Business, Innovation, and Managing Life (August 14, 2024)

Hi there. Welcome to another episode of Business Innovation and Managing Life. And I see a few questions here, though I'm happy to answer any new ones that come in. Let's see. Well, anon asks, this is from our submission form. Can you tell us more about your book collection or artifact or other collection? I suppose I have. Oh, gosh, must be, I'm guessing around 10,000 books that and that I've accumulated over a long period of time. I mean, I still have a bunch of books that I got at the beginning of the 1970s. I used to have a habit of writing. The date, my name and the date when I got each book in each book was a good habit because now I know that I got a particular book in June of 1972 or something like that. I stopped doing that at some point. I'm not quite sure why. And that's meant that I really don't know when I got a bunch of books, except for the fact that I've had a database of books. That has some information on when I got these books. What are all these books? Well, I started getting physics books around 1971 when I was 11 years old. Accumulated lots of physics books. I've carried them around everywhere I've gone. I lived in the same place now for 20 something years near Boston. And we built a house which has sort of a necessary feature of the house was that it had enough space for what's now close to 10,000 books, which means lots of bookshelves and enough levels of bookshelves that there's sort of an elaborate ladder structure that one has to kind of wheel around to get to the top shelves. And it's always been an interesting question for me which books go on the top shelves? And it's usually books where I was interested in that topic once, maybe I will be again, but it'll be a while. And then there are occasionally times when I'm like, okay, now I'm actually going to work on this topic. It's time to move those books down to a lower shelf. I think I might be about to do that for books that I have about economics, because that's a topic that I'm thinking I'm going to work on soon. So I started accumulating books that I have now in the early 1970s. And I would buy lots of books and I found it and then I had various schemes for arranging these books. They're now in categories, probably, I don't know, 40 categories, 50 categories, maybe that things like discrete math or mathematical logic or scientific Biography or other kinds of things. And within each section I had had various schemes for arranging the books. In the end, in most sections they're arranged alphabetically by author. Because my other schemes, like arranging by date of publication, it was basically impossible to reshelve books without a huge amount of effort. I had a whole plan back around the year 2000 to have kind of an image of every shelf of my bookshelf. This was at a time when it wasn't so easy to take pictures of things on phones or anything like that, but to have an image of every shelf and to know based on the color and thickness of the book, where the book will be reshelved. But anyway, I've ended up with a big collection of books about scientific, technical subjects, Some other subjects that I've been interested in. Lots of biography books and scientific history type books. Then at some point in the 1990s, I started buying kind of antiquarian science books and technology books. And I got pretty good collection, I think, right at the right time. Because at the time when I was buying these things, there wasn't a lot of kind of high budget interest in those kinds of books. And so they were of modest price. That changed. And a bunch of people in the tech industry, I suppose, had made a lot of money and decided they wanted to collect antiquarian science books and so on. And so prices kind of went up by factors of 10 and things like this. So no doubt the collection I have is quite valuable, although it's not my interest to sell it or anything. I particularly concentrated, I would say, on the minor works of major figures, partly because the major works, if you want to just read the thing, you can easily find it either these days online. But back in the day, I typically would have a copy of a recent edition of, I don't know, Newton's Principia or something like that, because that's something that's been widely reprinted. But if you take Newton as an example, the last book he wrote that came out just after he died was a book called the Chronology of Ancient Kingdoms, which was a very obscure book that he was interested in kind of doing history to figure out when certain events that were described had happened as a way to kind of date things for purposes of doing better astronomy. His kind of view of what had actually happened in history was in many cases quite wrong. But that book is interesting to read and it's something which maybe there's some other edition of it, I'm not sure, but that's something where the original edition is you can get the original edition and it's not clear there's a subsequent edition. So that's the kind of thing I also collected. Oh, lots of things which are kind of pseudo books like manuals for very early computers, things like this. So these make for interesting reading and they're fun to look at. And one thing that tends to happen, because I work on projects that involve history. I'll end up often realizing, gosh, I actually have a copy of such and such a book or whatever. And that's a very useful thing to be able to go through when I'm doing these history projects. So I've ended up with. There are other kinds of books, like sort of popular science books. Not quite sure quite what I end up doing with those. They end up usually going into shelves that are not as readily accessible as others just because it's sort of. I feel like I should own the book. And maybe it's written by a friend of mine or something like this, but I'm probably not. It's not a book that I'm going to particularly need to refer to except by virtue of sort of knowing what that person wrote about, so to speak, I guess. Other categories of books I have. I have many shelves of books about Wolfram language and Mathematica that people have written, that people send me. And that makes for an interesting collection. Sometimes it's a bit challenging because there's a book on a very nice book that happens to be very much based on Wolfram language, but it's about a subject and it's always kind of a little bit. Do I shelve that with Wolfram language books, or do I shelve it with the subject? And I have been not completely consistent in that regard. But I think what tends to happen, like the books that I have right near here are books this happens to be on the side of the main office I. I work in. And I've ended up, when I work on particular projects on sort of timescales of many months, I'll move books around into the sort of buffer area of things that I'm actually going to pay attention to. Now. Some books I have sort of because I think they're interesting to have nearby. And I'll occasionally look at them as some form of inspiration for this or that. I have the challenging problem that some of those books are really nice early editions. And I know that over a period of decades, the sunlight, despite ultraviolet coatings and things like that, the sunlight will eventually. Cause some of the dye in the spines to degrade it's kind of interesting. I happen to have. Actually, I might even have it here. So. See, this may be a disaster to pick this up, but maybe not. We'll see if this is a disaster. I happen to have a copy of New Kind of Science that I was using mostly as a thing, As an alternative to having some block to put something on. But if you look at the COVID of the book, there's my name in nice red print. If you look at the spine of the book, that red has decayed. And when we printed these books, we were kind of studying kind of the whole print process. And there was this question of which kind of ink do you use? What will survive? And the statement was, well, that dye will last for. That dye will at least last for about 20 years. And now that particular copy of the book is probably 22 years old. And it was out in bright sunlight for a while. And so that dye has somehow chemically degraded over that period of time. Well, let's see what else to say. I think I do end up continuing to buy actual books because I find the sort of. It reminds me that I have the thing. If it's just an online thing, well, I put it in folders. And when it's a folder related to a specific project, yes, I'll find it when I'm doing that project. But if it's something that I'm sort of more generally interested in and kind of want to remember that I have that thing. Having a physical book is a good way to have something where I can sort of remember that I have it. In terms of artifacts, Well, I have all sorts of artifacts. I would say that there are certain categories. There are a lot of natural objects that I collected. Mollusk shells. Let's see. I think I even have one within reach here. There's a mollusk shell. There you go. This is. I think that's an imperial volute, if I remember correctly. And I collected those things because I was studying