Cross-platform solutions

Recently when searching the internet for platform-independent solutions for particular requirements I got annoyed and angry about what people write on their websites. It seams that they mostly do not understand what cross-platform means.

In many cases software vendors who offer a tool or solution for Windows and Mac find that sufficient to call it a cross-platform thing. My opinion: This is pretty poor for labeling it "cross-platform". If you look at Wikipedia for instance you can see not only Windows and Mac, but also all kind of Linuxes (and there is really a lot - see distrowatch.com) including widely used Ubuntu/Debian, Redhat/Fedora or Solaris or BSD or ... (well we skip DOS, Amiga and the like here ;-) ).

So: Cross-platform is NOT just Windows and Mac! At least the major distributions of Linux should be also included in the list of supported platforms!

I simply cannot consider a software that is just running on Windows and Mac as being cross-platform... - Those vendors should write: "Blablabla - for Windows AND Mac!" instead of "Blablabla - cross-platform solution!".

Related posts: Why Linux?, Why I switched to Ubuntu.

Virus scanning on Ubuntu

You might need to download files on your Ubuntu machine (or any Linux running Gnome) that are intended for Windows machines or you might have fear getting one of the few Linux viruses (yes, there are some).

There is the free clamav software. On Ubuntu (at least 9.04 Jaunty and 9.10 Karmic) there are the following packages available:

clamav
anti-virus utility for Unix - command-line interface
clamav-base
anti-virus utility for Unix - base package
clamav-daemon
anti-virus utility for Unix - scanner daemon
clamav-freshclam
anti-virus utility for Unix - virus database update utility
clamtk
graphical front-end for ClamAV
klamav
KDE frontend for ClamAV
nautilus-clamscan
Antivirus scanning for Nautilus

There is one problem with the clamtk (the gnome GUI tool for scanning) - at least under Karmic used together with Thunderbird: Thunderbird tries to open all files with the clamtk and there if you change that it does not remember the change. And this although I don't want to scan everything automatically (even text files).

And there is a problem with with the nautilus-clamscan (at least under Karmic): It hangs in an infinite scan for the first file.

So whatever I tried, everything is annoying. Solved it the following way:

apt-get remove clamtk
apt-get install nautilus-actions (if not already installed)

Then I created a new nautilus action (via System->Preferences->Nautilus Actions Configuration) with the following command (enabled for files and folders and multiple selections):

path: gnome-terminal
parameters: -x /opt/clamscan.sh %M

And I have created the clamscan.sh with the following content (and set to be readable and executable for everyone):

#!/bin/bash
clamscan $@
read line

Last but not least:

chmod 755 /opt/clamscan.sh

Now I can scan files with clamav on demand using the context menu in Nautilus.

Releated posts: Stationary under Ubuntu, Change hostname on Ubuntu.

Change hostname on Ubuntu

In business it is a common pattern that you install one user workstation that becomes the template for others.

Basically after cloning such a template machine (e.g. using CloneZilla), the most important thing after starting the clone is to change the hostname - and here is how to do it:

There are two files to edit - and you should edit them in this order - just changing the hostname (in total should be 3 times occuring):

sudo vim /etc/hosts

127.0.0.1 nowthenewhostname localhost.localdomain localhost
127.0.1.1 nowthenewhostname

sudo vim /etc/hostname

nowthenewhostname

For those not familiar with vim - just press "i" to get into edit mode and when finished press "ESC" and enter ":x" followed by pressing "Enter".

And sure, you can use gedit instead of vim also.

See also more information on Cloning Ubuntu to Different Computer Hardware.

Related post: Virus scanning on Ubuntu, Stationary under Ubuntu.

Damn Fast Linux

I was visiting my parents on the weekend (as usually once a week) and as I stumbled over some small Linux Editions dedicated to run on older Computers, I gave it a try as I have some older Computers there. I even have a 486 with only a small amount of RAM. I did not know that there are either distributions specialized for 486 architecture also.

Unfortunately trying Linux on the 486 was not successful, because ..... because the BIOS did not support the CD drive in the boot order. I did not expect that. OK, it was the time of the 5 1/4 inch and 3,5 inch disks...

Then I tried another machine that was a lot younger - some Pentium with an XP installed and it was annoying slow already to start up. I tried Puppy Linux on it and playing around a little with the Live CD.

IT WAS SO DAMN FAST - even although from CD!

So my hint: Before you are thinking of buying a new computer just because your one is old and slow - and you do not much more than EMail, Browsing and writing some documents - then really think of giving Linux a try!

A list of Linux distributions that also should support older architectures can be found at DistroWatch.com. I would start with Puppy Linux, MEPIS, absolute Linux or Damn Small Linux for example.

