Things learnedOne of the interesting things I learned when I received my very first phone bill for my home abroad was something that made me understand better Bob's point. Besides the infamous 21% sales tax, on my phone statement I found 3 separated charges: 1) Total Calls, 2) Internet Access Line, and 3) Internet Service Subscription. I was amazed when I discovered that 1 and 3 were actually the most expensive ones. In my case around 65% of the total bill!! In other words I discovered that only 35% was for being **connected** to the network. The remaining 65% was to pay for a huge "man-in-the-middle-infrastructure" that exists only because it made sense in the past, and persists because it allows the "providers" (and Government aka "Regulator") to make money!! So I asked myself: Can we actually get rid of that unnecessary Leviathan that sits right in the middle and that costs me as a consumer at least 65% of my monthly bill? I think Bob helped me imagine an alternative model....Here is how I understand it: I can have my actual physical connectivity and pay for it, for instance, through my annual property tax (In California we pay for several existing community infrastructures this way!!!). Right? Then I can call an equivalent of a standard electrician to come and hook up my home to the "global-zero-barrier-universal protocol-network". And that's it!!...What we eventually need to ""invent" is the latter. Believe it or not we do have all the necessary tools and knowledge for doing it. And how my smartphone, smartwatch, streaming device, etc. etc. can actually get online and communicate? Simply by supporting that imaginary "Universal Protocol", that is, the global network "lingua franca" and voila'!!
Ambient ConnectivityRedefining 'connectivity' as a common resource simply means considering it at the same exact level of any other community infrastructure such as school system, parks and green areas, sewer system, bridges, streets, sidewalks, highways, etc...In such a perspective "Ambient Connectivity" stands for anywhere and anytime access possibility regardless of the specific application one might desire running. You ask: "how does reaching an ambient level create a business model for the average consumer's (the majority) who do not understand the tech side of this type of industry"? My answer: Why should it? A business model should always have one or more products or services at its center. The success of the former depends almost exclusively upon the success of the latter in reducing pains or increasing gains of the targeted market segment. "Ambient Connectivity" definitely helps but it doesn't guarantee a favorable outcome for either of "horizontal" or "vertical" models!
Ambient Connectivity & Business Model
Pre-Chasm Startups & Home AutomationMany IoT related companies appear to be typical 'pre-chasm' cases. According to Everett M. Rogers there is a chasm between the early adopters of an innovative product (the technology enthusiasts and visionaries) and the early majority (the pragmatists). In 'Crossing the Chasm', Geoffrey A. Moore explains how the visionaries and pragmatists have quite different expectations. The transition between visionaries (early adopters) and pragmatists (early majority) seems being the most difficult step. This is exactly what a chasm is about. If a successful firm can create a bandwagon effect in which enough momentum builds, then the product becomes a de facto standard. Roger, as you can see, the issue for Control 4 and similar companies is whether they are able to pass the Rubicon...I definitely wish all innovative companies good luck. But when it's about the so called "Home Automation" I would invite the courageous "hi-tech warriors" to take a few steps back and look into the bigger "battle field": They would be surprised to discover that *home* is one of the most *conservative* areas of the overall human experiences where the "red lines of century-old habits" could be crossed often over more than one generation. Ambient Connectivity, as I understand it, could at best be considered a trend-level facilitation but in no particular way it can directly impact the way we still live our homes both as social environment and physical infrastructure...Having schools and colleges per se' doesn't increase the demand for education; having green areas and parks doesn't generate automatically "Sunday walks"; having good roads doesn't necessarily translate in a significant private transportation demand; having a great sewerage system doesn't....Again Ambient Connectivity appears being the "magic potion" of the general virtualization trend and not a helping hand for companies involved in "Home Automation". The latter need to cross their inevitable chasm: A chasm is much like Dante Alighieri's Hell: Before crossing it you have to "Abandon all hope, ye who enter here. ("Lasciate ogne speranza, voi ch'intrate"). Once you are in you have to go through many cycles and rings...that's what makes the whole thing truly complicated.
Roger, we need to recall that a business model is always an ad hoc activity: There is no free beer, food is expensive and wine is served only at the arrival line.....LOL
Technicism FallacyYou ask: "...what is the common chasm to this market to overcome when you have two companies from very, very different approaches to this market not really doing much in succeeding to evolve to that unexplored future???"
You appear to assume that "solving the problem in the home first" is necessarily a technology-related issue. That's why you seem utterly amazed of the substantial market failure of "Home Automation" companies. According to your analysis they have done their best in technology offering yet both are losing the game...A recent past experience can help us understand: Webvan, named by CNET the largest dot-com flop in history. Ask yourself how did that happen? Only a cash flow issue? A simple faulty financial forecast mixed up with many bad infrastructural decisions? Yes and no! Yes, because they made terrible errors; and no, because the real reasons resided a lot deeper: They were totally out of sync with the reality of people's strongly rooted habits and behavior. They tried their best to innovate by offering a high tech alternative to the decades-old grocery store experience. They offered a smart and efficient solution to a "problem" that the overall society did not felt as such around 2000. Their quasi delusional over reliance or overconfidence in technology as a benefactor of society (Technicism) failed them. They believed Internet alone was enough to virtualize right away the whole grocery store experience. Webvan died out right before the chasm; crashed and burned hundreds of millions of dollars!! It took another decade or so till what we are watching as this writing: The increasing virtualization of the overall shopping experience . As I mentioned in one of my previous posts, *home* is one of the most *conservative* areas of the overall human life where the "red lines of century-old habits" could be crossed often over more than one generation. As Bob put it "Ambient Connectivity is about removing the pay wall around the unexplored future." It's a trend-level enabler, nothing more, nothing less!!