the patterns of growth in biological organisms. And so I wound up with a lot of those kinds of things. Particular type. I wonder if I happen to have one. Yes, I do. Look at that. How efficient these are. That's a cone shell. It's a critter. The actual critter goes inside here and has a very nasty stinger that comes out of the back that produces a neurotoxin often called the conotoxin. Those are very nasty things. But anyway, I was studying cone shells. They have this. If I can show that closer up. They have this. If it goes into focus I'm not sure it will. They have this interesting pattern on them, which turns out to be a pattern that one can model very well with a very simple program, a cellular automaton. And there was this moment sometime around 1982 or so when I noticed that, yes, cone shells have these patterns that look like cellular automata. I subsequently discovered that when I had been a young kid, I had collected a bunch of shells, including some cone shells. But it's one of those typical cases where you have something that's sort of sitting in front of your nose and you don't notice it and you don't notice the connection until quite a lot of time has passed. But. So I have a pretty large collection of things like mollusk shells. I'm always surprised when I go and see some natural history museum and they have mollusk shells, and I look at their collection and it's like, yeah, I have pretty other kind of notable kinds of shells with interesting shapes and patterns and so on. And similarly, I've collected a bunch of fossils and other things that are interesting if you're interested in studying biological form. There also are all sorts of kind of artifacts that I've gotten at different times. Whether they are kind of. I think I might be able to reach one. Let's see. There's one that's fun. That was a very. This is a terrible thing. That was a very early. It looks like a death mask, right? That was a very early 3D printout that somebody gave me of a scan of me turned into a 3D printout. And that's just one of a very large number of artifacts that I've collected from sort of mostly the technological world and so on of. I don't know what else is within reach here. Let's see. Well, there are. Oh, yeah, that's another. That's a 3D printout. That's actually one of the. It's a 3D printout of one of the hypergraphs that shows up in our physics project. But, you know, I've just got lots of these kinds of things. And really the challenge is, how does one sort of organize them in a way that one will ever find them again? People give me interesting artifacts, and I have these. A few drawers of where I put these things. You know, I've got other artifacts that I collected. Like, I have lots of. I have all the calculators I used to have from the. From the early 1970s and the slide rule and things like this. And I've sort of put that in my collection of artifacts. Someday I'LL I'll try and do a tour of artifacts. So it takes a little bit of effort to, to organize carrying the camera around and looking at all these things. But, but that's, that's the kind of thing I've done. I've never really been interested in sort of the incredibly valuable artifacts or books or whatever. There was a time actually in the early 1990s when I was buying a bunch of antiquarian books and somebody wanted, you know, was trying to sell me a Newton's Principia, which I think in those days was like $100,000 thing. It's now worth a lot more than that, but it was like, am I going to buy this thing? And usually the way this works is they'll send you the thing to check out. And the first question is, is the thing what it's supposed to be, so to speak. And I was lucky enough to be able to get somebody at, in that particular case, University of Illinois Library, where they had another Principia to kind of go through and flip through the pages of each one and confirm, yeah, it really does have all the pages it's supposed to have. Because in that world, kind of the price of a thing is incredibly dependent on its perfection level. And, but then there, one of the things that was, well there, there were two things that really put me off buying such an expensive object. One is what was I going to do with it? You know, I was going to end up putting it in a safe somewhere and never looking at it and just sort of, it, it's, it didn't seem like it really had. It wasn't like the antiquarian books that I have out on shelves and so on, where it's kind of like I feel just fine having them out on a shelf, taking them down, looking through them and so on. And I'm not like, oh my gosh, this is a really valuable crucial historical artifact that one sort of has a high responsibility to take care of. So that was one of the issues is what do you do with the thing that's that valuable? And you wind up putting it in a place where you can't really enjoy it, so to speak. The second thing was that in that world the value of a thing, as I mentioned, depends on a lot of details. And in particular in the case of Newton's Principia, and I haven't followed the research on this more recently, but there are sort of two versions of the Principia, one was the so called export edition and the other was the edition that was sold in England in those days when a book was printed, it was like this book is sold by this bookseller who is near this wall and this road in London or whatever else. And so there were some copies of the Principia, but I think what was it? Maybe 150 copies maybe were printed in the first print run, something of that kind? I think that's right. Few hundred at least. And the sum of those copies were, were going to be sent out of the country to people who'd ordered them from who knows where. You know, maybe Descartes ordered a copy, I, I don't know, I think he was around at this, at the appropriate time. But in any case, maybe Leibniz ordered a copy to check out his competitor, so to speak, I don't know. But there were copies that were going out, out of the country and there were copies that were going to be sold in England. And the particular one that I was thinking of getting, I believe was an export copy. And it had slightly different things printed on its front page. And maybe it had less information about how you go to the corner of this street and that street to buy the thing, because people didn't figure that would be a thing that would be relevant. But in any case, the issue was, was the export version printed after the domestic version or not? Because if it was printed after, it's no longer really first, first, first edition. And so it's much less valuable. So there was a bunch of scholarship where trying to answer that question. And in those days, people would print individual pages in a book. They'd set the type, lead type or whatever for each individual page. They'd go, you know, they'd print, print, print a bunch of copies and then they break down the type, reset it again for the next page, print that page, and so on. And the question was, if it was the case that all the pages, the export version, were printed after all the pages for the domestic version, then it's sort of not really a first, first edition. But if the pages were all sort of mixed up and they just made big piles of pages and they would pick pages off those piles that are at random to put into the domestic and export versions, then there's no kind of what was the first first edition versus not. And so how do you tell? Well, in the course of the print run that was just making impressions of successive pages, people would notice typos and they would fix them. They would move the lead type and take a little piece of lead type out and put another one back in and fix the typos. And so the question was, if you went around and looked at copies of the Newton's Principia that were in libraries around the world, could you systematically see, oh, typos had been fixed in the export version relative to the domestic version, indicating that it had systematically been printed later. So in the 1950s, people had done a certain amount of scholarship. Just I think one group of people had gone round and looked at sort of the Principias they could find. And according to that scholarship, there was no sort of ordering of export was after domestic and so on. But the value of the thing, depending by a factor of two or three. On this kind of thread, of this one academic paper that tried to define that, and one could certainly imagine that somebody would come and do a better, bigger study with more libraries and so on, and say, no, no, no, you're wrong. Essentially all the pages of the export edition were printed after the other. Then it's kind of like, well, not really a first, first first edition. So that was another reason why I thought it was sort of a risky thing. You spend all this money on something and then. Well, actually, the academic paper was a bit wrong and so it's really only worth a third as much. And that would seem like a shame. So a little bit on books and artifacts. Let's see. Memes is commenting. Spending a bunch of dough on something you're going to worry about doesn't seem like the best investment. Yeah, it's