K Lars Lohn

WEBVTT
NOTE
Transcription provided by Podhome.fm
Created: 7/6/2024 5:18:24 PM
Duration: 2541.855
Channels: 1
1
00:00:13.905 --> 00:00:15.365
Hello, and welcome to podcast.init,
2
00:00:15.905 --> 00:00:22.790
the podcast about Python and the people who make it great. I'd like to thank everyone who has donated to the show. Your contributions help us make the show sustainable.
3
00:00:23.650 --> 00:00:26.550
Linode is sponsoring us this week, and you can check them out at linode.com/podcastinnit
4
00:00:27.650 --> 00:00:32.375
and get a $20 credit to try out their fast and reliable Linux virtual servers for your next Python project.
5
00:00:33.235 --> 00:00:41.030
We're also sponsored by Rollbar, which is a service for tracking and aggregating your application errors so that you can find and fix the bugs in your project before your users notice they exist.
6
00:00:41.330 --> 00:00:42.310
Use the link rollbar.com/podcastinit
7
00:00:43.650 --> 00:00:49.095
to get 90 days and 300, 000 errors tracked for free on the Bootstrap plan. You can also visit our site at pythonpodcast.com
8
00:00:50.035 --> 00:00:58.910
to sign up for our newsletter, read the show notes, and get in touch. And to help other people find the show, you can leave a review on Itunes or Google Play Music and tell your friends and coworkers.
9
00:00:59.450 --> 00:01:12.385
We also have a discourse forum where you can find out about upcoming guests, suggest questions, propose show ideas, and follow-up with past guests. And your host as usual is Tobias Macy. And today, I'm interviewing K Lars Lawn about his career, his art, and his work with Mozilla.
10
00:01:12.909 --> 00:01:14.929
So, Lars, could you please introduce yourself?
11
00:01:15.869 --> 00:01:19.170
Well, hi. I'm Lars. I work for the Mozilla Corporation,
12
00:01:19.765 --> 00:01:21.465
and I'm speaking to you today
13
00:01:21.845 --> 00:01:27.145
from a yurt on an organic farm in the central or Willamette Valley of Oregon.
14
00:01:28.210 --> 00:01:31.110
And can you tell us a bit how you got introduced to Python?
15
00:01:32.210 --> 00:01:38.775
Well, there's a story behind that, just like there's a story behind everything I do. But I actually must thank Microsoft
16
00:01:39.555 --> 00:01:42.854
for leading me to open source software and Python.
17
00:01:43.635 --> 00:01:45.170
Back in 1999,
18
00:01:45.790 --> 00:01:47.330
there was a dotcom crash.
19
00:01:47.790 --> 00:01:52.130
And ironically, I weathered the dotcom crash by opening a dotcom.
20
00:01:53.255 --> 00:01:54.295
I, started,
21
00:01:55.255 --> 00:01:56.155
with a nursery,
22
00:01:57.415 --> 00:01:59.435
selling rose bushes online.
23
00:01:59.840 --> 00:02:06.340
And I hadn't done any research ahead of time and it was turned out that we were the only nursery online in 1999
24
00:02:06.640 --> 00:02:11.735
that had a buy now button. We sold everything we had in the first 4 hours that we were open,
25
00:02:12.115 --> 00:02:12.615
but
26
00:02:12.995 --> 00:02:15.095
the business was actually really quite successful.
27
00:02:15.395 --> 00:02:17.015
And fast forward to 2, 002,
28
00:02:17.980 --> 00:02:19.280
and I got a letter
29
00:02:19.820 --> 00:02:20.320
from
30
00:02:20.700 --> 00:02:28.775
people called the Business Software Alliance or Association or something like that. And it was a threatening letter telling me that I had,
31
00:02:29.475 --> 00:02:31.655
pirated software in my business
32
00:02:32.035 --> 00:02:46.165
and that they were preparing to sue me. But there was an amnesty program, and if I paid this certain amount of money for all of these licenses for these Microsoft products, they would go away. I was stunned. There were 7 computers
33
00:02:46.625 --> 00:02:52.670
in my business at that time. And the only software in them was the software that came from Dell
34
00:02:52.970 --> 00:02:54.830
and the software that I wrote myself.
35
00:02:55.130 --> 00:03:02.115
I, like I said, I was stunned. I didn't know what to do about this. And I know I couldn't find a single 1 of those certificates of authenticity
36
00:03:02.415 --> 00:03:08.440
that came with the machines. So I had I guess no way of proving that I owned the software that was on my machines.
37
00:03:09.060 --> 00:03:12.840
So after several sleepless nights and pouring over my budget,
38
00:03:13.140 --> 00:03:15.960
I realized that I was gonna have to lay off an employee
39
00:03:16.575 --> 00:03:17.715
in order to afford
40
00:03:18.334 --> 00:03:28.830
to pay the amount that I was gonna have to pay. I finally decided which person I was going to lay off and then came up with a plan on how I was gonna get through an entire shipping season
41
00:03:29.130 --> 00:03:38.905
short and employee. And I pulled out that letter and suddenly was shocked by something I hadn't noticed at the beginning. My business name was called The Uncommon Rose,
42
00:03:39.390 --> 00:03:40.370
and the salutation
43
00:03:40.830 --> 00:03:43.970
on the letter was dear miss Rose Uncommon.
44
00:03:44.349 --> 00:03:45.569
What had happened apparently
45
00:03:45.870 --> 00:03:48.769
is this business software group had purchased
46
00:03:49.455 --> 00:03:58.349
marketing mailing list, and they just spammed everybody with the same message. I realized they knew nothing about my business at all. They didn't know anything.
47
00:03:58.890 --> 00:04:00.030
I even had computers.
48
00:04:00.569 --> 00:04:03.549
So I realized it was just some sort of protection racket.
49
00:04:03.930 --> 00:04:09.094
I took the letter, copied it, sent it to the attorney general's office explaining what was going on,
50
00:04:09.394 --> 00:04:23.585
and, then decided to ignore it and decided that Microsoft has no business having their hand in my revenue stream whatsoever. The software had been written entirely using Microsoft C plus plus, the Rogue Wave C plus plus libraries,
51
00:04:24.045 --> 00:04:26.785
Microsoft SQL Server, and various other
52
00:04:27.165 --> 00:04:32.699
Microsoft technologies, I decided to eliminate all of it. All the Microsoft keyboards and Microsoft
53
00:04:33.240 --> 00:04:34.120
mice went to,
54
00:04:34.680 --> 00:04:39.495
electronics recycling. I started doing some research. I discovered this language called Python,
55
00:04:39.795 --> 00:04:45.255
which looked like a very good language to rapidly rewrite the entire software for the business.
56
00:04:45.955 --> 00:04:47.575
I chose the Postgres database
57
00:04:48.160 --> 00:04:55.060
and Linux and in 2 months, I entirely wrote the entire suite of software that ran the business
58
00:04:55.405 --> 00:04:55.905
from
59
00:04:56.205 --> 00:04:57.345
the greenhouse management
60
00:04:57.885 --> 00:05:00.065
all the way through the website generation.
