A few days ago I posted about Microsoft’s efforts to curb unauthorized distribution of its products by misrepresenting a piece of malware as a “critical security update”. However, Microsoft’s also arousing ire by refusing to offer a patch to fix a critical security flaw in Windows ME and 98. In short, unless you want to risk exposing your computer to criminals, you need to either (a) pony up $100+ for XP or (b) switch to GNU. Now, of course I offer the second choice seriously, but Ed Bott has another idea: Microsoft should offer a “starter edition” of XP and price it around $30. The starter edition would be crippled to the point of near inoperability, but, tongue-in-cheek, that means the it should work as well as 98 or ME.
Here’s almost a textbook example of where FOSS triumphs over proprietary ware. If ME and 98 were FOSS, then other programmers could patch the flaw (assuming there was sufficient demand). However, it’s not in Microsoft’s interests to do so, so they leave their users with little choice: Buy or die.
I really think that this poses a golden opportunity for GNU. What we have here is so incredibly fortunate, I suspect divine intervention. Here’s my plan—let’s reach out now, in a powerful and unified way, to Windows 98 and ME users. Let’s explain to them the danger of continuing to use their OS, why the development model is the cause of their being in this situation, and help them make the switch to GNU.
I would like to see a special distro, perhaps a version of Ubuntu, specifically designed to help ME/98 users make the switch. Let’s strike while the iron is hot!
Comments
The big problem
Well I am a strong advocate of FOSS, but let's face it:
Most (Win)users don't know about these "critical security flaws" and even if they do most just don't care. Even if the system breaks there is always a nice boy/girl in the family or neighbourhood who will happily fix(=reinstall) it and most users take this as normal.
I just recently had to learn, that some folks do indeed use www.google.com (their starting page) to punch in the addresses they wanna surf too (Phone call "I' am in google now." Me: "You mean you are running firefox!?" He: "What? I am in Google!" Me: "Then Internet Explorer?!" He: "No Google". Me "Google is a web-page, it is not a program. You are viewing it in a so called browser" He "No, no I am running Google ..." Me (giving up)
"Now punsh in the adress in the address bar" He: "What bar?" Me: "Where you punch in the address you wanna surf to. It is somewhere at the top" He: "I always use Google"....)
Desktop Linux (or *BSD etc) looks different and most users I met look for a couple of buttons they are most familiar with and they now look different or are located elsewhere.
Security, stability, freedom of choice is NOT an argument for them to switch.
The situation is somewhat different in companies. They mostly have the money to pay for Windows and thus - unless a FOSS-infected admin pushes them (mostly with little support from the administration as they are familiar with Windows too (and have their admin for fixing the problems) - will upgrade, as the switch to FOSS-based OSes will not pay off immediately.
We need to target the opinion leaders in small and large companies. If they switch successfully, the "normal" users will soon follow.
It would be even better if we could have a seamless upgrade
It would be even better if we could have a seamless upgrade. I mean, a simple windows application they can run, that will make a new partition on their hard drive, save their settings to it, then reboot the pc, install the new GNU/Linux distribution and then restore all those backed up files (perhaps even importing IE favorites to FireFox, mail to thunderbird, etc).
If it could be a point-and-click operation, you can bet a lot more people would do it.
Robin
Sounds Ideal!
Your solution sounds ideal, Robin. I'm not skilled enough to do what you're suggesting here, but it's the way to go. It really does need to be that simple. I figure most people using ME or 98 are, for the most part, users with pretty simple needs: Email, web, word processor, maybe editing digital photos and printing. If we could show them how easy it is to do all of these things with GNU/Linux, I'm certain we could get them to switch.
Or you could just wait...
I agree that a system like that would be terrific, and I'd cheer loudly for anyone working on it, but it's a huge job directed at a moving target which is aware and taking evasive action. And to acheive your goal you pretty much need total success -- a partial conversion may even get more flak than a system that makes no such promises.
That said, there are companies that make products along those lines. There is, for example, a boxed MEPIS GNU/Linux that contains a crossover product (I'm sorry I've forgotten what they called it, but it was a major selling point).
These are commercial proprietary applications -- but so what? So is Windows, and if the whole point of the application is weaning you off of Windows and proprietary software, then it's not so bad. It's like a nicotine patch -- it's still drugging you, but if you actually do quit smoking because of it, then it was probably worth it. And from a practical perspective, a proprietary development environment is better suited for this kind of 'moving target + limited horizon' product (sort of like tax preparation software).
But in the end it will not be 'interoperability with Windows' or 'ease of conversion from Windows' that makes the difference. It will be 'how many people started out with Linux in the first place' and 'how many of my friends are already using Linux?'. Despite all these complaints about the difficulty of switching, Linux use has been slowly growing. There are now just about as many people using Linux on the desktop as Macintosh, IIRC. Eventually, that number will start to tick up faster because the growth is intrinsically non-linear -- more users mean more opportunities for growth, which means more users, etc.
At that point, Microsoft-style 'lock-in' becomes 'lock-out' -- Windows, due to its engineered incompatibility with other platforms will start to lose big when users and developers no longer believe "it's what 90% of the market uses, so it's all we have to support".
Also, it's becoming clearer and clearer that Microsoft can't keep up. How long has Vista been in development? People are starting to lose patience with huge delays, terrible security, low performance, and buggy software -- yet that's what Microsoft gets, because they are stuck in the proprietary development regime, which has a complexity ceiling. Yet the demand for more and more capable O/S software means that it's impossible to stay below that ceiling. Above it, only free software bazaar-developed software has much of a chance of being maintained.
So that means that as time goes on, Windows will go on being slower and slower to deliver and worse and worse in what it delivers, while Linux just keeps on sailing more or less like it does now. Proprietary development is, by its nature, a sprinting activity -- it doesn't do well in a marathon.
Of course, this could take another decade, but I'm not sure if it matters. I think we're already past the point where Linux could be easily killed by proprietary pressure, so that it's now just a matter of time. It's not like we have to 'take over' from Windows, we just need to keep growing.
I'd really prefer to see Linux developers continue to work on making Linux better without reference to Windows. All this 'ease of conversion' and 'MS compatibility' work seems too much like 'chasing taillights' to me. I'd rather see us sail into unknown waters and let others do the following.