Tuesday, July 19, 2005

No More Nail-Biting Experiences!!!

This is one interesting information one of my colleagues – Hiri – sent me. Apparently Japanese have achieved data storage on human finger nail. Not small amount!! . We are talking reasonable volume here – about 5 mega bits. If you use the technique developed by Yoshio Hayasaki and colleagues at Tokushima University, you can store dat for up to six months on your nail (by which time your body would have anyway more or less replaced your finger nails).

I have some odd thoughts on the whole idea. Does the 5 megabits figure correspond to only one nail? Can my wife, who grows her nail really long (apparently long nails add to the beauty to feminine gender in this parts of the world!) store more data on her nails than me? Would people who have the habit of biting nails be disadvantaged in this scenario (No more nail biting experiences Tim!)? How are you going to plug in your data disk (or more precisely your nail) into the computer for data read and write?

Well, lucky me, some of the above questions already have clear answers available. To understand this data storage technique, you need to look at the technology involved. The team uses a femtosecond (10-15 seconds) laser system to write the data into the nail. The nail's fluorescence increases at the point irradiated by the femtosecond pulses. You just have to use a fluorescence microscope to read the data. Sounds fairly simple?!! Cross-talk between data stored at different depths is prevented by appropriately setting distance between the planes.
The bad news is that so far the experiments were limited to small pieces of (detached) nail. We need to wait and watch some more time before this technology enables us to use our nails (while being attached to our very own fingers) in place of the USB drive!!

Monday, July 11, 2005

Curtains for PSTN?..

I have been using a small VOIP application called Skype (http://www.skype.com) for the last few months to talk to my niece in USA. It is another comp-to-comp phone application to begin with. But it is not just another one by any standards. It gives me better clarity than the PSTN lines!!! I am serious!!! It is a lot better than ‘just picking the phone and calling my niece’. On top of it, it is free. Mind you, it has provisions for calling a US telephone number directly (that means comp-to-phone) as well. But that costs you a little money.

What is interesting about Skype is that it is a Peer-to-Peer (P2P) application. Further, it uses Session Initiation Protocol (SIP). (You can read about SIP at http://www.sipforum.org/ ). Skype supports voice mail, call in, call out and conferencing features. It is probably the product with the fastest adoption rate in last year.

Recently, I came across VOIP buster (http://www.voipbuster.com/en/index.html), a similar application which provides virtually free calls to telephone numbers in US and selected European Countries. Here too the call quality is as good if not better than PSTN lines.

There is another application on the same lines which caught my attention recently. This one- Ineen (http://www.ineen.com/) - not only supports voice conversations but is actually a video phone. It provides excellent voice quality and fairly good video clarity. The only limitation is, it is com-to-comp.

vSkype (http://www.vskype.com) is a free add-on to Skype which adds pretty much the same functionality to Skype. I am yet to test this one.

The way VOIP is going (though I have serious doubts about the revenue models associated with many of the applications listed above), I can hear the death knell of telephony and PSTN loud and clear. Do you also feel the same?

Thursday, July 07, 2005

Where is RFID going?

In December 2004, Larstan Business Reports conducted a survey among 669 supply chain IT executives to understand the pattern of adoption of RFID technology in various segments. In the retail industry vertical, as high as 77% percent of the executives judged an advanced RFID infrastructure as either ‘Very Important’ or ‘Important’. Infact 88% of them felt that their business would benefit if suppliers and business partners implemented RFID. Now contrast this with the state of RFID implementation in Retail today. The survey found that only 35% claims to have some form of the technology implemented in their organization. There seems to be a clear gap between the aspiration and realization. Despite the promise and the value proposition, there seems to have been precluding factors in the diffusion of this technology.

To understand the reasons behind this, one has to look at its inception and growth. Radio Frequency based identification cannot be considered as new by any standards. Right from World War II, this technology had been in use. But it was only in the last few years that it got the prominence it deserved, especially in business circles. RFID’s appearance on the business front could not have been more dramatic and forceful. Initially, media has, based on predictions by technical analysts, gone ahead to position RFID technology as the harbinger of next technical revolution through out supply chain. By the middle of 2003, the hype had given way to pragmatism. In 2004 Gartner has predicted RFID hitting the ‘trough of disillusionment’ of their ‘Hype Cycle’ by 2006. The industry started discussing ‘to do’s with respect to this technology to make it deliver the perceived value.

