Technology Themes

In my career of technology consulting, projects over the years have been driven by some high level themes regardless of:

• the technology (i.e., network design, management, security, etc.),
• the domain (i.e., local, wide-area, data center, etc.),
• or the customer (i.e., retail, healthcare, banking, etc.)

To summarize roughly by decade:

Decade	Theme
1980s	Technology Evangelism
1990s	Technology Deployment
2000s	Technology Evaluation
2010s	Technology Justification

I haven't been in information technology for that long,

IT Leadership

A CEO has a decision - who to lead IT now that the current CIO is leaving due to many failed projects and a general perception that the company's use of technology is falling behind.

The background: the company has business and department leaders housed in mahogany offices on the top floor of it's fancy headquarters. They've identified that collaboration is the issue - too many decisions are being made in a bubble without the appropriate input from across the business. This is why the last few technology projects have failed and they cannot afford another failure. Current collaboration technology includes Lotus Notes email, no instant messaging, an old analogue phone system and no video conferencing.

The CEO has three choices to fill the role:

1. The technology wizard - vast experience with the company's IT systems and all relevent vendor / industry certifications.
   
2. The MBA graduate - business undergrad with an MBA focusing on Information Systems management, hired two years ago as a director and being groomed for the executive fast-track.
   
3. The outside technology consultant with business acumen - loads of hands-on technical experience, but no systems or management experience with this particular company.
   

What should the CEO do?

Seven Thoughts on The Seventh Sense

I recently read The Seventh Sense: Power, Fortune, and Survival in the Age of Networks by Joshua Cooper Ramo and some reflection on my career as a network engineer and consultant.

I've been designing information and communications technology (ICT) networks for the better part of my career. The process, the technology, the methodology have all changed - in some cases dramatically. We no longer build networks for objects to connect to; we build networks to transport information. This may seem a semantic debate, but the two scenarios require very different approaches. And what emerges may be a very different architecture. Building houses for people to live in seems straightforward. But having information like, "all the people are disabled veterans" changes the architecture from a two-floor cape to a single-floor ranch, and alters many other design elements.

Following are the seven thoughts I need to document:

Extending Knowledge with Chrome Extension

A bit of Google-ing and a read through the examples on Chrome Extensions, and I was able to create my first very simple Chrome extension.

When I come across a movie that I'd like to see - in a newsreader, YouTube trailer - I like to add it to my Netflix queue. This involves opening a new tab, going to Netflix, typing in the movie name in the search bar and then adding it to my saved queue. It'd be nice to just highlight the movie name text in the page I'm reading, right-click and select "Add to Netflix Queue". So that's what my extension does.

I needed to understand the basic extension manifest, how to use 'contextMenus' and the Netflix search URL. Code can be found on GitHub.

I haven't thought about putting it on the Chrome Store because of the graphics using the Netflix logo. This is not an extension authorized by Netflix so I don't want to have any issues around that. That would of course make installation easier as well as make Chrome not complain and remove the extension on each relaunch due to it not being in the Chrome Store.

I could also look into the Netflix API to see if I could authenticate if not already logged in and do some other checking, but for the 0.0.0.1 version, I'm more than happy I could hack something together that actually works!

Counterproductive Productivity

After a discussion with a co-worker today, I got to thinking about the artificial walls we're building as we outsource while technology continues to get more integrated. It seems counterproductive.

Programmability and "software-defined" are the latest buzzwords, with vendors providing Application Program Interfaces (API) for integration. But integration points expose only the bare minimum for stable operations. This cause is two-fold: to protect their internal systems as well as limit the non-standard use cases and allow the provider economy of scale. The effect can dramatically limit the options for creative, innovative interoperability.

Can a VoIP vendor offer third-party instant messaging integration; "yes". Will a provider reselling that vendor's system as a service permit third-party instant messaging integration? Maybe not, if their model doesn't include that third-party or they can't measure and thus charge per IM. And that limits what the consumer can due with the service.

I always say the technology is not the limiting factor; we can do anything with technology. Especially someone like me - tenacious, wide skill set, master of none, but I have enough tools in the toolbox to figure something out. If I can't find a hammer I can use a crowbar creatively to pound something into place.

But with a managed service where the vendor is providing the toolbox, maybe all they offer you is a hammer, and then you know, "everything starts to look like a nail."

Remote Send from the Comfort of Your Packet Crafting Shell

I've been spending a lot of time updating the Perl Packet Crafter to a new version - a complete rewrite with lots of new features and plugins.

During the course of testing, I thought about remote sending and remote capture, that is, using the shell to create the packets and passing them to an agent on a remote machine for sending. Like a the old Network General / Associates distributed Sniffer concept, but distributed sender in this case.

Low and behold, WinPcap and Wireshark have the remote capture functionality, but remote sending? That required some research.