true. It's. I think, yeah, Let's see. Well, a couple of quite different questions here. One from LC about managing life and pursuits. How does a dashboard or portal web page, as I show sometimes with lists and links to projects, fit in your daily workflow and daily routines? Yeah, I end up having a page which is sort of useful links, and I use it all the time. It changes rather slowly. That's probably a good thing because otherwise I would forget what was on it. And if I want to find a link to some document that I wrote or some website that we're developing, or some product that we have, or for that matter, some piece of reference material that I have, there's a good chance that that's one of the several hundred links that's on this kind of personal homepage that I put up. And that's. I mean, it's actually the technology of that thing could be better. I mean, it starts life as a Wolfram language notebook and then it's kind of. You press a build button and it turns into Sort of pure HTML with a bunch of JavaScript stuff and so on that is sort of optimized for loading quickly. I'm not even sure that's necessary anymore. I think that the technology we have for the Wolfram Cloud is now good enough that I could probably just make it a pure cloud notebook and just, you know, whenever I have set my browsers so that whenever they generate a new page, up comes this page of personal links. And yes, there are lots of services out there that will do similar kinds of things. I just happened to build this, oh, I don't know, when the web was young, sometime in the, oh, probably in the early to mid-1990s. And so I've been sort of using it ever since. So that's an important thing. I would say that in terms of sort of the status of things. Well, yes, there are links on that page to lots of kind of status of things type pages. But also I end up getting emails sent to myself about kind of the status of things. So I'll have a piece of email that's the number of keystrokes I typed into what applications get sent every day I have a bunch of email, the number of steps I walked, resting, heart rate, time I slept, all those kinds of things that get sent to me every day by automated systems. I also have a report of files that got backed up. I also have a report of web pages I visited, these kinds of things. Those reports are mildly useful for me as a kind of way to be aware of what I'm doing and so on and sort of vaguely aware of my level of, oh, I seem very productive and I feel good that I'm productive and so on. But they're also something which is going into my email archive. And so that's something where I have sort of a uniform search interface to that. So if it's, if I, okay, I visited that website five years ago, great. I can now find it again because it's easily findable through the search. So that's something I find useful. I don't. The only kind of real time display that I have is that right is. Well, actually I have several. But the one that I use a bunch of is my email processing real time display, which just has time series plots of the amount of unprocessed and unread email as a function of time for the day, the week, the month, the year and all time. And so I can kind of see sort of how horrifying the kind of peak that's built up is. And it's rather satisfying when you can kind of grind down the peak from 5,000 unprocessed emails down to 500, let's say, takes a lot of time and effort to do that. I mean, one of the things that tends to happen, I get sent an awful lot of email by lots of people about lots of things and it's always challenging to process it. I mean, the largest component is internal email from within my company and other organizations that I deal with. I would say that of that email, maybe half is semi automated reports like monthly reports of people and things and so on. And the other half is kind of more custom stuff, email threads that I'm on, emails specifically sent to me asking questions and so on. And in the end it's probably 500 emails per day altogether, of which, as I say, a good fraction are sort of internal emails. And for me it's always very helpful from a sort of management point of view if like I see the monthly reports of a large chunk of people in our company and I glance at them, you know, some of them I'm interested in the project that's being done and I will read it in much more detail. Some people, I've been reading the reports they write for 25 years and they're always interesting and they're well written and they tell me things about what's going on in that part of the company. Anytime I can just read it in a monthly report, I don't need to have a meeting and ask what's going on because I kind of already know. And I'll often comment on some monthly report if I've noticed some particular aspect of it. I'll comment on something that might be deep in the report because I happen to notice that part. My most common comment is something along the lines of the thing you're doing is related to this other thing that's being done somewhere else in the company. And I'm kind of the only connector because if you look at the org chart or whatever, those two subtrees just don't meet until you get to me, so to speak. So it's kind of unsurprising that I'm going to be the one pointing out, oh, there's this thing on the other side of the company that's also relevant to that particular activity. So then what ends up happening, I also get, well, a, a bunch of. I mean, there's a. There's, there's mail where I can just delete it. I just glance at it, I delete it. There's the, the thing that is always Troublesome is mail where, yeah, I got to think about it. And usually when I'm on kind of a mail processing sprint, which kind of happens a bit every single day, it's kind of like if I have to think about it, it's going to break my sprint and I'm going to just put it aside and come back to it. And that's always the hard thing because I'll build up this kind of collection of sort of tougher and tougher nuts that are, you know, emails that are older and older where I kept on going through it and saying, yeah, I can't think about that now. It's going to take me half an hour to think that through, let me put it aside to do it later and so on. And I end up having to do these sessions that are very kind of take a certain amount of personal self encouragement or something to actually go through and go do the difficult emails. I have to say that that is often something which I consider that I need sort of babysitting for from various people in my office. One of the reasons that really helps is because often when there's a long and complicated piece of email to process, the reason it's long and complicated is because it requires pulling in threads from different places. And if there's someone there helping one, you can see, oh, there's this piece of email. I'm going to look at it. I've read it, I now understand what's in it. Now it raises a bunch of questions. Now we have to go research a bunch of those questions. But if somebody else can do that while I move on to another piece of email, process that they come back to me 10 minutes later and say, okay, I found this and this and this and so on. But I haven't lost context. It's not like I'm going to have to reread that original piece of email a separate time. I still got in memory roughly what was in that email. And so then I can go through and try and respond to it. And so that's always one of the difficult things. I mean, what tends to happen, I will get email from just a very wide range of people and organizations around the world. It's kind of, kind of almost shocking because I'm involved in science, I'm involved in business, and I'm involved in philosophy, and I'm involved in education, etc. Etc. Etc. Lots of different kinds of things and they lead to all sorts of different kinds of emails. And sometimes it's like, sometimes it's hard to know why somebody thought I would care, but you know, and then we have to ask, why would I care? And sometimes the answer is a very reasonable answer, and sometimes there isn't any answer at all. But I think the thing that I've tried to do over the years is to build these kind of buckets that I can put email that comes in into. I try to respond to anything reasonable that people send, or at least I try to have somebody in my office respond. I think that's unusual, but I found that I think it feels good to be able to do that. And even if the response is, we're not sure why you're sending us this email, or a fairly typical thing is people will say, you work on science about this. And I've got a theory about that too. And most of the time it's like, yes, but I'm running as fast as I can, working on the things that I'm working on, and it's unlikely that I'm going to go and chase what you're interested in as well. So it's kind of like if there's a way in which what you're doing directly relates to what we're doing, then explain it. And then maybe it's something that can be sort of helpful, or maybe we'll have something to say to contribute to somebody else from things we've done. But, you know, so there end up being these kind of. I've tried to invent these buckets of what