Project MAC, where we met S at MIT A the Software Arts building where we worked together T and the attic N where VisiCalc was written
Other writings on our personal sites:

RSS Feeds:



Comments from Frankston, Reed, and Friends

Wednesday, April 03, 2002

BobF at 12:17 PM [url]:

Dan Gillmor on ICANN

Dan Gillmor's column (3-Apr-2002) is about ICANN and the related issues. It provides additional perspective as well as referencing the Missing Internet essay.

Sunday, March 31, 2002

BobF at 11:26 PM [url]:

Marketplaces, Connectivity and Other Stuff

In second grade I was put into a speech class for stuttering. But I didn't stutter - I was just trying to say too many things at once. It makes it hard to know what to write about since so much has happened since I wrote the "Missing Internet Essay". The problem is compounded in that I tend to be a "fine fettle" writer in that I need to be in the "write" mood.

In the past two weeks I've actually done a lot of writing but it has tended to be in very rapid turn-around email discussions among people I know so I can be frank and build on previous points. But even then it is easy to be misunderstood. When I write for a larger audience I tend to try to do much better writing and try to write in a way that reaches a much wider audience. And that is far more difficult.

Excuses provided, I'll start by catching up on some shorter topics and observations without going into the topics in full detail. One complex issue is the battle between those who want to preserve existing businesses and those who understand the concept of a marketplace. We see this playing out in telecommunications which is still viewed as a high value service that is threatened by connectivity. The same attitude pervades the attempts to keep the current model of record and movie distributions intact.

Such attempts to keep stale models alive are nothing new. Sheet music was once the focus of similar concerns. The Doonesbury comic strip (that doesn't seem like the appropriate description for the editorial content) has focused on these issues recently. I'm particular impressed by the March 17, 2002, strip which said "true conservatives are pro-market, not pro-corporation". It also highlighted the assumption that content can be shared. I don't defend theft but the effort to control how people use "content" limits its value and only frustrates the search for better business models. It's strange to watch many TV commercials that assume you burn your own CDs while the inter-commercial space is filled with attempts to prevent what is now considered normal activities.

This issue of marketplace models is pervasive. The "Missing Internet" is about this same struggle to allow innovation by the users at the edges rather than giving all the power to those who want to be in the middle.

Last week we took a trip but since I was flying and wouldn't do the return driving we used my wife's car. I have gotten used to the Magellan car navigation system in my car and didn't want to travel without it. Since I hadn't been using my iPAQ for a while (I use my Kyocera phone as my PDA) I took the opportunity to upgrade it and to get the PocketCoPilot. This is a GPS unit that fits around the iPAQ. While I can hook a separate GPS unit to the iPAQ it's much more convenient and mobile to have a single unit. It was relatively simple to get it to work once I noticed that it didn't try to find the location when the car was stationery.

I was struck by the different in the two approaches to the car narvigation marketplace. While the Magellan unit works, as with the GigaSet phone system I discussed previously, it is a hardware company not a software company. There is a serial port on the unit but the only way to upgrade it is by mailing in the unit whereas I can download CoPilot updates myself. Magellan does have their own standalone units but they are closed.

This is still a marketplace that is working itself out. I look forward to having connectivity available. I could then have a GPS source that could be positioned elsewhere in the car and use the data in various applications.

The idea of sharing the GPS information is intriguing for airplanes too. Instead of relying on a single shared screen showing the position, I could use my laptop with detailed map data to make the trip much more enhanced. When you look out the window, you would know what you are looking at and "see" much more.

But airplanes are very problematic spaces totally controlled by their owners. Airports have the same problem, with 802.11 seen as a threat to revenues from a captive audience. The adoption of new ideas is very slow and even when there is a good idea like Boeing's Connexion service, it takes a long time to deploy it.

While flying to Phoenix, I tried to use the GTE Airfone to send email but couldn't make a connection after a number of tries. When I spoke to an operator, I was told to set my modem speed to 9600 and turn off dial-tone detection but that didn't solve the problem. This problem is not unique to airplanes. There are similar problems when using phones in airport clubs. The problem is that we have a huge pile of kludges that can be made to work individually. If I have my own phone and can simply dial any number I usually have no trouble.

Airplane phones barely handle voice and the phones in the clubs have no protocol for letting me pay for the call. I solve the latter by using 800 numbers.

But there is something very strange about these phones. They have jacks that are purposely designed so I can connect a modem and then they seemingly space out and forget it. So I am left connecting my modem to a codec that turns it into a digital signal as if I were making a voice call. Imagine if we implemented e-mail this way. You'd print out your message and fax it to the destination and then have your computer try to use OCR (Optical Character Recognition) to reconstruct the original letter. A great hack but not something someone would take seriously.

Yet that is just what happens with these phones. Why should I wait a decade for the airlines to deign to deploy Connexion when there is already a digital path between the plane and the ground? Sure, it's slow and expensive but it exists. And it is a service that can be provided without rebuilding all the planes. It just requires upgrading the phones and adding an RJ-45 jack with an Ethernet chip and a small amount of processing. Considering how much a call costs the cost of such a device is completely trivial. In fact, entire Ethernet boards for PCs are now so inexpensive that they are just thrown in the box if you buy other equipment.

Even if we don't put in an Ethernet chip it seems utterly silly to advertise the ability to use the phone to send and receive email and not just listen for the modem signal and turn it into data right in the airplane. Instead my modem signal must be carried in all its complexity to the ground and then passed on to my ISP. My ISP charges $6/hour for an 800 number call. That's about two minutes at current rates for a call from the plane - surely they can afford to provide IAP (Internet Access Provider) capabilities within that cost.