61
00:05:00.445 --> 00:05:02.945
So like I say, I must thank Microsoft
62
00:05:03.250 --> 00:05:05.910
for introducing me to Python and open source software.
63
00:05:06.930 --> 00:05:19.574
It's funny how many times I've heard people say that they came to Python because they had to completely rewrite an existing project in a short matter of time, and then they were surprised by how quickly they were able to actually achieve that.
64
00:05:21.090 --> 00:05:29.270
It was amazing. I I was really, really impressed with the Python language at that point. Having been a c plus plus programmer for
65
00:05:29.625 --> 00:05:32.845
many years, and I started c plus plus way back in 87,
66
00:05:33.385 --> 00:05:34.525
and the whole
67
00:05:35.225 --> 00:05:37.645
cycle of writing and testing
68
00:05:39.140 --> 00:05:50.375
was an order of magnitude faster. Yeah. Having the REPL for being able to do things interactively and in an exploratory manner has really been sort of foundational in how I've approached my programming career.
69
00:05:50.835 --> 00:05:57.095
Yep. When I was going through your blog to prepare for this interview, I noticed that you have an interesting pair of articles on your website
70
00:05:57.470 --> 00:06:11.795
trying to detail how you perceive code and why you think that code formatting should be able to be configured in a manner similar to CSS. So I'm wondering if you can explain a bit how your particular way that you perceive code affects the ways that you program.
71
00:06:12.975 --> 00:06:14.355
Certainly. It is only
72
00:06:14.770 --> 00:06:17.350
recently that I have really come to understand
73
00:06:17.650 --> 00:06:19.750
how my brain is different
74
00:06:20.050 --> 00:06:22.455
from most of the other programmers that I know.
75
00:06:22.935 --> 00:06:25.995
I have a minor form what is called a synesthesia.
76
00:06:26.534 --> 00:06:28.074
A synesthesia is when,
77
00:06:28.694 --> 00:06:29.835
certain perceptions
78
00:06:30.220 --> 00:06:35.360
are mixed up in the brain. People taste colors and things like that. Well, I see
79
00:06:35.900 --> 00:06:36.400
complicated
80
00:06:36.780 --> 00:06:38.880
twisted lines in colors
81
00:06:39.485 --> 00:06:41.025
Even if they're not actually
82
00:06:41.485 --> 00:06:42.544
done in colors.
83
00:06:42.845 --> 00:06:44.225
To me, code
84
00:06:44.604 --> 00:06:47.720
is the same sort of idea. I see it in colors
85
00:06:48.180 --> 00:06:57.134
and that became first apparent to me when I was in graduate school. I was in an artificial intelligence class using the language lisp. The professor was giving
86
00:06:57.514 --> 00:07:03.694
some sort of lecture on recursive pattern matching algorithms, and I probably was just doodling,
87
00:07:04.569 --> 00:07:08.830
and not paying close attention. But I noticed all of a sudden a flash of blue
88
00:07:09.210 --> 00:07:20.145
caught my eye, and he was writing on an overhead projector in black ink. And I stopped him and said, wait a minute, there's something wrong with your algorithm here. And then I said something vague about,
89
00:07:20.705 --> 00:07:31.949
you lose context in the recursion at this point. He smiled at me and said well I've been using this example for 10 years and we know it works, etcetera, etcetera. And I backed down. I wasn't gonna tell him it's too blue.
90
00:07:32.815 --> 00:07:37.155
So I backed down and he continued with his lecture. After a few minutes,
91
00:07:37.535 --> 00:07:39.635
he suddenly stopped in mid sentence,
92
00:07:40.095 --> 00:07:42.900
scrolled the overhead projector back to that algorithm,
93
00:07:43.440 --> 00:07:46.580
looked at it for a moment, and then looked at me and grinned,
94
00:07:47.039 --> 00:07:51.380
and said nobody has ever caught that problem. I didn't know it was there myself,
95
00:07:51.815 --> 00:07:55.675
and ever since then I can say I can look at code
96
00:07:55.975 --> 00:07:57.755
and if I see something blue,
97
00:07:58.135 --> 00:08:02.960
I know there's a problem. I may not know what it is and it's not always completely reliable.
98
00:08:03.420 --> 00:08:03.920
So
99
00:08:04.300 --> 00:08:05.520
the way I code
100
00:08:05.980 --> 00:08:13.605
is is kind of unique. I'm very and a very exploratory coder, and I believe very much in being very verbose
101
00:08:14.065 --> 00:08:15.285
in writing code.
102
00:08:15.585 --> 00:08:17.285
I believe that a variable name
103
00:08:17.640 --> 00:08:24.620
shouldn't be a word. It should be an entire sentence describing what it is. A collection of
104
00:08:24.945 --> 00:08:25.445
something.
105
00:08:25.825 --> 00:08:28.325
Variable names such as looping variables.
106
00:08:28.785 --> 00:08:31.205
I rarely ever use x or I.
107
00:08:31.585 --> 00:08:36.070
I want to say what the object is, and I'm gonna use an article in front of it.
108
00:08:36.690 --> 00:08:37.830
A blog post.
109
00:08:38.450 --> 00:08:43.670
A integer, a rather an integer. Something like that. Because then the code to me
110
00:08:43.975 --> 00:08:45.515
reads like English sentences.
111
00:08:45.975 --> 00:08:47.755
If a blog post
112
00:08:48.454 --> 00:08:50.795
is greater in length than etcetera.
113
00:08:51.140 --> 00:08:59.880
So that's how code looks best to me. If it's written out in long form like that, I can go back to it and read it again. Unfortunately, it doesn't fit with many people's
114
00:09:00.275 --> 00:09:02.615
or many organizations coding standards
115
00:09:03.155 --> 00:09:09.175
and, you know, most of the people I know know never to actually talk to me about PEP 8,
116
00:09:09.660 --> 00:09:12.480
Because for me PEP 8 is a disaster
117
00:09:13.340 --> 00:09:14.080
and honestly,
118
00:09:14.620 --> 00:09:16.000
it gives all code
119
00:09:16.300 --> 00:09:17.600
a bluish cast.
120
00:09:18.365 --> 00:09:19.745
It almost blinds me
121
00:09:20.125 --> 00:09:34.079
to finding problems in code if I can't use this synesthesia that I have. The main problem and and it's certainly has been addressed in PEP 8 is line length. I have always felt that limiting line length to 79 characters
122
00:09:34.540 --> 00:09:42.305
causes you to overload the meaning of white space. It no longer just shows structure, it also shows that you have,
123
00:09:43.325 --> 00:09:47.680
lines that are too long, And with my variable names being long sentences,
124
00:09:48.060 --> 00:10:08.620
almost every line is too long. And so I've had to modify how I I write code, and to me it becomes far less readable. Wouldn't it be great if we could find a way of looking at code and seeing it in the form of which we can read it the best? Now, the only thing I can think of that can do that sort of thing is the same way that HTML is styled in CSS.
