|Lounging around in the Telecosm|
WFI is a superbly positioned company providing turnkey greenfield to maintenance services for leading edge wireless networks. It specializes in the baffling complexities of 2-2.5-3G deployments. The Telcos do not want this role and backlogs are huge. On the team is Scott Fox, who allegedly fought at Bell South for CDMA (Bell South in Latin America is still all CDMA, though new Bell South wireless owner SBC is a TDMA bigot and prevented conversion of Bell South US to CDMA after the complex merger). Although the key people at WFI tacitly acknowledge the superiority of CDMA, they are officially "technology agnostic" and ATT is one of their customers. WFI is a paradigm company, mostly, and possibly ascendant in 3G CDMA, but because of its TDMA and GSM operations I have not yet decided to put it on the list. (I have been citing companies on the board that I do not put on the list, such as Sonic Innovations of Salt Lake City, a newly public hearing aid company based on Carver Mead's technology). As a stock (with the usual caveats relatively to timing and price) WFII is also a superb prospect. 7/10/00
We have ample words of encouragement for AVNX holders and we will have many more over the years. After what we have already written, though, what more could you want?
Although Raman is likely to be an important technology, high powered lasers will have many other uses in all optical networks. Meanwhile Raman faces a significant challenge from the simpler and cheaper Avanex Power Shaper line which extends the distance of unregenerate optical transmissions by improving the sharpness of the original signal through non-linear filters. 7/10/00
The question is not the royalties but the chipsets. Qualcomm's WCDMA chipsets are about a year behind their CDMA 2000 chipsets. However, noone has perfected WCDMA chipsets yet, though NTT is already trying to buy support for its WCDMA standard by offering free gear to the Koreans. 7/10/00
This is reasonably interesting stuff, but not part of the Network Attached Storage paradigm because the key factor in storewidth is not reads but writes. Microtest does read only storage--ROMs of various kinds, CDs and DVDs. The filer based NAS companies such as NetApp and Procom do manage the far greater complexities of read-write systems with different protocols. 7/10/00
No exciting news from Bookham as far as I know.
Xilinx is a PLD company (field programmable to be specific); Tredennick thinks FPLDs will rule the world.
I don't think anyone can predict semiconductor cycles since they track the entire economy, which is goaded or gulled by government mismanagement, which follows the unpredicable spasms of politics. In any case, the semiconductor companies on my list are focused on leading edge designs and processes and thus march to a different drum anyway.
Re:WFII's ability to add CDMA to thier repetoire
They already do CDMA deployments, as they must on the 3G front. The point is that they are "technology agnostic" and have large business from ATT. 7/10/00
David Payne is an important figure in the history of the industry, but he has been reading my speeches too literally (the rice on the chessboard). I have not been able to sustain the UUNet estimates of doubling traffic every 100 days (it would mean 10 times per year, 1000 times in three years and one million times in six years). However, the expansion of bandwidth will be larger than he estimates. Companies that learn to waste bandwidth and manage lambda circuits to deliver needed bandwidth at the right points will tend to prevail. The leading bandwidth waster of the moment is Metromedia Fiber (MFNX). Key to getting it to the right places are Global Crossing and GlobalStar. Some of you will be familiar with these companies, which do not decline in value because the market these days is showing a wonderful instinct for the capillaries. 7/13/00
Recently saw an offer of $699 for the phone and 95 cents per minute. 7/13/00
Let me bring my limited market wisdom informally to bear on the Qualcomm issue. This is clearly a low point for the company. The media are focusing on a series of setbacks: In China where CDMA has become a politcal token, in Korea where NTT is trying to buy share for WCDMA, in the US where RF Microwave has announced a WCDMA peripheral chipset (without the critical CDMA modem), in the North American market with Voicestream's GSM technology perhaps being financed and extended by DeutcheTelecom. Add the apparent possibility that the national PCS spectrum footprint assembled by the now bankrupt NextWave will slip away to AWE or a GSM carrier and Qualcomm's position seems precarious.
If you want to reshape your portfolio with a lower risk reward profile, you might want to lighten up on Qualcomm even at the cost of selling into bad news. Normally, however, bad news for a fundamentally strong company creates a selling trap and a buying opportunity.
CDMA is the only available technology that uses all the spectrum band all the time and thus can accommodate the bursty data of internet links and offer bandwidth on demand. It is inherently the lowest power technology. It is the only technology that can use incidental pauses and multipath to enhance the signal, and enable elastic cells that adapt to changing traffic patterns (expanding cells along highways at rush hour and devoting suburban cells to data after sundown, for example).
