Archive: February, 2023

There's a guy, he runs a small-ish tech company and is somewhat famous for the books that he and his partner have written (collections of blog posts, technically). The books challenge how we do things in business, and especially in software companies. I was all-in on what they had to say, but mostly the parts about how killing yourself for your job is kind of dumb. That they concluded this at a relatively young age is fascinating, because I find this to be more true in midlife as I have fewer and fewer keystrokes left to give.

But some of the things that they rally against are dogmatic in the same way the alleged entrenched process are. Context tends to matter as you apply your experience to different situations. And if your experience is almost all from one company that you've been at for essentially your entire career, chances are you have some blindspots. And your blindspots might be worse if you've had some popular books that seem to validate your ideas.

The latest rally by this guy is against the use of cloud resources in favor of privately owned hardware in a data center. Basically, their spend was crazy high and they believe that they can do better themselves. On the surface, this is not unreasonable, but there are two problems The first is the assumption that their new state will be equivalent to their cloud endeavor. It won't be, because you can't afford to build out the same level of resiliency yourself. And I'm suspect of what they were spending on in the first place. The other problem is that he seems certain that what's good for him is good for everyone, much in the same way he applied that logic in the work culture books. I've been through three cloud transitions (maybe four if you count the partial movement of the stuff my team worked on at Microsoft), and I can tell you without a doubt that we saved money every time. I have my anecdotes, he has one, but I think mine would be baseline in a larger, normalized dataset.

Now, I wanted to call the guy out by name because he just comes off as arrogant (and recently declared that diversity, equity and inclusion are a waste of time, despite research that clearly indicates the opposite). I am skeptical about anyone who is so sure that they have all of the answers, and even more skeptical if they simply haven't seen a wide variety of scenarios up close. Everyone wants shortcuts, but the hard truth is that while we can start with some guiding principles and ideas, contextual adaptation is the only consistent path to success. Yes, I'm saying that the sure thing is understanding that there is no sure thing.

I gotta say, I was really stewing about this guy, whom I've never met, and frankly don't care that much about. My weakness is letting people who have the attention of others, but aren't using that power for good, rile me up. I don't have the attention of many people, and the attention that I want is inversely proportional to my age. But also, I am acutely aware of how ineffective it is to write a bunch of stuff about some random dude that's kind of an asshole. No one cares. Nothing changes by writing that. It's an energy drain that I would lump in with keeping up the appearance of having an awesome life, or trying to convince people their religion is wrong.

But I still find it therapeutic to talk about the ideas of the person, and why I don't agree with them. It helps me center my own direction. Like I said, a lot of what your experience teaches you is what not to do.

Back in December I decided to take the plunge and buy a drone when DJI introduced the cheaper Mini 3 (non Pro). I ended up returning it since I live in the Disney no-fly zone. It was an expensive impulse buy, and I was irritated. But now, I have that project, and there are at least two instances where I will need to get some video from the air. And when I'm just wanting to have a little fun, I can drive a half-mile away to get out of the bubble. So I bought one again, but realized that it made more sense to buy the pro variant. It adds two important features, the first being obstacle avoidance, since I'm probably going to run into things otherwise. The other important thing is that it records in "cinelike" quasi-log mode, so it's a bit flatter and you can mess with the exposure more in post.

I've done two test flights. My first impression was, wow, this sure is easy to fly. To be clear, it should be, since this is a video camera first, that just happens to fly. It has a gentle, automated means to takeoff and land. In theory, if it loses contact with you, it will fly home (where you turned it on) automatically. It's supposed to have a range of 10km, but I can't imagine any scenario where I would even try to fly it that far away. There are a number of tracking maneuvers that it can do, but I haven't tried any of them yet.