125
00:10:09.245 --> 00:10:17.985
I would love to see some way an editor of some sort to enable me to format my code in a way that I can read it
126
00:10:18.750 --> 00:10:21.250
and then when someone else looks at it in their editor
127
00:10:21.550 --> 00:10:35.955
it's formatted in the way they can read it most easily. I don't know the feasibility of this. I've sat down and tried to, think about how I would program this, played around with a few things, and really didn't get anywhere with it. It may never happen, at least in my lifetime,
128
00:10:36.575 --> 00:10:44.145
but it's a great fantasy. Yeah. It's definitely an interesting problem to ponder because as you said, different people view code differently,
129
00:10:51.185 --> 00:10:59.830
Particularly when Yeah. Particularly when you're saying, for instance, the long variable names, you know, as far as the CSS system, 1 potential approach would be to have
130
00:11:00.235 --> 00:11:15.569
maybe a long and a short form of the variable so that you could have a toggle that will determine how those render so that people who want shorter line lengths can use the abbreviated form, and people who prefer the more expressive variable name can view it in the in its original context.
131
00:11:16.795 --> 00:11:22.575
And then once you get to that point, then do you want to have, well, the variable in a different language
132
00:11:23.275 --> 00:11:30.880
and etcetera like that, that, and then, it becomes impossible to write code at all. Right. And and and then you have to deal with I a t n n before you can even actually write your program.
133
00:11:31.340 --> 00:11:31.840
Exactly.
134
00:11:32.245 --> 00:11:36.425
Yeah. It's an interesting interesting problem to ponder at least if not solve.
135
00:11:36.805 --> 00:11:37.305
Right.
136
00:11:37.685 --> 00:11:42.389
I'm curious if the manner in which you perceive complicated expressions
137
00:11:43.089 --> 00:11:49.605
has contributed to some of your fascination with complexity theory, which we'll touch on a little bit later, but also your,
138
00:11:49.985 --> 00:11:51.365
your interest and
139
00:11:51.745 --> 00:11:55.365
proficiency in drawing, what I must say, are quite beautiful mazes.
140
00:11:56.270 --> 00:12:00.210
They're certainly quite closely related. Do you know of the language called APL?
141
00:12:00.830 --> 00:12:04.930
I am familiar with it. I have actually recently listened to a podcast on,
142
00:12:05.935 --> 00:12:11.795
for somebody who was very familiar with the language, it was on the Functional Geekery podcast. So I'll put a link to that in the show notes.
143
00:12:12.870 --> 00:12:21.050
I love that language. When I lived on the Blackfeet Indian Reservation, I happened to be working for a man who was on the original design team for that language.
144
00:12:21.425 --> 00:12:27.764
And we shared a love of it and its complexity and its ability to do an amazing number of things
145
00:12:28.144 --> 00:12:30.644
in just very terse and brief,
146
00:12:31.810 --> 00:12:32.310
statements
147
00:12:32.690 --> 00:12:49.790
is very exciting to me. Of course, it doesn't fit well with my long variable name idea, but still I still appreciate it in that respect. So the code that I write, I often have a difficult time with the rest of the people on my team because I write code that is complex.
148
00:12:50.250 --> 00:12:54.670
To me, many things that are complex are really quite clear. I love recursion.
149
00:12:55.050 --> 00:13:00.274
I love multi threading. They are both make things so much simpler in my mind, but
150
00:13:00.654 --> 00:13:01.154
complexity
151
00:13:01.695 --> 00:13:13.130
is kind of a bad word in programming today, that there's always this move to move towards simplicity because it's easier code to maintain, but to me it makes code less interesting and I often
152
00:13:13.715 --> 00:13:16.615
see code simplified to the point of being incorrect.
153
00:13:17.235 --> 00:13:24.980
No longer properly modeling what it is you're attempting to model because I guess I can say that if the business process that you are attempting
154
00:13:25.360 --> 00:13:27.700
to to model is more complicated
155
00:13:28.115 --> 00:13:57.460
than the code you have written, then it's likely wrong. You have taken let's say, you've discovered that in the business model, there are 2 cases, x and y, and you realize in your mind that, wow, these 2 cases are really very similar so I'm gonna roll them into 1, but the business model didn't get changed, only your program got changed, and I'm gonna say that your program is now wrong because it merged those 2 cases, but it made the program simpler and perhaps easier to maintain, but it didn't model the business process anymore. Now
156
00:13:57.839 --> 00:13:59.860
take this on to my
157
00:14:00.404 --> 00:14:06.904
interest in mazes. I drew mazes as a kid. I found many programmers or people with with mathematical
158
00:14:07.445 --> 00:14:07.945
backgrounds
159
00:14:08.269 --> 00:14:20.925
were very interested in mazes as a kid. I stopped drawing them when I was in high school, and just never returned to it. Until about a year ago, when I was in an office depot, and I saw that they still make flare pens.
160
00:14:21.385 --> 00:14:33.079
That's what I used as a kid to draw mazes. So I bought a set, and sat down and immediately started drawing mazes and thought this was this is amazing. I love this. After drawing about 20 of them, my husband suggested
161
00:14:34.214 --> 00:14:40.074
that I try drawing them on a tablet instead. And so I got my Android tablet out and I bought a a
162
00:14:40.660 --> 00:15:00.060
a pen for it and started drawing on that, and wow, it was amazing. I could really do some interesting things, and I realized with these drawing programs, I can do layering and make layer masks and put these things in the colors that I actually see them in in my mind. And from then, all of a sudden, these amazing things started coming out of me. These,
163
00:15:00.519 --> 00:15:01.019
represented,
164
00:15:01.639 --> 00:15:16.930
botanical drawings, and I stunned myself with these things. I've I've done I think I finished the 44th 1 just very recently. Some of them take up to 80 hours to draw, but it's as if maybe perhaps sometime a year ago, it's like I had a stroke
165
00:15:17.390 --> 00:15:21.810
or something. And all of a sudden, I have an artistic ability that I never thought I had.
166
00:15:22.270 --> 00:15:25.089
It's great to discover something new at age 56.
167
00:15:26.635 --> 00:15:35.055
Yeah. It's definitely an interesting story because viewing the images that you produce, you would think that you've been doing it all your life because I must say they're quite astounding and incredibly detailed.
