Patrick Moore has for many years made a living by offering himself to industry as an "environmentalist" who will argue industry's side of the issue. He is universally loathed as a sellout turncoat by people who truly care for ecological sustainabilty. I'm surprised that the usually savvy editors of slashdot headlines were taken in by him. If you want to understand Patrick Moore, go see the movie "Thank You For Smoking." I would say that George Bush is as much of an environmentalist as Patrick Moore is.
If you ever hear him speak or read his writing, you should ask how much he's being paid to do it. He really is dirty destructive industry's whore.
A non-regionalized internet is essential to the development of a global democratic governance system that the world badly needs these days. Let's see. Who do I want running global business rules, labour rules, environmental rules, and human rights rules? George Bush, who runs it now through the US military, or a globally elected governance organization.
As the internet globalizes information, communications, and transactions, and organization in general, it can help lead us toward global governance of the large scale issues of our human activity.
A regionalized internet is a bastion of desperate, and losing, nationalists.
Only though global democracy can we start taking people worldwide toward an economically just, socially just, and environmentally responsible highest-common-denominator way of living. The global internet, and globalized organized, but grass-roots-level processes taking place on it, can help lead us there.
Today, we are acting at global scale, in business, and in our economic activity. Our effects are also global. Our climate warming will be global, as will oil shortages, and fresh water shortages. Our governance solutions need to become global to manage these issues. And George Bush's empire will simply not cut it.
The US needs to recognize that the rest of human society is a) out there and b) could be more of a help than a threat if the US would try to facilitate intelligently instead of acting like the ignorant and arbitrary schoolyard bully. Global governance is happening, and it can either be wih, or without the US's cooperation.
By those nations that regionalize their internet, you'll recognize those fundamentally opposed to global democratic governance, and you'll recognize that those are the nations whose power must be limited to enable the new system to emerge.
The global interoperability of the Internet is so important going forward, for human social and economic and scientific progress, that we should treat it as a fundamental human right.
Keeping the Internet free from unreasonable constraints and damage imposed by individual national governments is essential to the internet's survival.
If any national government takes action that damages the integrity of the internet, that country's IP addresses should be firewalled off from the rest of the net. i.e. a reverse great firewall of china should be imposed until the Chinese government backs down from its internet-destruction plan.
I hope that the independent non-governmental voices on ICANN make sure this happens if China persists with its mischief.
ps. Yes this is mostly US government's fault for thinking they're in control of the net, but the important question is what's to be done to discourage this kind of partitioning of the internet.
Having used Java exclusively since 1995, I've decided to try a webapp in python (turbogears).
The turbogears python webapp framework is a nice bit of tool integration for rapid development.
Python itself is a mixed blessing compared to Java.
On the one hand, it has very nice compact, in-built syntax handling of multiplicity (lists, dictionaries, tuples) as all powerful languages should have and java still lacks.
It has other simplicities and flexibilities that are nice.
However, the "pythonic" philosophy of "anything is allowed if you try hard enough could be re-termed "moronic". It's just bad design to have multiple ways of doing the same low-level thing in a language. Complexity multiplies, as we all know.
Also, Python is not as platform agnostic as Java in issues such as byte-ordering in data structures etc, nor is it as secure as java in this respect, because java does not specify the representation in memory of its data objects, making them more difficult to sniff.
Java also has other code security features that python to my knowledge lacks, such as class bytecode verification, class loader security, etc.
Finally and importantly, while Java's (and its standard libraries') documentation is only moderately detailed, at least it is consistent and pretty complete.
The same cannot be said for python documentation, which is sloppy and incomplete, and inconsistent in places.
So is python "progress" from Java? In some O-O and functional programming language respects yes, but in platform quality, simplicity, platform-agnosticism, learnability, and security, no.
We still need a new language that combines the best of these breeds.
He demonstrates a lack of understanding of human factors design principles. More particularly, he demonstrates a lack of understanding that there ARE such design principles. Linux better get its act together if it wants to remain a serious contender going forward.
