The Entrepreneur Forum | Financial Freedom | Starting a Business | Motivation | Money | Success

Welcome to the only entrepreneur forum dedicated to building life-changing wealth.

Build a Fastlane business. Earn real financial freedom. Join free.

Join over 80,000 entrepreneurs who have rejected the paradigm of mediocrity and said "NO!" to underpaid jobs, ascetic frugality, and suffocating savings rituals— learn how to build a Fastlane business that pays both freedom and lifestyle affluence.

Free registration at the forum removes this block.

A physicist's guide to learning hard things (and using the scientific method)

lowtek

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
332%
Oct 3, 2015
2,164
7,186
42
Phoenix, AZ
So I was responding to a thread, and realized that one problem I see with some folks, newbies in particular, is that they lack a systematic methodology for teaching oneself. This isn’t surprising, as it’s not really a skill that we’re taught early in life. Realizing this, I figured I would share what I have learned about learning difficult things, over the years.

Of course, the scripted dogma is that we must go to school for 4 years, rack up a mountain of debt, and hope that we'll have marketable skills at the end. For STEM fields, this actually isn't a bad bet, but since most people don't major in useful things, it turns out disastrously for the majority.

I actually see some parallels between that scripted dogma and some of the posts here. Some folks will do extended learning challenges, work through dozens of tutorials, or seek out mentors to hold their hand... Perhaps it's the case that some people really need these constructs and arrangements, but I believe that there exists a better way to teach oneself.

Full disclosure: I guzzled the scripted dogma for the first thirty years of my life. I loved undergrad so much that I did a 5th year as a victory lap. Not willing to enter the real world, I then went on to a masters program and ultimately a PhD. However, from the ashes of these potentially ruinous decisions came an invaluable skillset that benefits me to this day: the ability to learn new things with minimal guidance.

Much like we have the CENTS framework for evaluating business opportunities, I believe there exists a mental framework for teaching oneself new skills.

What does it mean to learn something, really? I argue that learning means internalizing information in a way that allows you to apply it to new problems that are only marginally similar to previously encountered problems. After all, if you only know how to solve one problem, you’re not really a problem solver, you’re a robot.

Central to learning is having a solid mental model of cause and effect relationships. This is necessary because the world we live in is causal. To illustrate, let’s consider a simple example.

Let’s suppose you’re selling a widget on your website. For the sake of argument, we’ll assume that the widget is something people actually want. You’ve got some traffic, but only 1% of it is converting. You want to increase the conversion rate, because you have big baller dreams of lambos and lavish nights out in the big city. Problem is that with that 1% conversion rate you’re barely making burger flipper money, so hyper cars are out of the question, let alone hiring someone to fix the landing page. It’s on you to solve this problem. How to do it?

Most people will approach this problem by reading blog posts, books, or listening to podcasts on conversion rates. They will think they’re taking action, and assimilating knowledge, but in reality they’re wasting time. There are two reasons for this:

1) This delays the feedback loop between taking action and receiving results, thus obscuring the relationship between action and results.
2) It’s often event focused rather than process focused. Consuming prepackaged information often obscures the why of the solution.

The correct approach to learning how to increase conversion rates is this.

Clearly state the problem; the more precise and detailed the better.
Make a guess about what things you can control contribute to the problem
Formulate a plan for iteratively testing each of these possible root causes. Test most likely cause first, if possible.
If necessary, seek out information that tells you how to address each possible root cause.
Execute the plan.
Repeat until solved.

So if your conversion rate is only 1%, you could guess that the following things contribute to that low rate:
1) the page loads so slowly that people leave before it finishes
2) The color scheme makes peoples’ eyes bleed, so they can’t find the buy button
3) The copy reads like it’s written by an overseas scammer
4) The perceived risk is too high

These are just a subset of the possible reasons, and indeed a combination of them could be the problem.

We’ve already stated the problem, so that’s good.

Our analytics indicate loading speed could be an issue, and indeed any of the other 3 could be the problem so we have no choice but to test them all.

Brute force it is.

We therefore start at the top, load speed could be an issue. We then specifically seek out information regarding improving load speed issues on our site. Execute on that information immediately and observe the results. If the loading speed improves but not the conversion rate, then it’s on to the next thing.

