What’s Wrong With the Affordable Care Act

The Affordable Care Act, also known by many as Obamacare really set this country’s healthcare system back in a major way. Let me try to explain why I have such an issue with it.

The biggest problem is that it is not a single-payer (“government-funded”) healthcare system like in every other first world country, not to mention very many third world countries too. This is the only correct system. If you believe our system of health insurance is more or less functioning properly and nicely and efficiently and providing the best bang for our personal and government-contributed bucks, you are utterly misinformed.

Working in healthcare IT for many years has given me a small glimpse into the madhouse of medical billing in America. The system is fucked. Ask anyone in healthcare and they’ll tell you the same thing. It’s all dumb. The thousands upon thousands of different insurance plans, the multitude of types (government, HMO, PPO, Discount Card, Indemnity, POS, EPO, Medicare, Medicare Advantage, Medi-Cal, Medicaid, Premier, Worker’s Compensation…) have different rules about reimbursement and fracture the American people’s ability to negotiate good deals on drugs and services. There is wild variation in prices due to an utter lack of transparency. The scheme of employers providing healthcare for their workers makes no logical sense, hurts competitiveness and massively screws over anyone who loses their job, unless they want to pay COBRA to keep their coverage (now around $1000/mo for some). Medical bills are submitted to third parties on all sorts of different paper forms, often faxed around. Oh yes, faxes are considered state of the art when it comes to medical billing and health insurance companies. In short, the current system would only be considered acceptable by anyone who has no idea how much better pretty much everyone in the developed world has it than us. This is reflected by popular opinion, as of 2016, showing a majority of Americans just want a normal federal-funded single-payer healthcare system. It’s the obvious solution, everybody knows it. So why don’t we have it?

Well, we have this health insurance scheme instead. Instead of the government providing basic healthcare to everyone it only provides healthcare to some people through a bewildering array of disparate systems. Poor people, people with kids, poor people in California, seniors, veterans, congresscritters, that sort of thing. If you don’t fit into one of those you can buy healthcare insurance instead of maybe get it provided by your employer if you’re lucky. A plan can cost several hundred dollars a month and may or may not include vision and dental insurance too. There is a provision of the ACA that says everyone must have health insurance or face a steep tax penalty (2.5% of your total household adjusted gross income). As a result, everyone must have health insurance.

Now I, just a lowly taxpayer, wonder if everyone has insurance, what exactly are the insurance companies doing? What benefit do we as a society gain from a universal system of insurance? Well, supposedly the health insurance companies price risk appropriately and disburse funds for claims.

Probably the main indicator of your likelihood of filing a lot of expensive claims are your pre-existing conditions. Like if you have cancer or some rare disease, you’re going to cost a lot of money. Before the ACA this would make it very hard to get normal health insurance because you would be a terrible policyholder from the insurance company’s point of view. One of the truly fantastic things, and at the same time one of the most subtly problematic things in the ACA is that it disallows pre-existing conditions from driving up your premiums or being denied coverage. This is actually really great news for seriously ill people, and it deserves being said that it has given (relatively) reasonably-priced healthcare to millions of people who need it the most. The end result appears to be a major net positive for society, and it really is. Taken in a vacuum this is a noble and wonderful thing. But it also undermines the entire argument for the extant system.

Now if everyone has insurance, and the insurance companies are not allowed to properly price risk, what function do they provide? If they cannot perform any real actuarial purpose then they are essentially clogging the arteries of our healthcare system, siphoning off vast quantities of money, wasting everyone’s time and collecting a massive payoff while decimating the quality of our healthcare.

Healthcare providers just want to care for people and, well, provide health care. That’s why they got into it and that’s what makes them put up with the years of school and massive debt and internships and all the rest. Once they actually try to practice medicine though they are faced with the intractable and thankless task of trying to deal with health insurance companies. Filing claims, getting denied, re-filing claims, telling patients you can’t see them because their employer went with the wrong insurance company or plan type, telling seniors they can’t get the medicine they need, buying ink for your fax machine, dealing with your medical billing intermediary company, upgrading to ICD-10, and on and on. This takes time and costs a great deal of money. It is not good for anyone except for the health insurance companies because it gives them a reason to exist.

