Deep Green Crystals: May 2003 Archives

« April 2003 | Main | June 2003 »

May 28, 2003

More on why I don't believe in a carrier model for WiFi

True story from today.

I rode the west coast SARs express to SF today for a day of business meetings. During my travels through the Seattle airport, SF airport, the SF airport Marriott and a couple coffee shops, I was assaulted by Intel�s new <a href="http://www.intel.com/products/mobiletechnology/intelandwireless/index.htm?iid=Homepage+Focus_cmthotspots_03ww21e&">Centrino</a> �hotspot� ad campaign more than two dozen times (I counted). I actually passed through four different branded pay for service hotspots from Starbucks (<a href="http://www.t-mobile.com/hotspot/">Tmobile</a>) to SeaTac (<a href="http://www.wayport.com/">Wayport</a>) to SFO (TMobile) to SF Marriott (SANS). None of them worked. All of them showed up in XP as an available access point, and all of them connected with an �excellent� rating, but none of them gave me a valid IP address or connected me to a DNS server. So opening my browser just gave me the white screen of death and �cannot find file�. What you would expect these services to do is direct you to a sign-on screen where they try to extort a subscription out of you. I wouldn�t have paid, but I at least would have liked to get the offer.

I had a lunch meeting at the <a href="http://www.fogcitydiner.com/">Fog City Diner</a>. I arrived a little early with about 15 minutes to spare, so I decided to see if I could find an access point. Across the street from Fog City Diner were a bunch of office buildings with a very nice courtyard in the middle with comfy benches. Sitting down on one in the shade I powered up and immediately found three access points. One was WEP protected with the SSID changed to the company name. The others were �wireless� (a cisco AP) and �linksys� (linksys). Fully unprotected and open with excellent signal strength. So I hopped on, fired up the VPN, cranked up Outlook and did a send/receive of all the e-mails I had written on the airplane. Checked a couple stock quotes, answered a couple mails, then shut down and walked over to lunch. All for free. With just the software in Windows XP and the built-in hardware of my laptop.

My experience today is not atypical. Free access points are going to become even more wide spread. There was a recent article in Fortune about an <a href="http://www.surfandsip.com">upstart wireless ISP</a> who is going to the business adjacent to Starbucks and putting in access points. His service is cheaper. When you fire up your <a href="http://www.starbucks.com/retail/wireless.asp">WiFi at Starbucks</a>, you get two offers of access points. He gets to piggy back on all the marketing Starbucks is doing. Chains like Starbucks are trying to use WiFi subscriptions as additional product revenue opportunities at a location. But the <a href="http://zeitgeistcoffee.com">independents, especially in Seattle</a>, have caught on to the fact that free WiFi encourages people to stay longer and consume more traditional product (coffee and doughnuts). What happens when you want to have a coffee and surf the net with your laptop and it is cheaper to do so at an independent than at Starbucks? You guessed it, Starbucks actually looses a coffee customer. I wonder what Howard will be saying when the total per store revenue of WiFi enabled stores starts DROPPING against non-WiFi enabled stores. They could create the appetite that will be satisfied by someone else.

There are two fundamental reasons why I don�t believe there is a carrier model for WiFi. 
1. There are and will continue to be plenty of free options either by accident or on purpose by competitors who are using free WiFi as a competitive advantage. 
2. The commercial WiFi hot spot operator coverage map is too spotty, too technically complex to warrant payment today. By the time they figure it out and get down roaming and all the other stuff, the free coverage map will overlay the paid coverage map multiple times.

There will be a lot of WiFi activity, but I for one won�t be paying a new subscription fee to anyone for coverage.

Posted by Martin at 8:25 PM | Comments (0) | TrackBack

Beware of WiFi cards left on!

True story.

I just returned from a day down and back to SF on the Alaska SARs shuttle. My laptop is a Toshiba Protege with built-in WiFi. There is a little switch on the front of it with an "off" and "on" switch so you can turn the radio off to preserve power when not needing it. I found out today there are many other reasons to turn it off. Normally I leave it on all the time just to see if there are available WiFi hotspots to hop onto. Today on the way back from SF on the plane at 30,000 feet, I figured there wouldn't be any WiFi connectivity, but decided to check anyway (I am a geek). There was another laptop with a WiFi card beconing me to connect to "beauty1" as an ad-hoc network. What the heck I punched connect. It seconds, I was cruising another laptop somewhere in the 737. I checked out the outlook.pst file, copies of the contact database as a .csv file, lots of business .ppt presentations, some music files, lots of spreadsheets, and even a copy of the guy's resume! I could have copied anything I wanted. Or planted a virus. Gosh, where is a good virus when you need it? In the process of having all this fun with another guy's laptop I had the realization that maybe he was snooping me too! dooh!