In fact, the cost would be lower because they don't have to dedicate a path for just me and can provide shared connectivity (though at the low speed the degree of sharing may be limited). Not that the speed has to be low since the native data speed has to be higher in order to carry the 9600bps analog signal and the aggregate capacity for the entire plane can be larger.

And such a simple network has other purposes such as distributing the GPS data.

So why doesn't this happen? I created HomePNA in order to avoid depending on others for home networking (though 802.11 is now the preferred alternative - even in airplanes). But when I'm in an airplane or hotel I don't have the options of deploying my own capabilities and the modem is the instrument for transcending the existing infrastructure.

In the hotel (hospitality) business success is determined by the ability to negotiate for a franchise and those already in place predate all this computer stuff. While they are struggling with providing "Internet access" is not something they really understand. It just a side issue compared with assuring the availability of x-rated movies and clean floors.

And we've seen the gap between the old bell-heads guarding telephony and those of us arguing for connectivity. Who in an organization providing telephone services in airplanes has both the survival skills and feistiness for such a radical idea? And how can they prove beyond any doubt that there will be new revenue? The only way is to deploy it but you can't deploy it 'til you have the proof. But it is obvious that it is a major win for all parties involved and it makes time in flight much more valuable which should be important to airlines. At least I'd like it to be though it is probably seen as secondary to the choice of lunch fare on those flights still trying to feed their passengers.

My best approach is to write this essay and maybe find people who are interested. The sad part is that this essay contains many patentable ideas. Sad because as one who designs and builds systems these are the typical solutions I churn through on a regular basis. The patent system seems to be designed to reward those slow enough to think that simple ideas are breakthroughs. OK, that's not entirely fair and the Gem paper clip probably did deserve patent protection.

That said Internet connectivity remains problematic, as I discovered when I tried to use MSN to send email. I found that for some reason I couldn't connect to port 80 on SMTP servers (also see David's comments last week). Another meddling middle! At least ATT Globalnet allowed me through but ATTBI (ATT Broadband), unlike their previous existence as MediaOne, didn't seem to permit me to get to their mail servers when I'm on the road. All the more reason we need encrypted IPV6 to get rid of this meddling.

Just think, you're sitting in the airplane and just plug into the network. You may not want to surf the web but at least email could trickle in and out and it should cost a lot less than $3/minute since it would be second class traffic. Is that too much to fantasize?

The generation gap cited in Doonesbury is pervasive. Some of us old fogies welcome the newcomers to what is not really a generation gap but a conceptual gap. And the abstractions of the DNS and connectivity are at the heart of this disconnect as is the business vs marketplace issue.

What makes it frustrating is how obvious it is. Yet we see doomed efforts like XM Satellite radio which would have seemed wonderful twenty years ago. But now I can just buy a 160GB disk drive for the price of an XM receiver and easily carry around my 100,000 favorite songs and playlists. Yet the satellite radio people want to ban 802.11 because it gets in the way of their trying to control what I listen to. Bluetooth is another such doomed effort but I've already written far too much to write about the danger of confusing "wireless" with "cellular".

See what happens when the fettle strikes? I lose control and start writing � so I'll stop for the moment with too much left to say.

BobF at 8:20 PM [url]:

What is it about Aiports?

This was reposted by David Farber on his IP Mailing List.

Warning: Do not read this if you are easily complacent.

Whether or not I agree with the current screening techniques at airports, I can understand the assumption that you can keep people from carrying bad things onto airplanes. But why do airports get evacuated when an unattended package is found? The airport is just another public space and, compared with others, not very dense. In fact, as some have pointed out, evacuating terminals creates greater concentrations and opportunities for nefariousness.

If we're going to be afraid of unattended packages then shouldn't we apply the policy to every public space? Or is the assumption that airports have a special mystical significance to "terrorists" apart from the aircraft themselves?

But why should I be surprised? In looking at the map of Idlewild (oops, JFK) I notice that there are no gates numbered 13. And we have the rules for PED (Personal Electronic Devices) which have a similar degree of "reasoning" behind them. At least we can be thankful that cell phones alerted some passengers on 9/11 and prevented an additional attack (albeit with great sacrifice). The problem is that superstitions are not benign - they lead to very bad judgment as well as tainting more reasoned policies.

Perhaps I shouldn't think about such things after reading the announcement from the FCC that I've referenced below. I view this as a direct attack on the Internet and the whole notion of resilience. Am I wrong to be frightened? The word "optimal" is a warning - someone knows what is best and will assure that other possibilities are disadvantaged. At least he has experience in sales and advertising so that he can reassure us that things are just fine. I want to learn more about this announcement and hope it is not as problematic as it seems.

These observations remind me of other "common wisdom" such as the notion that you cannot go swimming for an hour after eating (a peanut?). Or the assumption that you must sterilize everything a baby touches other than the dirt that it keeps putting in its mouth as it crawls around.

Well, I better stop here; otherwise I might lose my faith in the those who we have so much faith in.


For more, see the Archive.

© Copyright 2002-2008 by Daniel Bricklin, Bob Frankston, and David P. Reed
All Rights Reserved.

Comments to: webmaster at satn.org, danb at satn.org, bobf at satn.org, or dpreed at satn.org.

The weblog part of this web site is authored with Blogger.