Reasons Unbeknownst

My hope with this post is teach people how to view their CT scans with Mango, a free tool used by doctors. My research initially turned up OsiriX, software that can also do 3D renderings but it’s limited to those with a Mac. For the rest of us Mango is the best bet.

I recently blew out a tendon in my ankle and the doctor sent me to a medical imaging specialist. They stuck me in the high tech bagel dog and an hour later I left with a CD and an unhealthy dose of curiosity. The following guide will walk(limp) you through the process of getting the images viewable on a computer.

Download and install Mango. Make sure you get the 64 bit version and enable 64 bit support in the performance options if your OS supports it.
Open DICOM folder
Choose the location of the files.
If more than one series of images are displayed you’ll see them here.
Click Open
At this point you’ll be able to navigate through the image in 2d.

Here is my foot in 2d. You can zoom through the slices in Mango to get a look at pretty much any part you want.

To render and view it in 3D click:
Image, build surface
Click Options
Check use range
Unless you have a really fast computer and/or a lot of time use a small range like 550 to 600
Depending on your computer this might take a few minutes.
Play with the ranges to hilight the various types of tissue. This will depend on your specific images but bone was in the 300 to 700 range on my scans.

This is my foot rendered in 3d with the range set to show all tissue.

And here’s my foot rendered with the range adjusted so you can see bone, tendons, etc.

Apparently I got a pretty bad scan so your results may be better. I’ve only scratched the surface, this the result after 3 hours of tinkering. Most of the guides for Mango are geared towards doctors so hopefully this guide makes CT scans a bit more accessible.

If the developers enable CUDA or OpenCL support on this software it will be a beast. A full screen view that can output to 3D TVs / monitors would be nice.

A Facebook backlash is raging this week on the internet. Wired just wrote a damning piece on their questionable privacy practices and Reddit has had a few rants over the last week as well. People are calling for an alternative but there isn’t much talk yet about what it will look like. So I brewed some thick coffee and this fell out of my brain…

The Past and Possible Future of Social Networking (oversimplified):
Friendster ->
MySpace ->
Facebook ->
Open Social (Google, Yahoo!, Microsoft, etc.(better privacy) ->
Open Social (user hosted (total privacy))

My thesis is that you will have a consolidated online identity in a few years. Your email, voicemail, social networks, blogs, videos, and photos will be manageable through a single dashboard running software (open source and otherwise) based on open standards. So imagine adding someone to your phone and having them instantly appear as a Facebook friend. Except it won’t be Facebook and you’ll have total control over the privacy settings. There would no longer be a need for sites like LinkedIn and Facebook.

Google and Yahoo are only jumping on this bandwagon because there is a ton of money to be made in the short run if they can crush Facebook. In the longer run, everybody will have their own email and web servers (though they won’t know it or care) so Google and Yahoo will have no good way to really know what you’re up to.

Why open social is a good thing:

• Users will have complete control over privacy. You’ll never have to worry about Facebook CEO Zuckerberg changing privacy settings late one night and allowing your boss to see all of your embarrassing photos because he wants to hit his Q4 revenue target.
• Innovation – The Android app marketplace for phones is open and it is challenging Apple’s closed app marketplace. The same logic applies to social networking, think Farmville on steroids.
• File sharing (for better or for worse) will be deeply integrated into an open social system. Facebook would piss off too many friends in Hollywood so you’ll never see this on FB.
• Recommendation systems will improve as people feel comfortable putting in more information about themselves (think recipes, TV shows, news, music, movies, etc.).
• Privacy will be the basis for the entire system, not an afterthought required by pesky regulators.

The evidence that this is happening now :

• Google just hired most of the brains behind the open social based DiSo project.
• Yahoo bought Flickr (image sharing)
• Google bought Picassa (image sharing)
• Google is experimenting with Google Buzz (see video below)
• Google is a huge backer of open social.
• Yahoo recently added profile information and basic contact status updates (screenshot below):

Roadblocks to a transition to open social networks:

• Transferring Profile Data
• Facebook Momentum
• A lack of concern about privacy

Some interesting links:
The Future of DiSo

Google Buzz video:

Google just hired this guy (Chris Messina). This is from 2008.

