Some cardinal rules for O/S contender hopefuls

I find it pretty hilarious to watch operating system creators tackle certain problems. The solutions they sometimes put in place make it clear to me that they are not just new to the graphical end user O/S, but in reality don't 'get' the concept of visual computing.

Here are some cardinal rules for wannabe O/S makers if you ever hope to get a foothold with the unwashed endie masses;

1. What starts graphical stays graphical. Did you think that by flashing a few dialog boxes and asking some questions of the user that it would then be okay to drop down to a csh/bsh whatever shell and have them executing sudo commands? It isn't. If you start something with pretty pictures, finish it that way.
2. Programs don't just 'go away'. They crash, sure, they crash all the time, but if anything be annoying in your followup. Too many times I try to load an app in a linux based system trying to be an end user O/S and the app tries....but then it just goes away. No dialog, no pestering error, no offer to report it. End users know they did something, so trying to trick them into thinking they have amnesia is just not going to work. Offer to send the error somewhere, offer a picture of Paris Hilton. Play a crying baby sound. For gods sake, do something.
3. When something is 'installed', its helpful to actually make it appear as though something new is on the system. I've installed so many Linux and Open Src programs only to find that they don't seem to have any concept that the 'start menu' or any graphical menuing exist. This is key to end users! You must place a physical, visual metaphor for your application somewhere on in the users view after install.
4. Users don't understand 'better'. They understand 'does the same thing I already have and more'. Different is not good, different is different, and end users have a really hard time moving to new things when they'd simply prefer to use what they already know.
   

Felling the Giant aka The Fall of Microsoft Windows

Let me be clear right up front; I've been a Windows guy since it came out. My buddy and I smoked 3 packs of cigarettes while we listed every single file on every single disk of the Windows 3.1 and Windows for Workgroups installs just because we wanted to know what every single thing did.

Before Windows, I ran a Mac, a Commodore, Apple IIe, Apple IIc, Apple II, Apple Lisa.

I've been around the block. I'm a WYSIWYG guy, but I tell you, it ain't cause I need the interface. I was weaned on the console, but its never been elegant - GUI is the way to go, period.

Thus, I skipped the whole early 'omg Unix/Linux is so much better, faster, more stable' arguments as the drivel that they are - and remain.

Still even back then Wintel product would make me cringe everytime they released something new. What's worse, any Tom Dick or Harry with a copy of Visual Basic was releasing software to run - or more often crash - Windows.

As the years went by, I watched the computing power of my precious Intel hardware increase ten, fifty, and finally 100 fold. Yeah, I was one of those people that freaked out when I obtained my first 16mhz processor.

The punchline is, the O/S really doesn't do that much more for me now than it ever did. There are obvious caveats to this, features and functionality that support things that didn't exist before like digital camera's etc - much more easy to use now - granted.

But back in 1997 we were listening to Winamp, creating and sharing web pages, mp3's, flash, etc.

In 1987 I was still hammering away on my Commodore. It's 2007 now. My opinion is that the evolution of the Intel/Windows creature is slowing.

It may have stopped.

Yet, the stagnation of one does not automatically mean the victory of another. Sure, Apple has moved to Intel (more on that later) and I LOVE the upcoming O/S, but it is still running on what they say you can run it on, and I personally think an O/S should run on it's own merits.

Though there are strong and very viable arguments to the contrary. Supporting an O/S that runs on a consistent set of hardware is easy, predicatable and economical. There is no futzing around when you are intimately familiar with every chip in your box. You can't say that about the WinTel State of the Art.

I've seen endies try to run the most outrageous software on horrifically underpowered boxes .

They angrily stammer 'wut do ya mean it won't run?!' at me when I tell them that the Fisher Price Farm Animal Sound simulator they are calling a PC is outdated and needs upgrading.


But, as Evil Emperor Sidius might have intoned - 'I have forseen the destiny of the desktop', and I'm not sure that Microsoft's rule will continue much longer. Too many lessons are being learned too quickly by too many up and comers.

The real irony is that a lot of them want to put you into an O/S that is free.

I'll be honest, I don't get Open Source. I mean, to me, Open Source means, essentially, free. They say it right on the sites. On http://ubuntu.com the makers of this open source linux distro literally states in black and white:

Ubuntu will always be free of charge, including enterprise releases and security updates.

That totally rocks. But wait a second. Why? I mean let's be honest. Out of all the development you could do of any software you can imagine, Operating Systems have to be the toughest, most expensive and downright complex things you can tackle.

There are so many moving parts, and so many bases to cover. Yet, Ubuntu and many many others are out there, release after free release, plying users to give the Linux based O/S a try.

At first, I didn't take them seriously. A lot of times you can't because these Linux guys 1) hate end users, 2) aren't big GUI fans and 3) are the hugest uber geeks to walk the planet. Ironically, #3 is a huge salute and not in any way a crack on them.

But I've been watching. Every 3 to 6 months I go on a spree where I start downloading the iso images to any neat sounding Linux distro I can get. I'm totally not an expert on Linux but I am a platform generalist and have broad expertise so I get by.

My goal is, and has been to watch for the 'Fall of Microsoft'. I have forseen it, it is my prediction, and now I'm willing to put a relative date on it - with the next 4 years, or, by 2010.

I'm not just spitting water - I know what I'm talking about. I come from a background where somehow, despite being a loving Windows user for well over a decade, I'm impartial. I'm looking out for the endies, because they are the consumers of what people in my profession produce.

If they have a good user experience in the O/S they are using, I have fewer issues to resolve.

And, so it begins. I'm going to begin a series of explorations, experiments if you will; with the goal being to replace my windows operating system with something that does the same things better, faster and of course, cheaper.

There are rules to the competition;

1. The install must be easy for an end user to perform. Any complicated partitioning, compiling, or such things will disqualify a contender. My Dad needs to be able to install it.
2. It must be able to run the same broad capability software as Windows. I've included some examples on the 'must be able to run' list;
1. itunes - Come on, without it you rule out simply millions of people.
2. World of Warcraft. Its the most popular online game in its genre and a good pick for high intensity graphical games. Plus, I like it so there.
3. Photoshop - or, it must have something that is just as feature rich but not almost impossible to use (sorry GIMP).
3. It must run on Intel or AMD hardware and must be able to detect most popular hardware.

Pretty simple. Each month, I'll try one out and compare my notes with you here on this blog.

First contender: Ubuntu: Fiesty Fawn