So there are two things you can do to reduce your exposure.

1. Turn off your WiFi radio, or remove the card when you are not actively looking to connect to a WiFi network (safest). 
2. If you have to leave it on in a public place, go to the configuration settings and check "do not share files and printers with others". This will limit the resources another person can see even if they connect to your computer.

Be ware out there.

Posted by Martin at 8:24 PM | Comments (1) | TrackBack

Spam doubling every 42 days!

Ouch... <a title="" href="http://www.ajackson.org/spamstats.html">Spam doubling every 42 days</a>

Posted by Martin at 8:24 PM | Comments (0) | TrackBack

May 26, 2003

My Future in Review comments

Martin Tobias<a href="http://www.futureinreview.com"> FIRE 2003 </a>hot Spot comments, May 22, 2003

Mark, thanks for hosting this great event. I really enjoy the SNS newsletter and doing it live has been even better. While my SNS member number is not the lowest, I do believe I was one of the first contributors to ORCA and have been very impressed with the science that has come out of that effort. Thanks for having the event in San Diego as I have been out surfing every morning at 5:30am which for a guy from Seattle is a real treat.

A couple years ago I spoke at one of these prognostication events called �The Next 20 years� (www.next20years.com) and was the tech guy that night. My remarks took out Moores Law, what is happening to storage prices, energy consumption trends and broadband penetration out 20 years. It lead to basically a system of being able to consume any media you want anywhere you want any time you want. During the Q&A a very nervous employee of CBS TV got up and asked if she should change jobs. I said yes and I believe she did, so sometimes these things can lead to unintended consequences. No-one please change jobs based on these thoughts today!

This week I have been listening to the other speakers give very interesting macro views of the next 5 years. I had been trying to figure out a different take on things. One gentleman I met during a break said he was waiting for �what he didn�t know� and what would be �actionable�. He suggested most of the macro views of the next 5 years were probably �price into the market.� This morning watching the sun rise on the surf break it became clear what to do. Stick my neck out and make some very specific statements about the near future. I have broken them up into ten things I think WILL happen and seven things I HOPE happen. Here they are:

What WILL happen

1. PC economics will come to networking gear. The ASIC will fade away into only the highest performance equipment. Companies like CISCO will be challenged to maintain their pricing levels. This will be driven by Intel on the chip side (with their 2400 and 2800 family of chips) and a new group of start-ups on the software side. Customers of networking gear have been paying too much for too long and now is the time for a major architecture change. 
2. The first meaningful products will be delivered to consumers from nano-technology. Probably in the material science area. 
3. The interlopers that came to the tech business during the boom will all go back to selling toothpaste. The next generation of tech start-ups will be staffed with hard-core engineers who love to solve hard problems and are committed to the long-term of the industry. 
4. There will be VC bankruptcies. Available funds to start-ups will go back to early 90�s levels. 
5. Real broadband applications will be delivered and used in meaningful numbers. This year we passed 20M broadband households which I believe is a threshold for many applications to make economic sense. 
6. SPAM will be a thing of the past. A combination of really smart software and legislation (this morning Sen. McCain introduced new Federal legislation). 
7. Instant Messaging becomes a real business application. Many IM conversations will be with intelligent IM enabled web services rather than people. 
8. There will be no carrier model for WIFI. Larry at Cometa will kill me for saying that, but I really don�t believe a large number of people will pay an additional monthly fee for WIFI access when there will be so much of it free around. There will be a big business selling hardware for WIFI over the next 5 years. 
9. Round 2 of telco bankruptcies started in part by my friend Greg Maffei over at 360 Networks. Telco�s that don�t restructure their debt and equity will not be able to compete on price with those who have a dramatically lower fixed cost after restructuring. 
10. There will be no IPO market for small unprofitable tech companies. You will have to have 3-4 quarters of profitability and be on a $50M revenue run rate in order to go public.

