Reach thousands of intelligent, expert readers interested in the latest in Wi-Fi News with your message of sponsorship.

The above could be a paid, sponsored link. Contact me for more information.

Subscribe to essays from this site via email. Email to subscribe, or sign up via your Yahoo account.

Ken Berger is a regular and well-informed correspondent on issues Wi-Fi-related. He's a consultant (LogX Technologies) and an active member of BAWUG (Bay Area Wireless Users Group). He files this report from this month's MicroVentures conference, which is relevant to those of us following Wi-Fi and unlicensed spectrum issues:

Once again, Technologic Partners (affiliated with the authoritative VentureWire newsletters) hosted a cozy Private Equity Conference where investors and entrepreneurs could mingle and gain insight into a technical field of interest. In May, I wrote coverage which appeared on this site for the WirelessVentures conference. This past Monday and Tuesday, the title was MicroVentures, and it focused mainly on the semiconductor industry.

Many are saying that the economy now feels like it's hit bottom. When recoveries come, innovations start at the chip level (semiconductor). Dick Shaffer, the event's producer, and chief of VentureWire, is definitely a guy who follows the money; he hasn't done a microchip conference in ten years, and by picking now to host another, it would seem a signal that he sees now as an inflection point time.

(My logic here is somewhat parallel to Glenn's (this site's humble host) thoughts about the 802.11 Planet conference, also going on this week: Alan Meckler has had an incredible track record of going where the action is (and leaving before the action leaves), and he seems to be betting on that technology by last year starting a legacy of conferences on the 802.11 theme.)

The general mood is much as I reported back in May: there's not a lot of hype or wild partying, and investors and entrepreneurs alike seem very cautious and defensive. Only now, folks are much more experienced at feeling humbled.

Much of the conference revisited Moore's Law. Are we approaching a point beyond which continued acceleration of progress at the chip level will be difficult? Or the reverse: will new techniques such as MEMS (Micro Electro Mechanical Systems) and nano-techniques such as molecular electronics warp us to a new frontier and accelerate this law? Whether or not people will care as much in the future (is what we have 'good enough' as Sun's Scott McNealy recently said) was also the subject of a few of the investor panels.

Panel on startup opportunities in systems on silicon: Seems to be a widely-held belief these days that for a venture-able company to survive, how much money is needed and how quickly it can arrive at break-even will determine its success or even possibility of being born. A figure was posited to be $45 million maximum in total funding to break even: if your business plan requires more than this amount to start turning a profit, the message here is forget about even starting. I’ve heard other VC’s recently state even lower figures.

Another theme was that of the first mover advantage, yielding some differing viewpoints: Andy Bechtolsheim of Cisco talked about the "myth of the first mover advantage", while Andy Rappaport of August Capital in a previous panel stated his belief that "first mover advantage is huge". Opportunities for startups were pretty much what you'd expect: there's still huge opportunity for innovation in general, and storage and security are rising opportunities.

The next panel covered WLAN and winning in the market for over-the-air digital communications. You must be really innovative, and consistently so, to stay in semiconductors (something I say all the time when meeting with chip companies). Having a startup whose main value prop is making the chip merely faster, cheaper, better may have been fine in days of past when there was room (or at least money to be made) for many players. Today, you had better expect that the incumbents are and will be doing that all the time, and you need to offer something really differentiated.

The good news here is that there is tremendous opportunity for real innovation. For example, currently, a typical 802.11b-enabled PDA will last only a few hours without recharging-- this is unacceptable if people actually plan to use the things frequently, wherever they are.

Another issue discussed was the potential threat of big players deciding to put 802.11 technologies on their motherboards, and Intel’s plans relating to this with its upcoming Banias/Calexico line. Rich Redelfs of Atheros and Greg Raleigh of Airgo both countered that this could be a problem-- a motherboard manufacturer would have serious problems keeping up with all the innovation that is surely coming in WLAN, and this part can literally choke a Pentium with what's going to be happening. Intel also has a long history of getting into initiatives to move into ideas like this one, then getting out not long after.

A particularly relevant example is a brief time where Intel got into the analog modem business and subsequently gave up. Of course, the consequences of burning the battle field and wrecking the pricing structures for the companies left in the field still remain. Most likely will be to have the MAC on the motherboard, but the radio and other parts elsewhere.

