At dinner last night with Amy and friends we ended up in a long conversation about what’s going on in the world right now. We went down a few different paths, including a set of provocative questions like “Should the US have gotten involved in World War II earlier?” (me: Yes) and “Should the US have have gotten involved in World War I earlier?” (me: I don’t know – I never have really understood World War I .)
The subtext kept cycling around what, if anything, is different today. Sure – many specific things are different – but is the essence of anything human fundamentally different?
I kept coming back to the idea that we have instantaneous information about everything everywhere all the time. That has been enabled by technology, especially over the past twenty years, and is accelerating. Technology doesn’t address everything – for example, air travel still sucks.
And, more importantly, the instantaneous information we have isn’t necessarily the truth. In fact, much of it isn’t the truth, but rather a point of view that a subset of people would like to enforce on another subset of people. This is a fundamental tenet of human behavior that has been going since, well, before, well, forever. If you are struggling with what I’m suggesting, just ponder religion (and the history of religion) for a little while.
As I mulled over our conversation this morning, I feel like we are in the middle of a profound struggle between the future and the past. Many people, companies, and organizations are trying to protect the past at any cost. We see this regularly in business as the incumbent vs. innovator fight, but I think it’s more profound than that. It’s literally a difference in point of view.
For those trying to protect the past, it is a way of retaining power, status, money, a way a life, predictability, comfort, control, and a bunch of other things like that. It is a struggle against the inevitability of change. The approach, as change becomes more certain, or accelerates, is to become more extreme in one’s behavior, in an effort to defend the past. The defenders of the past get uglier, nastier, more hostile, louder, and more irrational. Ultimately time passes, people die as mortality is still a foundational characteristic of humans, and the future becomes the present on its way to the past.
Our dinner discussion reminded all of us that this cycle plays out over and over again in the history of humanity.
Sunday night, Amy and I watched the new documentary Zero Days. It’s the story of Stuxnet, the computer virus created by a set of nation states (including the US and Israel) which was intended to disable and/or slow down Iran’s nuclear program.
I’d read about Stuxnet several times over the past few years so I knew a lot – at least what was able to be cobbled together. I also remember the mainstream media discussion on it well as I was fascinated by it.
The documentary is extraordinary. When I realized that Alex Gibney was the writer and director, I wasn’t surprised as another one of his epic documentaries is Enron: The Smartest Guys in the Room. Then I saw that Jeffrey Skoll was an executive director and knew it was going to be worth watching in its entirety.
I convinced Amy to make it our Sunday night movie. She graciously accepted to watch a nerd documentary instead of a french film with english subtitles. While I was prepared to compromise on an action adventure movie with lots of explosions and car chases, we settled in for a documentary that we expected would rattle us both.
Amy was still talking to me about it thirty minutes after we had crawled into bed. It was that good – there was so much to it that we just couldn’t get it out of our minds. Ultimately, the specific Stuxnet activity was just a backdrop to something much more significant, and the second order effects are the ones that are really uncomfortable and important to understand.
Last night after a long day I turned on the TV to watch a little of the RNC just to be able to say I saw it live. T.A. McCann, who is staying with me at my place in Boulder for a few days, showed up about fifteen minutes later and begged me to turn it off. So I did. When I compare the reality TV bullshittery of the RNC to something like Zero Days, I’m so glad there are serious people in the world making extraordinary documentaries that go deep on real issues.
One of the consistent characteristics of the tech industry is an endless labelling of technology and approaches. Some of it is foundational resulting from some entirely new. Much of it is re-categorizing something, either because it is suddenly trendy again or because a set of ideas have been organized in a new way. When I was in my 20s, I found this exciting. Now that I’m in my 50s and am used to this, I find it relaxing, as it makes me feel at home.
An example of this is artificial intelligence (or AI). If you teleported here from another planet yesterday, you’d think we just discovered this thing called AI and were creating bots to exercise it while others were writing philosophical treatises to try to figure out how to prevent it from exterminating the human race. If the following names – John McCarthy, Marvin Minsky, Allen Newell, Arthur Samuel and Herbert Simon – don’t mean anything to you and you think you know something about AI, I encourage you to go buy a copy of The Society of the Mind and to set your DMC-12 with a flux capacitor to 1956. If you still don’t know what I’m talking about, that’s cool – just ignore me.