On my second test flight today, I flew it in between various trees to see how it would do with the avoidance. It did seem to help steer around them pretty well, and I didn't crash. Even flying over my car, starting two feet off the ground, it seemed to make it a point to fly with a lot of extra room going over it. It's forward speed tops out at about 22 mph, and it can climb or descend at about 5 m/s, about 11 mph. It's not an FPV drone, for sure. Fortunately, because it is so light, not only does it not need to be registered, but you can fly on a single battery for about a half-hour. I have the extra batteries, so I drained one to 20% before swapping them out.

What's remarkable is the stability of the camera. I guess I shouldn't be surprised, seeing as how they make a pretty amazing gimbal (I've had a woefully underutilized RS 2 for two years). You can push it forward, make a hard stop, and the image stays level. You can hover at 400 feet, and it looks like it's completely still. The image quality is solid if you use the D-Cinelike color mode, though I haven't yet quite figured out how to color grade it in a way that pleases me. The problem is that you have to crank the shutter speed up to 1/3000 to use it daylight, so while 24 fps looks "cinematic," it doesn't look as good with a lot of quick panning. There are neutral density filters available, but I have to wonder why there isn't one already on it. I don't imagine it's useful to fly at night, so it's implied you'll mostly fly it during the day.

I'll keep playing and logging time, but the things that I know I'll need to shoot are going to be very easy to capture. Slow movement looks a lot like you're tracking on a dolly or jib. It's really fantastic.

Over the years I've expressed my disappointment over the decline of Mac laptops. I won't re-litigate that, you can read for yourself. I wasn't planning to buy a new laptop this year, as my previous one was a little short of two-years old. But it kept coming up in content circles that I pay attention to. Video people had been raving about the M1-based machines for more than a year, because they can crush 4K video even with color grading and compositing. Also, the tools that I use for software development, .Net-based frameworks, had Apple silicon-native versions. With the triumphant return of good keyboards and ports (and banishment of the stupid touch bar), and the insane battery life afforded to machines using ARM processors and SoC's, I couldn't ignore these anymore.

There's one more important thing though, and that's the realization that small, thin laptops are kind of slow for software development, and not a great experience for video work. It's not always true, but it's a valid generalization. The MacBook Airs have been an exception, using these new processors. But small screens aren't great when you're looking at a wall of code either. Mac or not, I was ready to go big and heavy. I haven't had a big, powerful laptop since I replaced my 17" MacBook Pro back in 2012.

I knew what I was getting into because I've had the M1 version of this computer for a year, issued at work. It's total overkill for a people manager, and I've only had 10 battery charge cycles on it in that year. Even though it has mostly sat on my desk, I know it's big and heavy. The new version is basically the same on the outside. The trackpad is large enough to land a helicopter on, and the keyboard is the kind of solid amazing experience of the pre-2015 Macs. They've somehow managed to squeeze richer sound of the chassis as well. The sharp edge of the keyboard deck is still uncomfortable and not great, especially coming from a Surface Laptop with the Alcantara skin, and I don't know why they're too stubborn to go to a tapered or rounded edge.

There are vents on the sides, and in the hinge, though I've yet to encounter any fan noise, even when compressing video. At worst, I've felt a little warmness on the bottom, but never hot. The thermal management and energy efficiency is something to behold. Indeed, battery life for me has generally landed somewhere between 10 and 12 hours. I'm guessing a little because it's hard to keep track over several days, and it might be more. After it learns your habits, it may park the battery down at 80% (I wish you could explicitly do this), which is good for its longevity, but it hasn't done that for me yet. I may need to keep it plugged in more often.

The screen is as good as anything Apple has put out with high DPI's, and lettering appears to be painted on. I still haven't completely figured out color management as it relates to DaVinci Resolve, but what little I've edited on it appears to look right on Windows and Android. Because the battery is so enormous, you don't have to settle for 30% brightness to stretch the time between charges. This is very useful when you spend a lot of time out on a sunny patio.

