Come to think of it,even in the proprietary world, there aren't too many affordable software for many of these(by cost I mean the cost of acquiring these legal software for all the required workstations in the office).
I feel this is very relevant to this thread, I am now coming to the contradictions on client side computing.
Some five years back I had a feeling that Linux will steam roll over M$ machines and starts assuming a dominant role in the client side computing. During those period the M$ application software were bearing slightly heavier price tags too. But at the same time Linux tools were all free and the variety was so exciting that it could easily put M$ people to shame. The window managers like GNOME and KDE were so beautiful and versatile that there was absolutely no rival at all on the M$ side. And also Linux had lots fantastic client software such as Mozilla, Image Magic, Evolution, Gnumeric, Open Office suites and the list goes without an end and the M$ equivalents are heavily priced. And lots of software development activities started happening with the help of GTK and QT libraries. The tcl/tk had threatened to shatter the Visual Basic, and the gcc, gtk and qt combine threatened to kill the Microsoft C++ libraries. Under this scenario even the window stalwarts started migrating to Linux because every thing in Linux is free, powerful and cost effective.
All of a sudden with the growth of Internet the software development activity moved from client/server to web based ones. Currently almost all the softwares are WEB based (all developed using free software libraries) and what we need on the client side is only a browser. What people are now trying do under Web 2 specs is to re-capture the rich client experience (with the help of AJAX etc.) they have lost with the disappearance of standalone client applications. Now people started forgetting about what kind of OS they are using. Even in the event of OS crash because of virus etc, people just need to reinstall their M$ OS alone, there is no head ache of application (client ones) re-installation and configuration at all. Earlier when the softwares are of standalones, if something goes bad in M$ platform, people had the nightmare of reinstalling all their application softwares and the consequent loss of data. And exactly this software re-installation and configuration nightmares coupled with virus attacks and consequent security holes in M$ that drove people increasingly towards Linux in the past.
Now almost all the softwares starting from e-commerce portals, document management systems to mail clients are all WEB based and hence the OS has been reduced to an insignificant place. Even, I used to see Linux gurus installing their software in Linux servers and going to a M$ client for testing (what they have installed in Linux). What is subtly taking place here is even the Linux gurus are unaware of the fact they are indirectly sharing/promoting the M$ platform. Add to all Open Office, Gnumeric etc are already available in M$.
If Linux community themselves remain unmindful of the client side, is there any way of increasing the Linux users? I have a feeling that the Internet has already come to the rescue of M$. The M$ may be loosing the server side to the Linux and Linux is increasingly loosing the client side to the Windows. If you put this in another way Linux is eating into the Solaris more than the windows market.
The peculiar feature of a new PC is that it is coming bundled with the M$ windows and people just need to use it. All they require is a powerful w3c compliant Browser. This being the present reality I would like to know what the free software community as a whole would feel,
1. If a Linux solution provider approaching organisations both in private and public sectors and tyring to sell his software solutions and services, if encounters a situation of all the client machines being under M$ and the organisation is not willing to migrate to Linux, is it ethically and morally correct to build his binaries under M$ using open source libraries and install them in the M$ machines?
2. Are this kind of compromises allowed, and without this it possible for GNU/Linux community to increase the number Linux users in the near future?
3. So my view is that instead taking an antagonising path towards other OS users, it is better to convince or cajole them to the side of Linux.
4. I would like to know if the Hall of Shame theory is going to be applied on the client side, particularly for item No. 1
-- Rajagopal CV