168
00:15:35.530 --> 00:15:36.910
They're very therapeutic
169
00:15:37.290 --> 00:15:40.270
to draw. I can sit for several hours
170
00:15:40.810 --> 00:15:43.550
and just in a mindless state, almost
171
00:15:44.065 --> 00:15:45.445
zen like meditation,
172
00:15:45.985 --> 00:15:49.845
draw these things. They're drawn in several stages. I first draw the lines,
173
00:15:50.625 --> 00:15:53.790
just on a whim exploring different sorts of patterns
174
00:15:54.170 --> 00:15:58.830
and then I use the feature of drawing programs of flood fill
175
00:15:59.290 --> 00:16:09.644
to find out what regions I have made connected lines in and then I engineer the maze from that using, I guess, my programming skills.
176
00:16:10.810 --> 00:16:14.190
Maybe I should say I should use use my my skills of deceit
177
00:16:14.890 --> 00:16:22.945
to try and to try and make these, mazes as difficult as possible. I don't think I don't know of anyone who has ever solved 1 of my mazes.
178
00:16:23.245 --> 00:16:24.785
I know they all have solutions,
179
00:16:25.405 --> 00:16:42.415
because I I prove it to myself before I publish them. I actually had an art show in my little town here in June. This show went for a month and I think for an unknown artist in an obscure gallery in a small town, I did extraordinarily well for selling 7 pieces
180
00:16:42.810 --> 00:17:00.145
that went for from 350 to $500 each. It was amazing. Perhaps I have a new career if I ever get out of this programming stuff. Yeah. If you get tired of slinging code all day, you certainly have the talent for it. And for anybody who hasn't seen them, I'm gonna put a link in the show notes. But, you know, if you look at them from a distance,
181
00:17:00.560 --> 00:17:06.580
you would never guess that there's actually a second layer of purpose behind them beyond just visual artwork. And
182
00:17:06.960 --> 00:17:11.165
the amount of detail that you can get out of it at the same time as incorporating
183
00:17:11.545 --> 00:17:18.125
such complicated mazes is just astounding. And the fact that you start with just sort of an exploratory
184
00:17:18.825 --> 00:17:19.325
approach
185
00:17:19.850 --> 00:17:31.655
of seeing what sorts of shapes and patterns come about and then using that to then direct the creation of the maze. And I'm assuming as well the actual content of the image is pretty remarkable.
186
00:17:32.035 --> 00:17:47.105
I decide what the content of the image is going to be ahead of time. I don't just it doesn't doesn't spew out just whole. I'd said, okay. I'm gonna do a poppy here. And now I I sit and look I look at a poppy and and decide that what shapes are interesting and then start drawing them.
187
00:17:48.465 --> 00:17:51.765
And playing again on the subject of complexity,
188
00:17:52.225 --> 00:17:58.000
at Mozilla, you've helped with the creation and stewardship of a project called Socorro, which
189
00:17:58.380 --> 00:18:14.090
you mentioned in your keynote. You utilized some aspects of complexity analysis for being able to correlate the stack traces that you receive into this project. And I'm wondering how you sort of came across or approached that idea of incorporating that sort of level of theoretical application
190
00:18:14.870 --> 00:18:16.409
to something like error monitoring?
191
00:18:17.350 --> 00:18:20.570
Well, the, the actual use of fractal dimension
192
00:18:21.029 --> 00:18:21.529
for
193
00:18:22.005 --> 00:18:22.505
categorizing
194
00:18:23.125 --> 00:18:30.345
our classification of errors with something that never made it actually into the production code. We built the system over the last 10 years
195
00:18:30.805 --> 00:18:32.930
and it's a complicated
196
00:18:33.710 --> 00:18:35.650
multi process, multi threaded,
197
00:18:36.110 --> 00:18:36.610
application
198
00:18:37.310 --> 00:18:47.419
with my signatures all over it. And, at the final iteration before I've left that group, I create what created what is called the processor as a rule based system, so that classification
199
00:18:47.799 --> 00:18:54.620
systems can easily be written and then just plugged into this system at configuration time. And so while writing the,
200
00:18:55.015 --> 00:19:04.555
keynote, I was thinking about different uses of complexity and I thought, well, you know, or rather fractal dimension, I could write a classification system for Socorro
201
00:19:05.240 --> 00:19:17.365
that used fractal dimension. And so I coded it up in just a few minutes and inserted it into my own personal copy of Socorro and then ran a a 1000000 crashes through it and discovered
202
00:19:17.985 --> 00:19:19.765
that I could do some correlations
203
00:19:20.145 --> 00:19:31.465
between crashes that the rest of the system hadn't detected that were similar crashes or from the same root cause. But then at that point, you know, I was doing that for the lecture not for Mozilla.
204
00:19:32.165 --> 00:19:33.304
I found some counterexamples
205
00:19:33.605 --> 00:19:45.280
where it didn't work as well, where it found some correlations, where there were not actual correlations. There may be something to it, but it would take a little bit more work, and I'm not with that group anymore. So it's kind of gone by the wayside.
206
00:19:46.445 --> 00:19:48.145
Well, it's still a very interesting
207
00:19:48.605 --> 00:19:51.105
application of that kind of, analysis,
208
00:19:51.485 --> 00:19:58.620
and it it'll be curious to see if that does evolve further and maybe some of the other areas that it might pop up. Mhmm.
209
00:19:59.400 --> 00:20:08.424
So can you describe a bit about how the Socora project is architected and sort of how it works under the covers and maybe what the evolution over time has looked like for it? Certainly.
210
00:20:08.965 --> 00:20:10.985
I started on the Socorro project
211
00:20:11.820 --> 00:20:14.240
in, I think, 07 or 08
212
00:20:14.540 --> 00:20:21.425
just, before the release of Firefox 3. It had already been started, by some of the platform engineers
213
00:20:21.985 --> 00:20:46.265
and then I was brought on to complete it. And, in some ways, it was kind of like after some nuclear disaster and all of the developers were just gone and all of a sudden the code base was given to me and there was no 1 to talk to about it or what it was. And the idea was that I had to make it work in a very short order and of course, Python is really good at that. And so I hacked the thing together to make it work and it worked mostly,
214
00:20:46.805 --> 00:20:51.300
but it was an awful hack. And then for the next few years, we
215
00:20:51.760 --> 00:20:52.260
did,
216
00:20:52.640 --> 00:20:56.420
a group kind of formed around me and we did what I call retroactive
217
00:20:56.800 --> 00:20:57.300
forethought.
218
00:20:57.875 --> 00:21:06.455
We refactored the code and took the hacks out of it to make it, more of an engineered thing rather than something just thrown together.
219
00:21:06.880 --> 00:21:08.320
So it consists of several,
220
00:21:08.640 --> 00:21:10.740
independent parts running on separate machines.
221
00:21:11.280 --> 00:21:15.460
A collector, which is just simply well, at the time was a Python,
222
00:21:16.115 --> 00:21:19.015
It was actually a mod Python, I think, running under Apache
223
00:21:19.475 --> 00:21:19.975
that,
224
00:21:20.435 --> 00:21:25.575
catches the crashes as they come in from the field and stores them, on disk,
225
00:21:26.420 --> 00:21:31.080
on the the web head. And then another process takes them off that disk
226
00:21:31.620 --> 00:21:33.400
and sends them into,
227
00:21:33.860 --> 00:21:38.815
long term storage. At that time, it was another disk and Postgres.
