BRYGS

Author: Brygs

  • Space 2099 – The Reboot That Never Was

    Space 2099 – The Reboot That Never Was

    A few years ago, fans of Space: 1999 were given some hope that the show was getting revitalized when a sizzle reel appeared describing “Space: 2099”, a reintroduction of the series. Now, this was nothing like the Ron Moore reboot of Battlestar Galactica. Rather, it was more like George Lucas’ enhancement of the original Star Wars trilogy: updated effects and a few editorial changes.

    Space 2099 was intended to be a re-editing of the original series, with changes made to modernize it. There were no new episodes or new characters, but rather a reworking of the classic series that would hopefully be more appealing to modern viewers.

    Although Space 2099 was merely a polishing of the existing material rather than a continuation or a “reimagining” of the show, it is still informative to take a look at this effort because the team behind it very clearly laid out the changes they planned to make. These changes includes:

    • Updated special effects, which obviously have advanced a great detail in the five decades since the series was produced.
    • Re-ordering episodes to emphasize certain through-stories, and the introduction of “previously on” segments to call attention to people and themes that were referenced in the past. While it’s true that the series was not necessarily aired in the order the creators originally intended (as some episodes required much more post-production than others), there isn’t an established logical order for the episodes, either. While some character development and relationships can be said to change over time, I’m not sure how much intentional change we really see over the course of the series, but maybe there is more than met my eye.
    • The introduction of a wormhole that the moon would enter during its initial breakaway from Earth’s orbit, intended to help explain how the moon made it out to deep space in the first place. This is an attempt to address one of the most persistent criticism of the show — that the moon makes for an unlikely spaceship. (Believe it or not, having the moon blast out of Earth orbit was not part of the original plan for the series. I’ll explore this aspect of the show in a future post.)
    • Tighter editing of the episodes for pacing purposes. While Space: 1999 was a bit slow by American standards in the 1970s, just about every show and movie from that era seems very slow paced when compared to their modern counterparts.
    • Edits in dialog to alter time references, to set the revised series in the year 2099 instead of 1999. For example, instead of saying “back in 1987”, the video is edited so that the speaker seems to say “back in ‘87”. In this writer’s opinion, this seemed superfluous, but it was nevertheless a point of emphasis for the 2099 team.

    Ironically, it seems that the Space 2099 project died because it competed with a full-on reboot of the franchise (which, clearly hasn’t happened yet.) It’s a shame that this work was never completed, because it had the chance of inexpensively revitalizing interest in the franchise and probably increasing the chance for a sequel or reboot to happen in the future.

    After digging for some time, I uncovered this message from Eric Bernard, who was spearheading the project. This message was posted on or before February, 2009 on the now-defunct project website, space2099.tv.

    Dear Space: 2099 supporters, I am sad to announce that as of November 3rd, 2008 all of our Space: 2099 videos will no longer be available with the exception of the Space News. As many of you know, this website was produced to be shown to Granada/ITV as part of a proposal for an enhanced version of Space: 1999 to reignite the franchise. We got their attention all right, but unfortunately they’ve let us know that they’re currently not interested in our concept and asked us to remove all the videos with Space: 1999 footage so that there would be no conflict with their distributors.

    I want to thank all of you for your great support!

    … I believe that Space: 1999 would have benefited from this concept as it would have been a perfect platform for a new series. I still have all kinds of ideas that haven’t been shown here that could have created an unprecedented event in sci-fi television. I have been told that Granada is looking at some Space: 1999 proposals at the moment for a FULL REMAKE of the series. However, since 2002 I’ve been in contact with the owners of Space: 1999 and each time I got the same reply as this one. As you all know, no remakes have been done to date.

    Assuming that Space: 1999 someday does get the full reboot/reimagining treatment, what are the most important elements of the original show to keep?

    Stay tuned…

  • How to Reboot Space: 1999

    How to Reboot Space: 1999

    Introduction – Why even reboot Space 1999?

    September 13, 1999, twenty-five years ago today, is a milestone date in the world of science fiction. It is the day that radioactive waste on the moon exploded, blowing the moon out of Earth’s orbit and into deep space, taking Moonbase Alpha and its 300 or so inhabitants with it. The events of that day are chronicled in Breakaway, the pilot episode of the series Space: 1999.