Home Virtualization vs Home AutomationI do agree with you that "complexity" is the consumer's "inside the home problem". The issue is that you see it only as a "technicality" related to "home networking" and "seamless communication," etc... In the reality of home consumer experience that's only the surface, and probably the least important element of the whole situation. Let me explain. First of all, ask yourself what is "complexity"? Here is the simplest response: "Something with many parts in intricate arrangement." From the consumer's viewpoint "complexity" is "something" that is not intuitive, understandable with relatively little or no effort; "something" that tends to stay out of control. The typical immediate feeling associated with "complexity" is *frustration* and ultimately "fear" that inevitably derives from "not being in charge" in a given situation. The latter is the crucial one: "Not being in charge" is perceived as the exact opposite of "feeling at home". We have to keep in mind that "home" is one of the deepest element of one's identity and that's probably the main reason why we all tend to be utmost conservative when it comes to our "home experience." "Home Automation" is not only about the physical elements of a house; it's about one's "feeling at home", "belonging to", "being in charge". You can't change such a centuries-old and deeply layered experience overnight--or ever! The worst comes indeed when "complexity" directly conflicts with the deepest consumers standing. The result? Simple rejection! That's why "Home Automation", as currently defined, is not particularly appealing to the market. That's way these companies are often doomed to stay in pre-chasm phase till someone finally decides to pull the plug and...to the regret of "home automator warriors." Roger, the problem is "Home Automation" itself as currently defined and practiced simply because it doesn't make the consumer feel *empowered* but the exact opposite...Are you looking for a way out? As I recommended in one of my previous posts try to adjust your perspective: Follow "Home Virtualization" instead!!
We vs Things
...Which kind of "first" do you mean? Logical? Chronological? Analytical? Nominal? Alphabetical?... You should clarify your question though, don't you?Please recall that "virtualization" is not "something" that we "use" and therefore we need "something else" to gain access to it! As opposed to "automation", "virtualization" is a replacement tout court of a portion of our physical world experience. In other words, while "automation" tries to "patch" the physical reality to ease it and make it less painful, "virtualization" recreates elements of the physical reality ex novo on a totally new basis. For instance, Amazon is a "virtualization" because it tends to literally replace the whole shopping experience on a totally new foundations. That is, you don't need anymore the "shopping mall" or "grocery store" to have "access" to Amazon.com and make your purchase. You still need a UI somehow and somewhere but that's not anymore specific to *shopping activity*. Right? The same goes with Skype, Netflix, Hulu, Nasdaq, Facebook, LinkedIn, etc. etc. "Automation" does not liberate you as a human being from the *tyranny of the physical world*: It just facilitate you but with a big trade off: Your control upon the physical reality diminishes!! Remember that with "automation" you are not the only one "in charge": You share your "powers" with "things"....That's why with "automation" you might feel less "pain" but at the price of an alienating "complexity", as I mentioned in my previous post... If you haven't seen yet Charlie Chaplin's "Modern Times" I would recommend you look for it and carefully watch it'....You would then realize the true meaning of "automation"... So....here is my recommendation: If you simply change your perspective you would not anymore throw yourself in a causality "swamp" like the one that your apparently logical question pushes you to...then we can continue this dialogue with Bob and hopefully other participants without getting boring and repetitive....
Internet of What?...It's interesting to see you and R. trying to bring some basic order in to the current chaos that appears to characterize IoT related debate. Personally I would try to be even more "radical" than you guys: These days I have the luxury of reading through the IoT related literature. It's astonishing to see a fundamental lack of systematic or even serious conceptualization with related naming conventions, and so on (so much for Semantic this, Semantic that...) I think even the very name of the emerging field is problematic with its will of pre-determination right from the starting point: "Internet of Things"! Right? First and foremost, why should we call it "Internet" with uppercase "I"? Why the hell should we call it at all "internet" even with lowercase "i"? What do we mean with "Things"? Physical "Things", social "Things", cyber "Things"? All of the above at once, like Cisco folks' "Everything" that sounds more like a PR stunt rather than a fully though out standard naming? And how about location (spatial dimension)? Is that also a "Thing"? How about "time" dimension? Should we consider "time" a "Thing" in its own right?
Even more: Are we sure "IoT" (let's call it this way for the time being...) is only about interacting "Things" in a sort of abstract parallel "universe"? Or "Things" in the real word reach out actually *on behalf* of *people*? And if that's the case then why should we call it "Internet of Things"?
And it takes us to the fundamental issue of "What's in a name? That which we call a ....". Should we agree with Juliet that a name is an artificial and meaningless convention when we talk of "IoT"? How about zooming out a bit and re-watch the scene? Are we still truly seeing an "Internet of Things"?
