Monday, 30 December 2002
A key difference between TCP streams and VoIP or other 'real time' streams is that TCP packets always tend toward the size of the maximum permissible packet size for the route (often around 1500 bytes due to Ethernet packet sizes). Real-time packets sacrifice efficiency to achieve lower latency, and typically send smaller packets (the efficiency loss is that the per-packet addressing overhead is a larger proportion of the overall stream).
Stuart has argued that the best QoS algorithm is the simplest - move RT packets to the front of the queue, minimizing their latency. This only needs a single QoS bit.
So here's my proposal: use packet length as a heuristic for latency needs. When queuing, move shorter packets to the front of the queue.
You're only queueing when congestion is involved, so this should have little or no effect on the uncongested case. It will increase the Round Trip Time marginally for TCP, which should make it back off accordingly, and will make TCP-type packets more prone to dropping due to queue overflow (which is the desired effect as TCP adapts to this).
Cash is the original example of a network effect - Metcalf's Law applies to currencies (If you disagree I have some zlotys here for you).
Starting a new currency is hard. Paypal basically did it by good usability implementation and a lot of bribery with VC money (the $5 spiffs to new entrants and their introducers, the total absence of fees for individuals until they were hooked and dependent as businesses).
Cybercash, eCash et al failed in some ways by doing too good a job of security at the expense of usability, in the same way PGP has, as well as by being too early to work the bribe trick and ride eBay's market.
Trusting Paypal is like trusting the US or UK not to inflate the currency too much. One kind of private currency a lot of people got burned by accepting payment in over the last few years was company stock.
AKMA comments on the distinction between different kinds of reputation - moral versus financial, and then says part of what David�s saying is that credit ratings and credit cards generally work pretty well, and I suspect it�s worth building out from there.
This is where I disagree. Credit Cards work OK mostly; credit ratings work really clumsily. People spend inordinate amounts of time and effort doing illogical things to improve their credit profiles. I notice this more than most; as a US resident alien I discovered that my lack of US credit history meant that I could not obtain credit cards at all except on remarkably onorous terms, (eg I deposit $1000, get a $1000 credit line, am charged $20 a month and get an APR of 30%).
I got credit card solicitations commensurate with my salary like everyone else (along with the frankly insulting ones), but if I took them up I'd be rejected by the credit bot, and my rating would then have 'rejected' on it too. If I took up 0% credit lines and behaved rationally, paying them off before they expired without penalty, that was a further black mark. The thing to do is take out a loan on whatver terms are offered, make regular payments and wait 5 years.
Jaron Lanier said:
Real, though miniature, Turing Tests are happening all the time, every day, whenever a person puts up with stupid computer software.
For instance, in the United States, we organize our financial lives in order to look good to the pathetically simplistic computer programs that determine our credit ratings. We borrow money when we don't need to, for example, to feed the type of data to the programs that we know they are programmed to respond to favorably.
In doing this, we make ourselves stupid in order to make the computer software seem smart. In fact we continue to trust the credit rating software even though there has been an epidemic of personal bankruptcies during a time of very low unemployment and great prosperity.
We have caused the Turing test to be passed. There is no epistemological difference between artificial intelligence and the acceptance of badly designed computer software.
How about a distributed reputation system that is the inverse of a credit rating agency -one that collates individuals' experiences with financial institutions and gives them a corresponding rating? That sounds interesting.
Finally, you should all go and read Cory Doctorow's book 'Down and Out in the Magic Kingdom' (to be published Jan 9th)- Whuffie is reputation based finance, and the novel examines the implications:
Whuffie recaptured the true essence of money: in the old days, if you were broke but respected, you wouldn't starve; contrariwise, if you were rich and hated, no sum could buy you security and peace. By measuring the thing that money really represented -- your personal capital with your friends and neighbors -- you more accurately gauged your success.
Saturday, 28 December 2002
Try this out at your local farmers market. Walk up to the chap selling fruit & veg, and offer him cash for them. He doesn't know you from (AKM) Adam, yet he'll readily take your cash, and you'll buy his fruit (assuming it isn't mouldy or over-priced in your opinion).
