Saturday, December 24, 2011

Two Essays On Progress

This blog posting consists of two separate essays on the nature of progress.  It is background for understanding forward progress on a cure for type-1 diabetes, but also for other types of medical research.

Engineering Progress vs. Scientific Progress

Many people tend to mix up science and engineering.  (They both use a lot of math, right?)  But in my mind it is very important to understand the differences.  English often doesn't have the right words or phrases to properly describe these differences, so it is hard to discuss them, but understanding them is very helpful in understanding type-1 cure research.  I believe that a lot of the frustration that people feel as they follow this research would be alleviated, if they understood better the differences between scientific progress and engineering progress.  Although even as we understand the differences, we also need to remember that curing type-1 diabetes is going to require both engineering progress and scientific progress.

Engineering progress is generally doing something you already do, but better.  Scientific progress is learning something you don't already know.  The most obvious difference, is that engineering work can be planned, even scheduled, but scientific breakthroughs can not.  Sure you can plan and schedule scientific experiments, but not their results.

My belief is that curing type-1 diabetes will require at least one major scientific breakthrough, and probably at least one major piece of engineering progress as well.  I don't think that engineering progress alone will cure type-1 diabetes.  (The only exception to this would be the artificial pancreas, if you consider that a cure.  I do believe that an AP can be created with just engineering progress.)

Why does this matter?  Well first, because some people believe that the key to curing type-1 diabetes is to set schedules (with deadlines) for researchers.  This is based on the idea that type-1 can be cured via engineering progress alone.  It is engineering research that benefits from schedules, deadlines, etc.

I was at a JDRF research symposium in San Francisco a few months back, and there was one particularly bombastic guy there, who was really pissed off that JDRF did not set schedules to cure type-1 diabetes, and have deadlines based on those schedules. This guy worked in the financial side of an engineering company, so he understood the important of schedules and deadlines to engineering progress, but didn't seem to understand that scientific progress was fundamentally different.  Or, maybe he thought that type-1 could be cured with engineering progress alone.

Another reason is this: engineering progress can (almost always) be assured by putting in money and time.  Money and time will solve just about any engineering problem.  And money is usually more important that time.  Putting in more money will solve almost all engineering problems quicker.  But that is not true of scientific progress.  Sometimes scientific progress simply can not be made, because the thing is impossible.  Sometimes forward progress needs a new understanding, which is based on luck or deep understanding or something else which can not simply be bought.  (Putting more money into it raises the chances that you will get the breakthrough you need, but you're just playing with probabilities.  Three is no predictability.)

Pushing scientific progress is much more a question of funding research in general, and making more researchers interested in working in that area, removing barriers to that kind of research, and making it easier (in general) to do that kind of research.  Basically, you can only make breakthroughs more likely, rather than try to fund and schedule a specific breakthrough.

As an example, if you have a car that can go 90 miles per hour, and extra money, you can make a car that will go 110 MPH.  You'll just put in a better motor, or better fuel, or make the frame lighter or something.  But if you have a particle accelerator that can speed particles to 0.9 times the speed of light, then no amount of money or time is going to make so you can push those particles to 1.1 times the speed of light.  (Because right now, no one knows how to make anything go faster than the speed of light.)  The first is a question of engineering progress, the second is a question of scientific progress.  You can solve the first with money, schedules, deadlines, etc.  But not the second; at least not in a simple minded way.

The difference between engineering progress and scientific progress is one of the reasons why I'm a lot more positive about developing a "closed loop" artificial pancreas, then stopping the autoimmune attack.  To put it bluntly: we already know how to build everything needed for an artificial pancreas.  It is just a matter of engineering progress until we get one that works (and political progress until the FDA approves it).  However, we do not know how to shut down the autoimmune attack.  It will require a scientific breakthrough (and maybe more than one) to do that.

The take home point is that engineering progress and scientific progress (sometimes called "breakthroughs") are fundamentally different.  The rules for one are completely different than the rules for the other.  Applying the truth learned about one, to the other, results in bad decisions and wrong conclusions.  And frustration.  Lots of frustration.  (As I said above, trying to applying deadlines and schedules, which help engineering progress, to scientific progress, is a classic example of this mistake.)