Random Somewhat Unrelated Predictions / Thoughts:
Server hosting services will be provided by your wireless phone carrier but you won’t run servers on your cell phone (connectivity is too unreliable, battery life issues).

Your email address will look something like jane@JaneSmith.com

In 10 years email will work like cell phones. If you want to change cell carriers from Verizon to AT&T it’s no problem, you get to keep your number. With email, you’re generally stuck with @Yahoo.com or @Gmail forever. My address follows me around when I change blog hosting providers just like a cell. This flexibility is uncommon now only because it’s complex. When it’s common Google and Yahoo are going to lose huge revenue streams because they won’t know the content of your emails and will only be able to advertise on the pages of search results.

* I could write a whole post about how to enforce data sharing on a distributed system. This is going to be interesting and might look like anti piracy systems used on Blu-Ray and PC games of today.
* Google buys Canocial (for their brains not Ubuntu)
* The first successful desktop Linux will come from Google but it won’t be Chrome. It’ll be called Android++ to leverage the success/brand of their phone OS.
* Sites that will eventually become unnecessary – Netflix, Digg, LinkedIn
* TVs will come with Android built in (this is not a guess, it’s happening).

So let me now take a swing at how people will watch movies at home in 10 years:
Sit down on your couch
Turn on tv with your cell phone.
Navigate to the Android App based movie rentals and library.
Check for new recommendations (“Your friends liked” (open social))
Click on Iron Man VII
The movie is paid for and begins streaming from the movie studio’s servers instantly.
Notice the lack of need for Netflix here.
The movie will never exists on your hard drive or a plastic disc during the whole process.
“Buying” a movie will simply be a license to rent it an infinite number of times.
You’ll still be able to buy a box at the store but it will be similar to those pre-paid Itunes movie cards.

I have some assumptions about economics, finance, and trends but I don’t spend much time looking at how things would have to change to accommodate my weird predictions. So that’s what I’m writing about tonight… Tonight I make omelets.

My condensed take on the economic camps’ assumptions about our predicament: The Austrians assume job growth resulting from lower taxes will organically release us from the Great Recession. The Keynesian camp assumes deficit spending will save us. I think both are wrong in this case. The housing boom masked the fact that the middle class is shrinking due to automation and global wage arbitrage. Neither of which show any signs of abating.

Both camps assume that for 300 years the middle class will remain stable. They do that because their models (mainly Keynesian) get too complex unless you assume all things are equal. So the complexity of their models preclude the notion that demographics and common sense can come into play. This is probably why they missed the housing bubble. My take assumes accelerating societal change.

The Austrian school isn’t so burdened by math (they did see the housing bubble coming) but its practitioners are burdened by political assumptions about fairness and justice (Bastiat and Rand). The social consequences of a shrinking middle class aren’t a concern for most Libertarians. But in the long run you start to have serious problem with a society with no middle class (so I’ve heard).

In short, they Keynsians think they can avoid the inevitable and the Austrians just don’t give a damn. So nobody is writing much about the problem.

In the back of my head I’ve remained optimistic that the brains in Washington can transition society to a future where there is no middle class because of rampant productivity growth which will (not ironically) make increased wealth redistribution palatable. This rising productivity simultaneously solves the deficit problem. Then I read this unfortunate bit of research today from Julius Wilson via The Atlantic:

“A neighborhood in which people are poor but employed is different from a neighborhood in which many people are poor and jobless. Many of today’s problems in the inner-city ghetto neighborhoods—crime, family dissolution, welfare, low levels of social organization, and so on—are fundamentally a consequence of the disappearance of work. “

So there are two problems for the future. The first is dealing with the potentially outdated financial framework that assumes a healthy, stable middle class (think 30 year mortgages). And the second is the psychological ramifications of joblessness even with bountiful welfare benefits.