We know nothing about color schemes, so we spend some time reading some information about it. After some careful consideration, we settle on a pre selected palette of colors and revamp the site. Observe the results, and move on if necessary.

We don’t know much about copy, so we pick up a copy of Cashvertising and skim the book looking for content that would be helpful. When we’ve found it, we implement it immediately. Observe the results and move on if necessary.

To fix the perceived risk, we can reach out to the 1% that have bought our product and ask for testimonials. So we look up some information on how to craft such an email, and boom send it out. We can also implement a money back guarantee, if we haven’t already.

After implementing all these things, we’re left with a vastly improved site. If the conversion rate doesn’t budge, we merely repeat the process. Come up with additional root causes and systematically test them, looking up information you need along the way.

So to summarize, this is how you rapidly learn new things:
Isolate a problem
Guess the root cause of the problem
Address each root cause in turn
Acquire any new skills to solve that specific problems
Repeat until you’ve got your lambo

This is in contrast to the approach of “oh gee, I have a poorly converting site. Guess I need to read a bunch of stuff on how to improve conversion rates… oh gee, guess I need to know web design, and copy writing, and … oh I’ll just give up”.

This is not to say that reading general business books doesn’t have its place, but I think this is primarily useful for those who are already in motion and can take the 1 or 2 nuggets from a book and apply it to their business straight away. I don’t believe there is large utility in reading for those that are not in motion.

In subsequent posts I’ll talk more about finding underlying principles, and expand on how I’m learning difficult things in my own journey.
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.
G

Guest61835

Guest
So I was responding to a thread, and realized that one problem I see with some folks, newbies in particular, is that they lack a systematic methodology for teaching oneself. This isn’t surprising, as it’s not really a skill that we’re taught early in life. Realizing this, I figured I would share what I have learned about learning difficult things, over the years.

Of course, the scripted dogma is that we must go to school for 4 years, rack up a mountain of debt, and hope that we'll have marketable skills at the end. For STEM fields, this actually isn't a bad bet, but since most people don't major in useful things, it turns out disastrously for the majority.

I actually see some parallels between that scripted dogma and some of the posts here. Some folks will do extended learning challenges, work through dozens of tutorials, or seek out mentors to hold their hand... Perhaps it's the case that some people really need these constructs and arrangements, but I believe that there exists a better way to teach oneself.

Full disclosure: I guzzled the scripted dogma for the first thirty years of my life. I loved undergrad so much that I did a 5th year as a victory lap. Not willing to enter the real world, I then went on to a masters program and ultimately a PhD. However, from the ashes of these potentially ruinous decisions came an invaluable skillset that benefits me to this day: the ability to learn new things with minimal guidance.

Much like we have the CENTS framework for evaluating business opportunities, I believe there exists a mental framework for teaching oneself new skills.

What does it mean to learn something, really? I argue that learning means internalizing information in a way that allows you to apply it to new problems that are only marginally similar to previously encountered problems. After all, if you only know how to solve one problem, you’re not really a problem solver, you’re a robot.

Central to learning is having a solid mental model of cause and effect relationships. This is necessary because the world we live in is causal. To illustrate, let’s consider a simple example.

Let’s suppose you’re selling a widget on your website. For the sake of argument, we’ll assume that the widget is something people actually want. You’ve got some traffic, but only 1% of it is converting. You want to increase the conversion rate, because you have big baller dreams of lambos and lavish nights out in the big city. Problem is that with that 1% conversion rate you’re barely making burger flipper money, so hyper cars are out of the question, let alone hiring someone to fix the landing page. It’s on you to solve this problem. How to do it?

Most people will approach this problem by reading blog posts, books, or listening to podcasts on conversion rates. They will think they’re taking action, and assimilating knowledge, but in reality they’re wasting time. There are two reasons for this:

1) This delays the feedback loop between taking action and receiving results, thus obscuring the relationship between action and results.
2) It’s often event focused rather than process focused. Consuming prepackaged information often obscures the why of the solution.

The correct approach to learning how to increase conversion rates is this.

Clearly state the problem; the more precise and detailed the better.
Make a guess about what things you can control contribute to the problem
Formulate a plan for iteratively testing each of these possible root causes. Test most likely cause first, if possible.
If necessary, seek out information that tells you how to address each possible root cause.
Execute the plan.
Repeat until solved.

