Although cross-platform applications may stop or slow some people from switching, they also make it much easier to switch if you want to. Now there is choice!

As an IT professional, I doubt I’ll ever be free of Windows so having cross-platform tools like Firefox, Thunderbird, Keepass, Truecrypt, Google Earth, OpenOffice, Komodo Edit, etc. is a real god-send.

Regards, Julian
http://linux.knightnet.org.uk

Offering FOSS apps for multiple platforms does not only make the users’ life easier, it also gives the FOSS projects the potential for more developers. E.g. the music notation feature in KOffice 2 was developed under Mac OS X. With the sheer number of Windows installations there’s an enormous potential for additional KOffice developers. Even if those developers never switch to Linux/*BSD/Solaris, users of free platforms benefit from their work. While the Mac community does not have that many programmers, it’s a community with many artists and usability experts. This is very noticeable at instant messaging applications. IMHO Adium is the best instant messenger out there. It combines the programming experience from the Linux community (it uses Pidgin’s libpurple) with the aesthetic experience from the Mac community to a polished and popular messenger. As a result, Kopete users can now enjoy the large library of GPL compatible chat styles created for Adium. It’s a win-win situation.

If we are looking for people to switch to Linux very soon after they use open source apps, I think we will be disappointed and think it was a bad strategy to port those apps.

If we are willing to wait (possibly years) for people to switch to Linux after trying open source software, I do not see how the strategy would be bad.
1) They are exposed to new, free software in a comfortable (to them) environment (OSS positive)
2) The software impresses them, and they use it instead of proprietary software (OSS positive)
3) They, eventually, discover community support concepts (OSS positive)
4) They get used to good quality, free software (OSS positive)
5) Future software decisions will be OSS-aware if not OSS-biased (OSS positive)
6) Proprietary software (apps and OS) will eventually cause problems – crashes, viruses, upgrade costs, company goes out of business, etc. (proprietary negative)
7) Good experiences with OSS will help breed more OSS-friendly decisions in the future (OSS positive)

Short Version
Linux = OSS and apps = OSS. I believe the more positive experiences someone has with OSS, the more likely they are to try more OSS.
Question: Does “porting” Firefox to a proprietary operating system help or hurt the adoption of other OSS software?
Another question: Does “porting” Firefox to a proprietary operating system help or hurt the adoption of other browsers?

In the worst case, you can see their Windows OS as just the *kernel*, and an ugly desktop, Explorer. But the rest of the stuff could be open source.

If you’d really want, there could even be a nice open source desktop environment on it, that replaces explorer completely. … and maybe some decent services which implement posixs (sorry cygwin… you just don’t do it.)

In this way, to the user would interface with only open source products. This could really mean that in the future, they might not care about their kernel too much. So, they could switch to linux.

If we keep on evangelizing open software, maybe we should also try to remain open.

Yet they don’t seem to be aware that restricting the target user base is potentially _damaging_ the whole FOSS ecosystem.
We (FOSS people) say that we want to promote freedom but it seems we are always trying hard to devise ways to force users to switch to what we consider the best platform.
It’s not a real “vendor lock-in” since it’s free software, open standards and all that, but the approach is not so different: “In order to use my software, use my platform”.

But, for those who do want to get Linux to switch – what’s the right answer?

Obviously, if the goal is just “get more users,” then it makes *perfect* sense to port to Windows and Mac OS X and any major platform with a lot of users. If the goal is “get more LINUX users,” well, then it’s a matter of whether FOSS on Windows / Mac will ultimately support that, or go the other way.

Each time I see OSS people saying that “porting a FOSS application to proprietary platform X or Y is not a good idea”, I wonder what’s the point of making great software if you don’t want anyone to actually be able to use it.

I would agree, but some folks write software in order to promote FOSS — and it’s a concern if the software runs on proprietary platforms that users will stop there.

Thanks for the perspective!

What really matters is bringing more great software to people. The OS really _doesn’t_ matter.
And, of course, for software developers, reaching platforms besides Linux means getting a _whole_ lot more users. And a bigger community means more bugs reports and (hopefully) more contributors

Each time I see OSS people saying that “porting a FOSS application to proprietary platform X or Y is not a good idea”, I wonder what’s the point of making great software if you don’t want anyone to actually be able to use it.