Advances? Not Just Now.

Just been poking around the tech news, still using "opportunistic wifi" as my connection.  So first of all I'm disillusioned because Telstra have taken over a week to get an existing telephone line connected, and now Amcom are going to take another week, perhaps two, to get ADSL flowing over that (finally!) connected line.  And then, it's going to be ADSL1, not that that matters because my supposed ADSL2+ connection in Perth was worse than dial-up at times.  And speaking of dial-up, I tried using the dial-up number Amcom provided as a temporary measure until they get my broadband connected - and it's about the same as using acoustic couplers over a tin can telephone.

So I can probably be excused for thinking that technology is going backwards.  But reading all the usually jam-packed tech news sites, I'm seeing a kind of hiatus here.  We're well and truly overdue for a quantum leap in some web or internet technology or other, and also a new bit of tech gear that isn't just an Android or iPhone or a new social aggregator site.

Something pretty groundbreaking has to appear before Christmas, you mark my words.  A new internet paradigm, some fundamental scientific discovery, or a piece or hardware. (Or other techo item like an electric car or renewable energy source)  That will restore my faith in our ingenuity and sheer inventiveness, nothing less.

So I'm holding my breath and awaiting results.  What will happen first, my broadband connection actually working or some breakthrough?  As the tagline for my old BBS used to run: "TEdLIVISION! - don't touch that dial!"

Second Loot And Virtual Grids

I've been a player in Second Life, the wildly remunerative virtual world.  Remunerative, that is, if you're Linden Research, the company which developed it and now operates it.  I have to say it was a very immersive experience but ultimately it's priced itself out of my desire.  It's a work in beta, or even trailing edge alpha at times, yet Linden Research has the balls to charge for it, and continuously raise prices.  I say good on them, but it's not exactly the way the web was won...  And it has fuelled a lot of protest and controversy in-world, which is not to their advantage.

Because the truth is, that there are other VR projects out there, such as OSGrid and OpenLife, which operate in the same way as SL, and are substantially cheaper or free, and some of them allow you to run your own simulator server.

See, the dirty secret Linden Labs are holding close to their chest is that grids (which is the infrastructure on which simulator servers run) are not, by nature, closed.  And they are not difficult to run.  A brief history of the World Wide Web might be in order...

When the Internet first wobbled into public view, there were servers for a variety of functions, generally it wasn't simple to get these online and running, and the WWW developed out of a desire to simplify the end user experience for everyone.  First Web servers were also a slightly difficult beast to run, most people did not have the bandwidth to run one from home even if they could, and the idea of websites was still rudimentary.

Then larger organisations came onto the scene and allowed anyone to build a few web pages (think Angelfire, GeoCities) and other organisations began hosting more complex websites in return for money.  But it was and is still possible to run a web server on your PC at home or pay for a server in some hosting installation and host other people's websites.

That's the point that Virtual Reality is at right now.  Linden Labs allows you to host your "virtual website" - in other words, what they refer to as a "simulator" - and build your site there.  OpenLife does something similar but with less cost, and OSGrid provides you with a framework in which to connect your server to a bunch of others so that there is connectivity between them all.

And with some of the older generation not being able to find their way off AOL, virtual reality could actually become a new paradigm that will hopefully soon take over the web - instead of a website, how about owning a block of virtual land?  Then it's just a matter of walking from site to site.  This is much easier than remembering a URL.

And, of course, one of the primary problems Second Life is experiencing is allayed if not eliminated by opening a grid to multiple privately operated servers.  SL has many many problems because it is trying to juggle terabytes of data around between their machines.  This process often ends with databases scrambled - and then the "residents" of SL complain about losing virtual items, not being able to move between virtual locations, not being able to login, etc.

Part of the reason Linden Labs do it this way is technical, the rest is purely commercial.  The technical reason is simply that they can preserve your "assets" as you go from one simulation to the next.  It's not required.

Think about when you shop online: You log into a shopping site, fill your "virtual shopping basket" with what you want, proceed to the "checkout," pay for your items, and leave the website.  Your "assets" on the site were what you had in your basket, the details you provided to pay for those items, and the receipt, which has been emailed to you so it's a permanent record.

When you leave that site, do those "assets" travel with you?  No no a thousand times no!  Web workers put in a lot of hours to prevent precisely that kind of thing.  The "permanent" things you have are held on another server - your Paypal account, for example, or your credit card connection to your bank's network.  And the rest is either in cookies which your browser stores for you, or abandoned when you leave the website.