Related posts: Why Linux?, Why I switched to Ubuntu, The sad thing about Linux, The operating system.

Stationary under Ubuntu

Nowadays instead of printing your business documents and sending by mail, in many cases people send PDF by email. What is missing here is usually the stationary paper that is taken from the printer tray where the comany headers and footers are pre-printed.

If you want to have a nice easy way for using or existing word processor templates adding a stationary PDF - here is what you can do:

Preparations:

sudo apt-get install nautilus-actions
sudo apt-get install pdftk

Then create a new action as:

Path: /bin/sh
Parameters: /home/username/stationary.sh '%d/%f'

And here is the content of script stationary.sh (don't forget to make it executable):
#!/bin/sh
pdftk "$1" background /home/username/stationary.pdf output "$1.new.pdf"
if [ -f "$1.new.pdf" ]
then
rm -f "$1"
mv "$1.new.pdf" "$1"
fi

And all what you need now is to put save your stationary as PDF to /home/username/stationary.pdf

Related post: Document file format, Virus scanning on Ubuntu.

Why Linux?

I do not consider it a good way how FSF started bashing Windows with Windows7Sins. Besides the fact that this is bad behavior, they should have asked the people why they keep using Windows.

Here are some arguments, I get told by people who don't want to switch OS:

1. "I don't want to invest time into learning a new system. Computers are already a big time waster."
   Learning something new is additional effort. Although it might be an investment with a short ROI, it depends how people are using the computer (and how much money they have to pay others for fixing their problems).
   
   
2. "I heard that it is pretty common to experience incompatibilities with the hardware and I am worrying about my foto printer, scanner or mobile device not working with it."
   Yes, this is indeed true. There were made big steps forward in the last years but there are still plenty of gadgets not working well. If you are planning to buy new hardware, you should prefer those items that are known to be working with Linux over others - just for the case you might want to switch in the future. Anyway, with some manual work I was able to get my Canon foto printer and my Canon Scanner to work with Fedora already back in 2006. Nowadays it should work better.
   
   
3. "I do need to share a lot of documents in Word, Photoshop and other proprietary formats with others I collaborate with. I worry about incompatibilities."
   This is indeed another thing to be taken into consideration. In the best case it is just some additional click but in the worst case you can't read or write the needed format any more. Although some Windows applications run fine under Wine (Windows-Emulator), many others do not. Although people should prefer open standards this is not always possible.
   
   
4. "I am a gamer and the most cool games run on Windows."
   Also true, although there are many free games on Linux, vendors produce new games quite only for Windows. Dual boot is an option but means to maintain two systems - and you still pay for the Windows license.
   
   
5. "I have plenty of friends around who can help me when I have troubles with my Windows-Machine, but I have no or very few options for the case I then have problems with the Linux box."
   This is also true for many as there are simply still more people around knowing Windows than knowing Linux. It does not really make sense trying to convince a person worrying about issues that are really there.
   

I think that most people who switched to Linux switched because of suffering and not because of philosophical or political reasons. - I personally suffered - and so here are my top reasons for switching to Linux:

1. Stability & Reliability
   I had several key experiences with Windows in a short time frame like these: Machine began with strange behavior being slow and having hangups. Nothing in the event logs and no indicator whether it could be the hard drive, motherboard or "just" a software problem; Burned CDs that were not readable on a different Windows machine (but Linux could); Blue screens on startup turning out to be a problem with the CD drive (hell, I booted from HD so why can't go without the CD drive); files getting corrupted over time silently - I could continue further and not mentioning the plenty of issues I had trying to get reliable (restorable) backups/images of the OS.
   
   There are also some stability issues on Linux - I already had some sudden restarts of X - but I am simply less in panic mode on linux. In case of emergency, the OS switches to read-only mode to prevent from further corruption. Apart from that: Firefox runs more smoothly and stable on Linux, Memory usage of the complete system is less, ...
   
   
2. All the important stuff is right there after install.
   After installing a Linux desktop distro, you can right away burn CDs and DVDs, write your documents (full open office suite included) and emails and more. With a few clicks you can start developing software, do mind mapping, manage databases, do sound and video editing and much much more. - Without paying a dime!
   
   
3. You are legal without entering a single license key.
   I hate the annoying entering of serial numbers as well as the annoying registration over the internet. I have seen too many people not finding their original CDs any more (even in companies) and I have seen too much problems registering over the internet - even losing the only developer license of a product after setting up a new notebook after the old was rotten - just because vendor saved hardware information.
   
   
4. Software management.
   On Linux there are managed repositories of software. That means basically, that you have a list of available software, can go through it or search and you can install a software by just putting a checkmark in front. Dependencies of files and packages are handled by the package manager. The result is that there is only very few chance to kill your machine with just installing some software. I only had a very few cases where two application were conflicting and that is already a longer time ago. Removing both an reinstalling the one you like more then works.
   