to do with email people send. I have to say one of those buckets is ask that question on a live stream. And so I wouldn't doubt that there are questions that are queued up here that came from people sending them as email. And then we say, look, I'm not going to write you an essay in response to that question, but I do these live streams, and if you ask that question on a live stream, I'll do my best to answer it because it makes more sense for me to do it in this kind of forum where the answer will be of hopefully broader value than if I'm writing an essay back to one particular person. So, but, but that's, it's always a sort of the, the processing of email, which is sort of the principal way that I suppose I end up interacting with the world perhaps is, Is always, for me, at least, a challenging thing. I, I never use sort of other forms of, of, I don't know, you know, messaging apps and things like this. I occasionally, I get sort of pulled into some very brief use of one of those things. But it really does not fit with my personal workflow because my personal workflow involves let a bunch of stuff accumulate, maybe only for a day, maybe for a week, maybe for a month. And in an organized way, sort of say, I'm going to respond to this, I'm going to respond to that, I'm going to respond to that. And I don't expect that I'm going to get a response back to my response in two minutes. I'm expecting that the response will come back in a day, a week, whatever else. And so it'll be perhaps a more leisurely form of interaction. If I really want to have the instant interaction, then it's kind of like get on a zoom call or something and do it rather than sort of pinging back and forth with a text message where I'm multitasking between doing that and doing three other things. And at least for me, doesn't really work very well. So anyway, a few thoughts about that. KD asks, are you also the CTO of Orphan Research? What are the characteristics of a good cto? I guess I'm also the cto. I think it's like one of these things where a company is very technology oriented and so many of our senior managers are technical people. And at various times I've said sort of we have 5 ctos but no cxo where X is some other more business oriented role. Yes, I'm the one who kind of makes the kind of product strategy, product design type decisions and ultimately decides sort of large scale technological things. Although I'd been lucky enough to have a whole bunch of very talented, in many cases, very long term kind of people who are doing, who are figuring out lots of, lots of pieces of the strategy and the technology that we're building. In terms of what makes a good cto, it depends what the CEO is like in large part. I mean, in the case of our company, I'm a very technically oriented CEO and there are other roles in our company where I'm not paying as much attention. And so it's much more important that somebody is able to be a person who makes the trains run on time, so to speak, without me having to sort of personally check off every train type thing. There are companies where the CEO is not very technical and where it's the CEO has some vision, maybe the CEO is oriented towards very much more in the commercial direction, in the sales direction and so on. And where there's a critical need that the CTO can just deliver, the CEO has a vision. The CTO will deliver and it's often very difficult. I know many people who run companies where the CEO is not nearly as technical as the products they build. I mean, it was a shocking fact back in the, oh gosh, beginning of the 90s when our company was quite young and I would go to computer industry conferences and at that time, and times have changed since then, at that time Most software company CEOs could not demo their products. In fact, it was a bunch of times Bill Gates and myself were the only people who could demo our products. At these early 90s computer industry get togethers, that was a time when there was a lot of people who were CEOs who were kind of professional managers, so to speak, that really changed and it really, you know, the geeks took over type thing and it really was technologists who ended up being CEOs of tech companies. It's not always the case anymore and it isn't the case as companies get older and so on. But there's this situation where there's a non technical CEO managing a company which critically depends on technical advancement. And that's a challenging situation because if you're a CEO and you're trying to manage technology, then a very common thing happens to me all the time is people will just come back to you and say, oh, that's impossible. I think it even happened today to me. I'm trying to remember, yes, did happen today. And to me and so it's like, okay, but I have a good position of being able to drill in. Why do you say it's impossible? What is it about it that's impossible? Oh, technical issue, technical issue, technical issue. Okay, well, I am probably knowledgeable enough and more than that, confident enough that I just say, just explain the issue to me in detail, don't hold back, oh, you won't understand this type thing. And for me, in my way of managing our company, for example, that works well because I am sort of technically knowledgeable and confident in my ability to understand technical things. When people are managing companies and aren't in that way and where the company really depends on technology, it's a tough situation. And in that situation there really has to be a CTO in place who can fully be trusted to know what's impossible and know what's not impossible, so to speak, and to push, push, push for the things where some engineer is saying oh, that's impossible, which can be translated into I can't figure out how to do it or sometimes simply I don't want to do it. And that's you know, to be able to provide that, that level of confidence. If the CTO says it's impossible, it really is impossible and it doesn't need the CEO sort of second guessing. Wait a minute, are you sure it's impossible? Et cetera, et cetera, et cetera. I think the other thing is there are always sort of large scale technology implementation kinds of decisions that have to be made. And it's like, do you do it right? Do you make the right judgment call? And this one thing is, are you sort of technically architecting things? Right? And the other is, let's say you're going to bet on some type of technology. Is this a technology which is going to survive and grow or is this going to be a dead end technology? Oh, we wrote all that stuff in, let's say Java and we regretted it. We should have written it in something else and now we're stuck because Java doesn't run on the iPhone or something like this. You know, it's hard to see around those corners. But that's an important piece of being, I would say, a good sort of technology implementation strategist. Another issue with CTOing is kind of managing the technical troops, so to speak, which is again for a technical CEO, it's much more kind of the technical CEO can understand the issues of the technical troops, so to speak. A non technical CEO has a much harder time with that and assessing them and so on. So I mean, I would say that it's true with any C level position, but particularly CTOs, that it's kind of like, can you get a level of trustworthiness about whether you're making the right judgment calls and sort of communicating what's possible and what's not to the CEO level if the CEO is not capable of kind of looking in themselves? Now there's also the big question for CEOs and so on. How much do you just say to the people who are sort of in executive level positions, just do your thing and let me not meddle in it? I have to say I don't like to meddle because I have a million things I want to do. And if there's an area of our company where I don't have to know what's going on, it's great by me. The only problem is that I know from experience that if there are areas where I really don't know what's going on, don't understand what's going on, they will decay over time. Sometimes there'll be people, I've had great people who've kept for decades parts of the company, sort of at a certain energy level. But I would say that there's some decay time and that decay time is variable. And if I really am not checking in, it will decay and there will be things that go wrong and things where people. Where things are, oh, we can't do that because of this thing that's over here in this part of the company that's doing this or that thing, maybe some structural kind of thing. We just can't do that. And I don't understand enough to be able to say, no, that's just not true. We could figure out how to do it. So there's this sort of question of how much do you meddle versus how much do you let people get on with things? And as I say, my predilection is to not meddle. But I'm also well aware of the fact that you have to pay enough attention to kind of keep the thing from decaying, so to speak. And the things that I truly don't understand usually have not gone well. And over the course of decades, I've kind of had to poke into every part of our company and think, I have a pretty