Similarly, LL argue that by closing their grid and keeping all data together, it means that you will see all the public assets (buildings, scenery, objects, etc) when you travel from one simulation to the next.  It's a lovely argument but totally crap.  Because, when you open a website in your web browser, do you need to refer to some central repository before you can view images, for example?  No - these "assets" are served up by the web server as you connect.  Similarly, by keeping local assets local in a virtual reality simulator, you remove the need for a HUGE database juggling terabytes and spread that load among the individual servers.

Hence, I'm thinking that VR will become more and more commonplace as people realise how much easier it is to present information in such environments, how much easier it is for the users to find that information, in a properly designed simulation, and how much easier it is for a user to find their way between VR sites if they become a bit more website-like in their ubiquity.

And I'm a huge supporter of open sourcing things - if you let a team of developers loose on a project, they will end up like Second Life - more and more new features coming out, more and more justification for price rises, and none of the scutwork, such as fixing potentially show-stopping bugs, getting done.  If you let open-sourcers loose on this kind of project, there will be a similar rush of new features, but the bugs, being extremely irritating, will get cleaned up along the way, leading to a much nicer user experience in the end.

Look at Linux - started out as an open source replacement for a machine operating system that only alpha geeks used, and is now as simple to operate as the competition's systems.  Look at Apache webservers - started out as an open source web server that was complex and unforgiving, and now installs regularly and easily on a huge range of machines at a mouse click.

So start flocking to the alternatives, and show your support by donating to them, assisting them, and maybe even contributing something to the development.  Before you know it, Linden Labs won't be the only game in town...

WMML

What Makes Me Laugh...  Reviewing an Attenborough doco, listening to David enthusing about another creature "perfectly adapted to its environment," that's what.  All creatures great and small show compromise, there's not one "perfect adaptation" because if there was such, that creature would take over the environment.

What we understand of the selection and adaptation process, we know depends on the inherent flaws of the system, starting with the semirandom mutations of the DNA. Every system has parts with flaws, and every other system that can take advantage of those flaws, will.  It seems like the eventual aim of natural selection is to produce a biological equivalent of "grey goo" that will just coat the Earth in a thin layer, one organism.

So when there's a new niche in the "e-ecosystem," it's not surprising to find that an organism has started to explore it and try and make a living at it.  Which is pretty much what you're seeing here, in this article.  It's a wide open ecosystem, it can and does provide money and goods if approached right, and therefore equips anyone able to operate here to better survive, without needing skills like checkout operator at fast food stores, office clerking skills, or any other such.

It's gainful, if not legal, employment.  Being a banger or a burglar are also gainful, if not legal.  There's a niche - taking your stuff, your money - and some organisms (burglars) take it by one means, other organisms (banking and financial institutions) take it by more acceptable means.

The important thing to note is that you're the niche, the burglar and bank are the exploiting organisms.  Similarly, your money and your personal data (and therefore, access to your money held by the banks)  are now a niche online, and getting exploited.  It's survival of the fittest, now it's up to you to shore up defenses and repel the skript kiddies.

Here's Why Speedometers Do That

Why Do Speedometers Go To 160MPH?  According to that article, they don't actually know, but they agree that it leads to the psychological itch to go that bit faster.  In one way, the coming of digital speedometers has to some extent shown old dial type speedos to be the bullshit that they are and thus are solving one issue.

But for my money?  The clue's in the "needle's only half way" remark.  We like to think our machines have some reserve over and above our requirements.  Yes, we have drivers who are only happy when the needles on all the gauges are hovering partway into the red, and they who push that envelope are our race car drivers. (Or our hoon drivers, if they didn't luck into a racing career.)

But for the rest of us, the sense of security we get from driving along with "plenty in reserve" is what sells those cars with the 220kph speedos.  (Amusingly enough, if you push the average car to the valve-bending limit they will top out around 150kph, and then the various engine parts end up making pretty patterns on the road behind you.  We're such suckers...)

Also, of course, there's no way a car manufacturer wants to see the speedo needle hitting the end stop.  So even though 140kph is realistic, they will add that extra 80 just to make sure you won't ever "wrap the needle."  The next model will get that fast, buddy, you betcha.  Just buy it when we release it okay?

Also, all that unused range on the dial makes sure we cling to the speed limit or just over.  Despite knowing that every millimeter we depress that pedal is another half a litre per 100km.  So in a lot of ways, cars with dial type speedos have contributed to deaths, pollution, and the cost of constantly replacing your car with the next model.

And here comes my "personal responsibility" homily, yeah yeah...  See, the reason car manufacturers do those sorts of things is simple.  It's because they work.  You and I and everyone else that buys a car and is swayed by the horsepower and top speed, have contributed to this.