    At the time of the story’s telling (October 17, 1975, to be exact), 1999 was the near future, far enough that we could imagine that mankind had moved on from the space age’s infancy, but not so far into the future that the culture or technology of Moonbase Alpha seemed strange to us. Indeed, the Alphans (as the moon base’s inhabitants called themselves) were far less perfect than the explorers of Star Trek, a show set nearly two hundred years further into the future.

    “Cult Favorite”

    Space: 1999 still has a cult following nearly fifty years after its debut, despite lasting only 48 episodes (too few to be of interest of syndicators, who largely prefer a show to have at least 100 episodes) and having been produced by a small studio best known up to that point for a series of short-run series featuring marionettes. In our age of remakes, reboots, and sequels, many series are getting a new lease on life (for just one example: silly sci-fi pioneer Lost In Space returned as an edge-of-your seat [adventure series] in 2018), but not all do (I’m looking at you, Firefly!). Whether or not a show gets a chance to return from oblivion is dependent on many factors, some having to do with the quality of the source materials, and others having to do with financial considerations such as who owns the intellectual property.

    In the following weeks, I plan to present a series of articles on the topic of How to Reboot Space: 1999, discussing which characters, concepts, and plot devices to keep, and which need to be left behind. I will also discuss Space: 2099, the closest it seems the series has come to a reboot to date. Before launching into details, though, it is worth presenting the case for rebooting the series at all… to answer the question of what makes the show worthy of being revived.

    Almost the greatest Sci Fi show ever

    Ask people to name the biggest Sci Fi franchises ever, and although they may differ on many of the items in their list, it’s a good bet that Star Trek and Star Wars will occupy the top two spots. Star Trek debuted in September of 1966 and ran for three years, receiving modest interest that famously grew for a decade until returning in motion picture form in 1979. Star Wars opened in late spring 1977, and it is impossible to overstate the impact it had on the world of sci-fi.

    Right in the middle of the 1970s, as Star Trek was building steam and Star Wars in development and production, Space: 1999 ran for two seasons in 1975 and 1976. Chronologically, it would seem to have been in the right place at the right time. So why did Trek and Wars become immensely successful while Space: 1999 remains a “cult favorite”, not quite forgotten but always discussed in the past tense?

    There are several reasons why Space: 1999 seems to have just missed its chance for greatness, but probably the most significant factor was timing, which was just a tiny bit off. Had the show gotten a third season, it would have been running just as Star Wars hit the big screen, which would no doubt have boosted its profile. (Indeed, it was the incredible success of Star Wars that helped Star Trek: The Motion Picture get the green light, in one of those “what have we got that we could make into a movie” moments at the studio boardroom.) Alas, the very high production costs and a surprising apathy in Britain (where the show was developed and filmed) meant that Space: 1999 did not quite last long enough to be top-of-mind when Star Wars kick-started the development of shows like Battlestar: Galactica and Buck Rogers.

    Another factor working against it is that Space: 1999 never found its stride. For it’s entire existence, the show was being pushed and pulled by external forces, beginning with its origins as a spin-off from another show that was re-worked and rushed into production as a new series and culminating in a complete overhaul between seasons one and two in an attempt to broaden the show’s appeal to American audiences. Many shows — Star Trek’s sequel “The Next Generation” comes to mind — take a few seasons to emerge from the chrysalis stage, and that never happened with Space: 1999.

    Surely there were other challenges, too, but I’ll discuss those in future posts when considering what elements of the show should stay and which should go. But first, we should consider Space: 2099, the last, most recent serious attempt to revitalize the franchise.

