Rubber Chicken

HBS via CNet has an article on strategies for niche players in larger business ecosystems, such as those surrounding Walmart and Microsoft. The basic advice is commonsensical: find a specialty where you can create value, leverage the value of the platform but avoid being overly dependent on the big fish. The analogy of loosely coupled software architecture is a good one. Design your business to be able to get value from other entities, but don't bind yourself to them too closely. Your interests may be aligned, but they are not the same.

Posted by Rob at 10:59 AM | Permalink

The database is the staple of the IT diet. Forbes has a good article (although I hate the title) describing how companies are moving to MySQL and the kind of money they are saving. The basic database (along with other parts of the technology stack of operating system and application server) is becoming a commodity whether Oracle likes it or not.

Posted by Rob at 10:44 AM | Permalink

I was just looking at the list of fixes that are provided by Windows XP SP2. Wow. I haven't installed it yet, but from what I've been reading, it will be a good update to get.

Posted by Rob at 10:18 PM | Permalink

Wireless security is really difficult to manage. I upgraded the firmware on one of my el cheapo wireless cards and had to reconfigure it to connect to my wireless network at home.

Being somewhat security concious, I use WEP on my home network. Of course, I mistyped one of the characters in my WEP key when reconfiguring the card and couldn't get a valid ip address. I got some crazy internal ip address (169.*.*.*) but had a solid radio connection. So, since I had no idea what the firmware upgrade had done, I spent a good hour or two poking around at various settings. Finally, I decided to retry the WEP key and it worked. The driver's way of telling me that I had an invalid WEP key was to give me an invalid ip address. Seems fair, doesn't it? Of course, nowhere in the documentation is it mentioned.

No wonder there are so many unsecured networks out there (I just found one in my neighboorhood this evening). There has got to be a better way.

Posted by Rob at 08:41 PM | Permalink

From a CNet article yesterday about SugarCRM, an open source company that got $2M in VC funding:

By tapping into the open-source world, fledgling software outfits can assemble their software products from freely available components. Volunteer programmers not only help develop the product, they also create a pool of potential customers for starting companies. In return, programmers develop new skills and get free software.

I don't really understand. How do you get the volunteers to create the compoents that you need? How do the volunteers create a pool of potential customers for starting companies? I can see the value of new skills (gotta have 'em) but what good are new skills in a job market if the companies that employ you are relying on volunteer labor?

Open source is a very cool model, and as a consumer of software, the cheaper the better, but it remains to be seen if people can actually make money doing this in the long term.

Posted by Rob at 04:31 PM | Permalink