iPhone 2.0 (?)
Thursday, June 28, 2007 at 04:31AM What are your expectations of a cellphone/mobile-device?
What I want to see in my next mobile-device is an RFID Reader that will track tags throughout my home, with content tied to a searchable online database. In short, I want digital search of non-digital content… in my pocket.
This is entirely feasible (books.google.com). It would take some initial label-tagging of one’s existing collection, but for those geek enough among us who have catalogued our books (Library Thing), and for that matter, our DVDs (Chasing the Frog), this should be second nature by now, and future books and magazines would be pre-embedded, so it’s only the existing collection one needs to worry about. Something like an adhesive RFID sticker, not dissimilar to what is used in many inventory tracking systems today, that lets an item identify itself to the network.
Using my mobile device, I do a search of my home’s content for every printed reference to Frédéric Bastiat. The device picks up the signal from the RFIDs and reads the identity of all labeled content (books), then searches the digital equivalent online, and delivers the results on screen: 6 magazine articles and 3 books; “Ah, this is the one!” Where? Located on the third shelf on the left, reference on page 57. Though it is really searching a proxy in an online database and returning results based upon the books it sees present, to the end user it appears as if it is actually searching the physical publications. That’s what I want from my mobile device— Digital search of the content of my non-digital media.
This is all existing technology. It’s just a matter of implementation.
You can read Matt Dickman’s commentary on Apple’s new iPhone at Techno//Marketer.
View Printer Friendly Version
Reader Comments (13)
Could the combination guide a shopper in a grocery store so they take the optimal route and don't miss anything? Sounds plausible. It could let them know if something is out of stock and suggest an alternative so they could plan accordingly.
Publishers would be crazy not to jump on this and put their publications at the fingertips of every person with a device.
This is going to have huge ramifications on mobile. The center of the personal network is going to migrate from the desktop to the pocket. I'm completely convinced of this.
There use to be all this talk 10 years ago about the "smart refrigerator" that was going to restock your groceries for you. I actually saw a whole presentation on the concept in about '99. I always thought, "well, that's a clever idea and all, but won't that take forever to roll out? How often do people replace a refrigerator? Once in a lifetime? When they buy a new home, or do a remodel?" But cell phones, people upgrade those every two years or so, sometimes even more often. I project that refrigerators will mostly stay dumb. Food will get connected via RFID, and your phone will get smarter and smarter. I see the RFID on your milk carton embedded with the expiration date telling your phone when it needs replacing, adding that date to your calendar as a reminder. You've taken it the obvious next step: At the store, the milk says "Hey, I'm your brand, there are 5 of us left on isle 12." Or even better, you're walking down the street (I live in New York, I walk) and the store you're passing by sends out a signal: "Hey, your milk expired today, we have your brand in stock. Come on in." Stores could use it to optimize their inventory: If everyone comes in with their mobile device saying they're out of a certain brand of Salsa that the store doesn't even carry, the store's inventory software immediately spots the market demand and stocks up, even on a product they had not even been carrying. Some would feel it's too Big Brother-ish, but individual users could stay anonymous, and this kind of buying pattern analytics has been going on for ages now on the things that you already buy, based on the UPC tracking data of purchases. But this would be even more frictionless— it ties inventory directly to real-time market demand, not just to past buying behavior.
The thing I love about the book inventory gimmick is that it ties together two already existing technologies (RFID and search) and it is a utility that truly adds value to the owner/user in their own home, rather than a marketing tool (not that I have anything against marketing tools). It just adds so much to your life (if you're a reader). Google and the internet in general have made me frustrated with the inefficiency of the printed page. But if your home library was RFID'ed and the book titles were then tied to Google's online book search, you could search all of your printed content just as you do the web! That's so powerful. If it took off, you know magazine publishers would get on board, what publisher would want their product left out of that kind of search? All printed content today is produced digitally anyway, and most of it is duplicated for online publication in some form already. All that is needed to implement, is to place the RFID on the publication and match it to the online content in text search. Even though the user is really searching a proxy copy in an online database, it would "feel" like they were actually searching the text of their physical library. Isn't that powerful?