So if your conversion rate is only 1%, you could guess that the following things contribute to that low rate:
1) the page loads so slowly that people leave before it finishes
2) The color scheme makes peoples’ eyes bleed, so they can’t find the buy button
3) The copy reads like it’s written by an overseas scammer
4) The perceived risk is too high

These are just a subset of the possible reasons, and indeed a combination of them could be the problem.

We’ve already stated the problem, so that’s good.

Our analytics indicate loading speed could be an issue, and indeed any of the other 3 could be the problem so we have no choice but to test them all.

Brute force it is.

We therefore start at the top, load speed could be an issue. We then specifically seek out information regarding improving load speed issues on our site. Execute on that information immediately and observe the results. If the loading speed improves but not the conversion rate, then it’s on to the next thing.

We know nothing about color schemes, so we spend some time reading some information about it. After some careful consideration, we settle on a pre selected palette of colors and revamp the site. Observe the results, and move on if necessary.

We don’t know much about copy, so we pick up a copy of Cashvertising and skim the book looking for content that would be helpful. When we’ve found it, we implement it immediately. Observe the results and move on if necessary.

To fix the perceived risk, we can reach out to the 1% that have bought our product and ask for testimonials. So we look up some information on how to craft such an email, and boom send it out. We can also implement a money back guarantee, if we haven’t already.

After implementing all these things, we’re left with a vastly improved site. If the conversion rate doesn’t budge, we merely repeat the process. Come up with additional root causes and systematically test them, looking up information you need along the way.

So to summarize, this is how you rapidly learn new things:
Isolate a problem
Guess the root cause of the problem
Address each root cause in turn
Acquire any new skills to solve that specific problems
Repeat until you’ve got your lambo

This is in contrast to the approach of “oh gee, I have a poorly converting site. Guess I need to read a bunch of stuff on how to improve conversion rates… oh gee, guess I need to know web design, and copy writing, and … oh I’ll just give up”.

This is not to say that reading general business books doesn’t have its place, but I think this is primarily useful for those who are already in motion and can take the 1 or 2 nuggets from a book and apply it to their business straight away. I don’t believe there is large utility in reading for those that are not in motion.

In subsequent posts I’ll talk more about finding underlying principles, and expand on how I’m learning difficult things in my own journey.
Wow! That's....a crazy observation. So crazy that most people (including myself) were OBLIVIOUS ENTIRELY to their own inability to learn. That's something I've struggled with insanely in alot of things. Except music.
MUSIC is where I found my ability to learn on my own with little research to play instruments. I can play guitar,drums,bass guitar, and such completely by myself. Self taught. Yes I had to research to find the solution and did it but I was unaware of myself being able to self teach myself (or at least unaware of the concept). Thank you!

Now I know where to direct MYSELF into the darkest tunnels with help when needed on certain manners. Thank you SO MUCH for updating me about this post.
 

lowtek

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
332%
Oct 3, 2015
2,164
7,186
42
Phoenix, AZ
When I read the title, thought you were gonna post the Feynman Method, but this is pretty good too :smile:

I leverage the Feynman method in learning AI, hence the YouTube channel. But I think what I've suggested is more broadly useful and gets more to the heart of learning, which is seeing cause and effect relationships.
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

garyfritz

Silver Contributor
Read Fastlane!
Summit Attendee
User Power
Value/Post Ratio
116%
Jul 16, 2011
694
807
Colorado
Central to learning is having a solid mental model of cause and effect relationships.
+1000! It always amazes me how many people have zero clue about the connection between action A and result Z. They don't really understand what affects result Z, so they don't know what to fix. They will make wild guesses, flail around, or try what worked for their buddy Joe, and they wonder why they can't get it working.

Problem-solving is something that comes naturally to me. I have an engineering background, because that's what I was drawn to because that's how my brain works. When I was younger I was the lead support guru for a large Fortune 50 product launch. These days I do high-tech corporate training, so learning complex new things and diagnosing problems (when my students are trying to do lab exercises) is what I do.