228
00:21:39.435 --> 00:21:42.175
And then another process running on several machines,
229
00:21:42.715 --> 00:21:45.230
pulls them off the disk and out of Postgres
230
00:21:45.770 --> 00:21:47.230
and performs a miracle,
231
00:21:47.929 --> 00:21:48.429
using
232
00:21:48.890 --> 00:21:51.150
the stack walker and remarries
233
00:21:51.610 --> 00:21:55.465
this binary blob of stack information with symbol tables,
234
00:21:56.005 --> 00:21:56.565
so that,
235
00:21:57.284 --> 00:22:01.465
we can find out exactly what the code was that crashed. And then there's a
236
00:22:01.799 --> 00:22:09.820
a web app in front of that that, allows the developers then to go in and look at these crashes, categorize them, look at aggregate
237
00:22:10.615 --> 00:22:19.115
information about them, and actually drill down into our source code system to find out exactly what lines and look at the lines of the code. It's all multi process,
238
00:22:19.679 --> 00:22:20.980
multi threaded stuff.
239
00:22:21.440 --> 00:22:28.100
I went from multi threaded from the very beginning back in 07 because I love multi threading and then over the years,
240
00:22:28.585 --> 00:22:29.965
I started to do generalizations.
241
00:22:30.825 --> 00:22:39.070
1 of the things that was most apparent about the project is it needed to be very quick on its feet for a changing environment.
242
00:22:39.530 --> 00:22:43.470
We decided we were gonna in 2010, we decided to move and use HBase
243
00:22:43.770 --> 00:22:59.100
to replace the file system storage. And so we built 8 HBase modules and and linked them into the system, but I wasn't satisfied with that because I knew that we were gonna add other things in the future. And so I came up with this library that I call ConfigMan
244
00:22:59.480 --> 00:23:02.780
that does dynamic loading of modules in Python
245
00:23:03.085 --> 00:23:06.945
so that like connecting Legos together at the configuration
246
00:23:07.325 --> 00:23:35.320
time, we can change the front end or back end of this system at will. By the the climax of this in well, I can't remember was 4 between 14 15 that we had dropped HBase and we're moving all of our data to Amazon S3. We were able to put Amazon S3 module in and just tell the applications to sig hop, and so they added Amazon S3 in parallel with HBase
247
00:23:35.755 --> 00:23:47.630
without shutting the programs down even. It just reload like reloaded configuration and kept right on running, not losing a step. Ran that way for a month till we verified that the Amazon s 3 and the HBase data was identical,
248
00:23:47.930 --> 00:23:53.455
and then rewrote the configuration to take HBase out, sent sick hub to all of the,
249
00:23:53.934 --> 00:23:54.434
applications.
250
00:23:55.134 --> 00:23:59.154
HBase was gone. Again, the applications didn't even have to be shut down.
251
00:23:59.840 --> 00:24:01.620
That's quite a remarkable feat of engineering.
252
00:24:02.560 --> 00:24:04.500
It I I was very proud of it.
253
00:24:05.040 --> 00:24:05.540
Unfortunately,
254
00:24:06.640 --> 00:24:09.195
it adds a significant amount of complexity
255
00:24:09.495 --> 00:24:12.475
to the system with the added benefit of the flexibility.
256
00:24:12.855 --> 00:24:14.555
You know, I say that
257
00:24:14.880 --> 00:24:15.700
in my career,
258
00:24:16.559 --> 00:24:17.940
any of the software
259
00:24:18.240 --> 00:24:21.620
that I hard coded any external resource,
260
00:24:21.925 --> 00:24:27.385
that software is long gone. It had a 2 year life at most. But software that I've written
261
00:24:27.685 --> 00:24:30.640
that allows to have dynamic configuration
262
00:24:31.020 --> 00:24:40.645
and dynamic loading of modules, in other words future proofing it, it still is running today. I did something similar to this way back in my career in the 19 eighties
263
00:24:41.105 --> 00:24:43.365
in the language Fortran to drive
264
00:24:43.745 --> 00:24:54.180
flatbed plotters. A couple years ago, I went back and visited that company that I did this for and found that my Fortran code is still in use 30 years later.
265
00:24:55.745 --> 00:25:03.845
Speaking of somebody who hasn't had a technical career anywhere near that length yet, that is my my hat goes off to you, sir. Thank you. Thank you.
266
00:25:04.540 --> 00:25:12.080
At this year's PyCon US out in Portland, Oregon, you presented the closing keynote, and I have to say that it was 1 of the most engaging
267
00:25:12.380 --> 00:25:12.880
and
268
00:25:13.225 --> 00:25:16.044
engrossing presentations that I've ever seen. And I'm wondering
269
00:25:16.424 --> 00:25:25.429
where you got the inspiration for that talk, how you conceived of the mixed media approach to it, and as a follow on to that, how much you had to practice?
270
00:25:25.809 --> 00:25:32.165
Mhmm. Well, there's a story behind this too, and it all revolves around the regional conference, PIE Tennessee.
271
00:25:32.785 --> 00:25:37.525
About 3 or 4 years ago, PIE Tennessee was looking for sponsors. They went to Mozilla.
272
00:25:38.465 --> 00:25:38.965
Mozilla
273
00:25:39.390 --> 00:25:46.290
sponsored them and they were look Mozilla was looking for someone to speak and I happen to have a whole bunch of friends that live in rural Tennessee
274
00:25:46.750 --> 00:25:48.865
and so I just said I will volunteer to speak
275
00:25:49.265 --> 00:26:00.600
and talk about Socorro. And so we arranged it and I spoke the conference and of course, you know, it's 1 of those multi track things. And I think I had about 5 people in my audience.
276
00:26:01.380 --> 00:26:10.495
And I think the organizers of Pi Tennessee felt kind of bad about that. So the following year, I decided I was just I would submit papers,
277
00:26:10.875 --> 00:26:12.095
proposals to PyTennessee,
278
00:26:12.475 --> 00:26:14.495
and I submitted 3 proposals.
279
00:26:15.299 --> 00:26:21.320
2 very serious very academic proposals. I think 1 was about multi threading and the other 1 was about configuration.
280
00:26:21.860 --> 00:26:25.035
And then I decided I would send in a joke.
281
00:26:25.415 --> 00:26:26.715
And so I made a proposal
282
00:26:27.095 --> 00:26:28.155
about parallels
283
00:26:28.775 --> 00:26:30.155
in software engineering
284
00:26:30.950 --> 00:26:33.130
with baroque music from the 1700.
285
00:26:33.750 --> 00:26:35.850
Well, they called my bluff.