I've switched to OSX because I respect good design and also I have serious work to do with my computer. I don't have time to spend endless time configuring half-completed core applications on my machine, or outguessing those open-source developers who are allergic to English, or are allergic to manual writing, or are autistic so they don't understand the ways in which other people (yes, even smart people) don't think exactly the same way they do and don't know exactly what they know.
One of my design mottos is "The Default shall be good." I guess I know which political faction that puts me in in Linus's view. Weird arcane non-default options should in some cases be possible to get to, but on no account should their presentation complexify the presented simplicity of the core concepts of the application. Simplicity with adequate functionality is by FAR the most difficult design goal. Why do you think Apple is succeeding these days. They're not perfect designers, but at least they realize it's a very important issue. Calling interface designers names sounds like jealousy to me.
Fight, in cyberspace, and in meatspace if necessary, the United States of America's illegitimate extension of its sovereignty and interests to the rest of the globe.
There's nothing stopping a group of publishers from establishing their own "darknet" on which to "publish" their copyrighted material, and then try to sell tickets for admission to the darknet, kind of like to a peepshow.
If you don't like how the regular, open, web works, then sod off somewhere else with your precious material.
The problem of software quality for commercial software boils down to this. In a market where some schmuck (developer or more likely, executive) is always willing to do an undocumented spaghetti, one-off solution, for less and quicker, and where most customers don't know the difference, good internal-quality, maintainable, extensible software is too expensive to build and sell.
The customer will lose in the long run, when they find that the software is not upgradeable or maintainable. If the software vendor is trying to make a living off releasing upgraded versions of their product, they'll lose out too, eventually, but the manager of the original project will not be blamed. They will be the CEO by that time, rewarded for the short-term profits they produced by whipping the developers in the original project.
The trouble is, EVERYONE, in the commercial market, heavily discounts the future. Nobody minds creating a long-term problem if doing so will "apparently" solve a short term problem.
What this all leads to is a market where the market CANNOT AFFORD the prroduction of quality software. Developers of good instincts, skills and conscience (who are loathe to create bad, unmaintainable, special-case code) are pretty much guaranteed to suffer in this environment.
It would be nice to believe that there are some software companies that are led by long-term-thinking management with a true technical clue. Management truly interested in developing software capital; software of such quality and simple elegance that its intrinsic value (and eventually, its market value) grows geometrically as it builds on itself, rather than declining as it builds cancers onto itself, as is usually the case.
Does anyone have a tale of such a holy grail of a software company?
ok, lisp lists pretty-printed to a file, if you insist. circa 1954. Can represent any data, from any ontology, you like, in a fully general, simple, standardized notation. yawn.
Why is it that the word "patently" only makes sense before the words "false","incorrect", or "ridiculous"? (hmmmmmm..)
From the following comments "I've been doing a relatively interesting job until now, but they've pushed me into management recently. Although the new position is pretty boring..." I would say that you are going to need to learn more to be a good manager.
A good manager has to be a good cognitive psychologist. Understand what motivates and demotivates people. Understand how people miscommunicate with each other. Understand how people are competent/incompetent at individual tasks or task types, not competent/incompetent in general. Understand different personality traits.
As well, you need to understand logistics, dependencies, triage, planning and constant
refactoring of plans, consolidating at milestones
to get a sense of achievement, team building through responsibility assignment and credit sharing and recognition of specific contributions.
And lots of other things.
If you are of the Aspergers-Geek persuasion, think long and hard before getting into management. If you can't see the challenge of it, you will be bowled over by it.
If you block peer-to-peer, you are acknowledging that we are all consumer drones in a corporate-owned world. P2P technology continues the Internet's original egalitarian architecture trends.
Architecture is politics.
If you start blocking services, you're going to turn Internet into TV(the sequel), with the only difference being more personalized commercials.
Resist the temptation to block protocols that allow people to share their information and their creativity. Resist the temptation to block protocols that allow people to co-ooperate in a less hierarchical and more global way.
The future of P2P encrypted, onion-routing, anonymized file sharing services (of which Freenet and TOR are mere initial hints), will probably require such features as truly distributed, anonymous open-source development team (unsueable, and had to assassinate in its entirety,) using distributed, redundant, moving virtual servers to host the code of the P2P app itself. The development on no account should be centered in the US or countries strongly influenced by US legal heavyhandedness. This truly is the new frontier for freedom.
