Opinion: Centralized networks are outdated; intelligent communication among devices is the wave of the future.
Activitys rising rapidly at blog.eWEEK.com, where youll find timely discussions led by eWEEK News and Labs.
I have to hold myself back, though, from posting there everything that comes to mindonce a week, I need to reserve a posting to share with you here.
For example, I was struck by the many reports from the International Consumer Electronics Show, which took place earlier this month in Las Vegas, that made plaintive cries for a network approach to home entertainment devices.
One story after another talked about the war for the role of "media hub" or "brains of the network" or other such illusory role.
I say "illusory" because I dont see any point in ever again embarking on the construction of a network with an identifiable center. Warring over who gets to be in charge is fighting a battle that has, by definition, no winner in my home or in my workplace.
The roles of master and slave, in any home network worth the name, are as out-of-date as distinct roles of "formula" and "data" in spreadsheets.
The spreadsheet is woefully poor at running calculations in either direction: It cant let me use one setup to either ask about outcomes of assumptions or ask whats needed to achieve desired results.
Both in network hardware and in analytic software, such inflexible models are perhaps embedded in some engineers brains. I have to wonder, actually, if the inadequate rates of retirement savings that keep people working longer are also keeping some outdated mindsets from moving on.
How can things be otherwise? I can show you counterexamples in calculation, such as Hewlett-Packards HP-12c financial handheld thats soon to celebrate its 25th year of manufacture. With an ingenious but intuitive system of internal flags, it infers what youve chosen as assumptions and what you expect it to produce as results in time-value-of-money calculations.
Click here to read about Apples new "media hub" iMac.
I want the same kind of intelligence in my network of media devices. Those devices should infer from my actions, for example, whether I currently put a higher priority on hearing the audio from my stereo system or my speakerphone. Unlike an integrated calculator design, its going to require coordination among many device makers to implement this kind of intelligence in a distributed form.
People were also complaining in those reports about the lack of a common vocabulary even among devices that did share common protocols.
Im reminded of a comment by Dave Methvin, one of the founding analysts of whats now eWEEK Labs, on the subject of protocol compatibility: "Saying that devices are Ethernet-compatible is like saying that your washing machine and your refrigerator are 117-volt AC-compatible. It only means that they use a common resource; it doesnt mean that they have anything useful to say to each other."
Its a social and political process, rather than a technical process, to teach a big-screen TV display the meaning of a "rotated" attribute that was tagged on an image file by a photo-editing program on your PC. The TV might be a smart display that understands an attribute called "orientation," with possible values of "0" and "right90" and "left90" to tell it how to display a picture, but it will probably ignore a semantically identical attribute with a different name.
Have design features become irrelevant? Read more here.
We see this latter behavior all the time when rendering HTML or using other modern representations of content. I think of it as a strength, in general, when a device can ignore an attribute that it doesnt understand rather than throwing an exception that has to be handled up the lineor, worse still, by throwing a cryptic error message to the user. Its clearly necessary, though, for devices to have some way to negotiate with one another or ask for help from the user in a nonintrusive way.
A reader recently reminded me of a column in which I said that there are only seven fundamental functions of IT devices: Create, store, move, locate, display, edit and delete. Perhaps device makers would do well to begin with this list and elaborate on these ideasbut always with an emphasis on understanding one anothers capabilities in the context of the users intentions.
Peter Coffee can be reached at email@example.com.
Check out eWEEK.coms for the latest news, reviews and analysis in programming environments and developer tools.
Peter Coffee is Director of Platform Research at salesforce.com, where he serves as a liaison with the developer community to define the opportunity and clarify developers' technical requirements on the company's evolving Apex Platform. Peter previously spent 18 years with eWEEK (formerly PC Week), the national news magazine of enterprise technology practice, where he reviewed software development tools and methods and wrote regular columns on emerging technologies and professional community issues.Before he began writing full-time in 1989, Peter spent eleven years in technical and management positions at Exxon and The Aerospace Corporation, including management of the latter company's first desktop computing planning team and applied research in applications of artificial intelligence techniques. He holds an engineering degree from MIT and an MBA from Pepperdine University, he has held teaching appointments in computer science, business analytics and information systems management at Pepperdine, UCLA, and Chapman College.