While the U.S. got an early lead in Wi-Fi, the Asian markets are growing fast. Atheros currently ships about 70% of their products to Asia, although a lot of that comes back here as components of gear built there. There is lots of interest in China, and in Japan most business people carry laptops.

This conference also brought back the fervid company presentation sessions, held in 10 separate rooms in close reach, each with a company's CEO madly trying to compel his company's overview and including quick (and occasionally brutal) Q&A in the total 20 minutes allotted, strictly enforced.

The second day lunchtime keynote address was by Paul Saffo, Director of the groovy Institute for the Future. Every ten years we get the Next Big Thing-- in the 80's it was chips and PC's, in the 90's it was the internet and also cheap lasers. This decade? Sensors! Per Paul, sensory devices and networks will take off, especially sensors used with biotech models, accelerated by critical uses such as UAV's (Unmanned Aerial Vehicles) in Afghanistan. This is prescient, as quite a few of the companies presenting here design chips to enable sensors. RFID, such as for tags in shelf products, will be big, but the real opportunity will be in doing things not possible before…

The next session was innovative for a first try at a familiar forum: a mock pitch session where 4 individual analysts 'pitched' their ideas to a panel of 4 VC's. The main substantive problem here for the VC's to do their critiquing was that the pitch really just advocated a sector, not a company, and VC's will always tell you (at least post-bubble) that they only fund companies, not sectors. Nonetheless, the session revealed some current thinking among the VC community. The four Big Ideas were MEMS, HDTV with an associated chip to coordinate and integrate all home functions, VLSI, and Biometric ID.

The final session picked a top ten list of winners presenting at the show. 802.11 is STILL being proclaimed a big winner because of the incredible growth rate that continues. The question remains whether any particular company can build a critical mass. The panel agreed that it is too late to start an 802.11 chip company (the same was also already declared at WirelessVentures back in May), and that Atheros has a 'meaningful fraction' of the market, and did on CMOS what many said couldn't be done.

Canesta could have won in the 'coolest stuff for a James Bond film' category, and their product line is fairly diverse for such a small company. Their electronic perception technology perceives things in 3D, lending to recognizing your face when you approach your door so you don't need a key, sensing if a truck driver's head slumps if he falls asleep (what happens if it's just a sneeze??), and just as cool, a 'virtual projection keyboard' that projects an image of a keyboard from your PDA onto the desk or table it's on, then watches your hands and knows what keys you typed.

Cavium was chosen as being early to the much-needed security chip market. Imaging sensing and processing is a real need, although much of it is currently in the consumer market and not much investment is happening there, and long-term it will be a commodity. Nucor is making headway in this field, especially in Japan, and the company's management is mostly Japanese.

ChipWrights focuses on low power chips for handheld devices and could be the next Nvidia in graphics controllers. Though they weren’t named to the top ten, I was very impressed with MicroChips's product: a MEMS-based chip that gets preloaded with prescription drugs, then planted in a patient, and the doctor can communicate wirelessly with the chip to fine tune the settings rather than remove it. Also not named, but relevant to Wi-Fi enthusiasts, SyChip will be releasing soon an SD drive mounted 802.11b card for your PDA.

Military and 5 GHz

John Markoff broke the story yesterday in the New York Times that the US military had concerns about the unlicensed use of 5 GHz spectrum because of conflicting radar uses. See yesterday's rundown.

Today, we see an uninformed Slashdot discussion with the word 802.11b in the headline -- guys, 802.11a covers 5 GHz, not 802.11b -- which I won't even link to.

But InfoWorld delivers the goods in this article. The two writers clearly explain why the military is concerned, how current and future technology might and do address the overlap, and how 2.4 GHz/802.11b isn't in the picture.

Other News

Kentucky has largest permanent Wi-Fi hot spot installation?: Folks, after Nigel Ballard's announcement yesterday that Tacoma's convention center and adjacent Sheraton were the largest Northwest hospitality hot spot installation, we get a topper. The Kentucky Fair and Exposition Center and the Kentucky International Convention Center in Louisville represent 1,000,000 square feet of fully Wi-Fi-able area. Doug Keeney, the CEO of US Wireless Online, sent in the information in friendly oneupsmanship. Is Louisville on top? Some airports might have more physical area, but are they fully covered? Bragging rights are up for grabs -- but documentation is required!