A second order of issues is about "Ambient Connectivity" (AC). Again, here I detect some basic misunderstanding. I guess Bob's notion is about "funding model" regardless of any specific users' purposes while you and probably Robert are focusing upon the "purposes", that is, the ways we use the connectivity. For instance, reelyActive defines a "model and a purpose", a sort of "electronic cocoon", a quasi-parallel "universe" where "Things" (probably meaning people and objects) have an id (here the *purpose*, right?), I would say some sort of "democratic or distributed Big Brother" (No offence, indeed!)...As I understand it, AC is purpose-neutral, it doesn't require to establish yet another "universe" by its own right; rather it's a way to remove the "garden walls" around the golden universes generated and maintained by the current funding and access model. If you want, it's about a re-appropriation of the connectivity by the users regardless of their disparate purposes...We have streets, sidewalks, parks, sewerage, schools, etc.... so should we have connectivity and fund and maintain it the same way...
IoT ante litteramI would like to discuss your interesting post in more than one comment. First, the naming issue. "You have to call it something and the names tend to emerge from what captures most peoples imaginations...": Is that completely right? There is an interesting post (http://vint.sogeti.com/internet-of-things-terminology-arms-race/ ) where almost all the currently used terminology is listed. Besides the German terms (sounding rather bureaucratic!!) all the others appear being marketing-driven, that is, sheer rhetoric stunts...Each big dog pushes its own agenda no matter how true are lies!
Robert, names are important: "Internet of Things" focuses on "How" rather than "What." To understand we need to help settle the dusty hype. We may then see the missing/hidden "What": Do we really need Bruce Sterling's "Spime" at this point? To me that simply sounds too complicated, overkilled...Simple is beautiful: Humanity has already invented the "IoT-What" many centuries ago! What seems to be new is "IoT-How". In facts, when our far ancestors made the first "gift" in exchange for name it, they unknowingly started a new human experience that through the centuries led them/us to create the first "Internet of Things", IoT-ante litteram: It's called "Trade" or, yet better to say, "Exchange of Goods"! First there was "goods--goods" (barter) type of exchange but it wasn't sufficiently "scalable" and needed too much of human intervention (user data input). So we invented rules and conventions (protocols) thereby the "money" (semantics). That was probably the first successful reification of an abstraction into a concrete data structure (Credit goes to the Chinese genius, according to Marco Polo...).... "Money" emerged with 4 attributes: Medium, Measure, Standard and Storage. Then the "BigData-ante litteram" was invented: It's called "Bookkeeping/Accountancy"! "Banks" (cloud storage and processing) were invented so there was no need for money's Storage property (Nonreferential, local memory), hence over time we came up with "fiat money" (Referential memory, Pointer)....
As you see, sometimes the sheer naming can simply hide the most obvious aspects of reality and makes us falsely believe in something which is not...Just imagine how "IoT" debate could benefit if we abstract out the "What" from the "How"...Believe it or not, economy and market can still teach us a lot...
I will comment on your other points later.
Bits and Atoms, Cyber and PhysicalAnd now about the "network"... You wrote: "My idea of the network or the cloud brain comes mostly from oblong industries..." I truly loved your reference to "Tangible User Interface" (TUI) and, more in general, "Tangible Computing" (TC). That's sure one of the bleeding edges with great impact upon "IoT" future.
Let's dig it deeper! What is MIT's Hiroshi Ishii's "Tangible bits" in a nutshell? I followed his work for awhile and I guess his main point is that it's possible --and he and his students proved it-- to combine seamlessly two very distinct domains of human experience: The virtual world of bits (Cyber Things) and the physical world of atoms (Physical Things). Even though expressed with a different terminology, this "filling the gap" appears to me the major promise of "IoT" too. Interestingly there seems to be already a first comprehensive abstraction for it out there: Cyber-Physical System (CPS)! What is truly new here? Human being, each single individual per se' is the most advanced instance of a CPS-ante-litteram known to us. Mother Nature has already invented CPS! So...where is the quantum leap, if any? And this takes us to the transforming concept of "network", indeed.
Bob warns us to be careful when we use the word "network".Great, but wonder what is exactly his notion of "network" as it happens right under our noses? Mine is a minimalist one: Any group of interconnected nodes form a network. Nodes could be "people" or "things". People *or* things, right? No, wrong! In facts, both "IoT" and TUI are striving to change that scenario: Network is increasingly becoming an interconnected collection of people *and/or* things. Nodes of self-managed devices are emerging as "neo-users" and mixing up directly with existing (human) users..."Web 2.0" is not anymore only about buddies, or at least *human* buddies: Thanks to ideas such as "sensing as a service" and "semantic interoperability", things would be finally able to like, post, update their status, look for friends/peers and generate their own timeline and searchable graph...That's not all. I think one of the greatest merits of John Underkoffler's design is his (Oblong Industries...) TUI implementation that closes a decades-long chapter of HCI: "one machine, one human, one mouse, one screen.Well, that doesn't really cut it anymore," he stated during his TED presentation. He followed up with an amazing demo! Sure you saw it. I think that's a major milestone in a fascinating journey from dumb terminal to desktop to laptop to palmtop to walltop to, finally, "inroom computing"! With the latter the UI itself becomes a truly live network of users in the physical 3D space. Multiplayer online game redux? No, because it's not about avatars interacting but actual, physical people collaborating...Hiroshi's bits and atoms are coupling to create the foundation of a new scenario where we can probably recognize that "Networks R US!" Meaning humans and things? Time will tell...