So politicians are going to have to look at how to make wealth redistribution work given the understanding that you can’t just throw money at the jobless and expect them to avoid dysfunction.

Woohoo! I’m writing again. It feels good. (and cooking, mostly high end omelets). My thoughts from the last week:

“In this case it’s the application of reason to the study of the financial
behavior of irrational, social, creatures (humans). AKA economics. ”

“20 years ago we could blame the lack of information for our inability to fix the world’s problems. Now we’re the bottleneck.”

On CA:
“Our taxes are high but they’re not nearly high enough to pay for all of the earth fixing programs we like to “fund”.

“California is basically a giant version of an idealistic student at Berkeley with a Visa who has never taken a class on personal finance.”

“Don’t forget California. The Republicans are supposed to come in and cut taxes and spending but they find it’s a lot easier to just cut taxes.”

On wholistic medicine:
“It’s called the placebo effect and it does work but, like Peter Pan, only if you believe.”

On Keynesian misconceptions and fiscal shenanigans:
“Krugman (liberal) thinks Obama isn’t doing enough to stimulate the economy. Their prevailing wisdom is that you do need to worry about deficits but not until we’re through this recession. Otherwise we could get caught in a deflationary spiral (see Great Depression).
A deflationary spiral would collapse tax receipts and would grow the deficit. This is why unfunded spending and low GDP both have a negative impact on the dollar.”

“Giving away mortgages on crack houses only makes them crack homes.”

On the moon:
“Some of us Luddites think we should pay down the deficit before space travel becomes a priority.”

“Baby boomers already saw a guy on the moon. At this point in their lives they’re going to vote for more drug benefits. The baby boomers get what they want.”

On Socialism:
“Socialism is basically the Apple model. Things cost more but you get a strong leader (Jobs) and nationalistic pride (the Apple brand/ lifestyle), and you don’t have to worry about evil monopolies because you assume the government monopoly is benevolent.

Some of us prefer an open model with fierce competition. Brands aren’t as important because if companies are dropping prices fast and innovating faster, nobody will remain king of the hill. This is Capitalism but it requires regulation.

Bill Gates was completely wrong when he likened Open Source to Communism. Socialism consists of government run monopolies. Bill Gates as a politician in other words.

Socialism is basically the admission that the lobbyists have won, no regulation will pass, and it’s time to pray for competent bureaucrats. We Americans have trouble admitting that, probably given our history of occasionally tarring and feathering corrupt politicians.

Though I think we should take from the rich and give to the poor (within reason, wealth divides are growing) it doesn’t mean the government has to expand out from the department of motor vehicles and start running our supermarkets. If you don’t see the problems inherent in letting the government run businesses then you’ll probably like Socialism.”

On law:
“If regulation is just legislation and you assume laws are enforced then leadership should only consist of representatives. A Representative Democracy is arguably the antithesis of strong leadership.”

On Toyota:
“I think you should engineer safe gas pedals rather than change the behavior of the throttle system.”

“Toyotal Recall” My highest ranked comment this week but it looks like some headline writers at Fox news thought it up a few days before me.

On dog lovers eaten by their own dogs:
“You would expect man hungry dogs to eat a man so that’s not ironic but the fact that he is the supposed leader of their wolf-pack is.”

On Graphine based processors:
“I doubt that other than supercomputers the effort will be in reaching that level of performance. For most of us they’ll try to slow them down to only what is required for the application to reduce power consumption. Cell phones lasting a week on a charge would be nice.”

“Dixon Ticonderoga Number Two processors.” My second highest rated thought for the week according to Reddit.