The potential saving in efficiency, the massive gain in time of providers and nurses and office managers, the insanely powerful bargaining power, the standardization of billing forms and reimbursements would be some of the incredible benefits of having a coherent national health system. My problem with the ACA is that it cements this horrible mess of insurance providers. Its smaller improvement in expanding coverage means that we can pretend our system works for longer now instead of getting the sane and efficient system that we desperately need. That is why I have a problem with it. It powers the vacuum machine stuck in our pockets by health insurance companies. It entrenches the current corrupt system. It makes it that much harder for us to right it. The current shit abyss of healthcare probably resembles that of our legal system. It’s something most of us healthy people don’t have to get too acquainted with, so we don’t demand much change because it doesn’t affect us yet. But when you fall into this pit you’re going to wonder why our country is so fucked up. Even having health insurance won’t protect you from bankruptcy, of which health care is now the number one cause in America.

If a majority of Americans want a single-payer system, and we’re the only country that hasn’t gotten it yet, why do we still have this joke system of rent-seeking insurance companies taking their fat slice of the $2.9 trillion dollars a year we spend on health care (projected to shoot to $5.2 trillion in 2023)? Well, for one thing the opinions of the bottom 50-70% or so of Americans have no discernible impact on policy that gets made at the federal level. As in, the majority of Americans are literally disenfranchised because politicians do simply do not care what they think.

In 2007 health insurance companies were major donors to Obama’s and Hillary Clinton’s campaigns to become president. In the 2012 campaigns the insurance industry (with Blue Cross as the largest individual) contributed “a record $58.7 million to federal parties and candidates as well as outside spending groups” according to OpenSecrets. Draw your own conclusions.

MarketWatch has a bunch more fun facts about specific major problems with our health insurance system.

projectM: Open-Source Music Visualization

If you remember the old windows music player Winamp, it came with an amazing visualizer named Milkdrop written by a guy at nVidia named Geiss. This plugin performed beat detection and splitting the music into frequency buckets with an FFT and then fed that info into a randomly-selected “preset.” The presets are equations and parameters controlling waveform equations, colors, shapes, shaders,”per-pixel” equations (not actually per-screen-pixel, rather a smaller mesh that is interpolated) and more.

Most of the preset files have ridiculous names like:

  • “suksma + aderassi geiss – the sick assumptions you make about my car [shifter’s esc shader] nz+.milk”
  • “lit claw (explorers grid) – i don’t have either a belfry or bats bitch.milk”
  • “Eo.S. + Phat – chasers 12 sentinel Daemon – mash0000 – multi-band time-distortion aurora granules.milk”
  • “Goody + martin – crystal palace – Schizotoxin – The Wild Iris Bloom – mess2 nz+ i have no character and feel entitled to one.milk”

Milkdrop was originally only for windows and was not open-source, so a few very smart folks got together and re-implemented Milkdrop in C++ under the LGPL license. The project created plugins to visualize Winamp, XMMS, iTunes, Jack, Pulseaudio, ALSA audio. Pretty awesome stuff.

This was a while ago, but recently I wanted to try it out on OSX. I quickly realized that the original iTunes plugin code was out of date by about 10 major versions and wasn’t even remotely interested in compiling, not to mention lacking a bunch of dependencies built for OSX.

So I went ahead and updated the iTunes plugin code, mostly in a zany language called Objective-C++ which combines C++ and Objective-C. It’s a little messed up but I guess it works for this particular case. I grabbed the dependencies and built them by hand, including static versions for OSX in the repository to make it much easier for others to build it (and myself).

Getting it to build was no small feat either. Someone made the unfortunate decision to use cmake instead of autotools. I can understand the hope and desire to use something better than autotools, but cmake ain’t it. Everything is written in some ungodly undocumented DSL that is unlike any other language you’ve used and it makes a giant mess all over your project folders like an un-housebroken puppy fed a laxative. I have great hope that the new Meson build system will be awesome and let us all put these miserable systems out to pasture. We’ll see.

Screen Shot 2016-08-02 at 9.59.55 PM.png
cmake – not even once

Long story short after a bunch of wrangling I got this all building as a native OSX iTunes plugin. With a bit of tweaking and tossing in the nVidia Cg library I got the quality and rendering speed to be top-notch and was able to reduce the latency between the audio and rendering, although I think there’s still a few frames of delay I’d like to figure out how to reduce.

I wanted to share my plugin with Mac users, so I tried putting it in the Mac App Store. What resulted was a big fat rejection from Apple because I guess they don’t want to release plugins via the app store. You can read about those travails here. I think that unpleasant experience is what got me to start this blog so I could publicly announce my extreme displeasure with Apple’s policies towards developers trying to contribute to their ecosystem.

After trying and failing to release via the app store I put the plugin up on my GitHub, along with a bunch of the improvements I made. I forked the SourceForge version, because SourceForge can go wither and die for all I care.

