Yes that's true, but I actually owned Steam when I was 100% Linux based, and I ran games under CrossOver for a long while. But I would gladly re-buy Steam just for the chance of running it natively.

Excellent argument. Probably 100% valid, too.

Wha..??! DLL hell vanished from Microsoft OS's? Ha! HA HA HA! *HA HA HA HA HA HA HA*!!!
Thank you, I didn't laugh like this for a while...

But I agree with you on the grafting part, and that's what I meant. Once it works on Ubuntu, it will be very easy to port it to any other Linux distro. Whether or not Steam will officially support it is another issue (they may choose not to). What I also meant, is that Steam will be able to target Linux based gaming consoles. Heck, the OS is already there, and doesn't cost a damn thing. Why wouldn't they create their own dream console? Building it in China will cost peanuts and they will then have their own walled garden, with total control over the OS and what comes in and out of it. Total predictability. Gaming studios will love that. And Steam really does grok gamers, gaming and game distribution (and the whole business behind it). They will be in a position to completely replace the old style gaming distribution channels. It makes sense strategically.

I agree with you 100%. I think id Software and Epic did not have enough "weight" to bring studios into their product. Plus, it's been many, many years that Steam have been waiting quietly for the right moment. Now they are diving in, and this is going signal others that Linux has finally matured enough for Steam to enter in a major way. And we all know how VC's are like cattle, they need a leader they can follow like a herd (no offense intended to cattle).

Valve have a golden opportunity here, in several ways:

1. Ubuntu is a first stepping stone. Once they have the Linux experience, they can target all kinds of Linux based platforms & set top boxes, as they become popular. It's just like UbuntuTV in a sense - It takes a stable operating system and tailors it to a niche market, adds the back-office sauce into the soup, and you suddenly have a serious iTunes/Netflix contender (technologically anyway).

2. I believe game producers are going to see this as a blessing: Valve becoming the major conduit through which serious games flow into the Linux world, paving the road for those producers into user's desktops, while providing billing, game discovery, content distribution, and community tools. Nobody else is doing this at the moment with Linux, except for Canonical who have created their own "App Store" application (which by the way is pretty good!). So imaging Canonical's "App store" on steroids, for games! Once enough games are built for Linux, why would anyone use Microsoft Windows for gaming?

3. When you are first to capture a market, you become the dominant player. The longer you're the dominant player, the more difficult it becomes to unseat you from your throne.

I recommend reading the Wikipedia page on T. Gondii, which also discusses the above research on suicide rates:

http://en.wikipedia.org/wiki/Toxoplasmosis

Actually the GUI-fication of Windows Server was one of the many reasons Windows based networks are so insecure and poorly configured. It creates this notion that any "kid" can configure a Windows Active Directory Domain, and small to medium sized (and sometimes larger) companies hired amateurs to save a buck. I met some pretty bad sysadmins in the last 20 years of my life, who left a horrible mess wherever they lay their hands.

Not to mention that the GUI kept changing every year, because that's just how Microsoft does stuff; anything you learn becomes obsolete a year later. They are very volatile like that. This is not so in the Unix world, where API's and tools are kept minimalistic, are cleverly crafted, slowly improved upon, with stable releases coming out every few years instead of betas coming out every month. Indeed, history shows that a Unix person retains his knowledge for decades. Show me one Windows person who can say that what he learned about Windows 2000 Server is still useful?

Enough with the Microsoft Bashing, here's one product I actually liked that took the CLI forward: Splunk. They took the CLI and put it in a Web browser! Splunk is basically a data collection tool that pipes everything through whatever you type into the web based "CLI" prompt, with the basic function being a combination of "grep", sed, awk, with powerful regular expressions, etc.

The beauty of it is that you can take your piped processes, and save them as a "View", and you can even create charts based on the resulting data. The result is a Unix admin's dream dashboard into the depths of his IT environment. I seriously recommend any Unix head look into that technology, especially considering the product has a free version (the free edition removes some "enterprise" features, but for many applications you won't miss those features at all).

I also like the combined approach: UI tools that produce either scripts or configuration files that you can read and understand and modify manually if you wish. Make a change in the configuration file? It will register in the UI. Made a change through the UI? It will register in the files and scripts. You get the best of both worlds this way.

It seems people are too harsh on Flash, for no reason really.

Personally I see it as a failure of the tech world to understand why some people were stubbornly holding on to Flash.

Flash was a very easy way for product designers to develop some pretty advanced client side technologies, with a plugin that had more than 90% adoption rates. iOS changed that, much to adobe's chagrin.

But like some commenters said, this technology is now being killed without proper replacements. You still can't do socket communications directly from within a browser without using plugins. Definitely not with UDP. This was one of the reasons Flash was awesome. It filled the gap of all those features missing in a browser (or available only in some and not in others).

And let's not even start with the authoring tool - I have yet to see a tool that was as friendly and intuitive as Adobe's for producing Flash apps.

Tell you what, I can understand their motives... people allow unsigned code to run, don't they? I am guessing they understood human nature is susceptible to social engineering, and wanted to avoid it all.

Too bad for them, human nature is more powerful than DRM, and the Dark Side shall prevail ;-)

And about that Bluetooth thing, watch for BlueSn0w which have only just recently released a demonstration of a working hack. This is going to open many exciting capabilities to iPhone users, such as Stereo Audio, sending contacts via BT, and many others.