IBM’s Blue Cloud and new Startups

Project managing scrum

Blogs – dying or in transition?

It seems that many bloggers are doing fewer posts these days, and I’m not the only one to notice the situation, which many would be quick to tag as a fad. The Austin American Statesman recently had a front page article about how the number of new blogs is slowing and the number of dead blogs is increasing. In the article, Gallaga suggests that easily unloadable photos, video, and audio are where future bloggers will spend most of their time. It’s hard to argue with this notion as Flickr integration has been a part of my blog since inception, and the popularity of YouTube is undeniable.

But, the question remains as to whether plain old blog posts will go the way of the dinosaur. There’s no doubt that I’ve had fewer posts over the course of the past 3 or so years, including a one year hiatus where I only posted 1 or 2 times. From my point of view, the blogging world appears to be in a consolidation mode with bloggers grouping together to keep readers coming or with a few bloggers who have figured out ways to make money. For many other casual bloggers, the motivation to keep going just isn’t there.

Old Steam Engine

The other slant on this story is whether the readers of blogs have become information inundated, causing a reversal of the chicken and egg. For instance, the motivation to write for many is knowing the content is being read, and as readers have become bored with weak content or too much content, they have slacked off on subscribing entirely. While a decent percentage of my techie friends actively maintain a blog and/or read blogs, 90% of my non-techie acquaintances have never blogged and claim that reading blogs takes too much time.

Blogging may not be going away any time soon, but as a Darwinistic consolidation increases, those who are left will most likely be the ones with the best content, which should lead to a resurgence in subscriptions. As for myself, I cannot yet image a situation where audio and video completely capture the market for freeform expression; thus, I predict a long life for blogs or some similar format.

The virtualization platform for software?

The technology trade rags are filled with all the latest new products allowing virtual machines (VM) to run on varieties of platforms and chipsets. For many of us, this means running many versions of Windows or Linux on one machine simultaneously; however, it could be just about any OS supported by the VM software of your choosing. While the obvious solutions this could fill would be centralized management of the workstations within a company, there are some more subtle ideas that appear to be taking hold. (Duh disclaimer: It’s possible that I’ve ignored these ideas in the past, but as my confidence slowly grows with VMs in the test lab, these ideas don’t seem as farfetched as before.)

Instead of shipping software to customers for installation, imagine shipping a VM with the software already installed on the OS. If the customer was running compatible VM server software, they could just load an instance of the new OS with software already installed, configured and ready to go. In fact, as this article eludes, the customer could ship a production copy back to the software company for maintenance and/or an upgrade, which could be tested before being shipped back to the customer.

Granted, there are some hurdles to overcome just as the introduction of multi-threading and multiple CPUs had issues, which means the hardware to support all of this will have to be faster and more robust than ever. Hard disk I/O and resource utilization will be an issue for the foreseeable future, but the more time intensive problems of debugging environmental software problems could be minimized. In some cases, the changes to a customer’s environment might be as easy as updating to a patched VM.

While those of us in the enterprise software industry know that the customer’s actual environment is more often a contributor of a problem than the software itself, the approach does provide some advantages in supportability. Does anyone know of software vendors using this approach yet? I wonder how many years away we are from seeing software vendors only support certain VM server software, which would be just the opposite for most vendors today? Are VM standards being set, or will we see the same compatibility problems that occurred with J2EE?