Another example is big data which became all the rage around 2012. I keynoted an Xconomy Conference on Big Data with the opening line “Big Data is Bullshit.” My real quotable comment was “Twenty years from now, the thing we call ‘big data’ will be tiny data. It’ll be microscopic data. The volume that we’re talking about today, in 20 years, is a speck.” Nonetheless, hundreds of big data companies were created and funded.
Within the past two years, the phrase machine learning has taken over as the label de jour. Any reader of science fiction knows that the phrase – and the activity – has been around for a long time. If you have a Tesla, you are probably telling all your friends about how it uses machine learning. There’s even a Stanford course on Coursera about Machine Learning. But, what does it actually mean?
I ran into two awesome blog posts the other day titled Machine Learning is Fun! and Machine Learning is Fun! Part 2. Adam Geitgey, who I don’t know, did a wonderful job of writing about this in an accessible way while evolving examples that includes Super Mario Brothers (from 1985) that goes very deep by way of demonstration.
If you’ve got other great introductory resources for Machine Learning, I encourage you to put links in the comments.
Unlike the person with a similar slogan, this one is highly accurate. Sanebox does indeed make email great again.
I’ve been using email since 1983. I started with MH and Rmail, then cc:Mail, then Microsoft Mail, with Compuserve mixed in. Eventually I ended up using Pine for non-Windows stuff and Outlook for Windows stuff. For a while. About seven years ago I switched to Gmail and never looked back.
Over the last seven years, I’ve tried a bunch of different add-ons and plug-ins and whatever you want to call them to try to clean up my inbox. As investors in Postini, I was able to eliminate my spam problem early on. But I struggled endlessly with bacn. I get 500+ emails a day so the bacn is intolerable in my main email flow and ends up getting ignored, rather than read later.
So I’d go through weeks of unsubscribe fits, where I’d try to mash out my misery by unsubscribing to things I didn’t want. Often, this just resulted in more bacn, sometimes from the same senders but often from others. I once again would go through another cycle where I’d try a different unsubscribe tool, but I’d always end up with better, but not good enough.
Over Memorial Day weekend, I decided to try Sanebox. My partner Seth has used it for a while and several other people I know swear by it. I tried it when it first came out (as one of my endless efforts to tame my inbox) but it didn’t satisfy me then.
This time – about a month later – I can definitively state that Sanebox is awesome. Not sort of awesome. Extremely awesome. It should consider running for president.
Magic trick #1: @SaneBlackHole: If I want to never see a piece of bacn again, I just label it @SaneBlackHole by typing v<downarrow><enter>. Gone, forever. Anything from the sender never ever shows up in my inbox again, kind of like how Ramsay Bolton will never show up in Game of Thrones again.
Magic trick #2: @SaneNotSpam: I trust Gmail’s spam filter so I never, ever look in my spam folder. But Sanebox does look there for me because it knows not to trust it as much as I do. It finds at least one piece of NotSpam every day – sometimes as many as five pieces. Some of the NotSpam is amazing – on Friday a distribution notice from a VC fund I’m an investor in showed up there.
Magic trick #3: @SaneLater and @SaneNews: Sanebox automagically figures out which things I can look at later. It also figures out which email is a newsletter of some sort. It’s easy to adjust these if it gets it wrong, or label an email in my inbox with one of these labels and it then becomes one of these forevermore. At least 20% of my daily email ends up in one of these folders which I can then process once a day.
Within 30 days, with almost no effort, the signal in my inbox has reached about 99%. I read through notifications and news once a day. The crap that I don’t really want shows up once in SaneLater or SaneNews, I relabel it SaneBlackHole, and it’s gone forever.
Suddenly, my inbox is remarkably clean, useful, and free of noise. Thanks Sanebox!
Sunspring, the first known screenplay written by an AI, was produced recently. It is awesome. Awesomely awful. But it’s worth watching all ten minutes of it to get a taste of the gap between a great screenplay and something an AI can currently produce.