Let's talk about performance. You really don't notice any differences across most devices these days when engaging in web nonsense. For software development, my usual situation beyond the browser is to run a few Docker containers, specifically for SQL Server, Redis and ElasticSearch, as well as the Azureite storage simulator. The IDE I use is JetBrains' Rider, which is absolutely fantastic and a lot faster than Visual Studio on Windows. Comparing the IDE's isn't really valid, but the one thing that I can compare is build time. A fresh build of POP Forums, the solution that I work with the most, takes just over 4 seconds. On my Windows desktop, which is about 3.5-years-old, it takes about 9 seconds. That's a crazy improvement. My laptop took nearly 12 seconds. The 700+ unit tests in that project also run in under 2 seconds, and again, about double on my desktop.

Do those few seconds matter? When you're making changes fast and want to see the results, yes. This new world of having hot-reload in your build makes it even more important to get that fast feedback. It's extraordinary not just for the speed, but the fact that it isn't on Windows. (Sidebar: My sites have all been running on Linux in the cloud for years, so the platform variation isn't a new thing.) I can even keep dynamic analysis running and looking for crappy coding on my behalf.

The other big use case is video. I graduated to 4K in mid-2020, about a year after I built my desktop PC. Video editing apps, specifically Adobe Premier Pro and DaVinci Resolve, take advantage of dedicated graphics processors. My desktop has an RTX 2070 card in it, which was not top of the line at the time, but it was still $460, way more than I ever spent on GPU's back in my pre-Intel Mac days. Because of this, both apps tend to export compressed video just slightly slower than real-time, which is to say a minute of finished video takes a few seconds over a minute to process. It's not very smooth in scrubbing across a clip or timeline though, especially if you've applied color grading. It skips around and stutters a bit, though not all of the time. The CPU is an i7-9700, overclocked, so it's not slow. But that 4K performance has been disappointing during editing and grading.

The new Mac is insane for video. Color graded 4K scrubs like buttah. Exports run at about half time, so a minute takes 30 seconds. I have the base level 16" M2, with 12 CPU cores, 19 GPU cores and 16 neural engine cores. I haven't even used features like noise reduction, voice isolation and other AI-based things that would benefit from the neural cores. But this overall capability changes video completely. The only problem is that I'm still not crazy about video work on a laptop, but the perf is too great to ignore. I wonder if one of these crazy Mac Minis are going to end up on my desk someday.

I haven't traveled yet with this big machine, but I've really enjoyed using it. I don't feel like I have to wait for anything. It really is noticeable, and I appreciate how far we've come for dev work. It's so fast.

The big question is whether or not the value is there given the performance. The thin and high-end laptops are generally in the $1,600-$1,900 range, and this one starts at $2,600. There are a few things you could bake into that, starting with an even more premium chassis and keyboard, one of the best screens on anything and better battery life. It's a little harder to compare on specs because the architecture is so fundamentally different than an Intel-based computer. The base starts with a half-terabyte of storage, and 16 gigs of RAM, which is just OK at this price. Adding in either category adds hundreds of dollars, which changes the value proposition for the negative. And the RAM part feels especially weird since it's all embedded into the CPU anyway, and there aren't additional parts. If you work backward from price, then a similarly priced Intel laptop has more "stuff," but performance is hard to nail down because you can't measure the same things. Well, you can measure gaming, but I don't think people are buying Macs for that. For my two big use cases, the new Macs are worth it, provided you don't go nuts on the upgrades.

I'm committed now. I mentioned yesterday that I was gearing up for a big film project, and today I cemented the first scheduled day for that project. I hope to have a second day scheduled before the end of the week (which would actually happen first).

I'm excited, but also a little nervous because I want to get it right. I've been messing around with the gear, especially the cameras, to make sure I get it. I come from a world of run-and-gun news gathering style of videography, and my camera is capable of commercial grade imagery. It doesn't quite get to the level of an Arri Alexa 35 (because I don't have $75k), but if you're careful with exposure it's more than good enough for any streaming service or theatrical exhibition. I want to achieve that quality.