If it doesn't come naturally to you, you probably still need to solve problems -- so you should think about how you can improve your problem-solving process. Give @lowtek's steps a try. You may find out you have more problem-solving skillz than you realized!
 

Bekit

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
494%
Aug 13, 2018
1,133
5,596
So I was responding to a thread, and realized that one problem I see with some folks, newbies in particular, is that they lack a systematic methodology for teaching oneself.
...I believe that there exists a better way to teach oneself.

@lowtek this is gold! +Repped!
Seems like learning how to learn is the fundamental skill that will unlock all the doors.

I like your approach. Solid. Keeps the focus on taking action.

It also gives the learner permission to utilize "brute force" as a solution mechanism. I think in my case, the reason I turn to books first is that I've been trained to perceive brute force as a bad way to solve a problem. Why reinvent the wheel when others have already create a proven, streamlined procedure?

The problem is, reading the book or finding the "7-step how-to" or listening to the podcast creates a false sense of having "done something," which then leads to inaction. I've satisfied the craving in my brain to "know how," but I haven't acted yet.

So a brute force method would be better in cases like these, because until I actually stumble upon the real solution, I haven't satisfied that craving in my brain to know how, but I have taken consistent action on all the guesses.

As a thought experiment, I applied your method to the thing I'm currently stuck on.

I don't know how to get clients.
OK, that means I need to learn how to learn this. So let's go...

Isolate the problem: No clients.
Guess the root cause of the problem:
  • I haven't talked to my potential clients
  • I haven't researched to find potential clients to talk to
  • I don't have an effective sales strategy to close leads
  • I don't have a system to fill my pipeline with leads
  • I haven't put serious effort into showing my clients the value of what I do
  • I haven't scheduled time in the week where I focus on solving this problem
  • [Look at that... just off the top of my head, I have all those solid guesses... this is working!]
Address each root cause in turn: TBD - but I think I'd start here:
  1. Schedule time in the week where I focus on solving this problem
  2. Put together some materials that will show my clients the value of what I do so that I have them on hand for #4
  3. Do research to find potential clients to talk to
  4. Talk to those potential clients
Acquire any new skills to solve that specific problems: TBD - will wait until I run into the need for an acquired skill.

*whew* ... this is not rocket science, but I literally did not see the simplicity of my next steps forward until I conducted this exercise just now. I am NOT STUCK AFTER ALL.

Seriously, thanks @lowtek for this awesome guide.
 

Niptuck MD

plutocrat-in-training
Read Rat-Race Escape!
Read Fastlane!
Read Unscripted!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
164%
Aug 31, 2016
1,421
2,330
NORWAY - POLAND - WEST EUROPE
great post as usual. Physicists are the ultimate modern-day thinkers of problem solving and this confirms the latter.
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

YoungPadawan

Miles to go before I sleep
Read Rat-Race Escape!
Read Fastlane!
Read Unscripted!
Speedway Pass
User Power
Value/Post Ratio
280%
Nov 7, 2015
498
1,392
30
This is great stuff! It's basically like the scientific method for solving problems.

I've used a similar approach in developing standard operating procedures. I write down exactly what I did the last time I did a certain procedure, and if there are any issues, I look at the procedure to identify the exact point that is causing the problem, and adjust. Good old cause and effect at work
 

Tryptofy

Contributor
User Power
Value/Post Ratio
230%
Aug 30, 2017
10
23
37
Germany
I actual really like that. Coming from a scientific background myself I never thought of applying it to business.

That's what I just did for my website:

Problem: No Visitors
Cause
  1. Nobody knows about the website
  2. Not enough games in the database
  3. No SEO
  4. Maybe bad user experience / usability
Solutions
  • Take time in the week to get to it
  • Create a list of links where I can advertise for free ( like reddit )
    • Do the copywriting
  • Add 100 games to the database
  • Do SEO for Landing Page and maybe others
  • Improve usability
    • Search
    • Let the user sort platform specific game lists
New Skills
  • SEO, what are the standards today
  • Usability in web context
  • How to copy write, get examples
 

lowtek

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
332%
Oct 3, 2015
2,164
7,186
42
Phoenix, AZ
I actual really like that. Coming from a scientific background myself I never thought of applying it to business.