What I HOPE happens:

1. Give back the money! As Mark has often pointed out, the wireless spectrum auctions were a sham and should be reversed. It is probably the largest single (non-tax) transfer of private wealth to the public sector in the history of mankind. Now there is no money to build out services. This money needs to go back to the private sector to continue the tech build-out. 
2. Someone should find a way to force an upgrade to IPV6. Most of the internet infrastructure would need to be replaced and this capital spending could kick-start a whole new round of tech growth. 
3. Biotech inventions will start coming off the Human Gnome databases at a fairly regular rate so that someone will be able to postulate a kind of �Moore�s Law� for genomics discovery. As scientist better about utilizing the data, discoveries will come faster. I believe there will be a positive feedback loop created. There will be some predictability to it so that we can say �in 5 years we will have a cure for XXX� and have it be based on some empirical results. 
4. Apply the fire hose of VC toward problems other than those of the F500. There will be money made on inventions in alternative energy, Allan�s 4 billion poor, and many other areas that have not yet received the benefit of VC investment. 
5. I want to figure out ways to decrease the footprint of technology in my life. All these gadgets are supposed to ADD to our quality of life. Too many times they devolve into support nightmares and I end up debugging something instead of playing with my kids. That has got to change. 
6. I want someone to invent an affordable power cache that will allow me to arbitrage day and night power rates from my power company. 
7. The next president should make energy independence the #1 national goal for the next 10 years. Just like Kennedy did with the space race. Americans will take up the challenge and probably beat the goal. This would solve so many American and world issues, I can�t even start the list.

Posted by Martin at 8:23 PM | Comments (0) | TrackBack

May 17, 2003

Why are NPUs hard to program?

I have been thinking alot lately about the software stack which will sit ontop of the NPU for development of packet applications. Here is a very clear analysis of some of the complexitities involved and why a layer of abstraction from the hardware architecture is needed. <a title="EE Times - Network processors need a new programming methodology" href="http://www.eetimes.com/story/OEG20020802S0040">EE Times - Network processors need a new programming methodology</a> Now the hard question is how to get a foothold on that system software and not get developer seat pricing for your solution?

Posted by Martin at 8:22 PM | Comments (0) | TrackBack

Got alot of time to track a Blog Soap opera?

Here is the latest hot topic Blog. Wired can't figure out if it is real or not. You decide... <a title=""...she's a flight risk."" href="http://shes.aflightrisk.org/">"...she's a flight risk."</a>

Posted by Martin at 8:21 PM | Comments (0) | TrackBack

Internet audio players

In the world of "play your internet on your stereo", Philips has taken that one step further and just integrated 802.11 and MP3 into a mini stereo system. Soon, every system will just have the connectivity built in so you won't need any of these intermediate gadget dongly things. <a title="Philips Consumer Electronics" href="http://www.consumer.philips.com/global/b2c/ce/catalog/subcategory.jhtml;jsessionid=HSKICDVFKFUVWCRQNE2RYVIKGBUCWHD0?groupId=AUDIO_GR&divId=0&catId=INTERNET_AUDIO_CA&subCatId=INTERNET_AUDIO_SYSTEMS_SU">Philips Consumer Electronics</a>

Posted by Martin at 8:20 PM | Comments (0) | TrackBack

May 15, 2003

WWF of video codecs!

All the years at <a href="http://www.loudeye.com">Loudeye Technologies</a>, we were endlessly debating the merits of one video codec over another. Each kept developing. There are now an even larger crop of technologies. Here is a very detailed and well documentd round-up. <a title="Doom9.net - The Definitive DVD Backup Resource" href="http://www.doom9.org/index.html?/codecs-103-1.htm">Doom9.net - The Definitive DVD Backup Resource</a> Have fun!

Posted by Martin at 8:20 PM | Comments (0) | TrackBack

Water rocket cars!

If you have a little time this weekend, build one of these....<a title="W A T E R - R O C K E T - P O W E R E D - C A R S" href="http://www.outsideconnection.com/rockets/rocketcar.htm">W A T E R - R O C K E T - P O W E R E D - C A R S</a>

Posted by Martin at 8:19 PM | Comments (2) | TrackBack

A quick look at security suppliers

While surfing for a master list of interesting security companies, I found this on the Aberdeen site. <a title="Aberdeen Group : Knowledge Centers : Security" href="http://www.aberdeen.com/ab_company/researchareas/security-suppliers.htm">Aberdeen Group : Knowledge Centers : Security</a> They have a bunch of one pagers on a number of strartups and public companies that effectively summarize the key facts about the company and the products. Good overview.