Because of this fundamental superiority, Ericsson, the world's leading proponent of GSM at the time, capitulated to Qualcomm and actually purchased its infrastructure division in order to the enter the CDMA market. Because of this fundamental superiority, the GSM group in Europe resolved on CDMA and only CDMA for third generation wireless in the face of passionate political opposition.
In evaluating all the mostly political setbacks, keep in mind that Qualcomm already, alone in the world, has tested chipsets and systems that can supply megabit data and double voice capacity for the next era. In the end, wireless is a scarce bandwidth arena and an undershoot technology. With combinations of HDR data and 1x voice, Qualcomm is already two years ahead of WCDMA which does not promise remotely the bandwidth of HDR for several years. I believe that the carriers who adopt this technology are going to gain market share and the nearly 70 CDMA customers today are going to experience superior service and Internet access. Surprises are coming in wireless local loop, in Latin America, and in satellite.
I can assure you that I am not selling any of my famous Qualcomm shares. 7/15/00
The key is that CDMA2000 exists. WCDMA does not. It is still undefined and vaporous. Breathing it produces hallucinations and is dangerous to your financial health.
But I learned that the Europeans think they are three years ahead in wireless and that Globalstar works splendidly from a boat in the Baltic. 7/15/00
Re: MRVC It just made a very intriguing acquisition but it slips my mind and the moment what it was. It commands leading edge electronic switching fabrics in an era of optical switching fabrics millions of times more capacious. Refresh my memory further. There are alot of companies out there. 7/15/00
Re: sunw, orcl, amat, csco, msft My guess is that the one with the most reliable upside is Applied Materials, which supplies critical path equipment for every semiconductor wafer fabrication plant in the world.
As the developer of single mode fiber, Corning has already changed the world. It was the single most important company in the evolution of fiber optics. Now it is returning to its key role in this industry. But when I list a noggin nine, JDSU and Avanex tend to take precedence as pure plays in this realm, with JDSU as the components titan and Avanex as the spearhead of a new optical paradigm. But Corning runs alot closer behind than its PE would imply. 7/16/00
With all the usual cautions on my noggin notions--these are merely my own personal ideas as a Forum extemporizer, not part of any canonical GTR endorsement--Coke and Berkshire are over the hill, Microsoft is defenestrable, the call for Phillip Morris is dwindling, and Kulicke is a cyclical play unless something extraordinary has been concocted down in Horsham behind my back. I've been waiting for TJ to contrive a breakthrough at Cypress for 20 years. I have nothing of value to say about most of the rest--Human Genome Sciences awaits our biotech letter, which is still in abeyance. I like BEA as a Java middleware play though I have no idea about its current valuation. Everyone should watch Avanex closely and not let it get away. It is our favorite. GBLX is still a winner as are GlobalStar, Corning, WFII (a next generation wireless play), and Qualcomm, which you have. TI, Xilinx, Nortel, Terayon, Atmel, and AMCC all should be considered as valuations dictate. Failure to list any other company does not imply disdain--just a "senior moment." 7/16/00
Patents on something so obvious as "push" do not strike me as interesting (what is broadcast or multicast or even Ethernet?). One of my key rules is to ignore companies that stress their patents; patents allow you to stop others from innovating but do not empower you to produce the innovation yourself competitively. Backweb (and Marimba, which at least had a Java story) have been after me for several years without making much of a dent, but on the other hand I have been busy.
Since I am a devout member of the board of Wave, I cannot talk about it except in the most general terms. I enjoy high risk high reward investments, and like Wave, Globalstar continues in that mode, with powerful fundamentals and a critical mass of short sellers to help orbit more satellites and other valuables.
The problem of targeting these money supply numbers as an index of potential inflation is that in them deflation mimics inflation. If people expect inflation, they collectively accelerate their transactions and thus require more money creation to sustain the same price level. If people expect deflation--or non inflationary growth--they increase their demand for money and hold it longer, similarly requiring more money creation.
At a time of zero inflation in Japan during the 1970s and 1980s, the Japanese bank created M2 at a pace more than four times the current US pace. At the beginning of the 1980s boom and collapse of inflation, most monetarists identified the large expansion of money as a portent of inflation and urged tight policy.
I am a supply sider for all the real factors in the economy, but a demand sider in regard to money. What matters is the attractiveness of money, not its quantity. The attractiveness is manifested chiefly by the value of the dollar versus gold and other commodities and against other currencies. When the dollar is increasing in value by these measures it is absurd to talk of the threat inflation regardless of the behavior of the monetary indices.