I look at encrypted, anonymized P2P as the digital equivalent of cash, in the following sense. If cash (anonymous financial transactions, hard to spy on, hard to tax) were being invented now, it would be declared illegal by governments.
But most of us would rather have a cash system available to us, wouldn't we.
Encrypted anonymized P2P is the same thing, though it doesn't get the grandfathered legality that cash gets.
This is going to be a battle royal, believe me. It wouldn't surprise me if Bush declares that P2P developers are akin to terrorists. They're agin us, not fur us. Well it depends on who the us is, don't it. Liberte Egalite Fraternite.
So everyone does understand, don't they, that, scientifically speaking, Kyoto is only the first small turn of the steering wheel on the Titanic, and at least 10 turns of the wheel would be required to steer clear of the "global warming" iceberg. (How's that for a mixed metaphor.)
In other words, we have to globally reduce to less than half of 1990 emmissions levels, and using real math, not funny political math, to stop the growth of CO2 levels in the atmosphere.
Here's to a good start on the turnaround anyway. Cheers.
Now let's hope the US is as embarrassed as it should be about being the one remaining unrepentant energy pig on the planet.
"So, how much for this putrid ale?" "That's 5 drachmas for each of you, (goatbreath)"
Case closed.
What can a programmer/architect do in a year?
on
Google Code Jam 2004
·
· Score: 2, Insightful
If I were google, I'd be a lot more interested in what a programmer/software systems architect can accomplish in a year than in a day.
A day is totally random.
Also, shouldn't the candidate's creativity at originating novel ideas count for something and not just their "well-studied code mechanic" skills?
Spoken from the perspective of someone who used to get ~100% on the portions of tests that I completed in the allowed time for the test. Unfortunately, I only completed the first 80 or 90% of the test because I thought more like an Ent than like a squirrel.
So what was my ability? 98%, or 85% ? I suppose for all those relatively trivial problems that I have to solve in 1 hour or less (strikingly few problems like that in s/w eng. in my experience) I'm only an 85%er.
Usually I only have to solve problems in real-time when someone else has screwed up their planning phase (e.g. let's do this live important demo before ever trying it first.)
I think out of the code jam, google will find very good plumbers. Maybe what they need to find are software people who are the equivalent of Rem Koolhas or Daniel Libeskind (or Sergei Brin and Larry Page). You won't find them entering 1-hour or 1-day coding sprints.
I'm surprised that in 35 posts no one has mentioned that pay-as-you-drive insurance would tend to decrease driving, and so would help reduce greenhouse gas emissions, global warming, and urban air pollution.
These would seem to be the major benefits of this idea by far, in the grand scheme of things.
Also. There's no need to track everywhere the car goes in Orwellian fashion. All you need is a new design of tamper-proof odometer that can be read once a year when you renew your insurance.
One of the lessons of software design is that 1 x 1 = 1 (x 1 = 1 ( x 1 = 1 (x 1 = 1... whereas 2 x 2 = 4 (x 2 = 8 (x 2 = 16 (x 2 = 32...
In otherwords, complexity, and choices, that interact with each other, cause a geometric explosion in the complexity of the software, and the complexity of understanding it as a user or a developer.
For a web application developer, the only thing you can assume a web browser is capable of is the LOWEST COMMON DENOMINATOR of functions and standards-adherence. Otherwise you're just fragmenting the market for your app or site.
So the best way to proceed in making progress in browsers is to get good-quality, solid, core standards (with FEW, NOT MANY choices in them) well implemented by all browsers.
At best, an unruly proliferation of arbitrary plugins can only be looked at as a darwinian proving ground. The more useful extensions MUST be incorporated into a standard CORE profile for the browser software, or they are useless (unusable) when viewed from the perspective of a web application user interface designer.
Heck, I've even had to stop using Javascript in my web apps if I want to have a reasonable chance of reaching everyone these days. Java applets were a terrific idea way back too; another great idea destroyed by the fact that the surrounding standards (security, UI libraries, java version support) were multifarious (unusable) and not single (thus usable).