good understanding now of all the parts of it. But any place, oh, I don't know, like, for a long time, the transaction processing system that we had, I never paid attention to it. I didn't care about it. It's a thing where I don't think it's that different from what it is in many other companies, although that's evolved over time. And it's just like, so long as it runs and works, I don't need to know about it. And at some point that had decayed to the point where it was just like, oh, we've got a new product coming out. Oh, but it's going to take two weeks to put it in through the transaction processing system. Well, that's completely crazy. So something is very, very wrong. Oh, that's because there are layers of crud that are built up here. And like five years ago, we just decided we're getting rid of the ERP system, the enterprise. What is it? Resource planning. I think that's what that stands for, basically, transaction processing system that we had and we're going to write our own in Wolfram language, which is now largely deployed and is quite beautiful and has fixed those kinds of problems. But that was a place where after a long time, one kind of poked one's nose in and tried to make it better. But then there are places where it's sort of an issue of to meddle or not to meddle. And the thing that I've realized, realized it took me a long time to realize this, is if you can be successful in solving a problem, then you might as well just go and solve it. In other words, if there's something where there's some horrible bug or some issue and people are sort of spinning around and it's taken months and it's kind of like we don't know what's happening, et cetera, et cetera, et cetera, for me, it's better to just say, okay, let's set up a meeting. Let's look at the code, whatever else it is, explain the detailed issue, and let me try and do my best to solve the problem. And those meetings tend to go really, really well. People will sometimes, if they're not used to these kinds of things, they'll resist them because they'll think that those meetings are going to be me saying, you're a bunch of idiots. Why couldn't you solve this yourselves? Which is never what happens. It's because that's not a useful thing to say and might not even be true. It's just that the experience people have. I've accumulated a lot of experience. I'm pretty decent at solving problems. It's like they've tried pretty hard, but they haven't managed to solve the problem. But when I do solve the problem, essentially always I'm solving the problem by reflection from the people who've already tried to solve it. In other words, it's not like I'm going in and starting from scratch and saying, let me start from the beginning and ignore everything you've done. Rather, what's happening is it's me trying to say, let me understand what you've done and ask a bunch of questions. And the most common way those discussions evolve is somebody says, oh, that's an interesting question. Oh, yes. And then they basically solve the problem right there, or they come up with, you know, let's try this type thing, and then it ends up solving the problem. So I've, you know, I had thought, oh, you should never do that. You should never kind of jump down into the, kind of the. Into the trenches of the organization and try and solve these things. It's not like I do that every day, but I probably do one of those on average, I don't know, every two weeks or something and in different, different areas of the company. And it's something, maybe it's about the right frequency. And I think it's been a really big positive doing that. I think it has several effects. One is that sort of folks in the trenches who are working on something which they might have thought the CEO could not care less about, they discover that actually one does care about it and one understands something about what's involved. And it kind of is. That's, I mean, that's an encouraging thing to see. Another thing is that frankly, if I can solve it or can contribute to it, people are like, yeah, okay, the CEO actually has a clue. And it's a good thing to feel that the CEO has a clue. Because if you feel the CEO doesn't have a clue, then when you get asked to do something and it doesn't make sense to you, you're going to immediately conclude the CEO doesn't have a clue. This is just a crazy thing to do. Rather than, yeah, maybe the CEO does have a clue. Even though I don't personally understand why I'm being asked to do this, it still probably makes sense. So that's been my experience. Let's see, there's a question here from Ken. Do you find video calls draining? Yeah, I don't like video calls. I mean, I do. I spend my, my life when I'm not just working on things on my own. I spend my life doing meetings with people. I've been a work from home CEO for 32 years now and 33 years now. And so that's been, you know, I've gone through many different technologies of this. What I do is audio plus screen sharing. I don't do video. I found that I do video when I'm having an actual sort of human conversation with one person and so on. But I don't do it when I'm doing kind of get work done type meetings. For some reason I do find it extremely draining and I've just not found it terribly useful. I mean, I found that sometimes with meetings, one of the issues is actually that you want people around in the meeting who might have something to contribute, but you don't necessarily want them paying attention all the time because actually most of the meeting is not relevant to them, but you want them there because you want to be able to say, hey, so and so, what do you think about this? And maybe you have to repeat two sentences, but you're not like, oh, can we go find this person? Oh, let's set up another meeting with that person in it type thing. It's something where you can get an immediate response, but you don't want that person sitting in this meeting. That's absolutely irrelevant to them mostly and not doing other work. But yet if you have them on video, it's like you see them completely distracted and it kind of takes. Sucks all the energy out of the meeting. So that's been an effect. But sometimes I have to say when I get a little bit annoyed because we're in some meeting and nobody's saying anything, people are not responding. Do people understand what I'm saying? Et cetera, et cetera, et cetera. I end up occasionally, maybe once every few months, threatening to switch the cameras on so that I actually can tell whether people are thinking about the answer or just off, totally spaced out type thing. I don't think it's ever actually happened that we've actually switched cameras on. I think it's been more like then somebody will respond and the meeting starts flowing properly again. I don't know what it is about the sort of human dynamics of having video on. But yes, I agree. I find it draining. I've occasionally had it on, like when we've done some components of our summer school when we were doing that, some of that remotely. And we have one component of it that's remote. I've ended up spending many hours in a day with video on. And yes, I find it. I find it strangely draining. I can't explain the psychology of it. Maybe it's related to an experience that I had a few weeks ago. I did an interview that you might have seen because it was a live stream. I did an interview with a humanoid robot that a friend of mine has built. And I ended up spending an hour, or a little bit more than an hour talking to this robot. And the robot has pretty realistic eye motions and so on. And I wound up realizing that I was looking straight at this robot for more than an hour. And it's not clear anybody has done that before with this robot. And it was a really weird after. After I did it, it felt really weird. And I think that that weirdness might be somehow related to this thing. Oh, I'm concentrating on this little box on the screen for my. For my video call. Let's see. Young asks rather nicely, what's a continuity plan in the event that something happens to me because I'm instrumental in development of software and the gap will be hard to fill. Yeah, I think it's not. The good news is that we have a bunch of people who really understand our technology and our strategy and so on, and in many cases are much younger than me. So I think there's a good collection of Folks there. One of the things that. What I tend to add is sort of pulling all the threads together. And it is useful to have one person who pulls all the threads together, because that way there's sort of a single. Well, you end up with sort of, if you do your job right, sort of complete coherence. If there were three different people pulling threads together, there'd still be the question of coordinating between those. The other thing that happens is, you know, when do you make the decision to do something new and what. Who. How do you. Within an organization, how do you say, yep, we're going to do this new thing? And that's a much harder thing to do with a group of people than it is to do with one individual person. So that's kind of a thing that I end up finding myself. Yes, we should actually do this