It would be interesting to put Bob's critique of the current network "access model" on a background such the one above: "Networks R Us" versus "path providers"! No wonder if it sounds like the title of SciFi movie...LOL
Brain is the NetworkYou wrote: "[...] Hand enough control to the "cloud brain" to enable whatever I am trying to do [...]." I am still not sure, as you seem to be, that I want to assume a "cloud brain" at all! That vision appears to me very much like a market economy controlled entirely (but "democratically") by the Central Bank as the Chief Executive in charge (ask yourself why the Soviet Union's economy went belly up...."Planing Brain" with too many variables to control in presence of competing "centers of power" and malgre' KGB, etc...). Like you, I envision parallelism in our future, and lots of it in the world of cyber-physical systems. But I see it as embedded at the lowest possible level of the network: My metaphor is not the human brain with its neurons but the human body at the cellular level, the most advance nano system of systems that Mother Nature has ever invented. I rather try to learn from the single cell's internal machinery (local parallelism, concurrence and amazing memory management capability with its solid state DNA...) and the way it networks with other cells through the whole human body's extremely distributed computing environment. An upcoming book by Francis daCosta (I guess sponsored by Intel) offers many clues of how a "central-brain-less" (this is indeed my expression and my very interpretation of what I read in his blog...) system of systems could benefit the "IoT" design. He seems particularly inspired by "simple" live organisms such as the ants and the way they structure their community. Should we go back to read Maurice Maeterlinck's "The Life of the Ant"? Why not?
In my vision being connected is the key; the real "brain" is the network per se', which makes possible the interaction among the nodes. Rather than a "cloud brain" I envision the necessity for a design based on the principle of "checks and balances." I guess I daydream of some variation of what we may call "mesh trust" based on very simple logic that can "grow" (as needed) through higher levels of abstraction and reasoning thank to shared ontologies and semantic interoperability (I closely follow what Michael Koster and his colleagues are cooking based on their smart idea of "Horizontal Platform"). I think in such a futuretense framework we may have way less concerns with privacy issues, or at least more tools to build the necessary defensive layers...Please tell me why you disagree!
Ambiguity as a Market ForceI had a chance to learn about Layer 7 folks when I had a customer who was planning to deliver API services in a Building Automation environment and needed to handle the security issues . That's how I had to learn about API strategy and governance through them.
I personally see the API-based "IoT" and, more in general, "sensing as a service" very promising. By adding "shared ontologies and semantic interoperability" to the whole soup we can probably do amazing things...
What is though fascinating is the implicit societal metaphor behind each single approach. Treating "IoT" as a sort of "marketplace of goods and services" appears being the lowest common denominator. One approach focuses on "things" as providers of "goods and services", another one builds up around the idea of "payment and money" while a 3rd one directly puts "things" at the same level of "human agents" trying to take advantage of existing "web 2" related knowledge...There are those who appear more inspired by the idea of "free market of things" and others looking into "regulated market of things" while a 3rd group follows "socialistic model" by trying to put the "brain" in the cloud, etc...As you can see, at the end we wind up implementing our social and cultural mental models even when we have the illusion to do strictly technology...like I said before, ultimately "Networks R Us", the people: "Things" are at best avatars and surrogates... And we should not forget this ever!
Regarding "ambiguous issues": I do believe that ambiguity per se' is a fundamental market force even though we often believe and are taught the contrary: It's a constant push to expand existing markets and to create brand new ones as long as there is someone out there (which is fortunately always true) thinking to be "on the right track" in searching and researching unexplored avenues...Market is like life: Both are swinging forward and backward between our "optimism of will" and "pessimism of intellect", as Antonio Gramsci would put it....So...Got ambiguity? Go after it! Who knows, it might guide you to Treasure Island, hopefully it turns to be the Stevenson's one... LOL
Beyond Vertical Silos"varied approaches" definitely yes but, if you exclude the big "vertical silos creators" (such as Siemens, Honeywell, etc.) with their consolidated install-base, there are many startups trying to make things work together...The latter per se' is a market expander force...As I have already mentioned in my other posts, I personally imagine the future of "home networking at large" market in what I prefer to call "virtualization": Every time we successfully find a way to absorb the standalone functionality of an existing or needed "thing" (phone, radio, tv, thermostat, oven, fences, light switch, etc. etc. ) in a higher level abstraction device we are simply virtualizing either upward or laterally. Thank to this very process the "vertical silos" of "things", that utterly complicate the home networking tasks, get increasingly out of the way. The end result is an overall simplification of the home networking scene. As opposed to "home automation" that tries to "patch" the "things" and ultimately winds up adding up to the existing "complexity", the virtualization proceeds in the exact opposite direction by disentangling the whole scenario. IMHO, a winning market strategy would follow the latter rather than generating further source of concerns for home owners and tenants...
