Thursday, November 4, 2010

Video is Not Killing the Internet.

One thing that really stuns me is how often it gets said that video is killing the internet, with the justification for this claim being that it's now a much larger fraction of the bandwidth than "anything else".

It's a much larger bandwidth, BECAUSE IT IS LARGER.

If I spend an hour reading text on the web, it's unlikely that I will use up more than perhaps 100k of bandwidth. If I spend that same hour watching video, 100M is more like it. If something takes a thousand times more bandwidth to consume, it would have to take up a tenth of a percent of the time to be taking up exactly as much bandwidth.

If it were taking up 90% of the bandwidth, that's less than one percent of people's online activity. A projection from Wired a ways back showed web use decreasing to a "depressing" 23% Oh yeah. Video is slaughtering the text internet alright.

It's a wonder you're reading this at all.

Wednesday, November 3, 2010

Superheroes Everywhere

One thing I've been noticing is that I'm friends with a lot of people I really have to classify as Some Kind of Superhero.

I feel, on a daily basis, that I must have somehow landed myself in the middle of one of those circles of friends you read about in the history books as being "the masterminds behind the Enlightenment" or "the Founding Fathers of the United States". And increasingly, I think perhaps I am. But that's not what's amazing.


In fact, it's happening hundreds, thousands of times, in different sectors and in different arenas as distances evaporate. Core teams are self-assembling into groups of visionaries who will change the way the world works, from one viewpoint. But instead of defining a whole age for the whole of "civilized" mankind, each one will be hammering away at some aspect of the whole.

Look around you. You're probably surrounded by superheroes too.


Very impressive showing by Ubuntu on the basic scientific use front. Octave is absolutely delicious in that you can just bloody do things. I *might* drop Scilab but Octave isn't quite so shiny on my Win7 lappy.

And of course, Scilab has an Ubuntu release that runs pretty well (once you get OpenGL working, of course.)

Don't get me wrong, there's still a small stack of stuff that I still can't get working under Ubuntu, but that stack is SHRINKING. And it's small enough now that I DON'T need Windows for everyday work.

Also, Ubuntu is totally rocking my sound card already. Bravo, Open Source, bravo.

Sunday, October 24, 2010

Wait, Really?

That crazy guy who was going to launch inflatable space stations into orbit has two modules the size of a dorm room each already up? And Boeing is co-planning their manned space flight to have his much-larger future modules as a destination?

Holy cow, dude.

Saturday, October 23, 2010

The Explosion of the Possible

There are now so many things which have become possible so recently that only a very tiny fraction of them actually get done. It's true for all of us, which is one reason we all feel so paralyzed so often by the constant flood of new and amazing things that everyone else around the world is already doing.

However, the infinity of that which is undone easily trumps the infinity of that which is done.

Sunday, October 17, 2010

Post-Privacy Humanity? Not on their terms.

So my thoughts on privacy have gone back and forth a lot recently and I've decided one thing for sure:

"Post Privacy" can not be allowed to happen by means of grinning "social media experts" pinning our lives up on their board next to their ads.

Post Privacy is fine in concept, but in practice it needs to be paired with a massive advocacy push to make sure of one thing: that the experts, politicians, lawyers, corporate shills, and robber barons get as little or LESS privacy than any of us. And the problem with the way things are happening now is that it works the opposite way. The above types buy proprietary walled-garden networking solutions that let them have all of their conversations away from prying eyes while the rest of us buy our social networking with the data itself. And with our own personal experiences and our data so pivotal and precious, and with server space and time so cheap, it's a raw deal.

I don't think we can really afford to give our data so willingly in the long run. We need both to be spreading our information peer-to-peer rather than through a hub owned by marketing jackasses, and to be undermining the privacy of those who have any significant amount of power.

Because that should be the tradeoff. Privacy or Power. Nobody should have both. Or neither.

Saturday, October 16, 2010

I'm a pathological name-giver. So it shouldn't surprise you I have a whole host of names for various internet metapundit types:

Royalists: these are people who feel like the internet is too free. Specifically, they feel that, because everyone can say everything, nobody is really able to say anything. I don't buy this for a split second: top-down content selection isn't how you get greatness. It's how you get television.

NeuroLuddites: your brain is probably buzzing with new information right now. These people hate that. Fearful that their poor cerebral cortices will burn out if they keep running them this hard, they warn us that The Internet Is Making You Dumb, usually citing studies relating to specific types of thinking which have less value when you have access to fast search.

MediaMarxists: there's a really stunning number of rich jerkwads trying to make the internet more like television and less like anything anyone smarter than a raddish would actually want. Their direct opposition are MediaMarxists, who go entirely overboard from this admirable starting point and start claiming that if we don't confront and directly destroy these influences, we're all doomed. Useful allies when you're trying to keep internet neutrality alive, but I wouldn't trust them to actually run anything.

Internet Anarchists: There's always someone pounding the table and saying that all content everywhere should be free, and that nothing should ever be secret for anyone ever. I'm actually in this camp, as long as we start at the top and work our way down, which all too often isn't the case.

