DEV Community

Cover image for APIs to humans - Curse or Blessing?
Jan Schenk (he/him) for Postman

Posted on

APIs to humans - Curse or Blessing?

I am a person with Type 1 Diabetes (T1D). Knowing certain up-to-date data about my body is essential for a tight schedule of multiple daily precise treatments. Hence my opinion on the question is, of course, biased. I consider myself a T1D Cyborg, as parts of my body functions have been replaced through technology. I wear a blood glucose sensor and an insulin pump 24/7. I let an algorithm on my phone decide whether to administer a hormone that could be lethal if incorrectly dosed. I benefit from both sensors (in this case, the CGM) and actors (the insulin pump) connected to my body, and being readable and writable through APIs. Regarding my chronic disease, my available human APIs are a blessing to me.

Recently, one of the CGM (continuous glucose monitor) manufacturer announced the availability of their sensors for sport. You could predict through monitoring blood glucose exactly when your body needs sugar to keep up performance and not fall into a dip triggered by low glucose levels. Another company dedicates its whole existence towards this hardware's API.

Your smartwatch knows your exact body functions, knows when you exercise, fall, sit too long (for whatever reason 💩) or are experiencing stress. All of this data is available through the respective apps or a central app on your platform, e.g. HealthKit on iOS.

It's definitely not a technical limitation to make the leap from sensors to actors. I did when I started using my insulin pump. People with a pace maker have an actor, that may be just as lethal as it is healthy. But these actors don't always need to be invasive to the body. You could think about skin patches for other hormones that reduce stress through dispensing a hormone, or maybe even only set free a certain fragrance that helps relief.

I understand people are ok with sensors, and shy away from actors in or on their body due to the invasiveness and potential security issues. There's a strong benefit for people with a (chronic) disease, and these patients are more willing to let technology invade their bodies if it helps ease the burden of their disease. For those people, security is often not an issue. But building trust would definitely help a general audience get more comfortable with the thought of using actors that could kill them.
(You could kill me if you knew the serial number of my insulin pump and did some reading. I don't spend too much time thinking about it. Just yet.)

How far away are we really away from human APIs in everyday life. And will it be a blessing? Or it it just another area of life that homo sapiens sapiens is trying to subdue? Please share your thoughts in the comments.

Discussion (4)

armousness profile image
Sean Williams

Chrysler added the ability to do over-the-air software updates in their cars (through the cellular network), and of course it got compromised. They've now locked the powertrain control system down so hard, you're even restricted from accessing components using a physically connected scan tool. To get anywhere now, you need to 1. have a scan tool with firmware that's compatible with their ludicrous security system, 2. connect that scan tool to the internet, and 3. subscribe to a service (for $50/year) to be able to get encryption keys.

The moral of this story is, I don't think it's a good idea to put your life in the hands of programmers, outside heavily regulated sectors like avionics. And even then, bad code killed a bunch of people.

This is also why I drive a '93 Miata.

jansche profile image
Jan Schenk (he/him) Author • Edited on

My insulin pump was never intended to be algorithm-controlled like this. It's from 2007, and its protocol and communications were reverse engineered. Compromising systems isn't a bad thing, in my opinion, it's the intention that may or may not kill people.

If it wasn't for a bunch of developers (search for #WeAreNotWaiting) back in 2015 or so, I could be dead today, as one of 25 people with Type 1 die from a severe hypoglycemia.

This is why I use a 15 year old insulin pump. Because it this model was compromised at some point.

armousness profile image
Sean Williams

The world contains both "black hats" and "white hats," and something being compromised and on the internet gives access to both. Because of this, a cybernetic world would be one where bad computer security is fatal—or rather, much more fatal than it already is.

I suppose the problem with this question is, you're talking about two very different things: first, the balancing of concerns with wearing a device that could kill you to manage a medical condition that could kill you, and second, whether this stuff should go discretionary. Medical decisions are between you and your doctor; I have no opinion on them. For me, though, discretionary cybernetics is a no.

Thread Thread
jansche profile image
Jan Schenk (he/him) Author

I agree with you on "compromised on the internet" part. This wouldn't be an option if my actors were exposed on the internet. They aren't.

Regarding discretionary cybernetics, I'm not sure I fully understand what you say. Do you mean non-regulated technological replacements for dysfunctional parts of your body shouldn't exist? What about building a custom prosthesis? Or a set of augmenting glasses that enables people with colorblindness to differentiate between colors?