Home Networking Platform
According to your statement you do lament the lack of a "singular networking platform for...home," and you consider this the reason why a hypothetical "home networking market" is not developing, right?Why do you go after a platform for something ("Home Network") that has no actual substantiation beyond being a sheer legal fiction literally made up by telecom carriers for their exclusive financial benefits?
If you use the expression "Home Network" only to designate what is behind the MPOE without any other technical qualifications then I ask you: What does make your "Home Network" *technically* different from a "Starbucks Network," a "School Network," an "Airport Network" or an "Office Network"? Should we lament the lack of "Starbucks Platform", "School Platform" or "Grocery Store Platform", too?
Let's assume there is something that we can technically define as "Home Network." Why do you think the Smartphone is not the "singular networking platform for home" that you are looking for? As I wrote in my comment in another discussion thread that you referenced, the smartphone is actually emerging as the winning UUI (Universal User Interface), USI (Universal Service Interface) and/or PSI (Personal Service Interface). As we continue to virtualize physical "things" by absorbing their standalone functionality in higher-level-abstraction devices, we simply transform them in *services*. We access these services through apps (sheer bits...) rather than separate specialized physical devices...We do that according to our (personal) needs and regardless of the physical place where we are, that is, no matter at home or outside...Where do we run these apps? On our Smartphones, right? That's why you should look into your Smartphone to find your beloved platform for home and else where...
Smartphone is not an "analogy" but the actual solution. It's the one that is emerging as a universal expression of virtualization, a sort of "Stargate" towards the world of ubiquitous services (BTW, this is the main reason why we should look for a new name for "Smartphone"... ). Virtualization acts like an irresistible solvent: It gradually dissolves our old world of physical "things" (including the "home walls") and convert them to intuitive and user-friendly applications....
For the future I envision services everywhere without the current artificial boundaries. In such a scenario, MPOE and its fictional derivations will be a memorabilia of the telecom companies' golden age...
Emerging Ad Hoc Mobility NetworkingYou wrote: "[...] "home network" concept has some shortcoming [...]" Thanks for the discount but IMHO, it's not at all about "shortcoming" but sheer abstraction, legal fiction defined by telecom carriers for their benefits...As I have already mentioned, "Home Network(ing)" at best is a figure of speech and nothing more!!
You continue: "[...] it is also helpful when it makes the distinction between "public" (hence routable!) IP addresses" and "private"."
Well, do we really need to establish a "Home Network" to mask/NAT a subnet behind a little routing box? Is that a distinguishable characteristic of a "Home Network" as such? No, at all! In facts, you can insert your routing device wherever you like, even in the middle of the street or in Mojave Desert, and reproduce the same results as long as you have for instance "ambient connectivity" available to you, right?
You stated: "[...] this means that any devices that want to connect to your "networking platform" can't see it directly, it needs to go thru ... some "public IP host" (a server in the cloud, your DSL router etc.) to communicate."
Well, how about "WiFi Direct" type of connection? It's there and being served in a silver plate on top of the latest generations of smartpones while other smart devices have already started to support the feature...It actually seems to be the optional replacement for Bluetooth wherever you don't experience resource-constraint environment.
As you might know, with "WiFi Direct", acting as AP is a matter of on-the-fly negotiation. Virtualization has stricken again: Sayonara my beloved hw Access Point fatally removed by Soft AP!!
"IP- client" and "IP-reachable" are already old language...
Your words again: "[...]so many alternative ways to identify things ... and other goodies that come into the picture (like default gateway)."
Why do we need to identify things if they can address it by themselves thank to Protected Setup-style transfer? What is the use of "default gateway" if it can become a matter of negotiation among devices? If our smartphone can act as *roving AP-Router* directly enabled to tap into 3G/4G/5G (and who knows what else...) as needed, why the hell should we still require extra default gateway or static route?
You conclude: "[...]"artificial boundary" is not so artificial as you think after all. Think of it as a tool instead of a limitation [...]"
Believe it or not "artificial boundaries" are always artificial and often superfluous... In a world of *ubiquitous services" as a result of increasing virtualization of the physical "things", any unnecessary boundary (such as the one literally made up thanks to MPOE, etc...) will spontaneously become deprecated and thereby marginalized and eventually removed from the scene..
Ask yourself: Why should we allow telecom carriers force their legal fictions upon us if it's for their exclusive benefits and "rent-seeking" goals? Why should we in the first place accept a "limitation" to find later ourselves forced trying to use it as a "tool"?
Why should we smile when we get penalized?
Ubiquitous Services Environment (USE)As I have already mentioned, the whole idea of "home network(ing)" is a (legal) fiction or, more exactly, the sub-product of a specific contracting model that by default regulates the relationship among carriers and their customers. The "home network(ing)" is a sheer (legal) scheme that helps the telecom entities to better adapt their services' charges to their best benefit. The so called MPOE is the physical element that, by establishing an arbitrary boundary, formally sanctions the "home network(ing)" perimeter.
Therefore, assuming the existence of a market with traction, etc. based on such a basis appears to be a typical result of treating as a concrete thing something which is not concrete, but merely an idea (More sophisticated people would call it "reification" aka "concretism", or "the fallacy of misplaced concreteness")...