That's what I just did for my website:

Problem: No Visitors
Cause
  1. Nobody knows about the website
  2. Not enough games in the database
  3. No SEO
  4. Maybe bad user experience / usability
Solutions
  • Take time in the week to get to it
  • Create a list of links where I can advertise for free ( like reddit )
    • Do the copywriting
  • Add 100 games to the database
  • Do SEO for Landing Page and maybe others
  • Improve usability
    • Search
    • Let the user sort platform specific game lists
New Skills
  • SEO, what are the standards today
  • Usability in web context
  • How to copy write, get examples

Love it. What's your background?
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

lowtek

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
332%
Oct 3, 2015
2,164
7,186
42
Phoenix, AZ
This is great stuff! It's basically like the scientific method for solving problems.

I've used a similar approach in developing standard operating procedures. I write down exactly what I did the last time I did a certain procedure, and if there are any issues, I look at the procedure to identify the exact point that is causing the problem, and adjust. Good old cause and effect at work

It's pretty amazing how useful the scientific method is, even outside of a laboratory.

It also sounds like what you're doing is part of the LEAN methodology. I had training in that while I was at Intel, which was funny because they were one of the most un lean organizations. It was merely lip service.
 
G

Guest24480

Guest
These are the types of posts this forum needs more of. People can go out and apply this right away no matter what they are working on.

I am learning Linux in my free time and have applied similar principles. I wanted to learn practically so I decided to configure an email server from scratch. Mind you, I know a decent amount about email but I am nearly clueless when it comes to Linux administration.

The first thing I looked for was a step-by-step tutorial that utilized the exact technology stack I was, but I was having a hard time finding a decent one. I almost gave up because of this, but decided I needed to take a piecewise approach to this instead of looking for someone to hand the answers over on a silver platter.

So I broke the task down into pieces:
1). How do I set up basic security measures on a Linux server? --> Learn about that --> Implement
2). Shit, my firewall is blocking some ports I need open.. --> How do I unblock them, which ones? --> Implement
3). I need to setup DNS so the Internet knows who I am --> Which records, how to set up? --> Implement

etc, etc, etc.

Now within a few days I have an email server setup that can send mail locally. Wait..the Internet can't send mail to me this is pointless..but the tutorial didn't say how to configure that..

Oh well, I am figuring this out. Which port do they need open to send mail inbound to me? --> Find that out. Okay I know that, now I need to see if I can connect to that port? I can't? Great, let's figure out how to open it up. Oh, it's this obscure setting in the config file that some dude on a Linux forum knew. Cool. Time to change that setting and test. Shit, my MX record is trash. Fix that. Boom, now I can receive mail from the Internet. But it's not encrypted..

Learning is fun when you do it this way.
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

YoungPadawan

Miles to go before I sleep
Read Rat-Race Escape!
Read Fastlane!
Read Unscripted!
Speedway Pass
User Power
Value/Post Ratio
280%
Nov 7, 2015
498
1,392
30
I would also like to add: you could build an incredibly powerful intranet or knowledge base for your business if you were to write out something like:

Problem:
Website not converting

Possible root causes:
1) the page loads so slowly that people leave before it finishes
2) The color scheme makes peoples’ eyes bleed, so they can’t find the buy button
3) The copy reads like it’s written by an overseas scammer
4) The perceived risk is too high

Then, after you solve each one, for the problems that you may face again in the future, you could write out what you did to solve it that worked successfully. For instance:



"How to solve slow loading pages"

Summary
How to solve slow loading pages that cause an increase in bounce rates.

Steps to take
Write how to solve it

Commentary
Any additional details that could help more fully describe the scope of the activity.

You could create a flow chart as well to document it so that it is as easy as possible for someone new to understand. You could write this data into Evernote and whenever you have this specific problem, you just type in certain keywords and it will find the exact working procedure you're looking for.
 

lowtek

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
332%
Oct 3, 2015
2,164
7,186
42
Phoenix, AZ
I would also like to add: you could build an incredibly powerful intranet or knowledge base for your business if you were to write out something like:

Problem:
Website not converting

Possible root causes:
1) the page loads so slowly that people leave before it finishes
2) The color scheme makes peoples’ eyes bleed, so they can’t find the buy button
3) The copy reads like it’s written by an overseas scammer
4) The perceived risk is too high

Then, after you solve each one, for the problems that you may face again in the future, you could write out what you did to solve it that worked successfully. For instance:



"How to solve slow loading pages"

Summary
How to solve slow loading pages that cause an increase in bounce rates.