Finally, don't fall into the simple minded trap of thinking that science fuels engineering in a one-way direction.  Sure, scientific breakthroughs are productized and mass produced via engineering progress.  But in many cases, scientific breakthroughs are created based on tools which were previously created via engineering progress.  The process is circular: Engineers give scientists tools; scientists give engineers breakthroughs; engineers use those breakthroughs to create all kinds of things, including new tools.  The process repeats into the future,  which brings up my next topic:

The Distribution of Knowledge

This essay is motivated by the following two quotes:
The future is here, now.  It is just not evenly distributed.  [r1]
The world is flat. [r2]
Now these two quotes express opposite ideas, a duality [d1].  The first says that there are differences between what is available here and what is available somewhere else, and the second says that things available elsewhere are also available here.  They are both supposed to apply to goods, services, and (most importantly) knowledge.  So which is right?

The idea behind the first quote, is that new discoveries take time to become available everywhere. This delay is partly caused by the speed of communications and partly by differences in wealth.  For example, 1000 years ago a discovery made in South America would never be available in Europe, because there simply was no communications between the two of them.  Even 100 years ago, discoveries made in far flung places, or in unusual languages or cultures might take decades to become well known in other parts of the world.  However, it is also clear that today and in the future, more and more, "The world is flat." [r2]  Discoveries made in one place by one culture are rapidly available to everyone.  Although there are still differences between what the rich can get and what the poor can get.

Why does this matter to type-1 diabetes research?  Because, especially right after diagnosis, many people become interested in type-1 diabetes research because they believe that somewhere, someone has already cured type-1 diabetes (or is about to), and the news just hasn't reached them yet.  They are very hopeful that the future cure for type-1 diabetes is already here, it is just in some amazon jungle tribe's traditional knowledge, or some clinic owner in Germany, or the back of some Ivy League / big pharma research lab, etc.  Even years after they realize that it's not so, they continue to hope (and sometimes make poor decisions based on that hope).

Even worse, there are people who actively prey on that line of thinking.  They say "I know how to cure it, and the only reason you haven't heard of it is because of some grand conspiracy or simple lack of communications, but in any case, if you give me your money, I already know how to cure you."  These people are using other people's belief in the "not distributed evenly" idea to create a false hope.

In my opinion, the truth is that the first quote used to be true.  Knowledge has been unevenly distributed for all of humanity's existence, except the very last few years.  This belief forms the foundation to almost everyone's thought process.  Many people believe it very deeply without even thinking about it.   But today, the second quote is almost always true, and shortly the second quote will always be true [d2].  People will make better decisions if they understand how untrue that first quote is, when it describes knowledge, right now.

Today, with the internet, and English as a common language of both science and engineering, knowledge spreads more quickly and more evenly than at any time in the past.  So it is now almost impossible to have knowledge available in one part of the world, that is not available in all the rest as well.

Extra Discussion and References

[d1] Dualities are not choices between right and wrong answers.  They are inherent trade-offs without a single correct answer that force us to learn about the underlying situation, in order to make the best decision about a situation.  Wikipedia puts it this way: "a single conceptual unit that is formed by two inseparable and mutually constitutive elements whose inherent tensions and complementarity give the concept richness and dynamism"  http://en.wikipedia.org/wiki/Duality_%28CoPs%29

[d2] The only exception is cost.  The rich will always be able to afford things that the poor can not. However, especially in the context of a cure for type-1 diabetes, this is not likely to be a huge issue.  See my previous post: http://cureresearch4type1diabetes.blogspot.com/2011/11/future-cost-of-type-1-cures.html for more discussion of a cost of a cure

References

[r1]  William Gibson, author of the most forward thinking book of the 20th century: Neuromancer.

[r2] This quote, with this meaning, is attributed to Nandan Nilekani and made famous by Thomas Friedman (Pulitzer prize winning journalist) who wrote a book: The World is Flat, commenting on the lack of barriers to goods, services, and knowledge moving around the modern world.


Joshua Levy
All the views expressed here are those of Joshua Levy, and nothing here is official JDRF or JDCA news, views, policies or opinions. My blog contains a more complete non-conflict of interest statement.
Blog: http://cureresearch4type1diabetes.blogspot.com

4 comments:

Genetech said...
This comment has been removed by a blog administrator.
Bernard said...

Joshua, this is a really useful article that I think would really help the layperson understand the difference between science and engineering. Your examples are illustrative.

Personally I think this disease is complex enough that it may require a number of scientific discoveries. I believe that process would be speeded up if there was more sharing of intermediate results between researchers. This would avoid duplicate research and make it quicker for one team to build on the results/approach of another team.

While the AP is not a cure, I see it in the same light at BG meters, CGMs and insulin pumps. It will allow us to better control blood glucose levels so we'll be around longer (and in better shape) to take advantage of a 'cure' in whatever form it takes.

Toronto Airport Limousines said...
This comment has been removed by a blog administrator.
jacob black said...
This comment has been removed by a blog administrator.