I ended up trying to get it running in a web page with Emscripten and on an embedded linux device (raspberry pi). Both of these efforts required getting it to compile with the embedded spec for OpenGL, GLES. Mostly I accomplished this by #ifdef’ing out immediate-mode GL calls like glRect(). After a lot more ferocious battling with cmake I got it running in SDL2 on Linux on a Raspberry Pi. Except it goes about 1/5fps, lol. Need to spend some time profiling to see if that can be sped up.

I also contacted a couple of the previous developers and the maintainers on SourceForge. They were helpful and gave me commit access to SF, one said he was hoarding his GLES modifications for the iOS and Android versions. Fair enough I guess.

Now we’re going to try fully getting rid of the crufty old SourceForge repo, moving everything to GitHub. We got a snazzy new GitHub homepage and even our first pull request!

My future dreams for this project would be to make an embedded Linux device that has an audio input jack and outputs visualizations via HDMI, possibly a raspberry pi, maybe something beefier. Apparently some crazy mad genius implemented this mostly in a FPGA but has stopped producing the boards, I don’t know if I’m hardcore enough to go that route. Probably not.

In conclusion it’s been nice to be able to take a nifty library and update it, improve it, put out a release that people can use and enjoy, and work with other contributors to make software for making pretty animations out of music. Hopefully with our fresh new homepage and an official GitHub repo we will start getting more contributors.

I recorded a crappy demo video. The actual visualizer is going 60fps and looks very smooth, but the desktop video recorder I used failed to capture at this rate so it looks really jumpy. It’s not actually like that.

Concerning Attribution of Hacking

Organizations are getting hacked left and right these days, and that’s just what’s in the news. Naturally most organizations and people that get hacked either don’t know it or don’t want to tell anybody. It should be no surprise to anyone that the DNC’s emails got leaked. As is depressingly routine though, the news coverage is both sensationalist and lacking in depth. In this case I’m specifically referring to all of the media saying that the Russians did this with big pictures of Dr. Evil Vladimir Putin to go along with them.

Maybe they have sweet refrigerator magnets like me
Maybe they have sweet refrigerator magnets like me

I’m not saying that some Russian people didn’t do it, but I really feel like the reporting on the matter is irresponsible. Here’s why:

Attribution of hacks to individuals or nation-states is hard. It’s very hard for many reasons, not the least of which is that techniques, tools, attacks and compromised machines are shared. Someone in Bolivia may be using a Russian tool from a already trojaned host in China, connecting from a Romanian proxy. In the past many attributions have been made on extremely flimsy evidence, like seeing some Russian strings in a file and then saying it is the work of Kremlin-sponsored hackers. Or coming from a Chinese IP therefore it’s the CPC (e.g.: Norse-style attack maps). Or a machine that has been compromised by a trojan believed to be used by a couple people in Russia, even though others could be using the machine or the trojan.

My point being that there should be some basic level of skepticism from the public and reporters when attributing hacks at all, and maybe even more when connecting them to nation-state sponsored hacking. The Economist very recently said regarding hacking financial institutions:

the limited number of actors thought to have the capabilities to pull off something like this are tied to nation-states

I’m definitely no expert on financial information security, but I doubt the basic premise here that hacking techniques or tools can exist only in the hands of nation-states. Anything can be copied, especially an attack that’s been used before. Suppose a nation-state has a super sweet 0-day and trojan kit or whatever. Once they use it, it’s fair game for other people to replicate and use themselves. Case in point: stuxnet, the SCADA attack that wrecked Iranian nuclear enrichment centrifuges. This is suspected to be developed by the Israeli and American governments using highly specialized knowledge. Cool. But now a detailed analysis of it is on the internet for anyone to read and copy if they want. So even if a lot of work is put into development by a nation-state, others can copy it. And in most cases a person with a lot of time on their hands and a computer could do the same research and development if so inclined.

Now regarding the DNC hacking, all news articles eventually point back to a single press release by one guy at CrowdStrike that says the attack was done by “COZY BEAR and FANCY BEAR.” whom they know to be sophisticated Russian operatives. This is a pretty important assertion and if it was to be printed everywhere with scary pictures of Putin and likely lead to diplomatic responses I would expect more evidence behind it than essentially taking their word for it. Again, I’m not saying I don’t believe them, but not being properly skeptical about such assertions and considering plausible alternatives could lead to very serious consequences that would be in everyone’s best interest to avoid.