new thing. We've been talking about it for a while, but now, yes, it's actually the thing to do. I'm planning on sticking around for as long as I can, but I think the culture that we built at our company and the actual group of people that we have, I think is robust enough that I'd like to feel like I'm not as essential to the story as one might expect. But not to say that I don't intend to stick around and do what I can to sort of move things forward for as long as I can. Ollie asks, did you have a mentor while creating your business? Do you find mentors useful? I have ended up being a mentor for many people who do companies. I. I view it as one of my strange little hobbies of CEO counseling, I think, is the thing I. I usually call it. And I end up. I've. I've ended up with. I think now it's much more weighted towards CEOs, but for. For quite a while, it was kind of a. CEOs and kids. Those were my two kind of mentoring targets, so to speak. They have. I. I realized at some point that they have one feature in common, which is tech CEOs and kids. @ least many of each of those categories have the common feature that they believe that anything is possible. Many people are at stages of life and so on where they are just like, well, we're stuck in this thing and that thing. Whereas those two particular groups are like, yeah, anything's possible. We could do anything, so to speak, at least to some extent. So, I mean, I like to believe that I've been pretty useful mentoring other people. I have to say that my own people mentoring me was less successful Maybe because I'm too obstinate. Maybe because the things I was doing were too different from what other people were trying to do. I'm not sure. I mean, I've noticed in recent times, well, one of the challenges in mentoring is there's how you do things and there's how somebody else should do things. And I think I'm pretty well aware of the fact that. And I'll often say, look, given my personality, this is what I would do, but your personality is different. And I can try and guess what you should do, but it's more challenging for me to do that. And so often people's advice is, well, you should build a company that's just like mine, or you should do things just the way I've done them. You know, often that's not the right advice. I mean, it. Sometimes. Sometimes it will be, but often it's not. And that's. That's an issue. So in terms of. Of my own trajectory in. In, for example, in. In companies, you know, I started my first company when I was 21 years old. Made many mistakes with that company. I would say I certainly didn't have any mentors. I mean, I had people I brought into that company, I brought in as management people at that company. And I suppose that was a kind of a learning experience. But it wasn't like somebody was that I was talking to them and saying, hey, what should I do with this or that? I suppose I've had many friends over the years who run tech companies and so on, and it's a little bit frustrating to me that at times over the years past, I'd be like, what should I do about this? But the response for many years has tended to be from them. You're so much more clued about what to do about things than the person I'm asking. You'll figure that out much better than I could give you advice about what to do. That's a little frustrating for me because that's sort of not what I was hoping for, so to speak. But, you know, and I think in terms of. Is it useful to find a good mentor? Yeah, it's useful. I mean, I can tell that when I've done this with companies and CEOs and so on, it's been super useful. I mean, I can directly plot the. This is what happened because I suggested this or whatever else. And sometimes it's. You suggest something and people ignore it, and the good people come back to you a couple of years later and say, yeah, okay, you were right. Pity. I Didn't do that type thing, which is useful feedback for the mentor, so to speak, rather than, you know, rather than like, well, you suggested that, but that would have actually been a bad idea type thing. So, you know, I think it is interesting. So one of the, another challenge is when you get into an area that you have not been in before, there's a certain intuition that you have to develop for that area. And I'm not sure and probably it helps to kind of see particular, kind of see people operating in that area. You know, what are the important things? How does one deal with this or that? I mean, I. Somebody asked here about Steve Jobs, for example, who I knew reasonably well. And yes, it was useful to me to see how Steve did certain kinds of things. Sometimes I'm like, he's being a jerk. And I wouldn't want to do quite that. But it was still useful to me to see he's being very pushy about this and he's just being outright pushy. There's no whatever. Or more to the point, in his case, he's kind of trying to drill down and figure out what's the essential point, what's the, you know, let's cut away all the, all the detail and let's home in on what's, what's the essence of what's going on, so to speak, or what the issue is. And I think, you know, it's useful to see those kinds of things. But in any area that one hasn't been involved in, let's say one, you know, has done lots of academic stuff, but one's never done business, or one has done something that's not high tech business and now one's jumping into high tech business, or one's done and never dealt with sort of consumer products or something like this, or one's never done complicated financial kinds of things. Each one of those things has its own intuition, its own rhythm. And it's one of the things, one of the meta things to realize is when you first are exposed to it, you won't understand what's going on. And if you are forced to kind of make a decision about something without having become enculturated into that area, it's very hard to make the right decision. And you kind of just have to get used to the rhythm of that kind of thing to have an idea. Somebody can say, oh, you're going to get into consumer marketing. Why didn't you buy a giant billboard ad in this place? And if you had a bit more experience at that, you just Know, that's just a. That's just a really bad idea. You know, it's not, you know, when you think it through, the people who are relevant to you, they'll never see the billboard, or if they see the billboard, they won't be in a place where they can, you know, well, these days it'd be a QR code or something, but, you know, where they can, like, copy down the phone number or whatever. And. And it just. It won't work. But you kind of seeing around all those corners is something you only get to be able to do if you're kind of steeped in that area. And so the thing that I find useful in any area that I'm trying to explore is, yeah, go talk to people about that area, ask them a bunch of questions, and try to get a feeling for how people think about things in these different areas. And usually, at first, at least for me, and this is particularly true of intellectual areas, it's like, oh, my gosh, this area is really big, really complicated. I'll never understand it. And after all these years, I still often think that, even though it must be hundreds of areas now that I've understood reasonably well, but it always is. There's this period where it's like, oh, my gosh, this is too big. It's kind of like, finally, this is an area I'm not going to understand, or my brain is just not built in such a way that I can understand it. And you talk to people, and at the beginning, every person you talk to is telling you about a new, different thing you've never heard of before. And then I always notice there's this moment when I'm trying to understand some new field where I start talking, where the next new person I talk to, I'm saying, yeah, yeah, I know that, I know that, I know that, I know that. And they'll tell me some new things, but I'll have a complete framework of a lot of the things they're telling me. It's not like I'm swimming, I'm drowning, because there's nothing to hang on to, so to speak. It's like, yes, I know most of those anchor points, and you're adding one or two new things for me. And. But, but at the beginning, it's like there are no anchor points. I'm. I'm just. I'm just drowning. I don't know. I don't know how to think about this. The. Let's see. Bob Comments. Faces distract from logic. This is about video calls. I think Distract from logic because we spend too long assessing people's emotions may very well be true. I mean, I think, you know, I'm somebody who, for better or worse, is not. I'm not great. Like, I. I don't recognize faces very well. I recognize names well, but I don't recognize faces very easily. And I don't know whether I think I may be not. Yeah, I remember one of my kids who's really good at this, giving me this test of, here are a bunch of idealized faces. What are the emotions in all these idealized faces? And she was like 100% great performance. I was kind of cruddy performance. And so I think maybe it's easier for some of us who are less attuned to these things and have less sort of facial empathy than others to not get distracted