Posted by Martin at 8:18 PM | Comments (0) | TrackBack

May 14, 2003

The net preserves the sureal

One of the great things about the net is its ability to preserve for posterity stuff that many of us would rather forget. A great example of this is thes 1974 Weight watcher cards with all sorts of whacky recipies on them and some really bad 70's icons. Flip thrugh them all starting with Frankfurter Spectacular! <a title="It's Franktacular!" href="http://www.candyboots.com/wwcards/spectacular.html">It's Franktacular!</a>

Posted by Martin at 8:18 PM | Comments (0) | TrackBack

ISO 7 layer primer, even better one

Here is an even more comprehensive one from Novell. <a title="NOVELL: Novell's Networking Primer" href="http://www.novell.com/info/primer/prim05.html">NOVELL: Novell's Networking Primer</a>

Posted by Martin at 8:17 PM | Comments (2) | TrackBack

May 12, 2003

My pontifications on broadband home media

I was interviewed by a web radio station a couple months ago, they now have an archive page with the audio and some other interesting links. <a title="Digital Music's Future with Martin Tobias, founder of Loudeye.com; also, Tobias talks about venture capital in 2003 from Ignition Partners - Web Talk Radio On Demand - WebTalkGuys Radio" href="http://www.webtalkguys.com/p-tobias.shtml">Digital Music's Future with Martin Tobias, founder of Loudeye.com; also, Tobias talks about venture capital in 2003 from Ignition Partners - Web Talk Radio On Demand - WebTalkGuys Radio</a>

Posted by Martin at 8:15 PM | Comments (0) | TrackBack

May 8, 2003

Ever wanted to understand the ISO 7 layer network architecture?

Your prayers are answered...<a title="" href="http://www.pcsupportadvisor.com/nasample/t04124.pdf"> ISO network layer primer</a>

Posted by Martin at 8:14 PM | Comments (0) | TrackBack

May 6, 2003

Offesive and diagnostic SPAM tracker tool

Ever wanted to track down those who Spam you and reek havoc on them? Now you can...<a title="SamSpade.org" href="http://www.samspade.org/">SamSpade.org</a>

Posted by Martin at 8:13 PM | Comments (0) | TrackBack

May 5, 2003

Intel's new personal server!

E<a href="http://www.baard.com/archives/2003_5.thml#000071">rik and Mark Baard: May 2003 Archives</a>
INTEL'S WIRELESS POCKET SERVER

Personal Server: Has Intel Built the Handheld Killer?
by Mark Baard (Baard.com exclusive)
Intel is developing a wireless, pocket-sized personal server that may replace laptops and PDAs altogether.
The "personal server," which is being developed at Intel Research by ubiquitous computing wizard Roy Want, is the size of a deck of cards, half the weight of an iPaq, and has no i/o, no screen, and no peripherals. The device never leaves its user's pocket or handbag.
The personal server mounts on any PC that can recognize wireless devices: "Any computer becomes your computer," said Want.
MP3 enthusiasts can sidle up to any computer that recognizes wireless devices, and exchange files using the host computer's keyboard, mouse and screen.
Other potential applications for the personal server include sensor data retrieval for science and biomedical purposes, and presentation and other mobile computing applications for business users.
The personal server runs as a Web server on Intel's 400MHz XScale processor, and connects to PCs and wireless "information beacons" via Bluetooth and wi-fi. The device also takes advantage of the motes radio technology developed at UC Berkeley.
The personal server also makes efficient use of power: it

Posted by Martin at 7:58 PM | Comments (0) | TrackBack

May 4, 2003

NPU vs FPGA

I have been wondering how and when PC economics will come to IP packet switching gear. There are those who believe will never happen... <a href="http://www.arl.wustl.edu/~lockwood/eh2001_lockwood.pdf">Lockwood analysis of FPGA vs NPU</a>

Posted by Martin at 7:57 PM | Comments (0) | TrackBack

May 3, 2003

Linux Patch Management cont..

I have been looking for Linux patch management software, so my friend Troy said:
The Red Hat Network is also popular -- https://rhn.redhat.com/ Basic setup is a program called up2date that runs whenever the system is
live, fetches updated packages from Red Hat, and installs them.
$60/machine/yr adds access to a priority download server, bug details, and access to CD images for latest upgrades. $100/machine/yr for custom
patch profiles and Web administration.
<a href="https://rhn.redhat.com/">Red Hat Network - Sign In</a>

He also believes Altiris has done well. In the end people will pay for convenience and ease of use. We are Americans afterall. Open source will only go so far and will tend to be a science project. That maps up well to many current Linux users, but will diverge as it penetrates the enterprise more deeply.

Posted by Martin at 7:56 PM | Comments (0) | TrackBack

Just the toy I have been looking for

I have alot of books and DVDs at home. Lots of my friends borrow them. I never know where they are. Finally, a solution! <a href="http://www.knockknock.biz/html/kits.html#">Knock Knock Kits: The Wishing Kit & Personal Library Kit</a>