Keith's Odyssey to Planet Fitness: There is a technology sweet spot

Someone gets an idea. They create a prototype or proof of concept. Then the first versions come out and are rapidly revised as bugs or flaws are found. At some point there is usually a sweet spot where everything goes together really well, and other technologies mature around it. The trick is to recognize it and buy it then.

Development continues, and sometimes there is another sweet spot with more advanced technologies. Sometimes the sweet spot is ruined with further developments, and the product falls into decadence.

Some examples, you ask. Sure.

Airplanes. First heavier than air flight in 1903. Not much happened at first as people didn't really believe there was a future for it. Then WWI happened, and the technologies made a big step forward. Post war development continued, but they were essentially still trying to figure out how the peacetime aeronautical industry worked and how to integrate it with everything else.

Then the DC-3 came along in 1935 and revolutionized air travel. This was such a sweet spot, such a perfect all round airplane that some are still flying today, hauling cargo almost 90 years later. Of course, development continued and aircraft got bigger, faster, and much more complicated. Some would say the 747 was another sweet spot, and I wouldn't argue with them. Further development has brought out refinements, like more efficient engines, lighter and stronger materials, and computer aided controls. No decadence, so far, unless the Boeing 737 Max fiasco counts. My understanding is they tried to push the design too far, and tried to game the regulatory requirements around training, rather than build a new design.

Now let's look at a software example. Excel. Let's start with a joke to set the mood for you. The recruiter asks the contractor "How familiar are you with Excel?" The response is "I loathe it with the burning passion of a thousand suns." The recruiter says, "Good! Highly experienced."

I remember the first versions in the early 90's. It could barely add up a column of numbers and create a graph. But the usefulness was obvious, and improvements happened so quickly it was hard to keep up. I've lost track of version numbers, but the sweet spot happened somewhere 2000 to 2010 or so. Then they kept adding features. More and more, regardless of actual need. Most of them didn't really make it function any better, just made the outputs look fancier. Useful features were buried behind the bloat, unless you went through and put them on a special menu bar. Which was good as long as you were working on your own computer. Go to a client's office for a presentation and you looked like an idiot searching for a function button.

At first I could routinely exceed it's row and column limitations and had to figure out ways around that. At some point they sort of added the ability to handle a huge number of rows and columns, but they goofed. The filter feature was good for 10,000 rows, and they didn't update that. I quickly learned not to rely on the filter when there were more rows than that in search results.

My go to tool was INDEX MATCH. Once I wrapped my head around it, it beat VLOOKUP hand's down. I could make it sing and dance for all kinds of data migration tasks. And then decadence happened. They "improved" it to use a matrix style functionality, and completely destroyed the usefulness of it. That lead to a great deal of swearing during my last contract.

This happens a lot with software. They simply can't leave good enough alone. I've lost count of the number of products I've used, then stopped using because the further developments buried the original purpose, or made it so difficult to use. Evernote is another example. It was great idea to synchronize text notes across all platforms. I loved using it to track the various job applications during the year I wasn't working. I'd capture the original job posting, my exact response, any recruiter discussions, interview notes, and all the relevant dates. When someone called, I could quickly look up the notes and make it look like it was top of mind. I'd get to an interview a few minutes early, and review the notes.

Then development happened and one day, fortunately after I was working, I looked at it and couldn't figure out how to add a new note. They'd added so much bloat that the basic function was buried. That did it. I deleted it and have never used it since.

Don't get me started on Word. If there's another piece of software more bloated and cluttered with flab to the point of being unusable, I don't want to know. There's a reason many people in the software industry use Notes, or Notes++ for, well, taking notes, creating documentation, and other data related tasks.

I could go on about cameras. They are fundamentally a very simple thing, a lightproof box with a sensor of some kind, and usually some kind of mechanism to control light entry, and maybe a lens to focus the light. The new mirrorless cameras on the market now make it much easier to produce amazing images under almost any conditions. The lenses are so sharp, almost too sharp. Now keep in mind that the photographer most people have heard of, Ansel Adams, once said, "there is nothing worse than a sharp image of a fuzzy concept."

These new cameras are also amazingly complicated, so much so that many people put them on auto and use them like a point and shoot. They are essentially highly specialized computers now, with controls that are much too difficult to figure out on the fly. The menu systems are arcane at best. The software 'helps' by tuning the image to what it thinks it ought to look like, regardless of what the photographer's actual intent is, or the actual conditions. Computational photography is the death of photographic art.

In fact they are so complicated that the bottom has dropped out of the SLR camera market, because cell phone cameras are more than good enough for almost everyone. As well, they offer a much simpler interface (for photography, don't get me started on other aspects of them), and ton of other functions, not least being able to send a photo around the world in an instant. None of my stand alone cameras can do that. That any of my real cameras can produce a better image than almost any cellphone is completely beside the point for most people. They don't need an image that good.

Cars. My first car was a 66 Falcon, bought used of course. I was a kid when it first came out. Technically it was a compact car, in that day and age. (It was much the same size as the 95 minivan we owned, and much less practical.) Technically, what it and it's ilk are called now is deathtrap. I think it was manufactured with seatbelts, but they had been removed. No airbags, no padded dash, no collapsible steering wheel, no frame crumple zones, no ABS, no traction control, bias ply tires, (yes really, though younger readers might need to google them, this was just as radial tires were being marketed.) essentially no safety features at all, let alone other amenities people expect now in a car, like a zillion cupholders, back up and lane cameras, cruise control, the whole 'infotainment system' (a decadence topic in itself!) heated seats and steering wheel, remote start, the list goes on and on. I had to install a radio with a cassette player.

Are cars now safer? Yes, unquestionably. But driving is still one of the most dangerous things a Canadian can be doing. People drive more dangerously, relying on the safety features. There are far more distractions now.

All that is before thinking about the various versions of self driving cars or driver assist features. I would like to believe that a computer controlled car is safer than a human controlled car, but I don't know what the current state of the art is. My understanding is that in a controlled environment, the computer driven car is much safer than a human because it's never distracted. The problem is that our roads are almost the definition of an uncontrolled environment. Almost anything can happen, and at least some of it appears to be beyond the capabilities of at least some humans.

Some people get all excited by the so called trolley car problem, and think that until that's solved we can't have self driving cars. BAH! In a century of driving in North America, I'd be surprised if a dozen drivers have had to face the trolley car problem. No, I want to see the self driving car deal with all the variations of road construction, emergency vehicles (moving and taking up part of the road dealing with an idiot driver), slow moving vehicles such as bicycles or farm equipment, sudden weather and road condition changes, and other drivers doing unexpected things.

So when was the sweet spot for cars? I'm pretty sure I lived and drove through it. Of course, everybody has a different idea of what a car should be. To various people, vehicles as disparate as the 60's muscle cars and the 90's minivans could be thought of as the sweet spot. But I'm thinking recent cars have fallen into decadence. They have gone so far beyond providing a safe and efficient means of transportation that it's actually detracting from the primary purpose of the car.

Here's a photo of a DC-3 for you. Unfortunately it wasn't open to the public the day we visited.