Steps to take
Write how to solve it

Commentary
Any additional details that could help more fully describe the scope of the activity.

You could create a flow chart as well to document it so that it is as easy as possible for someone new to understand. You could write this data into Evernote and whenever you have this specific problem, you just type in certain keywords and it will find the exact working procedure you're looking for.

Spot on. This is precisely what we did at Intel in a high volume manufacturing environment.

Rep+
 

YoungPadawan

Miles to go before I sleep
Read Rat-Race Escape!
Read Fastlane!
Read Unscripted!
Speedway Pass
User Power
Value/Post Ratio
280%
Nov 7, 2015
498
1,392
30

André Casal

Bronze Contributor
Speedway Pass
User Power
Value/Post Ratio
140%
Jun 1, 2018
99
139
This is very well stated. I'm a computer science tutor and I often tell my students to use this method to solve any problem they encounter (in computer science or in life):
1. Clearly interpret, understand and define the problem
2. Plan a solution that solves that problem (if programming, write commented pseudo-code)
3. Execute

But I like the idea of isolating the problem, that is a golden nugget!

Thanks bud!
 

lowtek

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
332%
Oct 3, 2015
2,164
7,186
42
Phoenix, AZ
Some really great replies here... I'm so glad that people are finding value in the content.

Apologies for taking so much time to get to part 2 of this thread. I've been overwhelmed with other things, so I haven't made time for this ... until now.

What to do when everything fails - appeal to principles

Story time...

My graduate work was in a field broadly called spintronics, which is using the electron's spin (as opposed to charge, in electronics) as a unit of information. My work in particular fell under the category of nano magnetics, meaning that we used advanced fabrication techniques to make magnets that were on the scale of tens of nanometers. We subjected these magnets to external magnetic fields and microwave radiation and observed their behavior.

Much like the magnet in a compass, these nano magnets would spin around the external field, and in that process they would emit microwaves in the GHz regime. By subjecting them to external microwave fields, we could synchronize them to resonant frequencies and observe some really fascinating behavior.

It was pure science, so not really engineering related. All these phenomenon occurred at 4 degrees above absolute zero, and disappeared at liquid nitrogen temperatures, so you won't see the tech in a device any time soon.

The process to fabricate these devices was quite strenuous. It required around 10 hours of highly focused and highly skilled labor, and the fragility of the devices ensured that mistakes meant that you wasted the entire days' effort. I worked along side my professor, as I was the only student in the lab. He was known for firing people who didn't want to put in the work, and I guess I passed the test. I got to see first hand what it takes to push a project from idea to completion and it was this set of experiences that I believe have set me up for success, but I digress...

During October of 2010, suddenly... the process stopped working. Every device we made was dead. We couldn't even pass a current through the devices (the resistance was in the mega ohm regime), which would have given us some basic information to diagnose the source of the problem.

To say this was "not good" was an understatement. This was our livelihood. Our only avenue of research. If it didn't work, we had nothing.

Anxiety set in.

Keeping a cool head, we followed the process I outlined above: what could be the possible root causes?

We made a list and spent the next two weeks working non stop, tweaking every step of the process.

Slowly.

Methodically.

But there was no respite to be found. Each change was only met with more failure. None of our mental models were correct. We had skewed all the variables to such an extent that at least a single device should have worked. Yet ... nothing.

Panic sets in. How could we get back on track?

One night we sat down exhausted. It was almost midnight; we had been fabricating all day and were met with more dead devices, yet again.

My mind began to wander... when did all this start anyway?

"A couple weeks ago", my professor replied.

I began to stretch for reasons... and came up with one of my most brilliant insights in grad school. I realized that the weather had recently changed. We went from warm and humid, to cool and dry.

What happens in dry air? Static discharge.

"I've got it. It's not the process"

"What, then?"

"It's static. We're killing the devices as we test them. It may be imperceptible to us, but even a tiny jolt to a nanomagnet is like a lightning bolt."