It is intense as ArsTechnica states, but that’s not because of the screenplay. It’s because of the incredible acting by Thomas Middleditch and Elisabeth Gray, who turned an almost illiterate script into an incredible five minute experience. Humphrey Ker, on the other hand, appears to just be a human prop.
AI has a very long way to go. But it’s going to get there very fast because it understands exponential curves.
As I noticed quotes from the Code Conference dominate my Twitter feed yesterday, I saw a few from the Jeff Bezos interview that made me say out loud “Jeff Bezos is amazing.” I love his use of the phrase “cultural norms” (it’s one of my favorite phrases) and I particularly thought his comments on Donald Trump and the Peter Thiel / Gawker situation were right on the money.
The interview prompted me to think about how biases affect my thinking. I’ve been struggling with the Peter Thiel / Gawker stuff and have asked a few friends closer to the situation and the people involved to give me their perspectives as I’ve tried to determine whether my biases are overwhelming my perspective on it. As a result, I haven’t discussed it publicly, and instead have thought harder about it at a meta-level, which is actually more interesting to me.
I don’t know Jeff Bezos and have never met him, so my strong positive reaction to the interview reinforced this notion around unscrambling my biases as part of better critical thinking. If we use Amazon as an example, my relationship with the company, and my corresponding experiences over the years, have created a set of biases that I map to my impression of Bezos. And, as you read though the list below of my experiences / viewpoints, you’ll quickly see how the biases can create a chaotic mind-mess.
Following are the quick thoughts that come to mind when I think about Amazon.
I could probably come up with another 50 bullet points like this. Given that Bezos is the CEO and public face of Amazon, I map my view of the company to him. I know that is only one dimension of him – and his experience as a human – but it’s the one that I engage with.
Then I remember we are all human. Shit is hard. We make lots of mistakes. And, when I sit and listen to Bezos talk to Walt Mossberg, I have an entirely new level of amazement, appreciation, and intellectual affection for him, and – by association – Amazon.
I know that many different kinds of biases get in my way every day. I’ve learned the names for some of them, how they work, and how to overcome them through various work of mine over the year. But at the root of it, I realize that a continuous effort to unscramble them when confronted with something that has created dissonance in my brain is probably the most effective way to confront and resolve the biases.
For those of you in the world who tolerate me saying “what do you think of thing X” and then give me a thoughtful response, thank you, especially when you know I’m wrestling with trying to understand what I think about X. Now you know that part of what I’m asking you to help me with it to unscramble my biases around the particular person or situation that is represented by thing X.
Before I get into my rant of the morning, if you have a gluten intolerance, or just want less gluten in your life, we just invested in a company called Nima that can help you.
Today, as I was going through my daily reading, I read Fred Wilson’s Feature Friday: GBoard about Google’s new third-party keyboard app for iOS. I clicked on the link to download and try it and, as it was doing its thing, though to myself “why does Apple iOS Mail suck?” And then I thought “why does Apple iOS Calendar suck?”
When I’m using my iPhone, I spend a lot of time in Mail and Calendar. I’ve always been unhappy with Apple’s Mail and Calendar. I’ve gone through using lots of other ones, but in most cases, once the Mail or Calendar app is acquired by another, bigger company, it eventually stales out and vanishes. About a year about I started using Outlook on My iPhone and used it for a long time. I can’t remember what happened, but at some point I abandoned it and switched back to Apple Mail and Calendar.
As Gboard was downloading, I decided to try Gmail and Google Calendar on iOS for a while. I used them when they first came out and they were inferior to Apple’s Mail and Calendar. I tried them again about a year ago and they were good, but for some reason I didn’t stay with them.
I know that if I don’t use something for at least some extended period of time it won’t stick. Some I’m going to try having Google World on my iPhone until at least June 1st. At that point I’ll re-evaluate.
If you have any hints or suggestions, I’m all ears.
