Eroom’s Law

It’s often cited that it costs more and more billions to bring a drug to market. Erooms law is Moores law backwards. Countervailing forces like regulation are blamed for this phenomenon. (The author of this piece seems ludicrously positive on the past success of pharma while I think the reality is that LDL lowering drugs are disappointingly non-efficacious and aspirin is just a masker of problems in the body. I say this mostly in response to the better than the Beatles argument which seems to treat human health as some sort of a solved problem.)

My own theory is more radical. The whole practice of drug discovery in the pharmaceutical industry is flawed on a fundamental level. The body is a finely tuned machine more complex then any man made machine. It is so complicated that we simplify it to a comical degree. We come up with ideas like “the bad cholesterol” in heart disease or “the bad COX” in pain relief. We then come up with chemicals that will hopefully impact our chosen villain. However both Cox-1 and Cox-2, LDL and HDL are manufactured by the body and function as components in tiny machines not chemical waste products or some spice like salt which needs to be reduced to make a tomato sauce taste better. Moreover we often do not even know why a chemical like LDL is reduced by a compound so it is difficult to anticipate side effects. As science has advanced we have gotten better at showing just how efficacious our drugs aren’t but also how frequently they have side effects. It’s not only that drugs are expensive and time consuming to bring to market but also that they are now more frequently withdrawn. It may sound like heresay to some but I doubt magic pills with strong benefits are even really out there to be found. It’s unlikely there’s a particular acid that you could open up the hood of your car and pour in that would have that great of a benefit. This is not a condemnation of all medicine just the general current paradigm of pharmaceuticals that are supposed to effect a disease process. Antibiotics have a use, so do vaccines and surgery and the reason they work is clear. But in general the approach of medicine needs a paradigm shift that treats the body as a machine. In the future we won’t treat depression by just adding serotonin or heart disease by lowering LDL. We will instead rewire the brain and repair the plumbing of the circulatory system. The earliest pharmaceutical drugs were things like opium and cocaine I’m sure they sailed through regulatory hurdles in development but what is really the point they were never very good. Modern antidepressants some how increase the risk of suicide.

I don’t like analogies but I think this is an apt one. We are working with a car we inherited which is our body. We are only beginning to understand how it works. Its as if your doctor was a mechanic for a space ship. Rather than saying you have a broken muffler the mechanic would be in the position of well we added this chemical into the gas tank and we found a statistically significant result that most people did see a reduction in noise from the muffler after this treatment. It later turns out that it was just the right small amount of sugar added to the gas tank to slow the engine down without junking it. Isn’t the “miracle drug” metformin just agitating the liver in a similar fashion? In a new paradigm we will know how to replace the muffler (stem cells) or even build a better one with biomedical engineering.

So in sum I have a libertarian bias and would prefer we went as far as to do away with the FDA. But I don’t think its fair to blame them for the rising cost and general lack of success in today’s pharmaceuticals. I also don’t think the garbage the industry marketing machine has been jamming down our throats is just too tough to top. I think we’re moving from unconscious incompetence to conscious incompetence.

One More Strike Against you on the Simulation Hypothesis

A lot has been written about the simulation hypothesis. I don’t think its a complicated argument but I think its pretty bullet proof. I’d be curious as to what the moderator’s unstated counter argument was in this clip. Bostrom goes into it in great detail if you want to be extremely rigorous. However, Musk explains it pretty well in this tiny clip. Based on the magnitudes of the size of the universe, its lifespan, the amount of matter, and its potential for computation I think its very likely that we are in a simulation. Musk has one more strike against him though in that he has such a uniquely interesting life. Being that far out on the bell curve of lives worth simulating has got to make the simulation argument even more of a mind fuck for him. Not by much though because I think its already very likely, for everybody, that their lives are a simulation. And of course the puppet masters did provide Mr. Musk with an especially lovely new project to distract him right after he expressed the opinion that this is “unlikely to be real.” That is of course, if he does give her tunnel a lot of consideration. I know hes married, runs two companies, and must be extremely busy.

If you’d like some rough math and a time frame for the video game comparison Musk makes in the above clip check out this short clip below from a talk by Phillip Rosdale.

Antifragile: Things That Gain from Disorder

Nassim Nicholas Taleb is one of my favourite writers and one of the true original thinkers and true philosophers of our time. So I was looking for an excuse to write about his latest book here. He gave it to me in the following passage:

I was just reading in John Gray’s wonderful The Immortalization Commision about attempts to use science, in a postreligious world, to acheive immortality. I felt some deep disgust- as would any ancient- at the efforts of the “singularity” thinkers (such as Ray Kurzweil) who believe in humans’ potential to live forever. Note that if I had to find the anti-me, the person with diametrically opposite ideas and lifestyle on the planet it would be that Ray Kurzweil fellow. It is not just neomania. While I propose removing offensive elements from people’s diets (and lives), he works, by adding, popping close to two hundred pills daily. Beyond that, these attempts at immortality leave me with deep moral revulsion.