Ding ding ding. We have a new mental model to work with.

The following day, we revert back to the old process and instead of keeping the samples in a plastic container, we put it in an aluminum boat. A literal Faraday cage. We added grounding to our soldering station, and voila....

Success. The devices live, our overall success rate improved relative to the baseline, and we can move on with the science.

We had made the assumption that the process was killing the devices, when in reality it was our testing procedure.

So what's the moral here? The moral is this:

When all else fails, and you have exhausted all possibilities, something you have assumed is false. This is a foundational principle of problem solving (which is really learning).
 

YoungPadawan

Miles to go before I sleep
Read Rat-Race Escape!
Read Fastlane!
Read Unscripted!
Speedway Pass
User Power
Value/Post Ratio
280%
Nov 7, 2015
498
1,392
30
Some really great replies here... I'm so glad that people are finding value in the content.

Apologies for taking so much time to get to part 2 of this thread. I've been overwhelmed with other things, so I haven't made time for this ... until now.

What to do when everything fails - appeal to principles

Story time...

My graduate work was in a field broadly called spintronics, which is using the electron's spin (as opposed to charge, in electronics) as a unit of information. My work in particular fell under the category of nano magnetics, meaning that we used advanced fabrication techniques to make magnets that were on the scale of tens of nanometers. We subjected these magnets to external magnetic fields and microwave radiation and observed their behavior.

Much like the magnet in a compass, these nano magnets would spin around the external field, and in that process they would emit microwaves in the GHz regime. By subjecting them to external microwave fields, we could synchronize them to resonant frequencies and observe some really fascinating behavior.

It was pure science, so not really engineering related. All these phenomenon occurred at 4 degrees above absolute zero, and disappeared at liquid nitrogen temperatures, so you won't see the tech in a device any time soon.

The process to fabricate these devices was quite strenuous. It required around 10 hours of highly focused and highly skilled labor, and the fragility of the devices ensured that mistakes meant that you wasted the entire days' effort. I worked along side my professor, as I was the only student in the lab. He was known for firing people who didn't want to put in the work, and I guess I passed the test. I got to see first hand what it takes to push a project from idea to completion and it was this set of experiences that I believe have set me up for success, but I digress...

During October of 2010, suddenly... the process stopped working. Every device we made was dead. We couldn't even pass a current through the devices (the resistance was in the mega ohm regime), which would have given us some basic information to diagnose the source of the problem.

To say this was "not good" was an understatement. This was our livelihood. Our only avenue of research. If it didn't work, we had nothing.

Anxiety set in.

Keeping a cool head, we followed the process I outlined above: what could be the possible root causes?

We made a list and spent the next two weeks working non stop, tweaking every step of the process.

Slowly.

Methodically.

But there was no respite to be found. Each change was only met with more failure. None of our mental models were correct. We had skewed all the variables to such an extent that at least a single device should have worked. Yet ... nothing.

Panic sets in. How could we get back on track?

One night we sat down exhausted. It was almost midnight; we had been fabricating all day and were met with more dead devices, yet again.

My mind began to wander... when did all this start anyway?

"A couple weeks ago", my professor replied.

I began to stretch for reasons... and came up with one of my most brilliant insights in grad school. I realized that the weather had recently changed. We went from warm and humid, to cool and dry.

What happens in dry air? Static discharge.

"I've got it. It's not the process"

"What, then?"

"It's static. We're killing the devices as we test them. It may be imperceptible to us, but even a tiny jolt to a nanomagnet is like a lightning bolt."

Ding ding ding. We have a new mental model to work with.

The following day, we revert back to the old process and instead of keeping the samples in a plastic container, we put it in an aluminum boat. A literal Faraday cage. We added grounding to our soldering station, and voila....

Success. The devices live, our overall success rate improved relative to the baseline, and we can move on with the science.

We had made the assumption that the process was killing the devices, when in reality it was our testing procedure.