by that. But yes, I take your point. And I think. I think there is probably a tendency to get sucked into kind of the detailed interaction that way. Let's see. Wow. There's someone named Stefan Wolf Rum with a U M. Who says that's their real name. Wow. Okay. Well, nice to meet someone with such a similar name. The comments that they learned something about Mathematica when it was running on DOS and text mode and just commenting on their personal trajectory in the jump to Windows and so on. Yeah, it was a funny thing back in the day. When Mathematica first came out in 1988, the landscape of computers was very different. And one of the decisions, an important technology decision, was that we're going to build a product that was going to do what we wanted it to do. If the computers of the time weren't quite ready to support everything we wanted it to do, we were just going to wait and we were going to not say, oh, it'll only do these small set of things, because those are the only things that will fit in the memory of the computers of that time. We were like, we're going to make it do all these things. And if that means you have to get a fancier computer or wait, so be it. That was absolutely the right decision. I'll give you a very simple example of that. When you're dealing with something like colors. Back in those days, the idea that you could specify a color by red, green and blue values wasn't really a thing. It was, you set up a color map where You've defined what 64 possible colors are, and then you just use those colors. And we were like, we're not going to do that. We're going to say colors are specified as red, green, blue. And we're going to map that into sometimes with lots of kind of hackery at the time, into some kind of color map that you could then implement on the hardware at the time. And that was the right decision because we've been able to build a language which we've been able to keep compatible over the course of the last 36 years because we didn't make compromises that were based on, oh, right now you can only use a color map with 64 colors in it or something like this. And the same was true with kind of data and how much data you could store in memory. So at the time there were sort of, I don't know, maybe 4ish classes of computers. There were the personal computers, there were the kind of workstation class computers and there were the bigger computers and which were not usually single user, they were usually batch processing computers. The sort of class of computers that was kind of almost the sweet spot for what we were doing was the workstation class computers. Companies like sun and Apollo and HP and IBM had one and all that kind of thing. These were computers that were. Many of them were based on the Motorola 68000 processor family. And they were computers that had a few megabytes of memory in RAM and they had virtual memory and they had bitmap displays. And that was kind of the sweet spot target. But those were only, mostly, those were only professional folk had them. They were not personal computers. There were also personal computers. There were two types, basically PCs, IBM PC compatible PCs, and there were Macs. And you know, the Mac had come out in 1984, the PC had come out what, two years earlier, I think, and PCs were running Ms. DOS and Macs were running the Mac operating system. It was a. We managed to just shoehorn the first version of Mathematica onto the original Macs with a few megabytes of memory. And it just about worked. And we sold Mathematica on floppy disks for the Mac. And at that time there were lots of neighborhood computer stores where you would sell those things and you ship them boxes and they had floppy disks in them. And people would come like a bookstore, they would browse the software titles and so on and buy them and such like. There was a certain degree of, I would say that industry became a little bit on the downscale side as all sorts of strange semi scammy things of, you know, you order a bunch of software and then you send it back the day before you're due to actually pay for it. And it all got very messy and a number of those companies ended up in blazes of horror, so to speak. But any case, that was another class. Then there were the PCs. And the problem was you basically couldn't run at that time. Ms. DOS had a limit of 640k of memory which just wasn't big enough to run what we had. So at the very beginning we just couldn't run on PCs, on IBM PC type PCs. And then there started to be these DOS extenders which were these kind of hack towers allowing some level of virtual memory and so on that extended the memory that the operating system could address. And so we started to be able to run things, but we still were very much sort of command line based. In fact, the original versions we had for workstations were largely command line based because there wasn't really as standardized a Windows system for those kinds of things. On the Mac we had the Notebook front end from the very beginning. And the other game in town was the Next Computer, Steve Jobs company and computer between Apple and Apple. And the Next was a very lovely computer that was not really a big commercial success. So that eventually got reabsorbed into Apple and sort of became the whole Mac OS story and so on. But it was, it was this computer, it was a nice industrial design black computer. It happened to be a black and white computer which many people criticized, but it was a well done workstation class computer. And in fact, the very first deal we made as I built Wolfram Research as a company was a deal with Steve Jobs that was back maybe a year or so before Mathematica came out to bundle Mathematica on every, on every Next machine that got shipped out. And that turned out to be a pretty smart move on the part of Steve Jobs and a thing that did well for us. It was financially decent but not outstanding, but it was very good for visibility. And it also led to a bunch of other deals and so on. And for Steve Jobs it was amusing in a sense the extent to which, well, he had intended the Next to in large measure target the higher education marketplace that he saw as being a very kind of accessible market. And it was sort of amusing because there were so many people who would buy a multi thousand dollar NEXT computer so that they would get Mathematica for free rather than buying a Mac for a lot less money, but have to pay for their Mathematica, so to speak, or buy a Sun workstation and have to pay for it separately. So Steve was absolutely right. It was a good driver for selling NEXT computers, even Though in some sense, the economics didn't really make sense for the people buying it, I have to say. One group that bought NEXT Computers for that reason was the theory group at CERN in Geneva. And that, I guess, I don't know all the details of the story, but basically that theory group, its systems person, was a chap called Tim Berners Lee, who I still know these days, but who had this network of computers and ended up setting up this thing that he called the World Wide Web on that network of computers, which had been bought because the theory group wanted to have Mathematica for free on this computer that cost a lot of money. But that happened to be the. Was kind of a little footnote to history, that that was the set of cause of those computers existing in that place and with that network and so on, which the Web happened to be developed. But any case. So there was sort of this time when we had the next computer. We had the Mac, we had PCs and we had workstation computers, and we had the big computers with all kinds of mini supercomputers and things like this. And at the time, we saw sort of the core language, what's now Wolfram language, as being this thing that the kernel just ran. Steve Jobs always used to refer to the kernel as the backend. And I always used to try to avoid that and call it the kernel on the grounds that it sounded a lot more dignified. But in any case, the main point was that we had separated the front end, the user interface from the kernel, the part that was actually doing computation. And that was a very big thing for us in those days because it let us be agile in terms of what kind of computers we could run on and how. Because in those days, the NEXT operating system and the way that it dealt with Windows was utterly different from the way that that got done under the operating system on the Sun Workstation or whatever else, or on the Apollo computer or what. And the thing that went. So we had a completely separate code for those front ends. And we initially viewed these front ends as being kind of sort of throwaway pieces of software. The real story was the kernel. The front end was just this kind of wrapper around that that was convenient for different kinds of machines. So the one we built for dos, for example, was completely different and much thinner than the one that we had for the Mac and so on. Interestingly, I mean, a bit later we ended up in the mid-90s. We ended up deciding, well, okay, then Microsoft Windows came out. But it was very different from, for example, the Mac operating