Bud Gibson, who teaches non-technical students at the University of Michigan's Ross School of Business, has started a learning blogosphere.

Tom's Hardware is reporting lots of layoffs of folks who have Linux experience at Sun Microsystems. I was talking with the kernel team here today and the Virtual Server/PC teams and both are looking to hire folks who understand operating system design/development and who can code in C/C++. In fact, I continue running into teams here that are hiring. The MapPoint team, for instance, just put out a plea for more great developers.

Yes, Microsoft is a great place for Linux developers to work. Give us a try!

Have your RSS feeds read to you by NewsAloud.

You've probably already seen this video, but it makes me smile everytime I play it.

More.

The G'day World podcast just put up a bunch of things from the recent Demo conference.

Yahoo listened to my pleas for letting developers play (they just announced an API). How about MSN? Who will be the most developer-centric and geek-centric search engine?

Google got to where it is today by being the most geek-friendly engine.

But Yahoo's API doesn't look like they really gave me what I wanted.

Here's the first thing I wanna try to build: a search engine without blogs.

Huh?

Seriously. Blogs are increasing noise to lots of searches. We already have good engines that let you search blogs (Feedster, Pubsub, Newsgator, Technorati, and Bloglines all are letting you search blogs). What about an engine that lets you search everything BUT blogs? Where's that?

Is Yahoo's API good enough to do that? It doesn't look like it. It looks like Yahoo just gave us an API to embed its search engine into our applications. Sigh. That's not what I want. OK, MSN, your turn. Are you gonna really give us an API that'll let us build a custom search engine and let us have access to the variables that determine the result set?

What is Crossfader?

Congrats to Bill Gates for being knighted tomorrow.

OK, I give up. It's pretty clear that we're gonna get linking utilities from the big companies. After all, the bloggers are asking for it. I'm glad they are on the record so we can look back at today and remember what happened. I still think you're gonna open a pandora's box that isn't gonna turn out nearly as nice as you think it will.

But, since it's gonna happen, let's move forward and think about some issues.

1) My friend "Ellie," who is not computer savy, already has five toolbars. So, let's use her as an example. What happens when all five toolbars are trying to link ISBN numbers? Let's say she turns on all five toolbars linking feature. Which toolbar should win?

2) If Ellie turns on all five of her toolbars, how will she know which "butler" has provided which link? Shouldn't we have a standards committee thinking about this?

3) Would you hire a butler that would only bring you McDonalds' food, even if you don't like junk food? Yet, that seems to be what those who are advocating these butlers wants. Instead, why aren't we asking for user programmability of the butlers? In fact, I have a set of "Burger King" tags that you might want to use with your butler and I have a separate set of "health food" tags. Why can't I give them to you for your butler to use?

4) Now that you accept the role of a butler in your life, what's gonna keep another company from doing a "butler 2.0" which is even better in bringing you cool places than "butler 1.0?" In fact, what's gonna keep another company from kicking butler 1.0 in the knees and running over all of his links?

5) Doesn't it scare you that hundreds of millions of people will have the exact same butler you do and will only bring you certain things? How are we gonna make sure this system of butlers doesn't turn against us? Especially since the users can't customize this butler in any real way (you only are gonna get the butler that the toolbar manufacturers ship)? Would you let a butler in your house that's paid for by a big tech company? I wouldn't let one in mine, no matter how many promises this butler made. Why not? He isn't working in my interests if he's paid by someone else.

How would I make this butler "safe?"

Easy:

1) Ship the butler with no default behaviors. Have the user load "behavior packs." Make these "behavior packs" produceable by anyone. One of the packs could even be "the behave in content owner's wishes pack." I, the content producer guy, could then opt in and do stuff with that pack. Other packs? "the Google walled garden pack." That would only take you to Google stuff. Of course, I'd quickly do one for Microsoft. You might like it. It'd take you to MSDN and other places inside our walled garden.

2) Put a logo next to each added link. That way users can tell who added the link, and which pack is doing it.

3) If two packs, or butlers, are fighting over a link, let them both display. That means there'd need to be a layer with links on it showing who is trying to link the ISBN number, for instance, to various places (or even the same place, albeit with different URLs).

4) Make it possible for the browser manufacturer to let the user know when a non-original link is being displayed. I hope future versions of IE, Opera, and Firefox, can have a mode of "show me only original links." That would guarantee that the link system would keep its integrity.

5) Make it so that end users and content producers can produce their own packs and make it easy to share those packs with other "butler 1.0" users.

6) Make sure that "butler" users have the ability to go in and change the behavior of any of their butlers. Let them, for instance, go in and remove Google Maps from AutoLinks and put in some new map that comes out after the butler ships. Also, let them add in their own links for things that aren't linked yet.

Does that sound sane? What would you like your butler to do?