We can indeed take the liberty to segment the marketplace according to any criteria that better satisfies our assumptions or analytical objectives. Market analysts often do that in order to better understand certain dynamics, but the experienced ones never confuse the realities with their sheer abstractions unless they are paid to do exactly that...In this case we shouldn't call them anymore "market analysts" but "market ideologists" or "market activists"!
Just imagine how many spam emails we receive about the future great stock market performance of this or that company about which nobody has ever heard any serious or minimally positive comments... Several years ago I personally fell in one of these apparently legal schemes and lost quite a bit of money...That has become the price of my personal hard-training in not-to-fall-anymore-for-easy "market reifications"...
As I have already mentioned, I personally do not foresee a market for "home network(ing)" per se' but an ascending trend for what I temporarily would call "Ubiquitous Services Market(s)" or USM, which I consider inclusive of anything related to our home environment or, yet better to say, Serviced Living Environment (SLE), too.
Let me briefly explain: As we move forward with virtualization of our physical environment (Here Ray Kurzweil's "Law of Accelerating Returns" applies) we convert the standalone "things" in *applications* while their functionalities is abstracted out and offered as *provided services*. The virtualization process can move *upward* or *laterally* according to the very nature of any given "thing" and its transformation in *application/service*. The result of such a trend is the emergence of what we may call "Services Oriented Environment" (SOE) or "Ubiquitous Services Environment" (USE). The latter will generate what I call "Ubiquitous Services Market(s)" (USM)...This will be a genuinely e-Commerce type of market where the online payment (in all its burgeoning formats) will inevitably dominate. I also strongly believe that what we still anachronistically keep calling "smartphone" will be the main PSI (Personal Service Interface) of choice until it will fall "victim" of the virtualization process, too...The latter will be better foreseeable once we start to fully leverage the potentials of "wearable" and "implantable" computing devices...
What is in a Smartphone?
The home environment to which you are referring is the current one full of physical devices ("things") with stand alone functionality such as radio, tv, light switch, light lamps, alarm system, pc, laptop, tablets, printer, land phone, fridge, microwave, thermostat, home and garage door, etc, etc. In such a world you indeed need forms of integration through a centralized box, for example, right? It's very much what is done through "home automation" according to the current concept. In such a scenario you can talk about a "home network" as long as you want to make possible communication among these standalone devices and create a central point for system management, etc. Therefore you need the usual access point-router, internal wiring, dsl line, etc... In such a context a smartphone can indeed act largely as a user interface (UI). This is the context that I have characterized as "vertical silos" of "things"...
The environment I envision is very simplified compared to what we have today: Many of the objects with standalone functionality have been virtualized and converted to application and provided-services. No more physical devices (at least as we know them today) such as tv, radio, land phone, light switch, light lamps, HVAC, oven, microwave, fridge, etc. etc. In such a physically "simplified" world a "home network" with router and AP, etc doesn't make sense. Services are either cloud-based and delivered through smartphone (movie, music, tv, ebooks, ehealth, payment, banking, security update, firmware update, etc.etc.) or enabled to interact locally directly with one or more registered smartphones (door lock, windows blind, HVAC, microwave, etc) and through this connect to the cloud (for example when the smartphone is idle or during night time) as needed (download firmware update, security config, upload logs, for instance). In my vision there is nothing that you can call a "home network" beyond the metaphor but what I would define as "Serviced Living Environment".
As you can see in a Ubiquitous Services Environment the central unifying platform at personal level becomes your smartphone. In a real world situation there might be more than one smartphone being used in the environment by family members. In this case the whole group of smartphones will cooperate and provide the computing and connectivity as a distributed platform during 24 hours... Here the smartphone won't be only a UUI but also a "unifying platform".
Service Avatar (SA)"Ubiquitous Services Environment" is _not_ a "solution"; it's just a conventional name to designate the horizon of an interconnected world where many of today's hw-based "things" have been virtualized, that is, literally re-invented as "application-service." Nor is USE some sort of "government plan" (Orwellian world stuff...) to be imposed to alienated cohorts of citizens...You are right: After all, "you may want/need bigger (screen), more power (desktop), storage (disk), etc." We still have people who keep using VCRs the same way as I sometimes use my father's old hat to time-travel to my "Happy Days"...
But seriously, why do you think a "mobile-centric" USE means the end of big screen, local storage, etc? There is actually a concept that I am working on these days and will hopefully discuss more in detail in my upcoming blog: "Service Avatar" (SA), that basically stands for ways an "application-service" (a former hw-based "thing") could materialize and become available for personal or collective fruition, access, etc. For example, imagine you have a Personal Service Interface Device (currently called "smartphone") with access to Netflix or Hulu and fully certified, for instance, for DLNA via WiFi Direct streaming. You have guests and like to watch a movie together and enjoy what HDMI (and name it...) could offer. What are you supposed to do? 5" wouldn't be the ideal solution, would it? That's where a "Service Avatar" comes into play: a "walltop" or "wallembedded" or "wallpainted" giant screen, let's say 120" (or more), which can receive and play the HDMI signals streaming out of your PSID. "Service Avatar", as I imagine it for the very near future, could be a mono-task "player" with very "low cost & simplified" electronics. 3D Printer is another example of "Service Avatar" (SA) the same way as the future virtual HVAC based on microwaves, etc... (Just imagine what gonna happen thanks to coupling Virtualization and Nano!!)