The great trick here is cash. A medium of exchange. This is the original economic technology hack, and it is a remarkably good one, as it avoids the whole trust conundrum, or rather, offloads it onto a certifying authority. All the authority certifies is the money, not the parties to the exchange at all. Try paying at the farmers market with a credit card, and see how far that gets you (do take another form of ID with you too).
The succesful online payment company, PayPal, takes this approach, certifying the existence of the money, not the worthiness of the individuals.
David Chaum's work on anonymous digital cash should not be forgotten here. His key patents expire in 2005...
Tuesday, 24 December 2002
When men have nothing else to occupy their minds, they take to thinking. Having nothing better to do until B. arrived, I fell to musing.
What a wonderful piece of Socialism modern civilisation has become - not the Socialism of the so-called Socialists-a system modelled apparently upon the methods of the convict prison-a system under which each miserable sinner is to be compelled to labour, like a beast of burden, for no personal benefit to himself, but only for the good of the community-a world where there are to be no men, but only numbers-where there is to be no ambition and no hope and no fear,-but the Socialism of free men, working side by side in the common workshop, each one for the wage to which his skill and energy entitle him; the Socialism of responsible, thinking individuals, not of State-directed automata.
Here was I, in exchange for the result of some of my labour, going to be taken by Society for a treat, to the middle of Europe and back. Railway lines had been laid over the whole 700 or 800 miles to facilitate my progress; bridges had been built, and tunnels made; an army of engineers, and guards, and signal-men, and porters, and clerks were waiting to take charge of me, and to see to my comfort and safety. All I had to do was to tell Society (here represented by a railway booking-clerk) where I wanted to go, and to step into a carriage; all the rest would be done for me. Books and papers had been written and printed; so that if I wished to beguile the journey by reading, I could do so. At various places on the route, thoughtful Society had taken care to be ready for me with all kinds of refreshment (her sandwiches might be a little fresher, but maybe she thinks new bread injurious for me). When I am tired of travelling and want to rest, I find Society waiting for me with dinner and a comfortable bed, with hot and cold water to wash in and towels to wipe upon. Wherever I go, whatever I need, Society, like the enslaved genii of some Eastern tale, is ready and anxious to help me, to serve me, to do my bidding, to give me enjoyment and pleasure. Society will take me to Ober-Ammergau, will provide for all my wants on the way, and, when I am there, will show me the Passion Play, which she has arranged and rehearsed and will play for my instruction; will bring me back any way I like to come, explaining, by means of her guide-books and histories, everything upon the way that she thinks can interest me; will, while I am absent, carry my messages to those I have left behind me in England, and will bring me theirs in return; will look after me and take care of me and protect me like a mother-as no mother ever could.
All that she asks in return is, that I shall do the work she has given me to do. As a man works, so Society deals by him.
To me Society says: "You sit at your desk and write, that is all I want you to do. You are not good for much, but you can spin out yards of what you and your friends, I suppose, call literature; and some people seem to enjoy reading it. Very well: you sit there and write this literature, or whatever it is, and keep your mind fixed on that. I will see to everything else for you. I will provide you with writing materials, and books of wit and humour, and paste and scissors, and everything else that may be necessary to you in your trade; and I will feed you and clothe you and lodge you, and I will take you about to places that you wish to go to; and I will see that you have plenty of tobacco and all other things practicable that you may desire-provided that you work well. The more work you do, and the better work you do, the better I shall look after you. You write-that is all I want you to do."
"But," I say to Society, "I don't like work; I don't want to work. Why should I be a slave and work?"
"All right," answers Society, "don't work. I'm not forcing you. All I say is, that if you don't work for me, I shall not work for you. No work from you, no dinner from me-no holidays, no tobacco."
And I decide to be a slave, and work.
Society has no notion of paying all men equally. Her great object is to encourage brain. The man who merely works by his muscles she regards as very little superior to the horse or the ox, and provides for him just a little better. But the moment he begins to use his head, and from the labourer rises to the artisan, she begins to raise his wages.