Humbugs: For every person extolling the virtue of any facet of the internet, there's someone saying it's a distraction from the Real World, which of course is going to hell in a handbasket because you stupid kids are all on the internet blogging instead of getting arrested like We Did Back In My Day.

And of course, many more. I might have a problem...

Friday, October 15, 2010

Shopping List of Doom

Just running off a list of things I'd need to cause a ruckus in a pinch. Not a complete list of the goodies I'd have in a fully set up lab, but the core elements I'd need to, given access to the surrounding Technium, start cannibalizing things and making abominations.

In part because it'd be nice to know what to order if I ever find myself somewhere without tools, and in part because it'd be great to have, say, a small suitcase with these items in it.

* Microcontrollers, preferably a nice brick foam stuck full of 'em.
* USB stick with all the OpenSource goodies, including Code::Blocks, Arduino, WinAVR, Processing, etc.
* Arduino or other AVR programmer
* The Warranty Voider (this is a small leatherman multitool with wire strippers)
* Spool of solid-core shielded wire
* spool of magnet wire
* Resistors, 10, 100, 1k, 5k, 10k, at least ten of each
* Capacitors, 100nF, 1uF, 100uF
* Hobby saw
* Soldering iron and solder
* breadboard, and some cards of perfboard.

Any other suggestions?

Program or be Programmed

Woa, I need to get this book:

In the emerging, highly programmed landscape ahead, you will either create the software or you will be the software. It’s really that simple: Program, or be programmed. Choose the former, and you gain access to the control panel of civilization. Choose the latter, and it could be the last real choice you get to make.

And this doesn't just mean learning a programming language (you should), it means learning how the network works, and how to tweak it. It means learning how to perceive both the impact of actors and the impact of programmatic actors, and to hear, at least to some extent, the whispers of the programs that move through the network, extended hands of actors who may, no, have certainly, forgot what they're doing.

Thursday, October 14, 2010

Magic Boxes (with handles)

I think a big reason Electrical and Computer engineers expect that we will understand or at least replicate the human brain in the electrical domain is that we've seen impossible-to-understand systems tamed and replicated before: transistors.

Transistors are bundles of impossibility, where quantum mechanics are needed to even understand the basic operating principles, and to really design them, one must confront problems of such awe-inspiring complexity that it's hardly a surprise we are often the ones who go a bit mad.

And yet every day, trillions of these impossible boxes are instantiated, reproduced, copied, pasted, and spun like it was nothing. Unconsciously. The impossible box becomes a normalcy. And if we can do that, they imagine, why not the same with a similar system made out of gels and fibers?

Thursday, May 13, 2010

Eternal Growth vs Sustainability

These seem to be the two big camps on the Future and they both seem a little loopy to me.

Eternal Growthers seem to think that we can just trust the natural order of things to deliver us 3% growth forever regardless of what we plan, or in the most extreme cases, only if we don't make plans.

Sustainability folks lay claim to being the "realists" but to me they seem way off base in their presumption that we can plan to do anything for more than 20 years. Has this been the case in the last 20 years? The 20 before that? Was the earlier 20 more or less fast-changing? Wrongo.

So what's the "right" answer? I think it's in recognizing that we're surrounded by irreducible complexity but at the same time taking responsibility as the species most likely to be able to figure out what to do. We need to understand that nothing we recognize as "normal" is what the Earth was like before we showed up. And we need to start measuring and caring for the biosphere as if it were something we owned and wanted to keep working, even if we can't find the user's manual.

The Technium and the Biosphere probably can't be isolated from each other. But we can examine the places where they touch and try to make them function better. And we need to do so without scapegoating one another or inventing cheaply manipulative metanarratives to drive poorly-executed anti-growth plans. The problems are mostly hard. Here are a few that don't seem so hard (at least from the standpoint of whether they're the right thing to do), though, just so I'm offering some policy rather than merely pontificating on how we should all think smarter:

* Stop fishing, in as large an area as possible. The oceans are so fished out that you actually catch more fish when you cordon off most of the fishing areas and allow the seas to recover.
* Get Fusion working. One of the reasons Fusion is "always 20 years away" is that we keep cutting back funding because it's always 20 years away. We know it works because everything already runs on a centralized fusion reactor. We just need to technologize it.
* Fix recycling. I mean REAL recycling. Get all the value out of every atom we productize. Technologies like 3D printers are the near edge of the fabrication revolution, and harnessing them as recyclers could put the trash problem to bed, permanently.
* Higher density living. Cram people together into cities, bulldoze suburbia, plant forests.

And of course any one of these things changes the others. Get fusion working and electric cars proliferate on supercheap electricity made from sea water. Does that save Suburbia? Maybe information tech kills it first. Would higher density living accelerate or slow down the recycling systems? We just can't know. Irreducible complexity. But we can't go backwards, and we can't pretend planning is poison. We have to try, one misstep at a time, to design a future that actually works.