With the inclusion of Camera/Video/Audio recorder, Chat, Web, Email, PIM services, SMS, GPS, MP3 player, TV, and I would hope soon an RFID reader, I'm more inclined to refer to them as a "mobile-device" than a "cellphone". It is inevitable that credit card technology, likely something like MasterCard's PayPass, will soon be incorporated into cellphone/mobile-devices. Very soon, and for many people already, the phone will not even be the primary feature.
Markus Mikkolainen, CEO of Kabuto, in Finland pointed out that cellphones in parts of Europe can read barcodes with their camera, which I previously blogged about here:
http://gigantico.squarespace.com/336554365346/2007/1/27/mobile-phone-codes.html
First, let's decide what is the target market for your device? If you are targeting high end market where someone is willing to pay >300USD and good amount on monthly subscription than it's ok. But how many can afford it? may be 1-2 % of current mobile users.
Second, We have to make the device and service affordable to majority of the consumers, E.g., FM radio, which is now available in almost every low end phone.
RFID reader is good feature to have in the handset and same with mobile payment, card reader, music players etc. but are we just trying to address the solution to a particular requirement/class?
Mobile search and mobile advertisement is also being projected as the killer apps for the future and someone will have to take these features to the masses and build the solution for the masses, for most of others, these technologies are already available thru internet on their laptops and other gizmo's but challenges is to take the solution to the first time users of these technologies thru mobile handset.
> But how many can afford it? may be 1-2 % of current mobile users.
I don't understand, on the Linked[in] Q&A you state:
> All the research data also shows that more than 80% mobile phones
> are being sold in the feature phone category...
Now you're saying only 1-2 % can afford them. Do you mean feature phones, or just iPhones? Or the RFID features I'm proposing?
I know that you're operating out of India. Is this a reflection of the market you see there? Is your 1 or 2% statement referring to mobile users in India, mobile users in emerging markets, mobile users globally? This may very well be the case. I realize I'm speaking from an Americentric perspective, but I am after all in New York. I understand that there are many cellphones on the market that are economical, for "the masses"— Vodafone, Nokia and Motorola all make models that sell for as little as $25.
If you mean to say, who is going to bring these high-end features to the masses, well the faster they are rolled out at the top of the market, the faster economies of scale kick-in, and bring costs down on premium features that will eventually make them common features.
I refer to 1-2% users who are willing to pay >300 USD for a phone device and feature phone reference is for non smart phone OS category, which has more than 80% market share today.
RFID can very well come with low cost device also.
Regards,
Sunil
Also, just to be clear, when I speak of having RFID built into the phone, I'm not talking about having the phone RFID tagged, like say a smart-card device. That has been done:
http://tinyurl.com/2l7v5o
WHAT I'M TALKING ABOUT is building an RFID "Reader" into the cellphone:
http://www.symbol.com/category.php?category=5336
The technology would obviously have to come down in size to something compact enough to be used in a consumer level product. But that's always the case when a technology jumps from a commercial implementation to consumer use.
http://www.mobilemag.com/content/100/104/C2607/
Although I do not agree with going backwards and rfid tagging all our old personal items, I do see the value in having everything tagged from some point going forward (including ourselves).
The practical issues today with having a cell phone that can read RFID tags are:
1. What kind of tags can it read.
2. What software does the phone have to available to it to do extensive searching on those tags?
Beacuse of the variet of tags on the market and their different frequencies there is a BIG problem of standardization today. The implantable tags are 125khz. This is what I have in my hand and my cats have in their backs. However, both of our tags are different brands. Their tags are made by a pet company (Friendchip) and mine designed for farm animals (EM4102). There are also Hitachi and Verichip tags in the same frequence. All of which require their own readers due to incompatibilities. The tags which are in your car keys, credit cards, ez-pass, and so many other devices are also propritary with their interfaces and running at frequencies like 900MHz, 2.4GHz, and so on.
The phone itself would need to have extensive scripting abilities to do your bidding. That or at least a killer app that can lookup tags that are being scanned for you on-line and bring back the information in some sort of report format.