   
5. Intention and business strategy.
   The intention of the makers of Linux is to create a reliable and helpful system that helps you getting things done - at least for me it feals that way. If features are missing then they are not ready yet, they are not missing because of business considerations. This makes a big difference in user experience. PDF is a good example. On Windows there is a huge market around PDF and there is no interest in integrating PDF features in the core OS. - Or on the other hand there are "features" like DRM in the multimedia area that would be better missing. In many cases it seems for me that I have just the "light version" and need to pay more for the final needed productivity improvement. Further business strategy of Microsoft and Apple for example shows that they do not miss a chance to drive the customer into a dependency and try to block out every attempt of communities to improve interoperability. Although Microsoft shows some cooperation, they put big rocks somewhere else. Microsoft did a better job than all the other vendors in the last about 20 years helping to improve productivity on the client. But currently I can only see that they fully exploit their market situation and customers are suffering from that. There are many messages in the news pointing out untrustworthy behavior of Microsoft and even Apple (e.g. Microsoft capturing ISO or common strategy on entering product categories; Apple removing Google voice).
   There is a saying: "Fool me once, shame on you. Fool me twice, shame on me."

However, I can understand people who stick with Windows as Windows has it's advantages - for instance the more options in the Explorer configuration offering many more columns to show in detailed view. The above list of pros and cons is not exhaustive and shows just some general thoughts over particular features. For me personally the Linux ln command for creating symbolic links is a really cool feature. It allows you to link a file or folder to a file/path somewhere else in your file system. This works really seamless (not as the lnk files in Windows). You can use this to link parts of your profile from a completely different location (e.g. external USB drive). But this is just one of the small nice things - as the default support for multiple desktops.

The more I see people litigating about what OS is best (Windows, Linux, Mac or even others), the more I come to the conclusion that those quarrels will not bring a winner - at least not nowadays. I think, everyone should accept, that there are different operating systems and sooner or later we will find ourselves within a multi-OS environment. We should improve interoperability and find ways living together in peace.

Related posts: Windows 7 RC, My application set on Windows, Installing your PC from scratch, Apple worse than Microsoft?, The operating system, Why I have chosen Fedora, Why I switched to Ubuntu.

Paying for free and Open Source

When talking with other people about free and open source alternative software I often get asked how those companies behind such products make money. - It is nice that people worry about how other companies make their living - but sometimes they have concerns investing time into a product that maybe does not exist any more within two years because there no more founders.

I want to emphasize what I think is the core idea of creating software declared as free and open source:

You should pay people for implementing the features you like and fix the bugs that you suffer from - you should pay for effective work - and not for ??? (maybe doing something new and innovative, doing nothing, doing something else, ...).

This is like planting a tree - you pay somebody for planting it and then you let it be on it's own.

Yes, there is a certain amount of money needed to provide the infrastructure that allows you to offer the appropriate software downloads and some amount of money needed for media (for your working data and backups as well as for the case, someone desires a CD). An option is to put that sum into the amount you are charging for your work or into the amount you charge for support contracts.

But anyway, the Open Source way is just another business model. Nowadays most companies providing software free and open source do earn their money (in addition to charging for the work) selling support or for additional (non-free) features. I see this as a form of software leasing. You pay a certain amount for support which you expect to need more in the beginning. When the product is well established and you don't want or need to invest more you stop buying support but you still have the product.

Another different model is the complete rental of software as a service. - Well there are also people that go fine with that.

With the ordinary licensing model you were used to pay for the existing thing, but the company can/could plan better and in general you have many people contributing with the license fees to a common pot of money that is - or could be - used for bug-fixing and new innovative enhancements. What today many people have a problem with is that they can't be sure what happens with their money because there is no contract that says how the money from license fees has to be invested/used. And I think this is one main reason for the run to open source. Another reason is the higher dependency from the other company because companies don't have the source code and in case the vendor goes bankrupt they can't do the maintenance on their own. - However there are also companies who sell the source code in addition to the normal license fee.

The lack of open source software is that people have to get well organized if you want to have - let's say 50 people contributing each with a small amount of money for implementing a larger feature that all 50 would like to have. But the advantage is that if you got organized you have a better influence on the product as everybody gets implemented what he/she pays for. And: With the open source strategy you usually get a bigger community with some implementing features themselves for their (and maybe also your) need. The amount of developers working on the product is simply higher.

So for you as customer or vendor - check what model fits best for you - if you trust the vendor (in being innovative in the future) and he also sells you the source code (open source is not necessarily free) their is either nothing bad with that model.

Related posts: The Open Source movement, The license keys, IT projects cost explosion.