The last thing that I need to order, I hope, is a decent drone. There are at least two spots I need some aerial shots. And also I think it'll be super fun to play with. That's hopefully the last big equipment item, because I'm exhausted buying even small stuff. I did have a budget in mind for this project, and I'm nowhere close to reaching it, but some of that I figured will be travel expenses. Not sure yet if that's going to be super necessary.

Most importantly, I don't want to rush this. I've given myself basically the rest of the year for this. It won't be done until it feels done. And I think it's gonna be fun.

Bernie Sanders has released a book called, "It's OK To Be Angry About Capitalism." I know that Sanders means well, and while I haven't read the book, the title feels like the political baiting that frankly the nutty right-wingers are known for. The distaste for capitalism is completely misplaced.

I am generally in the camp of people who wants more equality, less discrimination and hate in the world. The aforementioned righties tend to scapegoat all kinds of people that are different from them, based on race, gender, identity, religion, ethnicity and nationality. We know, because we engage in critical thinking, that the scapegoating is irrational and not rooted in any real fact. So it's surprising then when people at the other end of the spectrum do the same thing, directed at people who have more than they do. This hostility ignores two realities they are often not interested in considering. The first is that rich people are not inherently evil, nor have they necessarily acquired their wealth by immoral means. The second problem is that their very existence, and possession of their wealth, does not necessarily come at the expense of the poor. Let's break those down a bit.

I'm not going to sit here and tell you that the "American dream," with its tales of bootstrapping and tenacity, is real. It is absolutely a myth. Your birth lottery has a lot to do with how you get to that mythical success. Despite that, some people, often, but not always, in advantageous circumstances, achieve a level of wealth and economic prosperity that is disproportionately higher than average. It doesn't mean that they cheated or exploited anyone. It's pretty lazy thinking to generalize that the rich all got there by nefarious means. It's even worse to assume that they are immoral or terrible people. I can tell you from hanging out in certain philanthropic circles that very well off people tend to open their wallets pretty freely. They are not immoral people.

As for the rich existing at the expense of the poor, that too requires a fundamental misunderstanding of the choices that our society has made. Especially in wealthy nations, the existence of rich and poor people in the same place is the result of societal decisions. It is not a zero-sum game. We as a society choose not to address poverty. The people who often make the decisions that lead to these choices are not necessarily rich themselves, but they are certainly influenced by the idea of the dream myth, and probably by people and corporations that handsomely reward their election campaigns for their resistance against addressing poverty. That in turn happens because people who vote don't feel it's important enough to elect candidates willing to do something about it (and generally, they seem to vote for people who fuel rage over offering solutions).

So what does a society do about it? I don't have all of the answers. I think it starts with finding novel ways to either keep money out of politics, or bring it to a completely new level of transparency. I get it, because of the Citizens United decision, corporations are people, and entitled to free speech protection. It doesn't mean that you can't require extraordinary disclosure from SuperPACs about where they're spending money. Remember the time before candidates were required to say that they "approve of this message" and it was required to state, even in small text, who paid for ads? We could take that further by requiring to declare where the money came from. Free speech is protected, while transparent.

We can certainly close all of the loopholes in the tax code that allow the rich and corporations to pay relatively little, in terms of percentages, compared to ordinary people. A vast portion of the populace continues to vote for candidates against that, despite it being in their self-interest (because the same people are against the gays and people of color, of course). That's a somewhat more challenging problem, because getting the revenue side of government, which can head off poverty (as we saw during the pandemic), is only half the story. The other side is to be critical of spending. We exponentially spend on the military every year without consequence, and we're not even at war with anyone.

Healthcare is an enormous drain on the economy, and we don't do anything about that. Our outcomes are at the bottom among wealthy nations, while we spend more per capita than the next several nations combined. This isn't a zero-sum game either, but when a fifth of your GDP is spent on healthcare, clearly something could give.