We have identified no collaboration between the two actors, or even an awareness of one by the other.  Instead, we observed the two Russian espionage groups compromise the same systems and engage separately in the theft of identical credentials. While you would virtually never see Western intelligence agencies going after the same target without de-confliction for fear of compromising each other’s operations, in Russia this is not an uncommon scenario. “Putin’s Hydra: Inside Russia’s Intelligence Services”

To me this says that people with no connection to each other are using the same tools to compromise the same systems. Why must it be only these two attackers? How do you know it’s them? If there are detailed reports on the tools, techniques and traces of these attacks how hard would it be for someone else to make it appear to be the work of Russians?

These are important questions because there are debates now on how to respond to hacks (or “cyberattacks” in the somewhat anachronistic terminology of the U.S. government), possibly with military force. In the past high-level American diplomats have warned China about their cybertanks rolling over our cyberservers or whatever the hell they imagine is going on. China denies any such attempts. Maybe China attacked us, maybe not? Let’s be as certain as we can before rushing to any conclusions about who attacked a computer through the internet, and who “sponsored” them. I don’t really know what sponsorship a hacker needs other than a few cases of monster, some pizzas and a laptop.


Typical Kremlin-sponsored nation-state operative workstation


There are definitely some points to mention that do back up the assertions from CrowdStrike. Dell Secure Works supposedly verified that it was Russians independently. In the press release it certainly does sound like CrowdStrike knows what they’re talking about and has been following these guys for a while. I’m sure they have lots more information than they’ve released and know a lot more than I do. I just have a problem taking their word for it about attributing it to the Russian Federation government when such attribution seems extremely problematic and pretty impossible to confirm unless you actually arrest them and look at their computers and network traffic. Of course if we could track them down to that point, it would mean that they were shitty hackers and the work they did could have been done by any other shitty hacker just as easily. And if these shady Russkiis have been at it for so long and are so well known, what’s to stop China or Venezuela or Iran from sponsoring hackers to imitate the Russian’s attacks to stir up some diplomatic incidents and nationalist fervor?

Oh also, the person who hacked the DNC started a blog and said they were responsible and it wasn’t Russians and laughed at CrowdStrike’s incompetence:

Worldwide known cyber security company CrowdStrike announced that the Democratic National Committee (DNC) servers had been hacked by “sophisticated” hacker groups.

I’m very pleased the company appreciated my skills so highly))) But in fact, it was easy, very easy.

Guccifer may have been the first one who penetrated Hillary Clinton’s and other Democrats’ mail servers. But he certainly wasn’t the last. No wonder any other hacker could easily get access to the DNC’s servers.

Shame on CrowdStrike: Do you think I’ve been in the DNC’s networks for almost a year and saved only 2 documents? Do you really believe it?

So there’s that. CrowdStrike posted an update responding to Guccifer 2.0:

June 15, 2016 UPDATE:

CrowdStrike stands fully by its analysis and findings identifying two separate Russian intelligence-affiliated adversaries present in the DNC network in May 2016. On June 15, 2016 a blog post to a WordPress site authored by an individual using the moniker Guccifer 2.0 claiming credit for breaching the Democratic National Committee. This blog post presents documents alleged to have originated from the DNC.

Whether or not this posting is part of a Russian Intelligence disinformation campaign, we are exploring the documents’ authenticity and origin. Regardless, these claims do nothing to lessen our findings relating to the Russian government’s involvement, portions of which we have documented for the public and the greater security community.

Now I’m not an information security professional and I’m not claiming to know more than CrowdStrike or American diplomats or anything like that. I’m just a software engineer who tries to keep up on security issues so that I can better protect my systems and applications. All I’m saying is that hard questions should be asked when attempting to attribute a hack to a particular person, group or nation-state before plastering the news with headlines like “Why Would Vladimir Putin Want To Leak The DNC Emails?

Screen Shot 2016-07-26 at 1.08.16 PM


AWS Lambda Editor Plugin for Sublime Text

Editing the source of a lambda procedure in AWS can be very cumbersome. Logging in with two-factor authentication and then selecting your lambda and using their web-based “IDE” with nested scroll bars going on on the page is not the greatest. Even worse is if your function actually has dependencies! Then you cannot view the source on the web and must download a zip file, and re-zip and upload it every time you wish to make a change.

Naturally after a while of doing this I got pretty fed up so I created a handy plugin (documentation and source on GitHub) for my editor of choice these days, Sublime Text. After setting up your AWS access key if you haven’t done so already (it uses the awscli or boto config) and installing the plugin via the Sublime Package Manager, you can call up a list of lambdas from within your editor.