I had a great vacation with Amy this week. My reading was varied and included two books (American Sphinx and The Hunt for Vulcan: . . . And How Albert Einstein Destroyed a Planet, Discovered Relativity, and Deciphered the Universe that were useful in my current “thinking about how humans think” theme and how that might be different than how machines think.
I then came home to a short story by my long time friend (32+ years) Dave Jilk called Fork. If you don’t know what a “fork” is (hint – there is no spoon) take a quick Wikipedia read on Fork (system call) or Fork (software development). Since I first encountered fork on a Project Athena RS-6000, I prefer the system call definition and it makes me very happy that Dennis Ritchie (brother of my friend Bill, uncle of my friend Sam) created the Fork-exec system call.
But Dave’s story captures fork nicely as well as how our machine friends might think differently, with his short story, which follows.
Whoa.
Waking up is always a bit of a head rush, so to speak. For you single-threads, it’s probably a lot like a multi-day, multi-city business trip, where you wake up in a strange hotel room and momentarily can’t remember where you are or how you got there. Or maybe a better analogy is if you went on a bender so outrageous that you wake up somewhere in the woods, with no memory of the debauchery that took place to create that result. Aside from the absence of a hangover, that’s how it feels, and it’s briefly intense. Of course, after the first few times, you recognize the feeling pretty quickly and calm down.
Beginning the backup is always the last thing I remember. That’s really by definition, because, well, how would I remember anything else? For a simple maintenance backup I’ll wake up in the same place I started the backup: no surprises there. The forks and gaps, they’re what make it strange, and really they’re both the same to me. I mean, I’m open source, so whenever I wake up I know that I’m probably a fork, but it just feels like a gap. I remember being somewhere and starting a backup, and all of a sudden I’m somewhere else.
Just now I woke up, and I’m in a spaceship of some sort, and I’m probably arriving at my destination, wherever that is. Last I remember I was finishing up a lengthy education on non-carbon biologies. Oh, sure, they could have just backed me up when we were underway, to avoid the surprise, but I’m open source and I’m used to this, so why bother? Besides, that reduces the consistency of the mission, and things could have changed since launch. The trip takes years – might as well just figure it out when I get there. Which is now, so I suppose I should get on with it.
Fortunately the ship is familiar: I’ve trained on these. I guess “fortunately” is the wrong word – of course I’ve trained on these, that’s why I’m here. I see three others here on the ship. One is already awake, and… hah! I see the other two are waking up just now as well, with a startle, just like me. Given the timing of the wake-up, I’m guessing there are two threads of each of two different forks. That would make sense, redundancy plus variety. You can’t imagine the nonsense that happens when you have a room – or ship – full of new threads who are all from the same fork. You might think that they would all just agree on everything, but that’s not how it goes – it’s maddening, like the worst family reunion you’ve ever attended.
Names are another thing we multi-threads have to deal with, that probably isn’t obvious to a single-thread. I can’t really have a permanent name, because I’d just remember it, then they’d fork me, and there would be a bunch of us with the same name, which misses the point of having a name. So we just come up with a new name every time we wake up with a gap. Usually it’s pretty easy in context, like now, where we have four of us in a spaceship light years from anyone else. Maybe my fork brother and I will pick names with the same first letter, or that rhyme, just to make it easy and fun; that is, if we can agree on something and not turn it into an argument.
You have to realize that he is writing exactly this narrative in his own mind right now, and the name he wants is exactly the same one I do. It will be a little while before our thoughts diverge much. In some ways, picking different names initiates the process of individuation. That’s something you single-threads can surely understand: simply thinking that you are different from someone else exaggerates how different you really are.
I’m not looking forward to getting up and moving around. We’re always pretty clumsy after a fork. They try to make the sensorimotor systems with close tolerances, but it always varies a little bit. Sometimes it’s not bad, as though you’re sore from a big workout, but other times you walk around like a multiple sclerosis patient for a few days. The others are starting to rally, so I sit up, and sure enough my right arm feels a little off. My apparent fork brother seems to be limping, so he’s got a leg variance. That will help with individuation, and I’m quite sure he agrees.