Or more precisely, encrypted, fragmented, redundant, auto-roaming, self-healing, auto-coalescing, smart-data stores are inevitable. If you want to call that P2P, so be it.
The point is, CONTENT will be DIVORCED from LOCATION! It's as inevitable an advance as the very "interweb" itself was.
Can we handle this level of freedom and info-anarchy as a society? Who knows but we're going to be faced with it. Some puny US senator (or senate or whatever) or two ain't going to stop it.
Architecture is politics. Politics can't control the info-architecture. The P2P architecture of the near future is an emergent system. A meme with its own self-replicating power. It's like trying to stop the common cold.
Ok, so weird optical quantum cryptography is one
thing. It will take an age for it to become generally relevant due to the need for special
fibre lines and devices and known paths.
What's more significant is the question
"What will happen once quantum cryptanalysis (sponsored to the tune of GIGA$$ no doubt today
by Uncle Sam's shady agencies) breaks all known conventional strong encryption?"
Think about it. Quantum cryptanalysis will completely change the
future course of Internet and Web architecture.
While today we're clearly heading for nebulous
but indestructable steganographic, strong-encrypted P2P virtual private networks (for business use, for personal and
group use), that whole model breaks down if
all conventional strong crypto is broken, as
may very well happen with quantum cryptanalysis.
That is a REALLY SIGNIFICANT fork in the road coming up in the near future of the InterWeb.
Slashdot Mirror
Patrick Moore has for many years made a living by offering himself to industry as an "environmentalist" who will argue industry's side of the issue. He is universally loathed as a sellout turncoat by people who truly care for ecological sustainabilty. I'm surprised that the usually savvy editors of slashdot headlines were taken in by him.
If you want to understand Patrick Moore, go see the movie "Thank You For Smoking."
I would say that George Bush is as much of an environmentalist as Patrick Moore is.
If you ever hear him speak or read his writing, you should ask how much he's being paid to do it. He really is dirty destructive industry's whore.
A non-regionalized internet is essential to the development of a global democratic governance system that the world badly needs these days. Let's see. Who do I want running global business rules, labour rules, environmental rules, and human rights rules? George Bush, who runs it now through the US military, or a globally elected governance organization.
As the internet globalizes information, communications, and transactions, and organization in general, it can help lead us toward global governance of the large scale issues of our human activity.
A regionalized internet is a bastion of desperate, and losing, nationalists.
Only though global democracy can we start taking people worldwide toward an economically just, socially just, and environmentally responsible highest-common-denominator way of living. The global internet, and globalized organized, but grass-roots-level processes taking place on it, can help lead us there.
Today, we are acting at global scale, in business, and in our economic activity. Our effects are also global. Our climate warming will be global, as will oil shortages, and fresh water shortages. Our governance solutions need to become global to manage these issues. And George Bush's empire will simply not cut it.
The US needs to recognize that the rest of human society is a) out there and b) could be more of a help than a threat if the US would try to facilitate intelligently instead of acting like the ignorant and arbitrary
schoolyard bully. Global governance is happening, and it can either be wih, or without the US's cooperation.
By those nations that regionalize their internet, you'll recognize those fundamentally opposed to
global democratic governance, and you'll recognize that those are the nations whose power must be
limited to enable the new system to emerge.
The global interoperability of the Internet is so important going forward, for
human social and economic and scientific progress, that
we should treat it as a fundamental human right.
Keeping the Internet free from unreasonable
constraints and damage imposed by individual national governments is essential
to the internet's survival.
If any national government takes action that damages the integrity of
the internet, that country's IP addresses should be firewalled off from the rest of
the net. i.e. a reverse great firewall of china should be imposed until the Chinese
government backs down from its internet-destruction plan.
I hope that the independent non-governmental voices on ICANN make sure this happens
if China persists with its mischief.
ps.
Yes this is mostly US government's fault for thinking they're in control of the net,
but the important question is what's to be done to discourage this kind of partitioning
of the internet.
I hope so, because a stunning lack of
basic facts of cosmology comes through
in most of the posts above.