After selecting a lambda to edit, it downloads the zip (even if it wasn’t originally a zip), sticks it in a temporary directory and creates a sublime project for you. When you save any of the files it will automatically zip up the files in the project and update the function source automatically, as if you were editing a local file. Simplicity itself.

If you use AWS lambda and Sublime Text, get this plugin! It’ll save you a ton of time. Watch it in action:


Video instructions for installing the plugin from scratch:

Mac OS X El Capitan and OpenSSL Headers

Apple stopped including the OpenSSL development headers on recent versions of OSX, trying to get people to move away from the old 0.9.8 version that’s been deprecated for a very long time. Making people stop using this shared library is a Good Thing to be sure but you may come across older software that you want to build for yourself.

If you try to compile a newer version of OpenSSL you will likely find that programs will fail to build against more recent versions because a lot of data structures have been hidden. You may see errors such as:

error: variable has incomplete type 'EVP_PKEY' (aka 'struct evp_pkey_st')

        EVP_PKEY pk;


/usr/local/include/openssl/ossl_typ.h:92:16: note: forward declaration of 'struct evp_pkey_st'

typedef struct evp_pkey_st EVP_PKEY;

If you want to get such code to compile there’s a quick and easy solution! OSX still ships with the 0.9.8 library, you just need to provide the headers. Remove any newer versions of OpenSSL, grab the 0.9.8 sources, and copy over the headers:

$ sudo cp -r include/openssl /usr/local/include/

And then you’re all set.

Developing a cloud-based IoT service

In my previous post I describe my adventures in building an AWS IoT-enabled application for a proprietary embedded linux system and getting it to run. The next step in our journey is to create a service that communicates with our device and controls it in a useful way.

What can we do with a system running with the aws_iot library? We can use the MQTT message bus to subscribe to channels and publish messages, and we can diff the current device state against the desired device state shadow stored on the server. Now we need the service side of the puzzle.

My sample IoT application is to be able to view images on an IP camera from anywhere on the internet. I’m planning to incorporate live HD video streaming as well but that is a whole other can of worms we don’t need to open for this demonstration. My more modest goal for now will be to create a service where I can request a snapshot from the camera be uploaded to AWS’s Simple Storage Service (S3) which can store files and serve them up to authenticated users. In addition I will attempt to build the application server logic around AWS Lambda, a service for running code in response to events without actually having to deploy a server or run a daemon of any sort. If I can manage this then I will have a truly cloud-based service; one that does not consume any more resources than are required to perform its job and with no need to pre-allocate any servers or storage. It will be running entirely on Amazon’s infrastructure with only small bits of configuration, policy and code inserted in the right places to perform the relatively simple tasks required of my app. This is the Unemployed DevOps lifestyle, the dream of perfect lazy scalability and massive offloading of effort and operations to Amazon. There is of course a large downside to this setup, namely that I am at the mercy of Amazon. If they are missing a feature I need then I’m pretty much screwed and if their documentation is poor then I will suffer enormously. A partial description of my suffering and screwed state continues below.

I’ve been bitten before by my foolish impetuousness in attempting to use new AWS services that have clearly not been fully fleshed out. I was an early adopter of the CodeDeploy system, a super useful and nifty system for deploying changes to your application on EC2 instances from S3 or even straight from GitHub. Unfortunately it turned out to not really be finished or tested or documented and I ended up wasting a ton of time trying to make it work and deal with corner cases. It’s a dope service but it’s really painfully clear nobody at AWS has ever bothered to actually try using it for a real application, and all of my feature requests and bug reports and in-person sessions with AWS architects have all resulted in exactly zero improvements despite my hours of free QA I performed for them. As a result I am now more cautious when using new AWS services, such as IoT and Lambda.

In truth attempting to make use of the IoT services and client library has been one of the most frustrating and difficult uphill battles I’ve ever waged against a computer. The documentation is woefully incomplete, I’ve wasted tons of time guessing at what various parameters should be, most features don’t really behave as one would expect and the entire system is just super buggy and non-deterministic. Sometimes when I connect it just fails. Or when subscribing to MQTT topics.

Usually this doesn't happen. But sometimes it does!
Usually this doesn’t happen. But sometimes it does!

Why does it disconnect me every few seconds? I don’t know. I enabled autoReconnect (which is a function pointer on a struct unlike every other function) so it does reconnect at least, except when it just fails for no apparent reason.

setAutoReconnectStatus is only mentioned as being a typedef in the MQTT client documentation. One would assume you should call the function aws_iot_mqtt_autoreconnect_set_status(), but the sample code does indeed call the struct’s function pointer instead. No other part of the library uses this fakeo method call style.