Of course hers is a very imperfect method of encouraging thought. She is of the world, and takes a worldly standard of cleverness. To the shallow, showy writer, I fear, she generally pays far more than to the deep and brilliant thinker; and clever roguery seems often more to her liking than honest worth. But her scheme is a right and sound one; her aims and intentions are clear; her methods, on the whole, work fairly well; and every year she grows in judgment.
One day she will arrive at perfect wisdom, and will pay each man according to his deserts.
But do not be alarmed. This will not happen in our time.
Monday, 23 December 2002
Lo, He abhors not the Virgins womb
is dreadful. It is hard to parse, uses obscure words badly, brings up a rather nasty mental image, and doesn't even scan.
Saturday, 21 December 2002
Anyway, while checking that my new BigZoo calling card access numbers really were local toll-free with SBC, I came up with a solution. When the local carrier is switching you to a toll number, it should play a recorded voice saying 'local toll' (or even better 'local toll 10 cents a minute').
You should be able to turn this off by explict request, but if the local telco did this BY DEFAULT, the chance of people accidentally running up huge dialup modem bills by mistake would be greatly reduced.
Of course, they make money from these mistakes, so it may require regulation to enforce this, but as local toll calls are now competitive, (which is why I was billed the $600 by AT&T, not SBC) SBC could do it to warn you that AT&T was about to bill you.
The real scandal of course is why InterLATA calls (over 17 notional miles within the area code) cost more than my calling card charges me to call the UK per minute.
Friday, 20 December 2002
Firstly, I don't understand why commoditizing something is deemed to a be a bad thing. If something is a commodity, it basically means that a market is operating well, and that its price has stabilised at an equilibrium; it is the things that aren't commoditized that are problematic- their values fluctuate wildly through fashion. A commodity business is predictable enough that you can employ MBAs to run it.
If we could commoditize connectivity, that would be great. Joel's essay on 'Commoditizing your complements' is interesting here, and explains some of the forces that move things towards Open Source or Infrastructure.
Eric says the internet is *truly* economically destructive in the sense that it bends the assumptions of supply and demand to the point that making money gets progressively harder over time (since the public domain chews everything up).
This shows a profound misunderstanding of economic value. Value is created by exchange. If I buy a sushi lunch, I value it more than the $12 it costs me. Conversely the sushi chef values the $12 more than the sushi. If this wasn't true, one of us would not make the trade. Thus, by trading, we have increased the total amount of value in the world.
The net enables these kinds of exchanges to happen more easily, reducing friction and delays, as well as enabling other value-creating ones like this conversation.
Arbitrage is when you move things from one place where they are less valued to another where they are more valued, and keep the differnce (less transport costs). This 'taking between' is the literal meaning of 'entrepreneur'. The net, by easing communication, reduces the opportunities for these kinds of gains.
This article explains how things play out by using economic data derived from web sales. You get commoditized, low-cost operations, and premium, high-service operations coexisting, just as you do in the real world.
This is what David was getting at with his recording industry example. It is the distribution side that gets squished, as that is all about moving things around. The creating music and discovering artists part is still valuable, but it no longer needs to be tied up with the distribution based companies. Which leads me neatly to mediAgora.
Tuesday, 17 December 2002
Hence the Ident-i-Eeze. This encoded every single piece of information about you, your body and your life into one all- purpose machine-readable card that you could then carry around in your wallet, and therefore represented technology's greatest triumph to date over both itself and plain common sense.
mostly harmless, 1992
The legend is this:
`The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at or repair.'
mostly harmless, 1992
Tuesday, 10 December 2002
This links in a weird way with what was said at the Harvard 'Managing Innovation' conference I attended - that Business School management can't do disruptive innovation.
So, what we need is business school managment doing the commodity connectivity business, with their 6 sigma quality and 5 9's uptimes and leave services to crazy individuals who like the risk of being in the fashion business.
Thursday, 5 December 2002
He missed the obvious solution though - the TPC fax gateway (which Demon sponsor to any fax machine in the UK):
email Tony Blair by fax
Tuesday, 3 December 2002
Sony Music Entertainment [announced] the advent of its new "Label Gate" digital rights management (DRM) package. Beginning in 2003, all CDs released from Sony Japan will carry this protection.