At least google some astrophysics, people.
And yes, that is nearby, in intergalactic
terms.
Space is big. really, really big!
Having used Java exclusively since 1995, I've decided to try a webapp in python (turbogears).
The turbogears python webapp framework is a nice bit of tool integration for rapid development.
Python itself is a mixed blessing compared to Java.
On the one hand, it has very nice compact, in-built syntax handling of multiplicity (lists, dictionaries, tuples) as all powerful languages should have and java still lacks.
It has other simplicities and flexibilities that are nice.
However, the "pythonic" philosophy of "anything is allowed if you try hard enough
could be re-termed "moronic". It's just bad design to have multiple ways of
doing the same low-level thing in a language. Complexity multiplies, as we all
know.
Also, Python is not as platform agnostic as Java in issues such as byte-ordering
in data structures etc, nor is it as secure as java in this respect, because
java does not specify the representation in memory of its data objects, making them
more difficult to sniff.
Java also has other code security features that python to my knowledge lacks, such as class bytecode verification,
class loader security, etc.
Finally and importantly, while Java's (and its standard libraries') documentation is only moderately detailed, at least it is consistent and pretty complete.
The same cannot be said for python documentation, which is sloppy and incomplete, and
inconsistent in places.
So is python "progress" from Java? In some O-O and functional programming language respects
yes, but in platform quality, simplicity, platform-agnosticism, learnability, and security,
no.
We still need a new language that combines the best of these breeds.
He demonstrates a lack of understanding of human factors design principles. More particularly, he demonstrates a lack of understanding that there ARE such design principles. Linux better get its
act together if it wants to remain a serious contender going forward.
I've switched to OSX because I respect good design and also I have serious work to do with my
computer. I don't have time to spend endless time configuring half-completed core applications
on my machine, or outguessing those open-source developers who are allergic to English, or
are allergic to manual writing, or are autistic so they don't understand the ways in which other
people (yes, even smart people) don't think exactly the same way they do and don't know exactly
what they know.
One of my design mottos is "The Default shall be good." I guess I know which political faction that
puts me in in Linus's view. Weird arcane non-default options should in some cases be possible to
get to, but on no account should their presentation complexify the presented simplicity of the
core concepts of the application. Simplicity with adequate functionality is by FAR the most difficult
design goal. Why do you think Apple is succeeding these days. They're not perfect designers,
but at least they realize it's a very important issue. Calling interface designers names sounds
like jealousy to me.
Fight, in cyberspace, and in meatspace if necessary, the United States of America's illegitimate extension of its sovereignty and interests to the rest of the globe.
No jurisdiction without representation.
Come get me at 49.21992N 122.24338W
There's nothing stopping a group of publishers from establishing their
own "darknet" on which to "publish" their copyrighted material, and then
try to sell tickets for admission to the darknet, kind of like to
a peepshow.
If you don't like how the regular, open, web works, then sod off somewhere
else with your precious material.
The problem of software quality for commercial software boils down to this.
In a market where some schmuck (developer or more likely, executive) is always willing to do
an undocumented spaghetti, one-off solution, for less and quicker, and where most customers
don't know the difference, good internal-quality, maintainable, extensible software is too
expensive to build and sell.
The customer will lose in the long run, when they find that the software is not upgradeable
or maintainable. If the software vendor is trying to make a living off releasing upgraded
versions of their product, they'll lose out too, eventually, but the manager of the original
project will not be blamed. They will be the CEO by that time, rewarded for the short-term
profits they produced by whipping the developers in the original project.
The trouble is, EVERYONE, in the commercial market, heavily discounts the future. Nobody
minds creating a long-term problem if doing so will "apparently" solve a short term problem.
What this all leads to is a market where the market CANNOT AFFORD the prroduction of
quality software. Developers of good instincts, skills and conscience (who are loathe to create
bad, unmaintainable, special-case code) are pretty much guaranteed to suffer in this
environment.
It would be nice to believe that there are some software companies that are led by
long-term-thinking management with a true technical clue. Management truly interested
in developing software capital; software of such quality and simple elegance that its intrinsic
value (and eventually, its market value) grows geometrically as it builds on itself, rather than
declining as it builds cancers onto itself, as is usually the case.