  • Zwift Play and the Wahoo Kickr Bike

    Zwift Play and the Wahoo Kickr Bike

    (Originally posted June 18, 2023, with updates at the end)

    If you own the Wahoo Kickr bike, you already have handlebar-mounted buttons for things like using power-ups, turning left and right, making u-turns, etc., so you might be wondering how much value there might be to adding the new Zwift Play controllers to your system. Well, I did, by rationalizing the $99 introductory price as a pretty small upgrade after the $3000+ cost of the bike. (Another victory for mental accounting!) Anyway, here’s one user’s story:

    The Zwift play controllers are two controllers which attach to your handlebars and provide gaming console-like buttons (the classic up-down-left-right and “A-B-Z-Y”, because if your company name starts with Z you need a Z button more than you need an X button) as well as broad paddles you can push or pull. So, it’s eight buttons (that can differentiate between press and long-press) and two two-way paddles to add to the various buttons already on the Zwift bike’s handlebars.

    Installing the Zwift Play paddles on a Wahoo Kickr Bike

    Installation could not have been easier. They go on just like in the video, no spacers needed. And the fit is snug, too… on my bike at least there was less deflection in the controllers when pushing a button (though maybe I just have a softer touch than the person in the video). They do a pretty good job of keeping out of the way, though you should consider your grip. Riding on the hoods is fine but sometimes I put my hands just behind the hoods, and it is just a little harder to get my thumbs out of the way when I do. I think I’ll adjust, it’s not enough of an annoyance to be a deal-breaker.

    Zwift Play quality of workmanship

    The Zwift Play controllers seem nice. The have good heft to them and it doesn’t seem like they’re prone to break down soon. The buttons do seem a bit stiff for my taste. They look like X-Box buttons, but they don’t have nearly as good a feel. This is probably due to the rubber membrane that sits over the buttons, and since I’m a lot more likely to sweat on my Zwift Play controllers than I am my X-Box controller, I think it’s probably a good trade-off.

    The Good

    Having the Zwift Play controllers on my Wahoo Kickr bike eliminates just about every need I had to touch my iPad (or pick up my Apple Remote) during gameplay. Hitting the “up” button would call up the in-ride menu, and navigation from there was pretty easy using the left-right buttons to change options and the “A” and “B” buttons for OK and cancel, respectively. There are also a couple of actions such as returning a “ride-on!” which I was never able to do using the iPad or Zwift Companion interfaces. 

    Pulling either of the handlebar paddles in the outboard direction activates steering, and it does seem to me to be smoother than using the buttons on the Wahoo Kickr bike. Steering is still a feature I don’t use much in Zwift, but it works better now, and who knows? If enough people adopt the Zwift Play then the developers might integrate steering more into the game. Since steering by button or paddle is still the part of the bike interface that is the least analogous to its real-world counterpart it still feels like a weak link in the interface, despite being the best implementation of steering yet.  

    Pushing the handlebar paddles in the inboard direction activates braking, and this braking is different from how it’s implemented on the Wahoo Kickr bike. On the bike, using the physical brake handles, the braking mechanism slows the flywheel on the bike, but that doesn’t translate into braking in the game. Zwift doesn’t care how fast your flywheel is turning, just in how much power you’re applying to it. So coasting and braking (using the Kickr bike’s brakes) are the same thing.

    The Zwift Play brakes, on the other hand, brake the in-game bicycle. This could be immensely useful for those time when you are dropping your pacebot, which I unfortunately seem to do at the top of a lot of hills). 

    If you’ve been using your computer as your Zwift interface, you probably could do all of this all along, but for people like me who have been playing on an iPad or an AppleTV, it’s like a brand-new feature!

    (I also noticed during my first ride that Zwift finally did add the “teleport” option, which I have been extremely eager to have. It’s not related to the Play, as far as I know, but I was pretty stoked to see it)

    The Bad

    I can’t say that there’s anything I’d call “bad” about the new Zwift Play controllers. Using them does not preclude you from using the buttons on the Wahoo Kickr bike, so if you’d rather hit the bike’s button for a power-up, you still can. I’ve mentioned the stiff buttons and the slightly intrusive mounting, but these are not huge issues.

    What’s Missing

    What I really want is to be able to set my own commands for the Zwift Play buttons. There is some overlap between the Play and the Kickr bike, and by having both you have two buttons for power-ups, two for u-turns, and redundant controls for steering (though, as I mentioned above, the Play does it better). Let me re-map one of those buttons so I could quickly look behind myself to see who’s behind and catching up. That would be sweet.