Label Gate is a one-two punch: Every track on a protected CD will be encoded, and the customer must use Sony's proprietary software to play back the tracks on a computer. This prevents people from converting the music tracks ("ripping") into other formats, such as MP3 or WAV, and sharing them over the Internet. Sounds pretty good, right? Wrong. Not for consumers, and, ultimately, not for Sony.
Unfortunately, this sort of action on Sony's part only hurts consumers --those that legitimately buy their products -- by making listening to music a hassle. Further, it violates the doctrine of Fair Use -- the ability to use a product you have legally purchased in any legal way you see fit. Sony has effectively made it impossible for a user with a couple of computers and an MP3 player to make copies of songs he has legitimately purchased. Further, DRM software actually encourages people to search for "unrestricted" copies of tracks online, through peer-to-peer networks. Unrestricted tracks would allow you to listen to the songs wherever and whenever you please.
If the record companies want to compete with peer-to-peer networks, they're going to have to find a way to give customers fast, inexpensive access to music. Restricting access and hassling customers is bad business, no matter how you look at it.
Let me state here a personal conviction that appears, right now, to be profoundly unfashionable; which is that a planned economy can be more productive - and more morally desirable - than one left to market forces. The market is a good example of evolution in action; the try-everything-and-see-what-works approach. This might provide a perfectly morally satisfactory resource-management system so long as there was absolutely no question of any sentient creature ever being treated purely as one of those resources. The market, for all its (profoundly inelegant) complexities, remains a crude and essentially blind system, and is - without the sort of drastic amendments liable to cripple the economic efficacy which is its greatest claimed asset - intrinsically incapable of distinguishing between simple non-use of matter resulting from processal superfluity and the acute, prolonged and wide-spread suffering of conscious beings.
It is, arguably, in the elevation of this profoundly mechanistic (and in that sense perversely innocent) system to a position above all other moral, philosophical and political values and considerations that humankind displays most convincingly both its present intellectual [immaturity and] - through grossly pursued selfishness rather than the applied hatred of others - a kind of synthetic evil.
Intelligence, which is capable of looking farther ahead than the next aggressive mutation, can set up long-term aims and work towards them; the same amount of raw invention that bursts in all directions from the market can be - to some degree - channelled and directed, so that while the market merely shines (and the feudal gutters), the planned lases, reaching out coherently and efficiently towards agreed-on goals. What is vital for such a scheme, however, and what was always missing in the planned economies of our world's experience, is the continual, intimate and decisive participation of the mass of the citizenry in determining these goals, and designing as well as implementing the plans which should lead towards them.
Used to planning plotlines across many galaxies in his novels, Iain completely misses the point that in a market economy, the intelligent planning is done by all the individuals in parallel. Some plans work out better than others, which is fine, and clearly better than us all having to agree the same plan, and stick to that. You can't plan innovation collectively, becasue you don't know what form it will take.
As for 'treating people as resources' - that is exactly what centrally planned economies do. Banks should read the Gulag Archipelago or any of Robert Conquest's histories.
Monday, 2 December 2002
I worked at the Interactive Television Unit (the BBC department that was founded for the Domesday Project) for the last 3 months of its existence in 1989 before it was spun out into the MultiMedia Corporation in Jan 1990 (I then worked at MMC until 1997, when it became a shell company owned by the stockbrokers, but that's another story).
When we left the BBC, they had all the original video data on Broadcast quality masters, and all the digital data preserved on VAX tapes. They must have thrown those out in the intervening 12 years (which wouldn't surprise me).
I know of two former MMC directors who have CD-ROM backups of the digital data and working Domesday systems.
Which is not to decry the work in emulating it - that is the real long-term answer. This is the technical (as oppossed to the economic) reason why DRM is futile. The Church-Turing thesis (that any universal computer can emulate any other one), when combined with Moore's law (that Computer power doubles every 18 months) means that we will always be able to run old software in emulation. The corollary is that emulation is often the best solution, even compared to recompilation of the original source. As I've said before, MAME is the most impressive example of this truth, though the 68000 emulator running inside the 'Classic' emulator on OS X is another worthy example.