Today if you really wanted to pull off a little RFID fun at home, I would recommend making your own hand held reader. Buy a parallax RFID reader, combine with microcontroller of yoru choice (I like the avr atmel mega series) and pickup a nice serial bluetooth module from sparkfun. Now you can quickly whip up a device on a breadboard that can scan tags, and link up with your home computer or cell phone via bluetooth. Through your phone or computer you can then script searches based on tag reads and the fun begins. The EM4102 tags are probably the most expensive being well under $1 each.
for informing the conversation about the Nokia 5140...
for clearing up some of the issues regarding compatibility...
and in general for lending your expertise the to subject.
As for the Nokia 5140, this doesn't really solve for the implementation that I describe. The Nokia's read-range is a mere 2 cm. You must practically touch the phone's sensor to the RFID tag in order for the reader to interrogate the tag. The kind of implementation I speak of is more of the nature that is used to manage inventory in a warehouse. This, as I understand it, has more to do with the tag than the reader. Namely, the kind of antenna, and whether it is active or passive. These tags can have a read-range of 100 to 1500 meters.
It's an interesting proposition you pose. This is the kind of future that many RFID advocates are looking forward to. Bruce Sterling coined the term SPIME to represent an object that is traceable in space and time, and he recognized that one of the technologies that could create such an infrastructure is RFID.
However, there are a few issues with the various functionality you mention in your post. First, we'll start off with the range issue.
> As for the Nokia 5140, this doesn't really
> solve for the implementation that I describe.
> The Nokia's read-range is a mere 2 cm. You
> must practically touch the phone's sensor
> to the RFID tag in order for the reader to
> interrogate the tag. The kind of
> implementation I speak of is more of the
> nature that is used to manage inventory in
> a warehouse. This, as I understand it, has
> more to do with the tag than the reader.
> Namely, the kind of antenna, and whether
> it is active or passive. These tags can
> have a read-range of 100 to 1500 meters.
The problem is that the only way you are going to get any significant range out of any RFID tag is to make it an active tag. That means the tag has a battery, and a limited lifespan of 3-5 years (typically). Introducing a battery also means a much bulkier tag, and one that is much more vulnerable to the elements (temperature, humidity, air pressure, etc.). The only thing you have left is to use passive tags and build an extensive network of passive RFID readers throughout your environment. This is a very expensive and intrusive prospect. Imagine having to plug in your set of dresser drawers and set it up on the wireless network just so you can scan your sweaters.
One thing is for certain, a reader integrated into a phone will most likely never have the long range you're looking for. Aside from antenna size, you're also dealing with power consumption and the magentic field that must be generated by the passive reader in order to power up and communicate with the passive tags. To communicate with a tag 100 meters away (or more), the magnetic field would have to be massive, and I wouldn't want to see the power drain specs on your phone battery either ;)
Another problem I see with this is simply the fact that there would have to be some kind of RFID standard where every object used the correct RFID frequency, communication protocols, standardized security mechanisms, and standard object access data schema. Let's not also forget about the security and privacy issues that might arise with every object you own advertising itself to anyone with an RFID reader.
That leads me into another issue; location finding. In your post, you state:
> The device picks up the signal from the
> RFIDs and reads the identity of all
> labeled content (books), then searches
> the digital equivalent online, and
> delivers the results on screen: 6
> magazine articles and 3 books; “Ah,
> this is the one!” Where? Located on
> the third shelf on the left...
If you did somehow come up with a hand held device that could read all passive tags within a 100 meter range, well... the first problem you'd run into is that you'd most likely be reading your neighbors items as well as your own. The other issue is that RFID is not a location finding technology, even though many vendors are trying to stretch the feature set and ultimately the definition of RFID.
http://www.rfidtoys.net/forum/forum_posts.asp?TID=139
http://www.rfidtoys.net/forum/forum_posts.asp?TID=281
Typically, the reader will only tell you the IDs it has read, not where they are. Unless you used a smart shelf style reader that had distinguishable antenna feeds (difficult/expensive at best) and/or multiple readers who's fields did not overlap or interfere with each other (also difficult), finding the actual shelf a book is sitting on would still be up to you walking over there and finding it, or at least walking down the library isles with a hand held short range reader that would tell you once you got within close range of the book... which is rather pointless considering you're already walking around looking for the book in the first place :)
It's nice to see several responses, and it's interesting to see Mikey's post on the subject... even if I have to correct one thing:
> Their tags are made by a pet company (Friendchip)
> and mine designed for farm animals (EM4102).