So what's the moral here? The moral is this:

When all else fails, and you have exhausted all possibilities, something you have assumed is false. This is a foundational principle of problem solving (which is really learning).
This is brilliant! Rep++
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

YoungPadawan

Miles to go before I sleep
Read Rat-Race Escape!
Read Fastlane!
Read Unscripted!
Speedway Pass
User Power
Value/Post Ratio
280%
Nov 7, 2015
498
1,392
30

garyfritz

Silver Contributor
Read Fastlane!
Summit Attendee
User Power
Value/Post Ratio
116%
Jul 16, 2011
694
807
Colorado
When all else fails, and you have exhausted all possibilities, something you have assumed is false. This is a foundational principle of problem solving
"When you have eliminated all which is impossible, then whatever remains, however improbable, must be the truth."
-- Sherlock Holmes
:)
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

NMdad

Gold Contributor
Read Fastlane!
Read Unscripted!
Speedway Pass
User Power
Value/Post Ratio
224%
Aug 6, 2017
612
1,370
New Mexico
Well stated--not just the general idea/overview, but the specific example is super helpful in understanding how this process is fundamentally different than how most of us attempt to solve problems.

The focus on process instead of result--paradoxically--makes the result more likely.

I need to remind myself of the process & apply it more regularly to business. Ironically, I use that process all the time in my day-to-day technical work, but less so to business experiments & projects.

Once you realize it's a process, you'll see it everywhere. We had a plumber come out at our house to fix the heat (boiler & radiant heat system, with aging components); he followed the exact process: observe, hypothesize, act, repeat.
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

MJ DeMarco

I followed the science; all I found was money.
Staff member
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Rat-Race Escape!
Read Fastlane!
Read Unscripted!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
445%
Jul 23, 2007
38,025
169,133
Utah
I'm not even done with the thread but read enough to see GOLD.

As a reminder, not all of the mods (including myself) read EVERYTHING on the forum, including some of the things we get tagged in! (Sorry!) As such, potential GOLD threads escape or purview. If you read legendary content and believe it should be GOLD or NOTABLE, please tag myself or a mod (@Andy Black @Fox @Vigilante or @AllenCrawley ) so one of us can get to it and tag appropriately! Thanks @YoungPadawan for doing so.

@lowtek -- thanks for this contribution. And thank you for saying "hello" to many of the new members here at Fastlane -- I do see it!!
 

lowtek

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
332%
Oct 3, 2015
2,164
7,186
42
Phoenix, AZ
I'm not even done with the thread but read enough to see GOLD.

As a reminder, not all of the mods (including myself) read EVERYTHING on the forum, including some of the things we get tagged in! (Sorry!) As such, potential GOLD threads escape or purview. If you read legendary content and believe it should be GOLD or NOTABLE, please tag myself or a mod (@Andy Black Black @Fox @Vigilante or @AllenCrawley ) so one of us can get to it and tag appropriately! Thanks @YoungPadawan for doing so.

@lowtek -- thanks for this contribution. And thank you for saying "hello" to many of the new members here at Fastlane -- I do see it!!

Thanks MJ!
 

daru

Bronze Contributor
Read Fastlane!
Read Unscripted!
Speedway Pass
User Power
Value/Post Ratio
172%
Feb 11, 2017
166
285
Earth
Very good thread! Thanks @lowtek .

I don't know if it's just me but putting things on a blank paper the old fashioned way with a pencil seems to trigger some sort of better problem solving neurons in my brain?
It's fast. Cheap. Easy to edit. No rules, just texts, arrows and symbols.
 
Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

lowtek

Legendary Contributor
FASTLANE INSIDER
EPIC CONTRIBUTOR
Read Fastlane!
Summit Attendee
Speedway Pass
User Power
Value/Post Ratio
332%
Oct 3, 2015
2,164
7,186
42
Phoenix, AZ
Very good thread! Thanks @lowtek .

I don't know if it's just me but putting things on a blank paper the old fashioned way with a pencil seems to trigger some sort of better problem solving neurons in my brain?
It's fast. Cheap. Easy to edit. No rules, just texts, arrows and symbols.

I've noticed the same thing. When dealing with complex problems, it often helps to map out the relationships inputs and outputs.
 

Post New Topic

Please SEARCH before posting.
Please select the BEST category.

Post new topic

Guest post submissions offered HERE.

New Topics

Fastlane Insiders

View the forum AD FREE.
Private, unindexed content
Detailed process/execution threads
Ideas needing execution, more!

Join Fastlane Insiders.

Top