It is the same kind of deep internal disgust that takes hold of me when I see a rich eighty-two year old man surrounded with “babes”, twenty-something mistresses (often Russian or Ukrainian). I am not here to live forever, as a sick animal.

I don’t want to bother to address these “boring” arguments needless to say I do not agree with Taleb’s take on Kurzweil.

Taleb as I said has fascinating ideas. His neologism “anti-fragile” in this book is genuinely new concept with some import. However, personally Taleb seems to be something of a character. He spends a lot of space attacking today’s neomania as if to suggest that the apparent acceleration of change in our time is the result of fad chasing rather than genuine technological improvements.

Taleb doesn’t even consider the truth which is that much of the new really is much better but it comes at us faster than we psychologically are prepared to accept it. We don’t seek out the new. Most people scoff, hold out as long as they can, and accept eventually that they are going to use something that is so much better. That is to say things are changing fundamentally, not superficially. The self driving car isn’t today’s pet rock. The P.C. is not just some neomaniac’s re interpretation of the typewriter.

Taleb is sort of a macho full of himself person who rails against nerds and idolises the ancients. I can’t help but laugh out loud sometimes when the audio book of this breaks out into a foreign language in quoting some ancient. He takes himself quite seriously. But even at his luddite worst Taleb makes good points. He observes that the longer an invention has been around the longer it’s likely to stay around. This seems largely to be true. When people 200 years ago imagined the future they pictured people floating on hovering bean bag chairs and replacing meals with pills. They did not picture that the vast majority of change would relate to communication technology and it would be film cameras and wired telephones, relatively new inventions, would be on the chopping block and not chairs, meals, or shoes which have been around for thousands of years.

Related to the framework of the robust, fragile, and anti-fragile Taleb also introduces the concept of applying optionality and convexity to everyday life. He correctly observes that a person with terminal cancer is better advised to take extreme interventions than someone with a cold. He also advocates that the burden of proof should rest on proving an unnatural action safe rather than disproving it’s safety. To my delight he to an extent advocates the paleo diet and intermittent fasting. However he eventually goes to far and crosses into the naturalistic fallacy as can be seen in his attack on Kurzweil. Because our genes want to live on at our expense does not mean we should adopt this as our life purpose – perhaps by becoming a prolific sperm donor. If he wants to leave books behind as a legacy we’re all richer for it. I take the Woody Allen approach. I’d rather go on living in my apartment than in the hearts and minds of my countryman. Of course Taleb would actually need to become familiar with Kurzweil’s work to realise this is a remote possibility which I believe he is unlikely to do.

I think what I like best about Taleb’s books is he examines modern life and applies philosophy to it. There are genuine take aways from his books which is more than you can say about many actual self help books. Eat as your ancestors did, invest in a dichotomy of very safe and very risky investments rather than just middle risk investments, surround yourself with less successful people, major in a science rather than business, take extreme interventions only in extreme conditions in accordance with convexity. There are many others and I would recommend all of his books. I believe I’ve read them all including his book of aphorisms which to my co-workers’ chagrin I used to read at work. Many of the aphorisms were about the drudgery of a 9-5 hourly job.

The Last Question

I cannot recommend this free Issac Asimov story enough. I guess the “singularity” is before his time but he anticipates so much of what is associated it like AI, immortality and transhumanism in many of his books. What makes the singularity surprising is that it presents us with a counter intuitive rate of change. But even this I suppose we mostly care about because we anticipate amazing things in our lifetime. If AI were 500 or 1000 years away I doubt people would care or debate that issue. Anyway, this story is short but fascinating and I can’t do it justice.


I believe I understand Bitcoin but I cannot wrap my head around the concept of using the blockchain technology to establish ownership rights in something else. We can all decide a cryptographicaly secure secret string that cannot be double spent is worth something. Money is sort of a fiction anyway. However, can this sort of technology be used to own a company? Perhaps somehow the dividend could be insured but what would be even more reassuring to me would be that control and assets of the company were represented by the shares. For example, Google does not pay a dividend but you know it has cash in the bank, positive cash flow, and if it were bought out it should go at the very minimum for its cash minus its debt. Hypothetically, you could have a bitcoin based operating account that paid out at a set rate and you could put dividends and expenditures to a shareholder vote. Cryptosecurity companies could build a web of trust by loaning each other money or through hierarchical reputation like in HTTPS. Ultimately, I do not know how you establish control or liquidation interest of something real without the legal system enforcing your contracts.