What really gets me though is the scapegoating. We can't solve problems by looking for people to blame, we can only solve them by exploring solutions.

I've kind of hinted at it, but I'm in the planning stage for a video project. It's getting closer to reality as I nail down some specific plans, and it's pretty exciting. So with that in mind, I've been obsessively trying to figure out how prepared I am, and what the gaps are in terms of my equipment stash.

The core and most important thing is a camera, natch, and I'm good there. I've had the Canon C70 now for almost two years, and as I said before, it has only become more capable. The hardest part about these cinema cameras is that they're just not as straight forward to shoot with as shoulder mounted cameras back in the day. You can rig them out with shoulder mounts, but the weight distribution is always entirely too front heavy. What I ended up doing is getting an Easyrig Minimax, which is this wonderful piece of Swedish engineering that suspends it from overhead, with the right amount of tension that allows you to move it up and down effortlessly. Keep in mind that this isn't a stabilizer, it's just something to relieve the weight. I have a gimbal as well, but it's way too heavy to use for more than 30 minutes at a time.

My lighting situation isn't great, but I have a brighter light on order suitable for outdoor and bright situations. What I still lack is a scrim for diffusion (inside or out) and various forms of negative fill control. I'm going to go with Diana to the fabric store and see if there's a cheap fabric that's dark enough to work, and then have her quilt it. I can clamp that up to some C-stands and be good without spending hundreds on solid flags. In the pandemic I bought some cheap LED panels, and they've been great for doing LEGO time lapses (and for just seeing better while doing anything at the dining room table), but they aren't bright enough for use in daylight situations indoors, let alone out. They're good enough for fill or color washes on walls, but not great as key lights.

Beyond that, my audio game is good enough with the DJI Mic set, and a small shotgun on-camera that I've had on two previous cameras.

I experimented the other day with using my R5 as a B-camera, and it turns out that it's really easy to match it with the C70 if both are shooting in CLog3. That's great because I can use it as a second angle for interviews. The shutter speed at 1/50th is close enough to 180º (i.e., 1/48th), but depth of field is challenging with the lack of ND filters. I can probably fake it if I put the long lens on it and stop it down to whatever. The 24-70mm f/2.8 looks generally amazing on the C70. Some day I'd like to compliment it with the 70-200mm f/2.8, but even with a recent price drop, it's still expensive. Assuming Canon doesn't abandon the RF mount, these should be the last lenses I ever buy, but it's hard to look at the cost as something that lasts decades.

I'll talk more about the project soon. This is my passion project this year. I'm really excited about it.

In late 2021, on the heels of some kind of quasi-normalcy relative to the pandemic, we took a cruise aboard the Disney Dream. That ship, in some ways, almost feels like a second home, we've been on it so many times. There were still mask requirements in most inside places, but it was so good to be back.

An indulgence for me on cruises has been massages. I'm not going to pretend that there's some kind of deal, even when you book one for some departure special or whatever. But on that cruise, after being inside for so long and looking for variety, I thought, how about a pedicure instead? Especially given my habit of picking toenails until they bleed (gross). I guess I didn't even realize that was a habit, but in the moment, it was obvious.

So we scheduled pedicures. And since it was included in the price, you're goddam right I was going to get the nail polish. Black, for my inner goth. I'm a little ticklish on my feet, but I had to admit that I overall enjoyed the experience. I was surprised at the general positive appearance of my feet, especially given my age. If I exercise any kind of privilege, it's admittedly one that leverages the ability to feel taken care of.

About a month later, I remember I was sitting on the floor in Simon's room, and keep in mind that, because Florida, I rarely wear shoes or socks. Whatever we were doing, I picked a toenail on my "pretty" feet, and I realized, shit, I kind of un-did what I paid for. Because I'm so used to tucking a leg up wherever I'm sitting, and often picking a toenail, it was borderline involuntary.