system. And we had a whole separate code base that supported the user interface on Windows. And that code base, we were always sort of struggling to get it to feature parity with the code base for the Mac and for the next and so on. Well, eventually in the mid-90s, we kind of had this big project to just redo the whole thing and merge the code bases. It actually started with the X window code base. That was kind of a fresh start because it was a new window system, a fresh start in building the user interface. We built that and then we used that to sort of pull in all of the other kind of Windows systems. And it was an interesting, you know, decision. Literally throw away some of these code bases and just replace them with the kind of unified code base from X Windows. And so we managed to do that. And for many years we had this sort of unified code base for our front end for our Notebook interface and so on. The sad thing that happened was when the cloud came along, maybe a bit more than a decade ago, and we had to start building front ends in basically browsers, we had to build a new custom front end. When mobile came along, we had to build a new custom front end because we just couldn't use that same sort of desktop technology base. You know, we tried and we tried, you know, in the early days of, we tried using sort of early webassembly kinds of things. It was hopeless. The problem with things like that is it works for the demo. But the tail, when you actually try and get everything to work, the tail is so long and so ugly that it's just not worth trying it. And it's better to just build the code from scratch and actually optimize things for, let's say, browsers and the issues of JavaScript and so on. So it's been sort of frustrating that we've ended up with these three separate code bases now for our user interface again, even though for decades we had managed to merge them. It is interesting though that that same interface that you see in Ms. Dos, you saw in Ms. Dos, the command line interface, it's still there. You know, if you run, if you type, you know, if you install like Wolfram1 or Mathematica on your computer, there'll be a binary called Wolfram. You type that and it will come up as a command line interface. And that's relevant. I mean, you can use it through Wolfram Script to run scripts and so on. It's also kind of fun that if you plot a graph and you might have to load something I forgot how it works now. But if you plot a graph and it's the command line interface, you'll get something drawn out with stars and dashes and, you know, ASCII art and so on, which has been that same way ever since the very beginning. And it's very nice, you know, when you're bringing up some new kind of computer. I remember when the Raspberry PI was young, sort of bringing up the first version or various embedded processes of various kinds, bringing up Wolfram language and yep, you can still run that same. It's still in there, that same command line interface that existed way back from 1988 originally for Ms. DOS. Let's see, maybe one more question and then I should. Wrap up. Let me see. Well, Han Lin is asking, could you share your methods for generating and keeping track of ideas? Do you have favorite techniques for being productive? You know, I tend to mostly think in kind of buckets. That is, there are certain kinds of things I'm thinking about maybe 25 different topics. And once when I have an idea, I don't usually have an idea outside of one of those topics because I know there's nothing I can do with it. Now I've been lucky in that Wolfram language is this very big system that encompasses many things and its goal is to make everything in the world computational. And so that right there, there are many kinds of things about which I might have an idea that I can basically say, yep, I'm going to think about it in the context of that very big kind of container that is Wolfram language. Sometimes it might be that that idea will turn into something, a little function in our function repository. But I have a place to put it and that's important to me that there's sort of a place to put the idea. The disembodied idea is very hard to deal with for me now. I also tend to write a bunch and I'll write every month or two. I'll typically have produced sometimes a very long 50 page or something thing that people might call a blog post, but I don't think that's perhaps a good characterization. It's a piece of writing, so to speak. And I'll always have a directory of proto topics, pieces of writing that I might do. And they're probably, I don't know, 40 of those maybe right now. And whenever something comes along where I say, oh, that's interesting and I think, oh, that's relevant to this thing that I might write and maybe sometime the next five years I will actually write it. When I come to write it, all those things that have been sort of potentially relevant will be sitting there in that folder, and I'll get to go through them. And I find that a useful way to deal with things. And in general, I sort of have, as I say, a bunch of topic areas that I'm kind of tracking, paying attention to, and they all turn into folders. Actually, it's a little bit awkward right now, and I need to renormalize this. I have one set of folders that's sort of a research set of folders and one set of folders that's kind of a writing set of folders. And I have to say, in some embarrassment that occasionally I've been starting doing some writing and I've forgotten that there's a research folder about this very topic. And I'll suddenly discover it and I'll be kind of frustrated because there were things that I already had collected or thought about, and they didn't immediately get into the things that I'm sort of checking out as I start writing something. So I've kind of made some conscious effort in recent times. Like, I just noticed one of these recently, and I just moved the folder from my research area into the writing area as a subfolder of the thing that I had there, just because I didn't want to forget it as I went to do writing. But so that's been a. For me, this is really the only way of keeping track of ideas is they have to go into these buckets, the disembodied idea, unless it's big enough that it's really going to become a new bucket in its own right, I have a hard time keeping track of. And okay, I have another sort of unfair advantage, which is that I have the organization with 800 people or something in a company, and then we now have a Wolfram Institute as well. And that's a bunch of people. And so some things that I sort of say, oh, that's an interesting idea. I can say, oh, those are some people who will care about that idea and be able to do something with it. Let me just send it to them, and hopefully it will take root and something will happen. So it's sort of out of my orbit. I'm not keeping it in my brain anymore. There are plenty of things where I kind of need to. I think of this. I try and feed it to other people. It's like, I'd really like something to happen with this. And the question is, does it actually happen? And you wait a Couple of months or something. And then did it happen? I have a mechanism for kind of an email alias, for sort of a tracking alias that I copy things that I'm sending where I'm kind of wondering, I don't know if this is really going to get taken action from or not. I send to the tracking alias. There's some people who help me track the tracking alias, so to speak, and they go and poke at things after enough time has passed and say to the people, did you actually do anything on this? And then the answer to that ends up getting fed back to me. And sometimes it's like they'll hear, well, they tried and it didn't work, or oh, they didn't get around to it or they're the wrong person or whatever. And then one can iterate it. And I found that to be a useful sort of background for doing these kinds of things. But in terms of. Yeah, there's probably much more to say about sort of how to, you know, when you dive into new kinds of things. I mean, I think I was alluding earlier, when one's diving into really a very new kind of thing, there's always for me, this period of acclimatization, enculturation, kind of getting used to the thing. And if I try and do that too quickly, usually it doesn't work very well. I'm usually kind of. I don't feel comfortable with it and I don't think I'm doing the right thing. And I think one thing that at least for me has been pretty useful is I think over the years I've gotten pretty decent at knowing when I know, so to speak, and knowing when I don't know. And that's something that's very helpful to kind of calibrate oneself on because knowing that one doesn't know means, okay, I shouldn't go ahead and push forward and do this because I don't really know what I'm doing. And I can be comfortable once I know what. That I know what I'm doing. I can be pretty sure that I'm not going to make a horrible mistake here. So that's been. That's been helpful to me. All right, I need to go back to my day job here, but thanks very much for joining me and look forward to chatting with you another time. Bye for now. Sam.