Regarding "owned network": In a USE characterized by "last mile" made of ad hoc mobile connectivity (check our debate in the smartphone related discussion running in parallel), what would be the exact practical meaning of an "owned network"? "Owned" by who? What is a "Home/owned Network" in a USE context where it makes much more sense that your _home refrigerator_ stays in touch with the grocery store smart delivery system rather than your washing machine, lighting system or smart door pad? In USE the "boundaries", if any, will reasonably follow a "service group" paradigm rather than a today's "walled garden" model...
I think I said it before but I am afraid the problem with your analysis is typical of people who try to imagine, or worse, plan the future in the framework of the constrained current conditions by keeping the latter as a constant in their equations...
Personal Service Interface (PSA)"everyone's PSI being pretty 'dumb', just an input/output interface device - all the processing and storage being done in the cloud".
Well, I guess that's the wrong answer! Both "processing" and "storage" should be flexible enough to enable a Ubiquitous Services user to have always a *fallback* solution
(failover capability) for the rainy day when supporting infrastructure experiences a glitch, for whatever reason. That's why I envision a PSI (erroneously called "smartphone") as a little powerhouse rather than a dumb device (don't worry much about the manufacturing costs...). You may obviously object that this will be an overkilled solution. My answer is indeed nay because a Ubiquitous Services Environment (USE) could use the idle processing and storage of a PSI (aka Smartphone) as a roaming/distributed component of the infrastructure itself during major downtime to guarantee user-end's enduring access to essential services...
So to answer your question:
1) I would use a combination of cloud-based and peer-to-peer/distributed processing;
2) I would adopt a federation of "referential" (cloud-based) and "nonreferential" (on-premises and/or "roaming cache") storage architecture.
Please recall: Putting all eggs in one basket is the worst answer to your dilemma.
Service ContinuityWe should keep in mind that a Ubiquitous Services Environment (USE) could go mainstream only if it "knows" how to handle abnormal situations. For example, once a "brick & mortar"- based portion of our personal or social life experience (e.g. shopping) gets successfully virtualized and converted into an extensively available "application-service," (eCommerce) going back to the old world of "physical things" (shopping mall, etc.) won't always be an option anymore. Therefore *service continuity* (guaranteed SLA, etc.) becomes a pre-condition of the overall context...Right?
That's why, for instance, an adaptable degree of decoupling between the communicating parties and the ability to communicate in an asynchronous way in an ad hoc mobile environment (that includes people and IoT) are all essential components of a successful USE...
Of course we still need to resolve many many problems as you would imagine. Yet it's very important to fully understand in which direction the whole thing's evolving...and particularly make an effort to set aside "prejudices" that constantly stem from the current state of constraints in technologies and market demands.
We also need to elaborate new vocabularies, redefine/kill old concepts and even change consolidated naming conventions whenever needed, so we can communicate, compare and debate concurrent visions...
WiFi Direct and DLNAYou miss the very nature of WiFi Direct, hence the fact that the latter and DLNA are completely different technologies. Wi-Fi Direct is a new way for Wi-Fi devices to connect to each other. Wi-Fi devices are able to make _direct connection_ (forming ad hoc on-the-fly groups, for instance) quickly and conveniently to do things like print, sync, and share content, etc., _without_ the need for an access point or router...With Direct the traditional hardware-based AP-Router gets literally virtualized and becomes a roaming *application-service*. The connections based on Direct specification work at typical Wi-Fi throughput and range, protected by WPA2™-certified security protocols and including WMM QS mechanisms....
DLNA is instead a set of interoperability guidelines for media streaming (audio, video) among two appliances. Products that are DLNA certified are typically recognized, with little or no setup, as soon as you connect them to your _existing network_. DLNA certification means that the device plays a role in the "home network" and that other DLNA products can communicate with it based on their own roles. Some products store the media. Some products control the media and some products play the media. There is a certification for each of these roles. Very recently DLNA incorporated WiFi Direct for seamless P2P (no supporting "home network", etc. needed) streaming.
There are indeed several other protocols to handle a variety of communication needs such as NFC, etc.
We should imagine USE as a time-based process rather than a sort of sudden "moment of truth"...It requires indeed pervasive and fault-tolerant connectivity as the supporting layer. The "connectivity" doesn't necessarily mean (only) "Internet" as we know it (see Bob's arguments) but it's "last mile" have to provide full support for ad hoc mobile environment. We still don't know how all this will exactly play out but some emerging clues such as the above mentioned could offer useful indications...
Next Big WaveBattery drain is a today's technology constraint: If we are already able to wirelessly recharge an electrical vehicle's battery (see the test run in SK), I wonder why shouldn't we be able to do a better job with a little smartphone...Right? Let me be audacious: How about looking for a viable solution to harvest energy off the network traffic itself or by converting smartphone's physical movements or...?