” I think the concept of “good enough” may start to apply in the near future.
I would rather have a cell phone or PC that uses 5 watts than something that requires 150 but can render full length Pixar movies.
GPUs will probably jump on this for performance but it’s hard to argue many people are dissatisfied with current top of the line graphics.”

“Watch your cpu use, even during compiling, a lot of the bottleneck these days is IO. Most people are surprised what their CPU can do after installing a good SSD.”

A funny thing is happening to TVs; they’re getting interesting again. Prices of computers smart enough to do all the basics of what you get on a MacBook or Dell are making their way into the guts of TVs. Other than 3D, the interesting topics this year were the Boxee Box and the various apps available for TVs.

We have all sorts of technology on smart phones and desktop apps but the vast majority of citizens will not use tech until it’s so easy to use that it doesn’t risk offending one’s sense of control of their environment. These new TVs are interesting because the apps aren’t revolutionary, their accessibility is.

Once the technology is transparent the really interesting phase of this shift begins. I think Boxee is the first real vision of “TV” post cable box. You can’t buy it yet so I installed in on the PC plugged into my TV to get a feel for it. You have your standard media center options (movie, music libraries, tv shows) but you also get apps. This is a huge leap.

Right now I’m sitting on my couch writing this on my TV with a wireless keyboard running Boxee. I’m running the Boxee LastFM music app and it just occurred to me that I haven’t sat down at my laptop in a couple of days now. Maybe my couch is just too damn comfortable but I suspect it’s my new found freedom to get stuff done without feeling like I’m at the office.

Shows are the new channels if this new TV series interface sticks:

The only reason I pay for cable at this point is because I occasionally watch live sports. I’m seriously starting to wonder if it’s worth the $600 a year. If these TVs take off I fully expect Apple to create their own big screen TV with proprietary apps tied to the ITunes store for rentals, etc. Time will tell.

Asia Times is one of the best sites for articles on global economics but I have a beef with an assumption by one of its authors. I see this a lot and I think it’s just wrong:
“Were the adverse effects of excessive money creation and huge budget deficits not to appear at all, much of economic theory would have to be rewritten. It would mean that excess money creation could magically disappear, leading to no real world effects.”

If you assume that no change (currency stability) means no effect then yes, they’d have to rewrite economic theory. But if you assume the world is changing then you have to consider the possibility that incredibly loose monetary policy might offset natural deflationary forces. In other words, no change means something strange is going on.

In this case the pain felt by society would be the high prices relative to what they should be paying. So perhaps without all of the money printing gas would be $1 a gallon. We don’t see the fact that we’re paying $3 as a problem because we’re used to it. But that’s a real impact on society when jobs are vanishing.

If that is in fact happening it’s the perfect tax because standard of living goes down due to job loss, furloughs, etc. while prices are stable. The alternative is prices rising when wages and employment are stable. Scenario one is no less painful but it’s harder to point the finger because job losses probably feel like more of a personal failure than $6 gas.

Photo by Iveta on Flickr.

The economist takes a stab at the robot debate but like most of the commenters I was left wanting more…

“What policy makers need to worry about is the impact of autonomous technology in a world of seven billion people where nearly all of them do non-mental tasks to earn a living. Robotic engineers, graphic designers, architects, etc… are a long way off. Robots that can negotiate a light assembly environment and put the cell phone in a box along with a bunch of pieces of paper are maybe a decade or two away. They will also be able to assemble the cell phone, or sew the shirt, or pick the fruit, or work behind the McDonalds counter or stock the store shelves. The Industrial Revolution displaced massive numbers of workers. But they were displaced into new jobs that machines could not do. A robot that can do anything a low to medium skilled worker can do, can do any new low to medium skilled job. There are no new jobs for the displaced that can’t be automated. Western Europe, Japan and the US have had a huge number of these jobs go to China, Vietnam, etc… These Western economies will actually gain as automated light industrial and textile factories return. A robot costs the same to operate in China as Glasgow, but shipping from Glasgow to London is considerably less. But what happens to China and the rest of the world? All those workers who left the farm have no intention of going back to the farm. What will they do?