That Wall Street Journal story of laser smuggling dramatizes not only the possible pitfalls in Bookham's strategy but also the flaws in the strategy of Sycamore, Monterey, Ciena, and other vendors of optoelectronic switches. Sycamore's Eric Swanson was desperate for vertical cavity surface emitting lasers that operate at 2.4 gigabits per second (OC-48). Not only is this slow for many applications (OC-192, for example) but optoelectronic switches require a laser for each lambda or wavelength bitstream. When Avanex introduces its 2000 lambda powermux, there will be 1.7 million lambdas in a single 864 strand cable, meaning millions of lasers, which will give Swanson more trouble at customs. Meanwhile all optical switches will be able to handle millions of lambdas with passive devices--mostly mirrors of various kinds that care not what bitrate impinges upon them. The transmit lasers will be distributed on the edge and it will be more important to make them tunable than to integrate many of them on a single device. 7/21/00
In an amazing coup, Globalstar has accomplished the hard part, building a unique technology based on CDMA and capturing the troposphere for less than $5 billion. Afflicted with such non linearities and dispersion factors as marketing and sales, ergonomics and finance, the earth will take longer. But with the possible exception of financial risk, which tracks the overall economy and puts the stock in reach, everything in the G* system constantly gets better. Globalstar is scalable without any changes in its satellites. As Moore's Law proceeds, all the earthbound components of the system double in cost effectiveness every 18 months. When the shorts see the light, the shares may well join the satellites in orbit. For the more risk averse, Loral owns 49 percent. 7/22/00
Toshiba is a great manufacturing company and a substantial customer of Broadcom. 8/8/00
Except that Broadcom has no fabs! At the heart of Intel are the best wafer fabs in the industry. Broadcom has been on the list since the IPO but it is not an Intel. 8/10/00
Re: Potential tax liability of 2.85/share on XLA
We are pursuing these matters aggressively with Mirror Image and will report what we can find out. I am personally appalled by the possible tax perils and by my role in inducing many customers to incur them.
Mirror Image, though, is a year ahead of schedule in launching its Content Access Points (I believe 20 are near completion), acquiring customers and perfecting the technology. 8/10/00
This is the paradigm: All computing and packet switching will be electronic (the microcosm); all communications and lambda circuit switching will be photonic (the telecosm). 8/10/00
Applying to individual electrons in their bands, spin is the last frontier of quantum modulation. It comes in many phases and is afflicted with Heisenberg uncertainty. Silk Road also purports to use spin. Who knows whether anyone can make it work. 8/10/00
XLA sources say there are many possible solutions to the problem and they will adopt the one that is most advantageous. Shareholders in any case will be made whole with regard to any residual tax liabilities. 8/10/00
A biotech newsletter. Bet it will be $295/yr when it comes out.
Unfortunatly, seeing some negativity in the lounge. I guess people scraped up $295 expecting to get a QCOM return, but get a red lump of coal instead. GTR is much better read if it not touted as an investment newsletter, but those hyping keep mentioning the QCOM pick and 2600% rise. Loyalists keep mentioning QCOM was on the list for years and only went up in 1999.
Maybe the solution is a Telecosm Mutual Fund so the GDTRF/XLA picks are balanced with the GLW, JDSU. Apparently many are not diversified so a collapse in one stock is disaster. I see the GTR as one person's view into the future. This is typical anger.
"This Gilder report we are all subscribing to is a waste of money. By the time we can view the report and put our buy order in the stock he is praising, the particular stock will have already risen 50% so you won't be making any money on it." 8/11/00
"GEORGIE BOY,,,,,,DO U KNOW THESE GREAT PICKS OF YOURS COST ME AND OTHER SUBSCRIBERS MILLIONS OF DOLLARS BECAUSE OF U???NOVEL,,,,,GLOBALSTAR,,,,,ALIGENT,,,,,,LORAL,,,,,,CONEXANT,,,,,GLOBAL CROSSINGS,,,,,XLA,,,,,,,,THANKS ALOT." 8/11/00
I would like to see the page 8 list scrapped myself, but it wouldn't sell as well, would it?
I notice that when in the lounge, clicking the News & Update bar requires me to re-login.
There is an investor forum in the lounge now.
Non GG Quote that I am studying.
"Exodus invested in XLA was defensive, because 1) Inktomi is pricy and 2) Akamai's outbound distribution focus hurts growth in hosting by reducing the need for larger more expensive sites." This sector is tough to get a future read on. Present read on price movement: AKAM has dropped 50% in a month. INKT dancing with May lows. XLA is 90% off it high. EXDS seems to move like the Nasdaq.