combined with some elementary Unix scripting.

Or should I just forget these languages that are rather unorthodox in the marketing and business world and concentrate on learning Excel and Visual Basic...

GALPon MiniNo needs only 64 MB RAM

It's been a while since I actively updated this blog. At the moment it seems that I might be able to make a comeback with this blog that really needs some updated content.


During the hiatus of my blog, several new distributions meant for old computers and hardware with limited resources have entered the market. Thus there are plenty of new topics to blog about!


When I started this blog, old computers were very limited by today's standards. Thus lightweight linux distributions might today address hardware that would have been very modern a couple of years ago. Even my main desktop at home is probably seen very old by many computer hobbyists.


Because of this, I was happy to find out that some of the new distros would still run on hardware made in the 1990's. One of this for me new distros is GALPon MiniNo that requires only 64 Mb RAM, 1.5 Gb disk space and a 200 MHz CPU. According to the techical specifications published on the project's website, the optimal requirements are a bit higher: 128 Mb RAM and 600 MHz CPU.


This, of course, means there is no place for GNOME or KDE. Instead of these, GALPon uses a tailored version of IceWM as desktop and ROX-Filer. The custom theme is very GNOME like which makes it easy to use even for the non-geek Linux users.


The latest release 1.2 was released almost 18 months ago. The distro seems to be actively developed -- I'll certainly test the distro after the next release!

Links: GALPon MiniNo Home Page.