Is it really progress if GDP doubles while the unemployment rate skyrockets? What if the standard of living rises but only in the average, not median sense? I’m still clinging to the belief that this recession and joblessness are not solely the result of a Chinese savings glut and subsequent credit bubble collapse. They are also the result of a people not able to adapt quickly enough to a (quasi) capitalist economy that is now subject to something like Moore’s law.

There is a strong egalitarian slant to our thinking as Americans. I don’t know how we’re going to reconcile that belief with increasing unemployment. There is a fear of robots replacing factory workers but I work on software that automates dozens of jobs in the name of improved healthcare deliver systems. No mechanical arms are necessary to obsolete most jobs in America. Maybe in the future we’ll have PDF pushers instead of paper pushers but it seems unlikely.

I’m intently watching this play out. How web based media adapts to new economic realities will shape politics going forward. Media may not be controlled by a handful of media corporations in the future but it certainly will be shaped by media figureheads (populisits) and the angst of the masses.

This recession (which I tend to believe is the first real affront to the status-quo by creative destruction) is far from over in my opinion but I sincerely do hope I’m wrong. If the stock market continues to rise then my worldview is going to need a major overhaul.

[Edit: I just got a bunch of traffic from Nvidia and Sandia labs so I'm thinking I may be on to something here]

So I was sitting here thinking about how modern RAID cards are completely inadequate when it comes to solid state hard drives (as I often do), and I had an idea. Why can’t graphics card hardware be used to accelerate software RAID? We have CUDA right?

For a second I thought I was onto something truly original but after a quick search it looks like some guys over at Sandia Labs beat me to the punch by a few months:

“One example is the RAID software developed by researchers at the University of Alabama and Sandia National Laboratory that transforms CUDA-enabled GPUs into high-performance RAID accelerators that calculate Reed-Solomon codes in real time for high-throughput disk subsystems (according to “Accelerating Reed-Solomon Coding in RAID Systems with GPUs” by Matthew Curry, Lee Ward, Tony Skjellum and Ron Brightwell, IPDPS 2008). From their abstract, “Performance results show that the GPU can outperform a modern CPU on this problem by an order of magnitude and also confirm that a GPU can be used to support a system with at least three parity disks with no performance penalty.” I’ll bet the new NVIDIA hardware will perform even better. My guess is we will see a CUDA-enhanced Linux md (multiple device or software RAID) driver in the near future.”

The problem for using this sort of tech in the consumer space is that people want to boot from their ridiculously fast RAID arrays and software RAID makes that difficult (though not impossible with Linux if you don’t RAID the kernel). If you’re using a software RAID array and the OS crashes you can lose an array.

Here’s a link to the actual paper.

Just had another idea. If a RAID driver could check for CUDA support and only use it if present then an OS could boot in a slower CPU RAID setup until the graphics drivers(CUDA) loaded at which point things would get a lot snappier.

In the future I’m going to guess we’ll see hard drives on PCI-E cards with ONFI flash chips and NO SATA PORTS. With RAID handled by some standardized version of CUDA. Operating systems will have ONFI drivers baked in. This would effectively collapse the hardware RAID industry as we know it.

SLC NAND is only going to get cheaper. I imagine we’ll see 100GB drives containing 200GBytes of NAND. 150GB for a RAID 5 array and 50GB for wear leveling. At some point diminishing returns kick it. Are we really going to need three gigabytes per second of bandwidth from our drives to open word documents?

Wikipedia on
CUDA:
http://en.wikipedia.org/wiki/CUDA

ONFI:
http://en.wikipedia.org/wiki/Open_NAND_Flash_Interface_Working_Group

I have a system of beliefs I use to view the world and it has served me pretty well in making predictions but it has a big hole. So I drive around in my theory without a complete understanding of the engine, but it gets me from point a to point b effectively.