On the boto3 (python AWS clienet library) side things are not really any better. The device shadow support (called IoT Dataplane) documentation is beyond unhelpful at least as of this writing. If you want to update a device state dictionary (its “shadow”) in python, say, in a lambda, you call the following method:

Usually when you want to specify a dictionary-type object as a param in python it’s customary to pass it around as a dict. It’s pretty unusual for an API that is expecting a dictionary data structure to expect you to already have encoded it as JSON, but whatever. What is really missing in this documentation is the precise structure of the update payload JSON string you’re supposed to pass in. You’re supposed to pass in the desired new state in the format {“state”: { “desired”: { … } } }:

My dumb lambda

If you hunt around from the documentation pages referenced by the update_thing_shadow() documentation you may uncover the correct incantation, though not on the page it links to. It would really save a lot of time if they just mentioned the desired format.

I really definitely have no reason why it wants a seekable object for the payload since it’s not like you can really send files around. I actually first attempted to send an image over the IoT message bus with no luck, until I realized that the biggest message that can ever be sent over it is 128k. This application would be infinitely simpler if I could transmit the image snapshot over my existing message bus but that would be too easy. I am fairly certain my embedded linux system can handle buffering many megabytes of data and my network is pretty solid, it’s really a shame that AWS is so resource-constrained!

The reason I am attempting to use the device shadow to communicate is that my current scheme for getting an image from the device into AWS in lieu of the message bus is:

  • The camera sends a MQTT message that indicates it is online
  • When the message is received, a DevicePolicy matches the MQTT topic and invokes a lambda
  • The lambda generates a presigned S3 request that will allow the client to upload a file to an S3 bucket
  • The lambda updates the device shadow with the request params
  • A device shadow delta callback on the camera is triggered (maybe once, maybe twice, maybe not at all, by my testing)
  • Callback receives the S3 request parameters and uploads the file via libcurl to S3
  • Can now display thumbnail to a web client from S3

I went to the AWS Loft to talk to an Amazon architect, a nice free service the company provides. He didn’t seem to know much about IoT, but he spoke with some other engineers there about my issues. He said there didn’t appear to be any way to tell what client sent a message, which kind of defeats the entire point of the extra security features, and he was going to file an internal ticket about that. As far as uploading a file greater than 128k, the above scheme was the best we could come up with.

Regarding the security, I still am completely at a loss as to how one is supposed to manage more than one device client at a time. You’re supposed to create a “device” or a “Thing”, which has a policy and unique certificate and keypair attached to it and its own device shadow state. I assume the keypair and device shadows are supposed to be associated with a single physical device, which means you will need to automate some sort of system that provisions all of this along with a unique ThingName and ClientID for each physical device and then include that in your configuration header and recompile your application. For each device, I guess? There is no mention of what exactly how provisioning is supposed to work when you have more than one device, and I kinda get the feeling nobody’s thought that far ahead. Further evidence in support of this theory is that SNS messages or lambdas that are invoked from device messages do not include any sort of authenticated ClientID or ThingName, so there’s no way to know where you are supposed to deliver your response. Right now I just have it hard-coded to my single Thing for testing. I give Amazon 10/10 for the strict certificate and keypair verification, but that’s only one part of a scheme that as far as I can tell has no mechanism for verifying the client’s identity when invoking server-side messages and code.

It wasn’t my intention to bag on AWS IoT, but after months of struggling to get essentially nowhere I am rather frustrated. I sincerely hope that it improves in usableness and stability because it does have a great deal of powerful functionality and I’d very much like to base my application on it. I’d be willing to help test and report issues as I have in the past, except that I can’t talk to support without going in to the loft in person or paying for a support plan, and the fact that all of my previous efforts at testing and bug reporting have added up to zero fixes or improvements doesn’t really motivate me either.

If I can get this device shadow delta callback to actually work like it’s supposed to I’ll post more updates as I progress. It may be slow going though. The code, such as it is, is here.


Diving into IoT development using AWS

I’m more allergic than most people to buzzwords. I cringe big time when companies suddenly start rebranding their products with the word “cloud” or tack on a “2.0”. That said, I realize that the cloud is not just computers in a datacenter and the Internet of Things isn’t all meaningless hype either. There exists a lot of cool new technology, miniaturization, super cheap hardware of all shapes and sizes and power requirements, ever more rapid prototyping and lot more that adds up to what looks like a new era in embedded system hardware.