I don't talk about habits or "stims" as they call them, because they're weird neuroatypical and not things you would talk about. But as I see Simon deal with things like picking skin, I realize that there's a large involuntary component to things like this. My own autism and ADHD diagnoses has given me permission to acknowledge similar behaviors.

After that cruise, a few months later, I didn't want to trim my toenails because I was worried that I would get back to pulling off nails to the point of bleeding. So I went to a local nail place, and for a fraction of the cost of the cruise pedicure, did it again. I've been doing it ever since (with one of those pedicures on a cruise). It's not so much the appearance of clean, taken-care-of hygiene, as much as it is a preventative measure against a bad habit.

I went again today, and my feet appear perfect. I'm not inclined to pick them.

It's odd, or maybe that isn't the right word, that you can find things that counter relative negative behaviours. It's another example of how, for people with ASD and ADHD, those things aren't always habits we can control. But we can, surprisingly, manage them or engage in behaviours that mitigate the problem.

I am gonna review the new laptop eventually, but for now let me say that I continue to be surprised at its performance with software development tools, but also with video manipulation. But even before I pulled the trigger, doing the math about what constitutes "better" performance was weird.

I built that PC on my desk about three and a half years ago. The GPU, an RTX 2070, is still more than respectable in terms of its gaming capability. It maintains high frame rates in Halo, Forza and LEGO Star Wars, the only games I regularly play. It did more than OK with Planet Coaster and Planet Zoo back in the day, too. The current generation of GPU's cost about three times as much for performance normal people like me wouldn't see. Where I leverage that power is in video editing applications like Premier Pro and DaVinci Resolve. The CPU itself, with a bucket of RAM and solid state drives doesn't seem to matter as much, though the plumbing between them does.

Up until a year and a half ago, Mac used the same hardware, so comparing capability was not straight forward. What Apple did consistently well even then was manage power consumption and heat, and by extension, battery life. This was possible in part because they own the operating system and the hardware design. But then they started going the route of ARM-based system-on-a-chip (SoC) processors. One chip replaced a lot of chips, and that has a lot of advantages to it, specifically low power consumption and shorter pathways between all the things.

Benchmarks don't really measure the "right" things anymore, and lack context. They used to measure clock speed and number of processing cores, but if there is efficiency gained by the SoC design, these numbers don't mean the same thing. Figuring out how to use the GPU in tandem with the CPU matters too. Also, if battery life matters to you, as it generally should on laptops, you have to take that into account as well. Battery life is a performance metric.

So with the wildly different architecture, I mostly had to lean into specific use cases and how they "feel" to the user, and what people were reporting. Video editing was already not even close, and I can confirm that. Trying to edit video that you're color grading and/or applying noise reduction to, on my PC, is not ideal. Scrubbing across the timeline skips a lot, and if I do it too fast, it's not usable. There are work arounds to this, namely creating smaller proxy files for editing, but that's an extra step. On the M2 MacBook Pro, I can apply many layers of adjustments to the video and it scrubs without a hitch. I mean, I've never seen it that smooth ever, even in the old standard definition days. The typical software development load is also ridiculously fast, using the same .Net tools I would use in Windows. App build times are a third of what they are on my desktop. (Mind you, I'm also starting to realize that my "thin and light" laptops have been not ideal for dev work.)

I realized this problem about benchmarking when reading a recent article about how to spec out a computer for video editing. As the author pointed out, you can buy a $600 Mac Mini now and you're already in a pretty good starting place. If you buy the $1,300 version, you have a crazy good solution that might well outlast the software running on it. Performance has never exceeded software capability in video editing until now.

I've also come to realize the "but you can't expand or replace parts" thing is completely not important. I haven't added RAM to a computer in over a decade. The only thing I've added to my desktop in 3.5 years is more storage, which I could just have easily made external. And even if I wanted to replace the CPU or memory, I'd have to replace both and the motherboard because the standards and clock speeds have changed. The whole modular thing doesn't matter anymore, and we can probably thank cell phones for that.