I’m confused by the fact that something that is not really a need can have value. We pay people to make us laugh but you can’t survive on laughter despite that famous quote about medicine. I have a theory and I’m going to use ‘laughter’ in place of anything intangible that we want but don’t need that we might pay money for.

Our ability to pay for laughter is equal to GDP (Gross Domestic Product) minus something I’ll call NTNP. Necessary tangible national product. NTNP is the economic value of the production of things like food, housing, transportation, clothing. Basic necessities which are the foundation of the economy in the 3rd world but that also exist in the 1st world. The NTNP as a percentage of GDP is probably a lot higher in places like Ethiopia where people struggle to survive and don’t spend $100 a month on HBO.

So in my thought experiment GDP = NTNP + EMNP. EMNP is the reason Eddie Murphy is a fabulously wealthy comedian, it’s the value of everything produced people buy after they’ve paid for necessities. Funny people in Ethiopia probably aren’t very wealthy because in Ethiopia they eat tomatoes, we throw them.

For the sake of argument
US economy: 20% NTNP and 80% EMNP.
Mexico: 50% NTNP and 50% EMNP.
Ethiopia: 90% NTNP and 10% EMNP.

The interesting point is that even if you’re having a bad year in Ethiopia the economy will never collapse due to a lack of confidence like it can in America. Eddie Murphy National Product on the other hand is ephemeral, subject to the whims of collective grumpiness.

One of my beliefs is that accelerating technological change is squashing the middle class. When everybody is driving electric cars in ten years there will be a million or so mechanics out of work because electric cars are so much easier to maintain. That’s just one example. The newspaper business is on the brink of collapse due to the Internet, etc.

I’m going to coin a new term to describe what our middle class is facing. Schumpeter came up with creative destruction. Kurzweil has accelerating change. Mash them together and you get the ugly term Accelerating Destruction.

EMNP will continue to shrink in the future as the rich get richer. Why? Because wealthy people only have so much time to go to the movies and Olive Garden. But those businesses employ a heck of a lot of middle class people.

We need to learn to live with deflation. The number one criticism of deflation is that it strangles the economy because people stop spending and wait for things to get cheaper in the future. If that was really a problem nobody would ever buy a laptop. Fear of deflation is probably due to the knowledge that our bubble economy sat precariously atop debt funded pillar of happy thoughts.

I get the sense that policy makers know that confidence is a prerequisite (hence the blind optimism) but they don’t seem to get that the confidence was paid for with unsustainable debt.

Wealth redistribution is not technically fair in my opinion (Bastiat) but it may be the only way to prevent the complete collapse of capitalism given accelerating destruction. Someone has to keep the people paying for laughter.

Three thoughts on solid state drives given the new realities/products out there today.

There are rubmlings of the inadequacy of the 300MB/s limit on SATA 2.0 due to the ridonkulous speed of SSDs. While it’s true that some SSDs are hitting the limit during sustained reads, random writes and even random reads are typically nowhere near the limits of the SATA bus.

The point? Focus on the bottleneck. The bottleneck is MLC flash memory not the bus. The solution? SLC flash will give better performance for many uses on a PATA bus that you’ll get from SATA2/MLC. SLC is more expensive but at some point the price will drop make MLC obsolete . The other benefit of SLC is that it tends not to stutter like MLC.

Older laptops with SSD upgrades are the new Netbook. A lot of the older laptops use the PATA interface (instead of SATA) but it’s really not much slower than the original SATA 1 specification. Some more info here.

And finally, OCZ just released their Z-Drive. It’s a PCI Express based drive with pretty good performance but it just feels poorly engineered based on my reading. Instead of creating a new chip to handle more channels they’ve just created a power sucking beast comprised of at least three RAID controllers.

An ideal solution would be
Memory -> Controller -> PCI-E

OCZ appears to have created:
Memory -> RAID controller -> Controller -> PCI-E
Memory -> RAID controller