Skip to main content

End To End

  • How many times have you read the following statement? "This acquisition will complement Company A's recent acquisitions of Company W, Company X, Company Y, and Company Z in the creation and delivery of an integrated, end-to-end delivery system for multiple network architectures."
  • It's pretty bad when you have to buy a whole string of companies just to get stuff to together as a system. It's like looking at the wall of ink cartridges in an office supply store-did no one even think of making this stuff compatible? We all realize the printers are dumped on the market and the money is in the cartridges, but there is a point where a business practice becomes absurd.
  • We're at such a juncture in our own industry as well. You'd think this far down the road that most boxes would just generally work together reasonably well. Sure, you might need to download a driver or two, but personal computers are no longer a novelty item. Yet it's my contention that many of our boxes and programs do not work well, especially when viewed or operated as part of a system.
  • Some programs take years to master and provide meaningless solutions. Other are modified weekly and require their owner/operators travel with a suitcase of adaptors, cables, and connectors and are about as intuitively obvious to operate as a class on particle physics.
  • At times it does look we're making a serious attempt to find the largest number of incompatible ways of doing exactly the same thing. The general topology of our systems has not changed radically in four generations: source, processing, distribution, amplification, transduction, and monitoring. Yet there's no decent MAC bridging to make two boxes from different manufacturers routinely talk to each other, no AV industry SNMP standardization for even line voltage, let alone failure reporting. There are industry standard platforms to perform such functions currently available, such as HP's OpenView and all of IEEE's 802.1G standards.
  • Yet today we have modeling, content generation and conversion, system design, source control playback, software processing, amplification, and monitoring programs. Off hand I can't think of one program that bridges all of these aspects. Perhaps there are a handful of programs that bridge a couple of these aspects.
  • Yet Company X's boxes do not talk to Company Y's boxes. The signal flow input and outputs are compatible, but not the networking or control. And that, my colleagues, is wrong. One would have thought with Microsoft pretty much forcing everyone's hand over the last generation we could at least have the basics right by now.
  • Ironically, the personal computer platform is taking over more and more of the tasks that used to be under our industry watch as a result of our refusal to get organized. Witness the digital signage engines commercially available. We don't even have a framework for industry plug-in software, which is commonplace in many industries. One suspects that this situation was not intentional.
  • No one believes programmers ever sat down and planned out months' worth of work to intentionally frustrate their user base to the point of emotional trauma. Yet this is our current state of the state from the user's perspective.
  • The expanse between user expectation and the delivered reality is often quite wide. Typically the best product designs require little operator training or input, as their functionality is intuitively obvious.
  • Most users don't really care what's under the hood; they just want something that works well and looks pretty. Once you pass a certainly cost threshold barrier, many do not buy on price alone. But perhaps most importantly, the less amount of effort required by the user to attain those benefits the better.