Small changes in social apps engender completely different social behavior

It’s funny how apps like Twitter, Facebook, Whisper, Reddit, Email, ordinary message boards and Instagram have a tremendous amount of overlapping functions but produce completely different social behavior.

Facebook causes people to post accomplishments and safe uplifting stories because its everyone you know and tied to you

Twitter causes people to announce news because of the one way follower dynamic

Whisper causes people to post repressed fantasies because its anonymous

Reddit causes people to post interesting things because it’s segmented by interest and voted on.

Ordinary v2 bulletin boards cause more back and forth debate because of the ease and visibility of a public back and forth where people go line by line when they disagree with someone but don’t say anything when they agree

Eric Schmidt had some interesting comments on the way we should handle email:

“Respond quickly: There are people who can be relied upon to respond promptly to emails, and those who can’t. Strive to be one of the former. Most of the best — and busiest — people we know act quickly on their emails, not just to us or to a select few senders, but to everyone.”

People attribute their actions to choice but the tiniest change in font size on a page can alter it drastically. Stack overflow was designed to focus on answers and not discussion in part by making comments tiny. This very blog post is only occurring because I installed Evernote. I used to email myself posts to post to WordPress later. That worked but using ever note is probably 10% better which has pushed me into doing this rather than just sitting there in line or on the train.

I find this interesting from the standpoint that we believe we have free will when in fact we’re just responding to subtle environmental prompts. I believe human cognition is limited not in our inability to remember or perform difficult cognitive tasks but our pitiful attention. We underestimate how influenced we are by advertising or friends or whatever were currently reading because their influence is not that they persuade us but rather that they put things on our map and cause us to give attention.

This will have some implications as apps like LinkedIn, Meetup, Uber, Groupon, and Airbnb are bringing the digital world into the real world. Blog posts for their part seem to engender overly long solipsistic boring essays I think because it feels to authors like nobody is reading them, people use them to think out loud, and it becomes a one sided conversation.

New Apps Just Like the Old Ones

When a new product comes out a lot of people will immediately tell you how its not necessary because it can hypothetically be done some other laborious way.

The question is whether it will make something you already do slightly better. The harder products are things that try to create a new behaviour. This might seem like the kind of incremental ism decried by Peter Thiel but I don’t think it is because small changes will eventually lead to completely new behavior.

I’ve heard anecdotes of investors passing on twitter because it was just the title of a blog, waze because we already had GPS navigation and bafflement expressed at the success of instagram when we already have photo apps.

Thiel says a new product should be 10 times better. But I think a product can be just 10% better if it’s a free app that hooks into existing behaviour and costs nothing.

Stocks fall on fear of ebola?

I saw this headline on Google Finance the other day. How could that be possible? All deaths are tragic but how can a difficult to spread disease effecting a few people in a country of millions possibly cause a rational person to sell their stocks. Anyway, the naive correlation between stock movements and headlines by news people is well known.

Mixed feelings about editing

I think when an essay is edited it tends to get worse before it gets better. I’ve heard a few anecdotes lately about how authors end up writing the second half of their book in one sitting with no editing. They tend to attribute this to flow but I have another explanation. Editing can hurt a piece a lot. When someone is speaking you generally get the flow of their thoughts and understand them without them having to pick over their words ahead of time for hours. Receiving comments in a group essay can encourage an author to try to accommodate every view point and possible counter argument. Even adding things in can weaken an essay. I find that when I go back and add one sentence it hurts the flow of the adjacent sentences. For a blog I think editing is only useful to look up facts and citations but it’s hard to break the habit. There’s clearly a point where laboriously editing a small piece will make the piece better than the original but I wonder how often we, as a practical matter, actually reach it. I had a professor in college who insisted all copies of papers be turned in because he said without that he’d change back things he had already changed. Saying this would make me question my purpose in life, but he didn’t bat an eye at it.

P.O.S.S.E. an “IndieWeb” concept

I love this concept. Publish Own Site Syndicate Elsewhere. According to Zuckerberg’s Law the amount of data we share about ourselves doubles each year and according to Zuckerberg it makes sense for companies to try to capture as large a share of that is possible. This is why Facebook bought Instagram, WhatsApp and tried to buy Snapchat.

P.O.S.S.E. is a simple concept. If you’re going to post to Facebook, Reddit or anywhere else put it on your own site first. That way you own it and control it no matter what happens. In my crazy ideal future we’d have every thought we’d ever had on our own blogs with ads around them. Our most brilliant thoughts would be SEO’d to the top and our dumbest would be lost. Because the world will “search don’t sort.” Either way this makes sense because if we’re going to be compelled by the apps to share anyway we might as well try to monetize it ourselves.