Some of the major shortcomings identified with respect to this technology which resulted in low adoption rates were:

Ø Privacy issues – imagine walking around with RFID tags all around you (on your dress, bags, wallet, loyalty card etc.) broadcasting who you are, what your tastes are and where you normally shop etc., to the whole world!!
Ø Radio waves do not travel well through metal and liquids
Ø Unit cost of tags
Ø Reading accuracy issues

These point to two important aspects. On the one hand, it emphasize that, with the current state of technology, RFID is not suitable for every business scenario where it can be applied. As a corollary the technology can work wonders and exceed expected ROI if it is applied diligently in suitable environments. This demands enterprises to enlist organizations with sound business background to work on identifying and implementing RFID solutions for them. In other words RFID is no longer a technical solution to a business problem but a business solution using one of the emerging technologies.

In 2004, organizations with expertise in industry practices, like ADEA solutions, entered the RFID arena providing sound business solutions to various clients addressing many of the concerns raised earlier. This was a welcome change from the initial domination by pure technology players in this segment. ADEA for example, concentrates purely on the Retail and Consumer Packaged Goods segment with well designed frameworks and tools to solve business problems in that sector.

RFID technology’s impending spread is evident from the growth it has shown in the first quarter of 2005. By March 1, Walmart is stated to have RFID implemented in 104 Wal-Mart stores, 36 Sam's Club stores and three distribution centers. The enormity of this is evident when we consider that it involved installation of 14,000 pieces of hardware, 230 miles of cable and is live with more than 100 suppliers. Walmart is now looking forward to implementation at 600 stores and 12 distribution centers by year's end.

From a retailing perspective, Walmart’s success has convinced the industry that RFID could help them track goods throughout the supply chain. It could ultimately lead to fulfilling every retailer’s dream: ‘right products in the right stores at the right time’. It'll also help locate specific products anywhere in the supply chain, rendering recalls easier to manage. Actually this is just the proverbial tip of the iceberg. Appropriate adoption of this technology could not only contribute immensely to the supply side but can also revolutionize the demand side leading to improved customer satisfaction. A customer loyalty solution ADEA developed for a large retailer in Europe stands testimony to this.

Next generation UHF Gen 2 tags, expected to be available by the end of the year could further hasten the industry adoption of RFID.Specification for UHF Gen 2 tags is already finalized by EPCglobal Inc., the nonprofit organization spearheading RFID adoption, in December 2004 enabling this. We are also seeing newer and newer uses for the technology coming to the forefront. Finally RFID has grown beyond the shadows of barcode tags. Considering the increased acceptance of the technology after Walmart success, we are confident that the gap between aspiration and actual implementation seen in Larstan Report would reduce to single digit levels by the end of 2006. The key to this success would be involvement of technology vendors with domain expertise in driving the RFID juggernaut.

Tuesday, July 05, 2005

Amazing – The gadget, the Technology, the Business Model and the development Process!!!

There is a site I had visited almost a year back. It looked interesting that time. It had a small application which would try to identify any object you imagine by asking 20 questions. It was an application with a learning engine. I tried it…tested with some uncommon objects and was fascinated by the accuracy by which it deduced them. As I said, this was quite some time back.

Recently while discussing with Rakesh Ravuri (You can visit his BLOG at http://www.eternalillusions.com/blog ) some new ideas for applications using AI and learning engines and I all of a sudden remembered that old little app. I did some googling and found that the application has gone places already.

Before we get into what its new avatar is, let us look up a little history. Apparently the brain behind the app, Robin Burgener programmed a simple neural net on a DOS machine way back in 1988. All that he taught the engine was 20 questions about a cat. He then passed the program around to friends on a floppy and had them challenge the app with their yes/no answers to the object they had in mind. The application (unlike we humans) kept learning from mistakes (and correct guesses as well.) So the more people tested it, the more it learnt. In 1995 Burgener hosted the application on the web for public to play with. As Kevin Kelly correctly pointed out, Burgener’s genius was to turn the hard tedious work of training a neural net into a fun game for humans. (Imagine the millions of Dollars it would have cost him otherwise!)
Last year, after a million rounds of 20 questions online, Burgener compressed the 20Q code to run on a chip, and with a select 2,000 of the most popular 10,000 objects it then knew about. In other words, he created a gadget (a toy which looks like a small sphere) which is a handheld version of his Twenty Questions web site.
So the new avatar is a $14 toy named ‘Radica 20Q’!!. You can buy it from Amazon. Before that, you might want to try out ’20 Questions’ application at http://www.20q.net/ and get fascinated while teaching the engine probably about new objects or new attributes of objects.