"Additional fixed infrastructure"? It depends on how we define that. In any case we will need extra support and much more if we want to enable a Ubiquitous Services Environment (USE). For instance, in my daydreams I clearly see neighborhood wireless towers as a viable replacement for the current "last mile/MPOE" rigid infrastructure. We definitely need to think of last mile connectivity in a lot more flexible way...
One last point: I think it's a very good idea to dedicate more thought to network traffic _typologies_ while making decisions about protocols, etc... Again, in my daydreams I see an increasing need for _ad hoc mobile_ connectivity: My forecast is that in the near future a big chunk of the current Internet traffic will be just "offloaded" to stay local right inside the last mile boundaries: WiFi Direct is just out there to anticipate the next big wave...
IoT's Needs and DifficultiesThe future role of what we still keep calling "smartphone" (for how long?) won't be determined by what is deemed best for this or that specific protocol. I think this is the overall evolution of the context that will assign role(s) to each single gear of USE (Ubiquitous Services Environment) and will determine the fate of existing protocols and devices...
Having said that, there are a few issues we need to address before throwing ourselves in an open-ended guessing game:
1- What is being connected: sensors, every-day objects, or appliances?
In other words, "things" are not the same. IoT's flat image of the "world of things" is just an abstraction: The reality is differentiated and appears chaotic: The question is how we can structure the scene by keeping the connectivity efficient, flexible and fault-tolerant.
2. Are the 'things' interactive, or do they deliver ambient intelligence via the network?
Again, "things" are diverse: Interactivity is a must but we may still need the "network." But network doesn't mean the _fixed and rigid_ one that we have today: "Ad Hoc" is the Name of the Rose! The question is how we can design by reasoning in terms of "liquid structures", "ad hoc mobility", "service group paradigm," etc.
3. Do the 'things' connect directly or via a gateway?
Again, "things" are not the same and vary in capabilities, etc. "Direct connection" is definitely a requirement yet we need _gateways_ too. But what is the exact meaning of a _gateway_ in a context where old fashioned deterministic hw gets literally liquefied, virtualized; where (quasi-)all is done in sw? The question is how we can enable "things" to negotiate their role(s) according to contingencies (including downtime) and adapt to ever changing data flow patterns...
4. How is data from sensors & devices secured and kept private?
Yes, "things" are diverse, therefore privacy and security could have different meanings in varying contexts and for disparate participants. The question is how we can design having _adaptive_ privacy and security in mind?
5. What applications of the Internet of Things will actually be valuable to users?
In other words, where is the exact value proposition of IoT/IoE, etc.? If the connectedness per se' is not a value to users then how can we establish the value pathways?
Elephant in the RoomThe reason I put a lot of emphasis upon "smartphone" is because it has already become, across the world, an actual _way of life_ with its own culture...."Smartphone" is arriving where a PC/Mac has never ever dreamed to reach.
Recently I have been travelling across S. America and visiting for business big and small towns. It's amazing to see in small towns of 50K inhabitants there are literally 10 to 15 stores selling mobile phones; to see almost everybody walking/playing with a mobile phone in their hands...even the kids have their little phone and learn how to rapidly move their fragile fingers writing, sending and receiving text messages...
Mobile phones are gone way beyond being a status symbol: Having a mobile is like being dressed, wearing shoes, being participant of a _virtual community of people and services_ that is already deciding the fate of the "real" one...
What we still keep anachronistically calling "Smartphone" has already become our personal UI towards the world and all the increasing number of services made available to users....
What ever will be the future of IoT/IoE, it simply can't ignore such an elephant in the room. Again, it's not about you and me and our little gadgets; but billions and billions of people out there whose second attribute after their humanity is actually their mobile phone, we like it or not!
Cart Before the HorseIoT doesn't go mainstream because it's still too complicated for the average user? Is that the moral of the story that you folks are trying to put across? Is that all? Hmmm...the problem is that your explanation is too large and generic that it doesn't really explain anything!
Simple is beautiful! Ok, 100% agreed. Then what?
How about shifting the perspective toward the following statement:
IoT doesn't go mainstream after decades first and foremost because the average hypothetical consumer still doesn't see any specific value proposition in it! In other words, aren't we still in front of a classic case of "wannabe a solution" looking for "a problem" to resolve?
If you ask people how they would move forward if one puts _the cart before the horse_, almost everybody would provide you the obvious answer: No way!
The problem is that in the real life we try (particularly in our "quick & dirty" analysis) to put the cart before the horse more often than what we would imagine or like to accept we do!
In facts, in a very recent report on IoT by Rachel Thompson of the Research and Markets Ltd there are 6 problematic areas listed. One of them is exactly the _value proposition_ issue as I mentioned above, which is relevant from a potential IoT user perspective.
I have briefly discussed several of them in one of my posts right in this thread but I report the complete list for your nightly reflections:
1. What is being connected: sensors, every-day objects, or appliances?
2. Are the 'things' interactive, or do they deliver ambient intelligence via the network?
3. Do the 'things' connect directly or via a gateway?
4. How is data from sensors & devices secured and kept private?
5. What analysis can be done on collected data?
6. What applications of the Internet of Things will actually be valuable to users?