People at the embedded linux conference can't wait to tell you about IoT stuff
People at the embedded linux conference can’t wait to tell you about IoT stuff

But what will drive this hardware? There is a lot of concern about the software that’s going to be running on these internet-connected gadgets because we all just know that the security on most of these things is going to be downright laughable, but now since they’re a part of your car, your baby monitor, your oven, your insulin pump and basically everything, this is gonna be a big problem.

So I’ve embarked on a project to try to build an IoT application properly and securely. I think it’ll be fun, a good learning experience, and even a useful product that I may be able to sell one day. At any rate it’s an interesting technical challenge.

My project is thus: to build a cloud-based IoT (ughhh sorry) IP camera for enterprise surveillance. It will be based on as much open source software as possible, ABRMS-licensed, mobile-first and capable of live streaming without any video transcoding.

I think I know how to do this, I’ve written a great deal of real-time streaming software in the past. I want to offload as much as the hard stuff as possible; let the hardware do all the h.264 encoding and let AWS manage all of the security, message queueing and device state tracking.

At the Dublin gstreamer conference I got to chat up an engineer from Axis, an awesome Swedish company that makes the finest IP cameras money can buy. He informed me that they have a new program called ACAP (Axis Camera Application Platform) which essentially lets you write what are essentially “apps” that are software packages that can be uploaded to their cameras. And they’re all running Linux! Sweet!

And recently I also learned of a new IoT service from Amazon AWS. I was dreading the humongo task of writing a whole new database-backed web application and APIs for tracking devices, API keys, device states, authentication, message queueing and all of that nonsense. Well it looks like the fine folks at Amazon already did all the hard work for me!

So I had my first development goal: create a simple AWS-IoT client and get it to run on an Axis camera.

Step one: get access to ACAP

Axis doesn’t really make it very easy to join their development program. None of their API documentation is public. I’m always very wary of companies that feel like they need to keep their interfaces a secret. What are you hiding? What are you afraid of? Seems like a really weird thing to be a control freak about. And it majorly discourages developers from playing around with your platform or knowing about what it can do.

But that is a small trifle compared to joining the program. I filled out a form requesting access to become a developer and was eventually rewarded with a salesbro emailing me that he was busy with meetings for the next week but could hop on a quick call with me to tell me about their program. I informed them that I already wanted to join the program and typed all the relevant words regarding my interest into their form and didn’t need to circle back with someone on a conference call in a few weeks’ time, but they were really insistent that they communicate words via telephone.

After Joe got to give me his spiel on the phone I got approved to join the Axis developer partner program. As far as ACAP they give you a SDK which you can also download as an Ubuntu VirtualBox image. Inside the SDK is a tutorial PDF, several cross-compiler toolchains, some shady Makefile includes, scripts for packaging your app up and some handy precompiled libraries for the various architectures.

Basically the deal is that they give you cross-compilers and an API for accessing bits of the camera’s functionality, things like image capture, event creation, super fancy storage API, built-in HTTP server CGI support, and even video capture (though support told me vidcap super jankity and I shouldn’t use it). The cross-compilers support Ambarella ARM, ARTPEC (a chip of Axis’s design) and some MIPS thing, these being the architectures used in various Axis products. They come with a few libraries all ready to link, including glib, RAPP (RAster Processing Primitives library) and fixmath. Lastly there’s a script that packages your app up, building a fat package for as many architectures as you want, making distribution super simple. Now all I had to do was figure out how to compile and make use of the IoT libraries with this build system.

Building mbedTLS and aws_iot

AWS has three SDKs for their IoT clients: Arduino Yún, node.js and embedded C linux platforms. The Arduino client does sound cool but that’s probably underpowered for doing realtime HD video, and I’m not really the biggest node.js fan. Linux embedded C development is where it is at, on the realz. This is the sort of thing I want to be doing with my life.

Hells yeah!

All that I needed to do was create a Makefile that builds the aws_iot client library and TLS support with the Axis toolchain bits. Piece of cake right? No, not really.

The IoT AWS service takes security very seriously, which is super awesome and they deserve props for forcing users to do things correctly: use TLS 1.2, include a server certificate and root CA cert with each device and give each device a private key. Wonderful! Maybe there is hope and the IoT future will not be a total ruinfest. The downside to this strict security of course is that it is an ultra pain in the ass to set up.

You are offered your choice of poison: OpenSSL or mbedTLS. I’d never heard of mbedTLS before but it looked like a nice little library that will get the job done that isn’t a giant bloated pain in the ass to build. OpenSSL has a lot of build issues I won’t go into here.