The EM4102 is typically NOT used for animal tracking... the reason you can get glass encased EM4102 tags is that they are meant to be used in encasement/embedded situations, chemical environments, etc. More info about types of tags can be found here:
http://www.rfidtoys.net/forum/forum_posts.asp?TID=278&PN=1
Like Mikey says, the Nokia 5140 came out in mid 2004. There are also other phones with BOTH a reader AND a tag in them, and they ARE currently being used throughout Asia for mobile phone based payment. I think there are great things happening with mobile RFID, but the most interesting things I've seen in a while have nothing to do with RFID (that's mostly because of RFID's limitations). All in all, I think the future you've posted about is something that could come about without RFID. One of the things I think will make a major impact in the coming years is a hugh increase in the security and reliability of biometric identification. Another area that just amazes me are the leaps being made in visual object recognition that go far beyond facial recognition. Simple software applications are making use of amazingly powerful video processing algorithms to allow these applications to "see" things like books and pencils and apples and even know the difference between a picture of an apple and an actual apple using stereoscopic vision to assess the depth of an object. This results in "seeing" robots, and products like Microsoft Surface, which uses video cameras and RFID together to create an astounding human interface.
Amal,
I will caveat my response, and acknowledge that by all accounts you are the RFID expert (you literally wrote the book), and I'm just a guy with some ideas.
However, I'm not easily dissuaded.
The privacy issue is something I'd think could be solved with encryption. I really do appreciate the forethought, but I'm going to table that for the moment to concentrate on technological feasibility.
I want to address this as a technology problem—
"Can it be done?" (and soon)
I came into this with some understanding that such an implementation would required active (powered) RFID tags with antennas. I was also aware of the battery life issue, though I never expected the conversation to get so in depth that I would need to cross that bridge (I should be so thankful that any of my posts get this kind of attention).
I suspect you're familiar with EnOcean's energy harvesting technology. One of the publicized uses is as an RFID tag application to replace the need for perishable RFID tag batteries.
http://www.powerpulse.net/story.php?storyID=16961
Could this technology perhaps lead to a solution?
-Chris
What Mikey said about not tagging our old items is, unfortunately, what's probably going to have to happen... at least at first. The first question I have regarding the ever popular milk-going-bad scenario is; what incentive do milk manufacturers have to add RFID tags to their cartons? The only real way for this to happen is for consumers and int turn the giant chain stores to demand they use the new standardized SPIME tags... and we've all seen what happened when Wal-Mart decided to demand just that... it didn't work. The reason it didn't work is that the manufacturers saw no reason to help Wal-Mark make more millions by footing the bill to tag their products just so they could sell their wares in Wal-Mart. A milk producer doesn't have to know or care when the milk goes bad... they know that when they stamp the "goes bad" date on the side of the carton and when they stack the crates together by bottling/cartoning date. Then the units are shipped out or they are junked if they've sat around too long. Because of the assembly line nature of milk production, the only real benefit of putting RFID tags on cartons would be for the consumer and possibly the retail storefront. Milk is already a pretty low margin item, and to add RFID and IT infrastructure to the bottling process and slap on a tag just so a consumer can get a text message from their smart fridge to remind them their milk is going bad... it doesn't add up for the milk producer.
Add on top of that the fact that the Wal-Mark debacle was still dealing with the relatively inexpensive passive tags... if you introduced a requirement for active tags, the cost skyrockets into the insane. Couple that cost with new power technologies like nanopower harvesting, and I really see corporate cost accountants everywhere pulling their hair out.
But let's get down to brass tacks... "Can it be done?". It really is chicken and egg thing here... if everyone else was doing it, then the milk producer that didn't would probably be at a disadvantage. If you really want to figure this out, the steps I think would have to be taken would be:
1) figure out a standard air interface (frequency, comm protocol, active/passive) and data schema (what bits mean what) that works for all types of products, books, plates, milk, cell phones, shoes, cars, hard hats, condoms, etc. My bet rests on passive LF or HF tags because both tags and short range readers are low cost, which leads me to #2.