There are PC parts that could blow away the current crop of Macs, but to get there, you need to buy CPU's that are over $500, GPU's that about a grand, and that's before you buy something to put it in. That setup will also use two or three times the power. Remember, Apple originally went from Motorola to Intel to catch up, but this change is something we haven't seen really in my lifetime. It's possible that Windows on ARM could make a similar leap, but it will probably require Microsoft to figure it out for the PC makers, much as they did with the Surface line.

The new Mac laptops are still awfully expensive, but given what they deliver, it's kind of justifiable. The MacBook Airs and Mac Minis are extraordinary in the performance-cost curve. Budget creators can afford the kind of power that used to be reserved for people with budgets.

I imagine it's mostly just coincidence, but I've encountered identity in a number of different jobs, and I'm seeing it again at work. "Identity" in the software sense is the way that you identify people as real, and use that information in another system. For example, we've all used various services on the Interwebs by logging in first to Google or Facebook or whatever. At work, you may login to some central thing before bouncing to some external service.

For years, even prior to someone filing an issue on the project, I've wanted to work external identity into POP Forums. Over the years I've had so many developers try to integrate the forum by way of changing the underlying data. In other words, they would try to replace the forum's user records with their own data. That almost never works out, unless you use the forum itself as your source of truth for user information (this is what I do on CoasterBuzz). But inter-service identity has been a solved problem for a long time. The protocols and standards are well known and fairly easy to use. I didn't attack this on my previous release, which was pretty huge, but it's a pretty big deal that justifies another release, sooner than later.

For whatever reason, I got a bug up my whatever on Friday and decided to start working out how to do this. I'm using my Azure AD as a test source, and so far, so good. I can login through it and the forum will automatically provision an account. If the user is in the right groups and serves up the right claims, it will also map those to moderator and admin roles. It will periodically check to make sure the user is still legit with refresh tokens. It all just works, and I got there a lot faster than I expected.

I still have to work out how to reconcile same names (a company might have more than one John Smith, for example). I was going to do the same for email, but I would think the odds of any organization having recycled email is pretty rare. The forum wants unique names and email, so in the event of a duplicate email I still have to do something about that.

I've been way more into thinking about video lately, because of another project, but it was fun to dive into a code project.

In recent years I've been noticing the "look" of movies in a more specific way. By that I mean, I've been seeing choices about color, dynamic range, lighting. While a lot of this is set up "on the day," the finished product arrives by way of a colorist who grades every single shot. I mean, nothing in the shot is left to chance. Now multiply that times several hundred (or thousand) in a typical movie. It's a lot of work, a lot of creative choices.

I'm partly more aware of this because color grading is a first-class function of DaVinci Resolve, the thing that I previously mentioned is my hot new video editor. This is an area of learning that's fairly new to me. Remember that the foundation of my video knowledge is from the days of 3/4" tape. When we edited, at best, we might have a timebase corrector that allowed us to make tweaks to a signal that wasn't particularly robust compared to what we can get today. These days, my hobby camera is a Canon C70, which has pretty extraordinary dynamic range and detail for what it costs. It's a new world for me. The camera can, if I'm willing to eat up crazy disk space, record the raw signal, the way still photos work, meaning you can recover an incredible amount of detail out of something that you exposed wrong. Even in the compressed mode that I normally record in (10-bit 4:2:2 long GOP at 24fps, 4K, if you were wondering), there's a lot of latitude there.

One of the ways you do this is to record in a log format, logarithmic, which is super flat and looks pretty terrible until you mess with it. But you apply what they call a look-up table, which maps the image information to the typical color space that you're used to, and from there you can make adjustments. For example, the sky might be over exposed, but you can pull the detail out of it and see the clouds and blue sky. Again, for still photographers, this is hardly new, but it's pretty great for video. So for my most recent LEGO time-lapse and review, I shot everything in log, as an experiment. There are literally more levers to pull than just exposure, and there's a curve (see what I did there?) to learning how to manipulate the image. It's fascinating. In this video, I see that the yellow-ish bricks look a little washed out, even though the rest of the image looks pretty good. I could isolate that color and adjust it, I just didn't see it at the time. And yes, handheld shooting is a choice.