To set up your app you create a key and cert for a device and then load them up in your code:

 connectParams.pRootCALocation = rootCA;
 connectParams.pDeviceCertLocation = clientCRT;
 connectParams.pDevicePrivateKeyLocation = clientKey;

Simple enough. Only problem was that I was utterly confused by what these files were supposed to be. When you set up a certificate in the IoT web UI it gives you a public key, a private key and a certificate PEM. After a lot of dumbness and AWS support chatting we finally determined that rootCA referred to a secret CA file buried deep within the documentation and the public key was just a bonus file that you didn’t need to use. In case anyone else gets confused as fuck by this like I was you can grab the root CA file from here.

The AWS IoT C SDK (amazon web services internet of things C software development kit) comes with a few sample programs by way of documentation. They demonstrate connecting to the message queue and viewing and updating device shadows.

#define AWS_IOT_MQTT_HOST              "B13C0YHADOLYOV.iot.us-west-2.amazonaws.com" ///< Customer specific MQTT HOST. The same will be used for Thing Shadow                                                                                                                       
#define AWS_IOT_MQTT_PORT              8883 ///< default port for MQTT/S                                                                  
#define AWS_IOT_MQTT_CLIENT_ID         "MischaTest" ///< MQTT client ID should be unique for every device                                 
#define AWS_IOT_MY_THING_NAME          "MischaTest" ///< Thing Name of the Shadow this device is associated with                          
#define AWS_IOT_ROOT_CA_FILENAME       "root-ca.pem" ///< Root CA file name                                                               
#define AWS_IOT_CERTIFICATE_FILENAME   "1cd9c753bf-certificate.pem.crt" ///< device signed certificate file name                          
#define AWS_IOT_PRIVATE_KEY_FILENAME   "1cd9c753bf-private.pem.key" ///< Device private key filename                                      

To get it running you edit the config header file, copy your certificates and run make. Then you can run the program and see it connect and do stuff like send messages.


Once you’ve got a connection set up from your application to the IoT API you’re good to go. Kind of. Now that I had a simple C application building with the Axis ACAP SDK and a sample AWS IoT application building on linux, the next step was to combine them into the ultimo baller cloud-based camera software. This was not so easy.

Most of my efforts towards this were spent tweaking the Makefile to pull in the mbedTLS code, aws_iot code and my application code in a setup that would allow cross-compiling and some semblance of incremental building. I had to up my Make game considerably but in the end I was victorious. You can see the full Makefile in all its glory here.

The gist of it is that it performs the following steps:

  • loads ACAP make definitions (include $(AXIS_TOP_DIR)/tools/build/rules/common.mak)
  • sets logging level (LOG_FLAGS)
  • grab all the source and include files/dirs for mbedTLS and aws_iot
  • define a static library target for all of the aws_iot/mbedTLS code – Screen Shot 2016-03-20 at 2.19.16 PM
  • produce executable:
    Screen Shot 2016-03-20 at 8.39.58 PM.png

The advantage of creating aws-iot.a is that I can quickly build changes to my application source without having to re-link dozens of files.

I combined the Axis logging macros and the aws_iot style logging into one syslog-based system so that I can see the full output when the app is running on the device.

Uploading to the Camera

Once I finally had an ACAP application building I was finally able to try deploying it to a real camera (via make target of course):

Screen Shot 2016-03-20 at 2.18.58 PM

Screen Shot 2016-03-20 at 2.14.21 PM

Getting the app running on the camera and outputting useful logging took quite a bit of effort. I really ran into a brick wall with certificate verification however. My first problem was getting the certs into the package, which was just a simple config change. But then it began failing. Eventually I realized it was because the clock on the camera was not set correctly. Realizing the importance of a proper config, including NTP, I wrote a script to configure a new camera via the REST API. I wanted it to be as simple as possible to run so I wrote it without requiring any third party libraries. It also shares the package uploader config for the camera IP and password so if you’ve already entered it you don’t need to again.

With NTP configured at least there are no more certificate expired errors. I’m able to connect just fine on normal x86 linux, but fails to verify the certs when running on the camera. After asking support, they suggest recompiling mbedTLS with -O0 (disable optimizations) when building on ARM. After doing so, it connects and works!

Screen Shot 2016-03-20 at 2.14.51 PM

🌭🍕🍔 !!!!! Success!

To summarize; at this point we now have an embedded ARM camera device that is able to connect and communicate with the AWS IoT API securely. We can send and receive messages and device shadow states.

So what’s next? Now we need a service for the camera to talk to.