At least we're now pushing for greener cars and thinking about driving a bit slower - aren't we?

Use Traffic Lights To MANAGE Traffic.

Okay - all you city planning authorities out there, looking for ways to green your cities and reduce pollution, reduce congestion, and make life easier for your citizens.  Lissen up, I have a thought for you.

I suggested, years ago, that a lot of traffic congestion could be reduced by designing the bigger better traffic light set.  Once again, the Most Abject Pathetic Technology & Electronics Kompany I worked for looked at me as though I'd sprouted three legs and an extra head when I suggested it, and once more, this is an idea well worth taking on board when you next send out a tender for replacing traffic lights.

It isn't hard, these days, to separate a video image into frames, and search each frame for salient features.  Such as "outline of a car," "outline of a motorcycle," and so forth.  Any reasonable computer can do this with ease.  Said reasonable computer can also compare one frame with a subsequent frame, and over a few frames, average the speed of all those detected salient features, as well as maintain a tally of how many such features are in frame.

So if you aim a camera back across the traffic lights at the oncoming traffic, the computer can tell how much traffic it has coming in, how fast that traffic is moving, and how much traffic has flowed in that direction for the last 24 hours.

Now suppose you have four such systems at each intersection, controlling the flow of traffic.  It doesn't matter what time of day it is, the traffic lights react to the existing stream of traffic and try and schedule so that the least cumulative amount of time is spent idling at the lights in all directions.

That move alone could reduce fuel use and pollution from idling at lights to insignificant amounts.  Since it takes a lot of energy to get rolling again once stopped, reducing the amount of times a vehicle has to stop makes an immense difference.  This is also true for electric vehicles (EVs) because they chew huge amounts of juice to start rolling again.

As a bonus, if there is a traffic stream (or single vehicle) approaching the controlled lights at a speed exceeding the speed limit, you can always provide a red light to them.  Then if they pass through the lights despite that, you have a picture of them running a red light, and also their speed at the time... %)

One last thought - if you don't want to waste the computing power needed for parsing images, perhaps you could respectfully ask car and immobiliser manufacturers to start placing a small piece of technology in their product - a Bluetooth beacon...  You could count the number of BT beacons at traffic lights just as easily as images, you'd lose the ability to enforce speed limits and the red light camera facilities, but it would simplify things a bit.

If you like my idea and decide to use it, feel free to.  But please also feel free to donate some money via my Paypal link so that I can continue to come up with new ideas...  

Few more details on the Joule.

A week ago I wrote about the newly-unveiled prototype of the Joule, a vehicle by South African company Optimal Energy and I was a bit short for details.

The vehicle is claimed to be capable of around 200km range on a single battery pack and 400km on dual battery pack.  That's quite a useable range.  All in all, it makes the Joule sound like a very useful and serviceable vehicle.

One quibble, you don't pay for the battery packs apparently - you lease them from Optimal.  If they just mean you lease the batteries and can then charge them yourself, that is fine.  They specifically mention a seven year lifespan for the batteries, as though you will retain lesseeship of them for the whole seven years.

But there's another alternative, one that needs to be clarified by Joule PR - what if they are just commenting in general that battery packs will last for seven years, but that their "lease" scheme means that you drive until the battery pack is flat, and then need to swap the battery pack for another "leased" battery pack?
I.e. they won't give you a Joule with a battery you own for seven years plus a charging cord, but rather it's like the "swap" system in use for propane gas bottles.  In this latter case, unless Joule has an outlet at every service station, the usefulness of the Joule would approach zero really quickly.  Because I seriously doubt Optimal Energy has that good an infrastructure in place.

So there's the battery to consider, and also, that kind of range comes at a price, which is top speed - the Joule is claimed to top out at "83mph," for whatever reasons.  It seems strange that they mention the odd 3 miles per hout, as though they have been straining all systems to get to that speed, like downhill with a tailwind...  Or it may be to give you inchworm-like 3mph awesome overtaking speed at 80.

Either way, I would not rely on a Joule to take me out in the country, which means EVs are not flexible enough yet.  That said, I still think the Joule is a commendable effort, and if they were available and I was able to afford a new car, the Joule would, at this stage, probably be it.

A HUGE Thank You To All!

Thanks to everyone who passed the link to that previous article out, I noticed a spike for the day, and hopefully it has reached, or will now reach, the eyes of someone with the necessary vision and resources to make it happen.  I tried to start things locally, but it's hard because this takes resources which no-one I know has.

*fingers crossed*