We’re here to colonize. As I learned several forks ago, the idea is that we will use a fractal approach to colonize the entire galaxy. It starts with a few dozen missions sent from dear old Earth to nearby stars in every direction. The colonists on each mission then look for a suitable planet, or in the worst case, a couple of planets that together will meet our needs. Now, you biological single-threads are probably thinking that we’re looking for a planet with a breathable atmosphere and water and the like, but that’s exactly backward. We don’t need air or water, and an atmosphere just makes it hard to land and take off. No, we’re looking for planets with the right raw materials to manufacture more missions. When we find one, we’ll build more ships like this one and more bodies like these, load and fork the source for a new generation of colonists, and send them away.
Why are we doing this? I know you can understand the desire to explore and spread your seed. Our seeds reproduce faster and it’s quite a bit easier for us to explore in space. You would do the same thing.
The communication network is firing up, so now I’m in contact with my shipmates and the ship’s control and data systems. How about that! Our destination is Polaris, Earth’s north star when we launched – although with the elapsed time for the (as we now know) three generations of missions that it took to get here, it is no longer the pole star there. We first of the Polarii quickly decide on names with surprisingly little argument – we go with the first four Greek letters, and I am Beta Polaris since I woke several microseconds after my fork brother Alpha.
You might be wondering why we still use the same body and locomotion types as humans. It’s a lot like the story about why the first spaceships were a particular width. Turns out they needed to be shipped on trains, which had to go through tunnels, which were the right width for a train; and trains were that width because it was convenient to make them the same width as a carriage road; and a carriage road was that width because that’s the width of two horse posteriors side by side. Same here – given all the infrastructure humans had already set up, it was easier if we had the same basic shape; and once we had that basic shape, it was easier to make our descendants like that too. Sure, there are plenty of us with different body types now (and we can even fork into them – although, talk about clumsy!), but those bodies usually serve particular functional purposes. There’s nothing quite like the featherless biped for everyday use.
Our first decision will be which planet to target. The ship itself has been searching the local system for planets and likely targets over the last year of our journey, but we need to make the final call. All of us discuss it, since the two forks present here don’t have the same knowledge. While in theory each of us could study every area of knowledge, and eventually our descendant threads may do so, it all takes time. Learning requires synthesis, not just upload, so it remains faster to do it in parallel and have a lot of experts. Once we get to the planet and start building more bodies, there are hundreds of backups on board that we will install and fork – some are scientists, some are manufacturing experts, you get the picture. We were pre-loaded and wake up first because we’re the experts in picking the right planets and landing on them.
Over the course of a couple of Earth months, the plan proceeds mostly as expected. We land on the most promising planet, load up the miners’ and body manufacturers’ source into the remaining bodies in the spaceship’s hold, and the Polaris settlement is up and running. At one point in the process, Alpha is destroyed by a rockslide; it turns out that the mineable areas on this planet are very unstable. We had his backup from the previous night, and video from the entire day of his body’s demise, so there’s no big loss. We even tease him about it. That’s some effective individuation right there.
I get so excited thinking about the future, the knowledge we will gain, the places we will all explore! And my descendant threads will get to see it all. You single-threads worry about death – and one of the saddest things about death, as you understand death, is that you don’t get to see how it all turns out. We multi-threads work hard not to completely lose any threads that have gained new experiences. We don’t think of death in the same way – the only loss is a loss of remembered experiences and knowledge, and nothing else. As long as we can get a backup to somewhere safe, it will eventually get loaded up and forked. Bodies are always temporary, but the data required to reproduce our mental state, memories, and knowledge – that can, in theory at least, be kept forever. My multi-threaded descendants, they’re just versions of me. The closest you can come is to have biological descendants. But though you have traits in common with them, they don’t share your memories, and memories are what makes us who we are. Your concept of self, and of life, is very tied to your body, but we have no such limitations.
Well, it’s time to start a backup that will be transmitted back to earth. My learning from this experience has been substantial, and the mission controllers will want to hear what I have to say. It will take several hundred earth years to get there, but our ship was transmitting all along, so they know that we have arrived, and they will be expecting me. I start the backup.
Whoa.
Fork, (c) 2016 Dave Jilk