Does anyone have a tale of such a holy grail of a software company?
ok, lisp lists pretty-printed to a file, if you insist.
circa 1954.
Can represent any data, from any ontology, you like,
in a fully general, simple, standardized notation.
yawn.
Why is it that the word "patently" only makes
sense before the words "false","incorrect",
or "ridiculous"? (hmmmmmm..)
of the extra cost.
From the following comments "I've been doing a relatively interesting job until now, but they've pushed me into management recently. Although the new position is pretty boring..." I would say that you are going to need to learn more to be a good manager. A good manager has to be a good cognitive psychologist. Understand what motivates and demotivates people. Understand how people miscommunicate with each other. Understand how people are competent/incompetent at individual tasks or task types, not competent/incompetent in general. Understand different personality traits. As well, you need to understand logistics, dependencies, triage, planning and constant refactoring of plans, consolidating at milestones to get a sense of achievement, team building through responsibility assignment and credit sharing and recognition of specific contributions. And lots of other things. If you are of the Aspergers-Geek persuasion, think long and hard before getting into management. If you can't see the challenge of it, you will be bowled over by it.
If you block peer-to-peer, you are acknowledging that we are all consumer drones in a corporate-owned world. P2P technology continues the Internet's original egalitarian architecture trends.
Architecture is politics.
If you start blocking services, you're going to turn Internet into TV(the sequel), with the only difference being more personalized commercials.
Resist the temptation to block protocols that allow people to share their information and their creativity. Resist the temptation to block protocols that allow people to co-ooperate in a less hierarchical and more global way.
Namaste
The future of P2P encrypted, onion-routing, anonymized file sharing services (of which Freenet and TOR are mere initial hints), will probably require such features as truly distributed, anonymous open-source development team (unsueable, and had to assassinate in its entirety,) using distributed, redundant, moving virtual servers to host the code of the P2P app itself. The development on no account should be centered in the US or countries strongly influenced by US legal heavyhandedness. This truly is the new frontier for freedom.
I look at encrypted, anonymized P2P as the digital equivalent of cash, in the following sense. If cash (anonymous financial transactions, hard to spy on, hard to tax) were being invented now, it would be declared illegal by governments.
But most of us would rather have a cash system
available to us, wouldn't we.
Encrypted anonymized P2P is the same thing, though it doesn't get the grandfathered legality that cash gets.
This is going to be a battle royal, believe me.
It wouldn't surprise me if Bush declares that P2P developers are akin to terrorists. They're agin
us, not fur us. Well it depends on who the us is, don't it. Liberte Egalite Fraternite.
This one made my day.
Haven't had a good laugh like that in ages.
(NEQ "IsNot" Original)
(NEQ Microsoft InFromLunch)
Lisp has had a NEQ ( = NOT EQ )
operator for ages too.
Good deduction.
Surprising from someone who obviously comes from the heart of the Greater Republic of Jesusland:
http://www.kenlayne.com/2004/11/secret_jesusland.h tml
So everyone does understand, don't they, that, scientifically speaking, Kyoto is only
the first small turn of the steering wheel on the
Titanic, and at least 10 turns of the wheel would be required to steer clear of the "global warming" iceberg. (How's that for a mixed metaphor.)
In other words, we have to globally reduce to
less than half of 1990 emmissions levels, and
using real math, not funny political math,
to stop the growth of CO2 levels in the
atmosphere.
Here's to a good start on the turnaround anyway.
Cheers.
Now let's hope the US is as embarrassed as
it should be about being the one remaining
unrepentant energy pig on the planet.
90% of software patents are ridiculously
obvious to any comp sci grad.
So ignore them. This is junk law.
This is the rich trying to abuse the law
to get richer.
It's bullshyte. Ignore it.
One-click this, a**hole!
Sniff sniff
Scotty we hardly knew ye.
Conversation overheard in ancient times:
"So, how much for this putrid ale?"
"That's 5 drachmas for each of you, (goatbreath)"
Case closed.
If I were google, I'd be a lot more interested
in what a programmer/software systems architect
can accomplish in a year than in a day.