286
00:26:36.870 --> 00:26:38.390
They rejected the 2,
287
00:26:38.870 --> 00:26:39.370
proposals,
288
00:26:39.755 --> 00:26:41.294
the academic ones, and chose
289
00:26:42.395 --> 00:26:44.575
the music 1. So I had to follow-up,
290
00:26:44.875 --> 00:26:48.655
and that became my lecture called The Well Tempered API,
291
00:26:49.420 --> 00:26:55.040
subtitled, I can play 300 year old music, but I can't open a Word document from 1990.
292
00:26:57.225 --> 00:26:57.665
And,
293
00:26:58.105 --> 00:26:59.245
it was multimedia.
294
00:26:59.545 --> 00:27:00.045
I,
295
00:27:00.665 --> 00:27:05.410
hadn't played a musical instrument since high school, and so I learned to play this,
296
00:27:05.810 --> 00:27:07.810
electronic woodwind of mine and,
297
00:27:08.690 --> 00:27:13.110
gave the presentation. There's a couple different recordings of it out there if you want to find it.
298
00:27:13.495 --> 00:27:25.070
You can go on to either Air Mozilla or on YouTube and find Well Tempered API and you know there are a whole bunch of interesting parallels between music and software engineering. I was
299
00:27:25.530 --> 00:27:26.750
realized that
300
00:27:27.130 --> 00:27:28.270
cooperative multitasking
301
00:27:28.890 --> 00:27:31.470
was done by organists in the 1700.
302
00:27:31.850 --> 00:27:35.135
In fact, if you look up baroque cooperative multitasking
303
00:27:35.675 --> 00:27:51.360
in, YouTube, you'll find a, an excerpt from my original talk at pi Tennessee that year. Then just as I was I'd finished that and figured that I was gonna be done speaking for a while, the Py Tennessee people invited me to do the keynote, the opening keynote for,
304
00:27:51.805 --> 00:27:53.265
this year in 2016.
305
00:27:53.725 --> 00:27:59.265
And I decided, well, let's just see how long I can stretch out this music thing. And
306
00:27:59.630 --> 00:28:01.730
at that point in at Mozilla,
307
00:28:02.350 --> 00:28:03.010
I was
308
00:28:03.549 --> 00:28:04.770
getting very frustrated
309
00:28:05.309 --> 00:28:06.130
with complexity
310
00:28:06.894 --> 00:28:14.195
and the inability of so many of my co workers to understand my code because it was too complex. And then I decided to demonstrate
311
00:28:14.850 --> 00:28:18.470
how if you don't think about it, everybody loves complexity.
312
00:28:18.929 --> 00:28:27.185
If you look at a a soap opera, they constantly you have these like 7 or 8 second scenes that constantly jump between story lines.
313
00:28:27.645 --> 00:28:36.070
That's cooperative multitasking, and we eat it up in entertainment, but we seem to not like it in programming because it's too complex. And so I made a,
314
00:28:36.530 --> 00:28:38.870
a talk for their keynote that I called,
315
00:28:39.250 --> 00:28:40.870
Fractal Dimension Music
316
00:28:41.235 --> 00:28:44.934
and the Art of the Left Turn. And that was the genesis of my,
317
00:28:45.395 --> 00:28:49.815
PyCon keynote because just like the day before I gave the keynote at PyTennessee,
318
00:28:50.320 --> 00:28:54.179
I got the invitation to do a keynote at PyCon. And, so
319
00:28:54.639 --> 00:28:58.100
I modified that talk to become the PyCon keynote.
320
00:28:58.804 --> 00:29:05.865
I took out, the reference to to music. I changed a whole bunch of things in it. I shortened it a little bit.
321
00:29:06.520 --> 00:29:15.380
I took out some very nice things I thought but I only had a certain amount of time to fill. I could well, as you're finding out in this interview, I can just talk on and on for hours.
322
00:29:17.455 --> 00:29:19.635
And the idea in the the presentation
323
00:29:20.175 --> 00:29:20.995
was to make
324
00:29:21.375 --> 00:29:22.435
a as complex
325
00:29:23.220 --> 00:29:27.639
a presentation as I could possibly make. Jumping back and forth between stories
326
00:29:28.100 --> 00:29:42.149
and yet engage the audience as much as I can and and everybody was gonna get something different out of it. Everyone was gonna notice something different, but there was gonna be something for everyone and every skill level in there. When I was,
327
00:29:42.610 --> 00:29:44.950
invited to do the the, PyCon,
328
00:29:46.049 --> 00:29:46.789
I asked,
329
00:29:47.090 --> 00:29:49.590
Brandon Rhodes about what should I talk about?
330
00:29:50.065 --> 00:29:58.470
And he said, talk about motorcycles, talk about mazes, talk about music. And so I said, okay. I'll talk about everything. And that's what I ended up doing. So
331
00:29:58.950 --> 00:30:07.850
as for how long did it take me to rehearse all of that? Having the Pi Tennessee experience behind me and doing parts of that presentation,
332
00:30:08.185 --> 00:30:10.765
I'd say that I ran through the whole presentation
333
00:30:11.545 --> 00:30:12.365
3 times
334
00:30:12.745 --> 00:30:14.845
ahead of giving it at PyCon. However,
335
00:30:15.470 --> 00:30:16.210
the music
336
00:30:16.590 --> 00:30:17.250
in there,
337
00:30:17.790 --> 00:30:19.970
I practiced to death.
338
00:30:20.669 --> 00:30:21.070
I,
339
00:30:21.710 --> 00:30:24.915
that is the the most stressful part of the presentation.
340
00:30:25.615 --> 00:30:32.275
Playing music in front of a whole bunch of people. I am not a professional musicians. I have I have a high school music education
341
00:30:32.970 --> 00:30:38.270
and well, I like my maze drawing just a very short time as an adult,
342
00:30:39.130 --> 00:30:53.820
playing music. I know very little about music theory except what I learned for when I wrote the Well Tempered API talk and so that part of the presentation I rehearsed an awful lot. Still made a lot of mistakes. I can't stand to look at the video of PyCon because I
343
00:30:54.120 --> 00:30:57.020
cringe at so many of the errors that I made in the music.
344
00:30:57.995 --> 00:31:01.054
Well, speaking as somebody who isn't as,
345
00:31:01.674 --> 00:31:09.820
adept at musical theory or musical analysis, I thought it was very enjoyable. Thank you. And it seems that you've got a bit of incidental virtuosity
346
00:31:10.279 --> 00:31:12.679
in as you, continue on in your life,
347
00:31:13.240 --> 00:31:16.325
putting a lie to the fact that you can't teach an old dog new tricks.
348
00:31:16.625 --> 00:31:17.924
Absolutely. Absolutely.