2) source out tags and create reader kits that are easy for techies (and laymen?) to assemble and use. Open source software may also have to be created to operate these kits, or at least get them to reliably interface with #3.
3) create a universal database that links item codes to manufacturer specs, or pass off queries directly to manufacture's web services, or if the SPIME standard allows for it, let the user query a URL stored in the tag's memory space. This database could start out as a simple, open source database that is powered by users in a Wiki style fashion. Users could decide on standardized item codes, submit specs or content for the item or link directly to manufacturer's URLs, etc.
4) once you feel the database and SPIME standard have reached a sort of critical mass, submit an ISO standards application and get an ISO certification for the SPIME standard.
5) Now time for the big leap, you would then go out and try to convince the rest of the consumers that this would benefit them, while simultaneously focusing on manufacturers and inventory management outfits, convincing them that: A) tagged items are wanted by the consumer. B) new items should be tagged during the manufacturing process. C) readers should be incorporated into countertops, refrigerators, desk drawers, and devices like the iPhone 2.0
What you said about using encryption to protect privacy and security works to a point. Encryption is a power hungry process, and right now passive tags have great deal of trouble working with anything over 40 bit encryption, and that's usually a proprietary algorithm being used. Standard algorithm 128bit encryption tags do exist, but they take 4-8 seconds to read a piece because of the lack of power for fast processing speeds. Active RFID tags have it easier when it comes to encryption because they do have a power source, but here's the big question with encryption: How does the average consumer, the same consumer that can't be trusted to read the "sell by" date on the carton of milk, ensure that when they sell their crap on ebay that they reset the encryption on the item they are selling so the purchaser can integrate that item into their own inventory. I bought a PDA phone off ebay and the guy forgot to unlock the phone before selling it to me and couldn't remember the password. I had to fight with the carrier and ultimately send it in to get it unlocked (I even tried a hard reset, wouldn't work).
I really like EnOcean and the nanoPower forum's focus on energy harvesting technology... it's a subject I've always been very interested in. I think at this point though that they've got a long way to go before having enough power to drive an active RFID transmission pulse that will reach the ranges you're talking about. The military tags you're talking about that get 100 to an amazing 1000 meters (and some reach up to a couple miles) are driven by giant batteries. I think with enough power harvesting thrown into the mix, and adequate power storage medium (super caps or something else), and a very long pulse delay (like several minutes instead of several seconds), you might be able to pull off an active tag that supports basic encryption and can interact with an interrogator (reader) to get at the memory contents.
Normally, an active tag that contains memory blocks does not transmit the entire contents of it's memory during each pulse, it only transmits it's unique ID. If the interrogator wants more information (like sensor data or memory data), then the tag and reader talk to each other in a lovely "dance of the bits" ;) This process is power hungry and if you toss into that the use of strong encryption, I think you'll still run into the need for a good solid battery. Power harvesting is a great concept though... and one with lots of promise. I've even heard of tiny motors/generators that could be attached to a person's vein pull power from the flow of blood. I myself submitted a power harvesting grant proposal to the DOE many years ago.
I really don't want to be a poo-pooer here and rain on the parade... I just think your best bet to making this a reality before you are 95 is to set out down the DIY path and see where it goes from there. Like Mikey said earlier, small, almost autonomous readers could be created as well using various wireless technologies (I like Xbee's mesh network technology) to interface the reader with your home inventory control system and the universal SPIME database. This would help negate the read range issue a bit.
If you wanted to do this for yourself though, and really wanted the range, then active might be your option... but it will cost you. Passive tags like the very basic EM4102 can be found for less than a dollar, but typical prices for active tags range from $15 to $45 a piece! The readers are also much more spendy.
Actually, before I wrap up this windy post... one thing came to mind. We talked about location finding before and this company just popped into my head. They've taken an active RFID setup and modified it using some fancy direction finding technology to create an active tag system that will lead you to your stuff. The product is called Loc8tor: http://www.loc8tor.com/ Maybe this could help you in your quest for long range active inventory tagging? Right now their product is hand held and limited to a certain number of tags, but they might entertain a fixed "home" system concept where all your things are tagged and the unit is "installed" in your home, perhaps with multiple antenna feeds for each area/room of your home? Chuck 'em over an email and see what they say... couldn't hurt.