One of my favorite recently discovered channels is this guy Darren Mostyn, a professional colorist, using Resolve. When you watch through one of his videos, he points out things that are not obvious to a novice, but they make a difference. A dozen small changes end up making a huge difference between some schmuck's iPhone video and professional results.

Never stop learning, y'all.

My first real non-linear digital editing experience was with Media100 back in the late 90's. It was a solid alternative to Avid, which was charging crazy money at the time. A few years later, I bought a $1,500 license for Avid for personal use, which was nuts. It even required a USB key to work. When the world of HD came around, I was using the old Final Cut Pro for a little bit, then they had that big change that made it lame and I switched to Adobe Premier Pro. I've been using that ever since. I held out for years with the last non-subscription version of it, then a few years ago had to cave because it just didn't work anymore on modern hardware.

But for the last year or two, I've noticed that a lot of the YouTubers and enthusiasts were taking DaVinci Resolve more seriously, ever since it was acquired by Blackmagic Design. The software has been the go-to system for colorists for a long time, with some pretty great control surfaces to pair with it. I'm not sure when the editing became stronger. I was skeptical mostly because it was free, missing only some of the best features like object tracking, noise reduction and not reading most of the more exotic or camera-specific file formats. I messed with it using some plain old edited files, and I was hooked. I could already see it was better, and the UI made it more obvious to discover how it worked. It's still feature dense, but well understood by the user community.

Then what really sold me is that they have this little controller called the Speed Editor. This has been the thing I've wanted since the days when I edited tape. Having a nice, weighted jog wheel and robust buttons that you can bang on is something that I always wanted translated to the digital world. I have a search dial that was never quite the same thing. But this little machine allows you to spend less time touching the mouse and keyboard, and it makes everything so fast. Trimming stuff is intuitive and fast, you can adjust audio levels with the wheel, you can do multi-cam editing... it's so good.

I'm thinking about dropping my Adobe subscription down to the $10 Photoshop level. I'm currently on the whole suite, but only at $30 because I don't use it enough to justify $55, and apparently they agreed enough to let me keep that price. The studio version license of Resolve came with the controller. I only have to pay for it once.

Editing is fun again, and I look forward to using it more.

This guy that I enjoy watching on the YouTube, Gene who goes by "Potato Jet" on his primary channel, posted a video about his new home studio setup. What I usually like about him is that he gets out and makes video pretty much everywhere, but he takes such care in its creation that it almost always looks great. So while it's a little surprising that he would put a lot of energy into his home setup, it's really inspiring to see what he did, and it's really pretty good. It's a solid bit of inspiration.

But as much as I would like to make stuff like that, the truth is that I'm so not interested in any content being primarily about me. It's not a self-esteem issue, it's that I feel like it would be inauthentic, and not interesting. As much as I loved the performance aspect of radio, the outcome was never about me, I was just there to get you to the music. Maybe it's my journalistic purist bent, where if you tell stories you should avoid becoming a part of the story.

The problem with the world of independent content creators is that it's all very personality driven, and I don't think that's me. This thing with having an online presence seems exhausting to curate, and if you know me, I'm not one to curate or even filter myself. Well, let me qualify that. I've spent much of my life I think trying to be less weird in the eyes of most people, but to know me is to know and celebrate my weirdness. Which person would I have to be online?

There's also the problem that most of the people who are really good at this, and are doing something more interesting than streaming video game playing or dick and fart jokes, are generally playing to a particular niche or interest. I think necessarily, you have to do that. The science and maker people are particularly good at it.

The bottom line here is that I really, really love making video content, I just don't want it to be about me. I should make a documentary or something.