A day is totally random.
Also, shouldn't the candidate's creativity
at originating novel ideas count for something
and not just their "well-studied code mechanic"
skills?
Spoken from the perspective of someone who used
to get ~100% on the portions of tests that I
completed in the allowed time for the test.
Unfortunately, I only completed the first
80 or 90% of the test because I thought more
like an Ent than like a squirrel.
So what was my ability? 98%, or 85% ?
I suppose for all those relatively trivial
problems that I have to solve in 1 hour or
less (strikingly few problems like that in
s/w eng. in my experience) I'm only an 85%er.
Usually I only have to solve problems in
real-time when someone else has screwed up
their planning phase (e.g. let's do this
live important demo before ever trying it
first.)
I think out of the code jam, google will
find very good plumbers. Maybe what they
need to find are software people who are
the equivalent of Rem Koolhas
or Daniel Libeskind (or Sergei Brin and Larry
Page). You won't find them entering
1-hour or 1-day coding sprints.
I'm surprised that in 35 posts no one has mentioned that pay-as-you-drive insurance would tend to decrease driving, and so would help reduce greenhouse gas emissions, global warming, and urban air pollution.
These would seem to be the major benefits of this
idea by far, in the grand scheme of things.
Also. There's no need to track everywhere the car
goes in Orwellian fashion. All you need is a new
design of tamper-proof odometer that can be read
once a year when you renew your insurance.
One of the lessons of software design is that
1 x 1 = 1 (x 1 = 1 ( x 1 = 1 (x 1 = 1...
whereas
2 x 2 = 4 (x 2 = 8 (x 2 = 16 (x 2 = 32...
In otherwords, complexity, and choices, that
interact with each other, cause a geometric
explosion in the complexity of the software,
and the complexity of understanding it as a
user or a developer.
For a web application developer, the only
thing you can assume a web browser is capable
of is the LOWEST COMMON DENOMINATOR of functions
and standards-adherence. Otherwise you're
just fragmenting the market for your app or
site.
So the best way to proceed in making progress in
browsers is to get good-quality, solid, core
standards (with FEW, NOT MANY choices in them)
well implemented by all browsers.
At best, an unruly proliferation of arbitrary
plugins can only be looked at as a darwinian
proving ground. The more useful extensions
MUST be incorporated into a standard
CORE profile for the browser software, or they
are useless (unusable) when viewed from the perspective of a web application user interface designer.
Heck, I've even had to stop using Javascript
in my web apps if I want to have a reasonable
chance of reaching everyone these days. Java
applets were a terrific idea way back too;
another great idea destroyed by the fact
that the surrounding standards (security, UI
libraries, java version support) were
multifarious (unusable) and not single
(thus usable).
Or more precisely, encrypted, fragmented, redundant, auto-roaming, self-healing, auto-coalescing, smart-data stores are inevitable. If you want to call that P2P, so be it.
The point is, CONTENT will be DIVORCED from LOCATION! It's as inevitable an advance as the very "interweb" itself was.
Can we handle this level of freedom and info-anarchy as a society? Who knows but we're going to be faced with it. Some puny US senator
(or senate or whatever) or two ain't going to stop it.
Architecture is politics. Politics can't control the info-architecture. The P2P architecture of the near future is an emergent system. A meme with its own self-replicating power. It's like trying to stop the common cold.
Ok, so weird optical quantum cryptography is one thing. It will take an age for it to become generally relevant due to the need for special fibre lines and devices and known paths.
What's more significant is the question "What will happen once quantum cryptanalysis (sponsored to the tune of GIGA$$ no doubt today by Uncle Sam's shady agencies) breaks all known conventional strong encryption?"
Think about it.
Quantum cryptanalysis will completely change the future course of Internet and Web architecture. While today we're clearly heading for nebulous but indestructable steganographic, strong-encrypted P2P virtual private networks (for business use, for personal and group use), that whole model breaks down if all conventional strong crypto is broken, as may very well happen with quantum cryptanalysis.
That is a REALLY SIGNIFICANT fork in the road coming up in the near future of the InterWeb.