349
00:31:18.544 --> 00:31:20.645
So another part of the presentation
350
00:31:21.105 --> 00:31:25.120
that I think tied it all together quite well was the fact that you managed
351
00:31:25.600 --> 00:31:28.820
to integrate a rather personal and impactful story
352
00:31:29.360 --> 00:31:29.860
as
353
00:31:30.160 --> 00:31:32.820
a, sort of it was almost a
354
00:31:33.280 --> 00:31:36.085
secondary plot throughout the overall presentation
355
00:31:36.945 --> 00:31:38.885
as sort of a subtext of complexity.
356
00:31:39.265 --> 00:31:50.470
And so I'm wondering what what inspired you to bring that element of the presentation and mix it in with the complexity analysis and musical presentation and fractal dimensions.
357
00:31:51.775 --> 00:31:52.835
It was the realization
358
00:31:53.375 --> 00:32:09.000
about how different my own mind is from other people's. I've spent a lot of time in the last year asking people how do their minds work? What do you think about? How do you program? How do you think? And in the same manner that it has become so evident
359
00:32:09.505 --> 00:32:12.565
in the tech world about our lack of diversity,
360
00:32:12.945 --> 00:32:17.924
not only in race and gender and gender identity, but there's also a lack of diversity
361
00:32:18.580 --> 00:32:19.960
in our own minds.
362
00:32:20.260 --> 00:32:21.059
If you are,
363
00:32:21.700 --> 00:32:22.820
familiar with,
364
00:32:23.299 --> 00:32:29.065
the professor, I think at Colorado State University named Temple Grandin, She has given some wonderful TED Talks,
365
00:32:29.685 --> 00:32:34.265
about the way her mind works. And I think that we very much need
366
00:32:34.830 --> 00:32:41.169
to embrace the different ways that people's minds work. We seem to push in software toward a monoculture.
367
00:32:41.470 --> 00:32:53.900
We want everyone to code the same way. We want because it's easier to maintain. We want everyone to think the same way and sit in a cubicle, and there are lots of different ways to live one's life and a lot of different ways that our minds
368
00:33:00.940 --> 00:33:01.440
the
369
00:33:02.425 --> 00:33:09.085
the recursive partition sort. I wonder what how his mind worked and how he was different from his peers.
370
00:33:09.690 --> 00:33:16.990
How he was different to the way mine my mind worked. And then it was such a tragic story that I never found out. And it was during
371
00:33:17.295 --> 00:33:18.115
the Pi Tennessee
372
00:33:18.495 --> 00:33:24.995
writing the the present at the keynote presentation for Pi Tennessee that I got out my copy of that book and discovered the translations
373
00:33:25.295 --> 00:33:41.215
of what had been written in there, which was just stunning. I had I had no idea. It still brings tears to my eyes to think of how much regret I've had over my whole life and not knowing that the my absolution, so to speak, was in this book that I have been carrying with me for 30 years.
374
00:33:41.995 --> 00:34:01.840
Yeah. It's a very it's a very powerful story, and I think that it went well along with the presentation that you made with the motorcycle video Mhmm. Which I thought was well presented as well as the musical presentation, and it just it it blended all together really well. And I actually came home from the conference conference with, and they've said, you know, I I had to
375
00:34:03.100 --> 00:34:04.320
show this presentation to this other person,
376
00:34:10.025 --> 00:34:18.125
they've said, you know, I I had to show this presentation to this other person who isn't necessarily technical, but they were still able to appreciate what you put in there.
377
00:34:18.660 --> 00:34:21.480
Well, I was totally amazed by the the talks reception.
378
00:34:21.859 --> 00:34:26.180
I knew it was a pretty good talk, but I had no idea that it would,
379
00:34:26.819 --> 00:34:28.119
be that well received.
380
00:34:28.565 --> 00:34:32.585
Plus, I must also say that I am proud of the fact
381
00:34:33.045 --> 00:34:33.545
that
382
00:34:34.005 --> 00:34:34.905
I can bring
383
00:34:35.205 --> 00:34:36.425
that kind of emotion
384
00:34:36.860 --> 00:34:38.560
to a group of technical people.
385
00:34:40.460 --> 00:34:43.280
Yes. There were That is an achievement in and of itself.
386
00:34:43.660 --> 00:34:45.760
Without violating the code of conduct.
387
00:34:46.085 --> 00:34:46.585
Yes.
388
00:34:48.964 --> 00:34:51.065
So in addition to your
389
00:34:51.365 --> 00:34:57.940
extensive technical career, you're also very focused on living in a manner that is sustainable and in tune with the environment
390
00:34:58.640 --> 00:35:04.605
and your philosophies. And so I'm wondering what kinds of synergies and conflicts exist between your professional
391
00:35:04.905 --> 00:35:06.204
and personal ideals.
392
00:35:07.224 --> 00:35:10.684
Well, I get pretty disgusted with the technical world
393
00:35:11.310 --> 00:35:13.570
in how much energy and hardware
394
00:35:13.870 --> 00:35:14.610
is wasted.
395
00:35:15.070 --> 00:35:17.490
The fact that our computers and our phones
396
00:35:17.790 --> 00:35:27.085
get turned over every 2 years and people who are lower on the economic scale, who have old computers because that's all they can afford,
397
00:35:27.705 --> 00:35:29.085
but they get no support
398
00:35:30.089 --> 00:35:33.710
and can't have the levels of security that we have.
399
00:35:34.010 --> 00:35:36.109
I have some friends who live in rural Tennessee,
400
00:35:36.890 --> 00:35:42.225
who use Apple Computers that are 10 years old. They can't get software updates anymore,
401
00:35:42.765 --> 00:35:46.065
but it's all they can afford. And it really bothers me
402
00:35:46.480 --> 00:35:48.660
that we have this stratification
403
00:35:49.600 --> 00:35:50.660
of of technology.
404
00:35:51.760 --> 00:35:52.900
Now as for
405
00:35:53.280 --> 00:35:54.260
living sustainably,
406
00:35:54.715 --> 00:35:56.335
I live on an organic farm,
407
00:35:56.635 --> 00:36:07.200
I live in a yurt. 1 does not live in a yurt to be energy efficient because it is just a tent. Fortunately, my climate here is mild and our heating needs are fairly minimal,
408
00:36:07.500 --> 00:36:10.320
but I do like the fact that my home
409
00:36:11.135 --> 00:36:13.315
will last many many many years
410
00:36:13.775 --> 00:36:15.955
and has a significant lower
411
00:36:16.335 --> 00:36:24.160
significantly lower footprint than a house does. And at the point that I have to retire 1 of these tents, the yurts.
412
00:36:24.940 --> 00:36:29.075
I can repurpose almost everything in it to some other other use.
413
00:36:29.695 --> 00:36:32.195
I've started living in a yurt in 1995.
