Lessons Learned

I've substantially updated one of the linked articles from last week's blog. I've added lessons learned so readers can understand the journey. This blog explains my purposes in those changes.

PhD Hires

Before starting Bible Time work 25 years ago, I worked at a large well known computer company. At 1 point I worked in a department that amongst other things was responsible for all PhD new hires, company wide. I had management experience before joining that department, but was not in management at the time.

The total pool of PhD graduates in the USA was so small that we kept a list of everyone working on a PhD. We knew which students were likely to stay in academia, and then we monitored the rest as possible new hires. One of my peers at the time worked the phones and tried to be on a first name basis with every PhD candidate.

We were, of course, looking for STEM degrees, but not always. Many other degree programs were on our list. For example, we hired someone with a PhD in Music. His thesis was built on a software model that proved some long thought hunches on the mathematical nature of what makes great classical music great.

There was 1 single skill that we needed to see in a PhD student. If we saw that, then they would have a standing offer from the company. If they accepted the offer, they could work pretty much anywhere within the company that they wanted. Guess what we wanted to see?

The candidate had to prove to us that they had solved some real world technical problem that nobody had ever solved before. That is what we wanted to see.

I remember in particular 1 interview trip, we traveled to Brown University in Rhode Island. Note we did the traveling, not the other way around, which is more common with less credentialed job applicants. We wanted to see if the candidate could prove this particular point.

Had he solved some problem that had not been solved before? Of course he did, but he had other career plans, competition for these skills was high, and we rarely won these battles.

Why is solving an unsolved problem so important?

Almost all interesting problems at the edge of the high tech world are of the same nature. The leading edge of innovation is solving problems that nobody else has ever solved before.

At that edge, there is nobody else to copy. There is nobody to ask for help. There is no amount of research that will help. Knowing how to thoroughly research a problem is a Masters level issue. At that level you show you can solve problems by finding a known but hidden or often transferable solution.

People on the leading edge of discovery must learn how to forge ahead without anyone else to follow. There is a set of experiences gained in this process that few people will ever understand. People who have those skills can then use them to solve other similar new problems.

By the way, this was long before the woke takeover of academia. These days you can get a PhD just by spouting your own opinions in your thesis. The truth that a PhD once represented has been supplanted by power. I doubt there is much reason to value a PhD degree as was once done.

By example, consider Elon Musk. He regularly demonstrates he operates in this same level of skill. Has anyone else ever built reusable rockets? Has anyone else brought an all electric family of vehicles to profitable mass production?

Musk has the basic skill set of solving fundamentally new problems. He also brings financial discipline and team leadership skills to his problems. Also, you might note, the media and public often misunderstand him. This is also very typical because the problems he regularly solves have never been solved before, so the techniques he uses are always new. Those with less skill will copy him later.

My time in that department seems to have been a setup for what Joshua would later call me to work on. What sort of problem is Bible Time? Or recovery of a lost manuscript?

These are both real examples of what I have been discussing here.

Getting Our Own PhDs

Years ago, Ryan and I spent a few hours with the dean of a small Christian seminary. We were discussing how to turn some of our early work, mostly Bible Time and Book Order, into actual PhD degrees.

The dean told us we had several PhDs worth of work already done. Rediscovery of the Bible's calendar, alone, is PhD level work. Identifying all the lost tribes is another. There are several other possible PhDs in our set of early projects.

For any of these projects, the thesis would need to be written, and reviewed. Perhaps we would need some prerequisite classes too. For $10,000 each, and maybe a year of class time, he'd take care of the rest and get us those diplomas.

We debated this idea for weeks after. The hilarious aspect of this was that we had already made the fundamental discoveries. We were already working at the level of PhDs. So why would we need any more course work?

We also did not have that sort of money, so the issue was moot.

But also, paying for credentials is something the text seriously frowns on. See Acts 8:18-20 if you want but 1 example.

Some of our friends heard this story. They eventually gave us honorary PhD certificates as a testimony to the nature of this work. In some ways they were also mocking the formal systems of diplomas.

Many employers in the tech world recognize this problem too. They often test candidates for actual ability commensurate with their supposed attained formal education. This is in part why Musk personally interviewed the first 500 employees at SpaceX.

So imagine you are on a trip to interview a PhD candidate. They are supposed to be solving some problem no one else has ever solved before. What do you ask them?

My peer that worked the phones had a trick he taught me.

Lessons Learned

One of the ways for outsiders to understand discovery process is through lists of lessons learned. If you are interviewing a PhD candidate, asking them for lessons learned is a way to gauge their discovery skills by understanding how unique was their discovery.

The discovery process is often a chain of lessons learned. The chain will lead from what is known about a problem out into the unknown. Of course everyone involved hopes for great lessons learned. Most often lessons learned reveal failures along the way. These are often valuable lessons, even if they are dead ends.

Some PhD level problems are never solved. Students eventually need to walk away and do something else. They can complete their degree program's written thesis by pointing future students to places that look fruitful. They do that by writing their thesis and emphasizing what did not work and where to go next.

We are not at the point of walking away, but we have often felt this way. This has been especially true after putting lots of work into failed attempts at manuscript recovery.

In thinking about these things, I decided to update the theory article that I first posted last week and expand it by including lessons learned along the way.

I also added a couple more sections. It now ends with a formal statement of our current working theory for how the text will be recovered. Here is the link to the updated article.

Paleo.In Theory (paleo.in)

Last week this article was about 2500 words, it is now over 6000. If you want to understand better what we have done to understand the problem of manuscript recovery, this is now a better place to start.

More Later,

Phil