414
00:36:32.734 --> 00:36:45.280
That particular yurt has been retired and there are pieces of it all over the farm serving all sorts of different kinds of duties from the parts that that didn't entirely wear out. We run our farm here as
415
00:36:45.745 --> 00:36:53.925
essentially an incubator for people who wish to organically farm learn to organically farm. We lease the property out for 2 or 3 year blocks
416
00:36:54.280 --> 00:36:55.500
to optimistic people,
417
00:36:55.960 --> 00:36:56.780
and they
418
00:36:57.240 --> 00:37:02.060
use our property to learn to farm. At the end of that 3 year period, they either graduate
419
00:37:02.360 --> 00:37:05.305
and go off and buy their own properties and become farmers,
420
00:37:05.605 --> 00:37:09.890
or they walk away glad they didn't spend more money learning they're not farmers.
421
00:37:10.450 --> 00:37:11.250
And we have,
422
00:37:11.730 --> 00:37:13.750
actually started at some very successful,
423
00:37:14.529 --> 00:37:15.269
farm businesses
424
00:37:15.650 --> 00:37:19.109
and had some pretty spectacular failures happen here too.
425
00:37:19.965 --> 00:37:27.460
And as somebody who is technically inclined, do you have a lot of agricultural automation or any sorts of, I guess,
426
00:37:28.020 --> 00:37:29.400
sort of sensor driven,
427
00:37:30.020 --> 00:37:30.520
agriculture?
428
00:37:31.780 --> 00:37:40.335
Not currently. When we were running the rose nursery, I sold the rose rose nursery in o 7, fortunately, right before the Rose Industry crashed. And,
429
00:37:40.715 --> 00:37:43.055
I was just then starting to explore
430
00:37:43.355 --> 00:37:43.855
using,
431
00:37:44.955 --> 00:37:45.455
tagging,
432
00:37:46.075 --> 00:37:49.810
of RFID tagging in plants to help track them in the greenhouses.
433
00:37:50.190 --> 00:37:57.494
The system that I used, I I wrote for the for the uncommon rows in Python did some pretty amazing things that I still haven't seen reproduced
434
00:37:57.955 --> 00:38:01.655
in other nurseries where customers would order plants,
435
00:38:02.210 --> 00:38:12.095
to have them shipped to somewhere else in the country, and they generally in the rose industry, plants are ordered in the fall, they're shipped in the spring. The software that I wrote would go out to the websites
436
00:38:12.715 --> 00:38:14.095
or the weather websites
437
00:38:14.555 --> 00:38:22.190
and scrape data to find out what the conditions would be at the, destination locations for shipping. And if
438
00:38:22.490 --> 00:38:41.090
the shipping route was likely to have bad weather or the destination was bad weather, it would automatically send an email to the customer suggesting we would like to delay shipment by 4 days because it's gonna be freezing in Denver, the likely route the plant will travel through in a truck and our customers love that and so we had that kind of sophistication
439
00:38:41.390 --> 00:38:52.025
in our operation here, but that's all gone since we don't have the nursery anymore and the people who are doing farming here aren't particularly interested in that level of technology.
440
00:38:53.090 --> 00:38:55.510
They're just trying to get plants to grow in the ground.
441
00:38:56.450 --> 00:39:00.285
That's great. Are there any other topics or questions that you think we should cover?
442
00:39:00.684 --> 00:39:05.905
Oh, I think we've talked often quite a bit here. I can't think of anything offhand.
443
00:39:06.444 --> 00:39:09.345
I probably will, of course, 10 minutes after we
444
00:39:09.700 --> 00:39:10.359
part, but
445
00:39:11.380 --> 00:39:22.915
I can live with that. Alright. Well, if you do think of anything else, you can feel free to join our discourse forum and continue the conversation there along with anybody else who cares to partake. Okay. That sounds great.
446
00:39:23.455 --> 00:39:29.520
So for anybody who does want to follow what you're up to and keep in touch, what would be the best way for them to do that?
447
00:39:30.060 --> 00:39:30.560
My
448
00:39:31.100 --> 00:39:31.600
website,
449
00:39:32.140 --> 00:39:32.880
2 braids.com,
450
00:39:34.255 --> 00:39:34.755
twobraids.com,
451
00:39:37.135 --> 00:39:41.154
is where I do my blogging. That's where my portfolio of my art is,
452
00:39:42.250 --> 00:39:50.109
and most of the stuff I do. I I tweet on Twitter at 2 braids, but it's the digit 2, braids.
453
00:39:51.215 --> 00:39:51.955
Couldn't get
454
00:39:52.335 --> 00:39:53.875
the spelled out 2 braids.
455
00:39:54.655 --> 00:40:01.540
Unfortunately, the the poor woman who has that gets a lot of stuff directed to her that was supposed to be directed to me.
456
00:40:03.280 --> 00:40:10.055
And and for anybody who isn't familiar with your appearance, can you describe a bit about what the genesis of that name is?
457
00:40:11.315 --> 00:40:11.815
Back
458
00:40:12.195 --> 00:40:20.779
in well, when I lived in the Blackfeet Indian Reservation, my beard is very long down to waist level, and I generally keep it braided into braids.
459
00:40:21.559 --> 00:40:22.059
And,
460
00:40:22.520 --> 00:40:24.140
you know, when I was teaching there,
461
00:40:24.920 --> 00:40:31.395
I I gotta say I had 2 types of students, the men and the women. The women would say my name. The men
462
00:40:31.695 --> 00:40:38.580
never would say my name. They'd always just call me white guy 2 braids. And so I embraced that and took the name 2 braids.
463
00:40:40.720 --> 00:40:46.765
That's great. It's a great story to go behind that. Alright. So with that, I will move us into the picks.
464
00:40:47.145 --> 00:41:10.950
For my pick today, I'm going to choose Terry Pratchett because he is a wonderful author who has brought a lot of humor into this world and a lot of enjoyment for anybody who has read his books. I'm currently reading his last book that was published posthumously. So for anybody who has never discovered him or who has but has recently forgotten, I definitely recommend taking a look at anything that he's written. And with that, I will pass it to you, Lars.
465
00:41:11.750 --> 00:41:13.130
My pick for the day
466
00:41:13.589 --> 00:41:17.849
is an old 1. I'm gonna say Johann Sebastian Bach's
467
00:41:18.265 --> 00:41:20.924
tokata and fugue in d minor.
468
00:41:21.224 --> 00:41:24.204
That's a great piece. If you listen to the fugue,
469
00:41:24.744 --> 00:41:26.125
you will hear examples
470
00:41:27.220 --> 00:41:28.599
of multi threading
471
00:41:28.900 --> 00:41:30.359
and cooperative multitasking
472
00:41:31.059 --> 00:41:31.720
in music.
473
00:41:32.420 --> 00:41:38.464
Alright. Well, I really appreciate you taking the time out of your day to join me and tell me about some of your
474
00:41:38.765 --> 00:41:40.545
work history and philosophies
475
00:41:41.085 --> 00:41:53.380
and, the wonderful stories that went into your and, the wonderful stories that went into your presentations and everything that you've done. So I appreciate it, and I hope you enjoy the rest of your evening. Thank you very much. It was an honor to talk with you, and have a good evening.