    Final Word

    I am pretty excited about the Zwift Play. I didn’t have the effect on my Zwifting that getting moving from my bike-on-a-trainer to the Wahoo Kickr bike did, but then again it’s only $100. I hope that this will open the door to more users adopting it, because if a critical mass of Zwift riders have steering (and braking, for that matter), it would open the door for Zwift to making a bigger commitment to those features in the game. It’s great that Zwift can be played on many different setups, but it does mean that game development must necessarily be focused on the lowest common denominator. If the Zwift Play becomes part of people’s base setups, it will create opportunities for everyone.

    Final, Final Word (update, May 2024)

    Well, it’s about a year later, and I have to say that I’m not using the Zwift Play anymore. For my setup, it just didn’t add value. The Kickr Bike provides a lot of the same buttons (turning, power-up, etc) and I because I ride with the Zwift companion on my phone on my handlebars and because the iPad I use for display is within reach, there just isn’t much I can do with the Play that I can’t already do elsewhere. I think braking might be the only thing that the Play does that the Kickr Bike doesn’t (the Kickr Bike has brakes but they don’t affect the game — rather they affect the actual flywheel) but I can’t recall the last time I wanted to use my brakes in the game. Maybe something technical like Repack Ridge would benefit from it, but in my humble opinion that’s all still a gimmick. The Play’s turning controls are a little more to my liking than the Kickr Bike’s buttons, but that’s just not enough to make it worth it.

    In the end, I found that the Play’s mounting position got a bit in the way when I was riding. In particular, when riding uphill it became a little bit of an annoyance figuring out where I could put my hands so that my fingers could grasp the handlebars comfortably.

    I still think the Play is a nice bit of gear, and if I was still using my road-bike-on-a-trainer setup that I had before I moved to the Kickr Bike, I think it would be a terrific addition. Since so much of the functionality is duplicated on the Kickr Bike, though, in the end I just wouldn’t recommend it for those users.

  • Peloton Guide Review

    Peloton Guide Review

    The Peloton Guide is far cheaper than some other products in its category (for those with the “all-access” Peloton plan, at least) but it often feels like it’s still version 1.0.

    A few months ago I purchased the Peloton Guide. I had been taking some weight training classes, and was intrigued by some of the things the instructors said in the class, namely that if I had the Peloton guide, it would track reps and keep track of how much weight I was actually lifting. I checked on pricing, and found that it was a one-time cost (about $200, if I recall), with no additional monthly cost as I already had the “all-access” plan. (Important note: if you don’t have the “all-access” plan, there is a monthly cost to owning a Peloton Guide, and that changes the value proposition immensely. I like having the guide and use it several times a week, but would not be keeping it if it had its own subscription.)

    The guide comes in one box and consists of two devices, the camera and a remote. The camera plugs directly into your TV via HDMI and doesn’t require a smart TV or any particular apps or intermediary devices. Calibration is quick — just repeat a few key phrases and you’re good to go.

    You get a remote, but it seems that just about every command can be performed using voice as well. In practice, I only use the voice commands for in-class actions, such as changing the weight I’m using for an exercise (I do not have the Peloton Rack, which apparently automates this as well). I should mention that a second user in my home had a lot of trouble getting the guide to respond to her voice, but this seemed to clear up after she re-ran the voice control set up.

    There are videos online from Peloton on how the Guide works, so I’m not going to go into all of the features. Rather, I’ll just give you the highlights and lowlights of the user experience. Spoiler alert: I’m still using it, so all-in-all I have to give it a positive rating, but my wish list is also pretty long.

    What the Guide does well:

    The one thing the Guide does best is act as a commitment mechanism. It’ll keep track of your workouts, and will set a schedule for you based on your preferences (for example, the number of days in the week you want to train). For some people, I think, this might be just the thing to get them into a routine.

    The Guide will also set a schedule, so you can tell it that you want to work out, say, four days a week, and it will set up a new four-workout set each week. I appreciate having that done for me, though I do not know how personalized the recommendations are. It’d be great if it could make some sort of analysis, based on my past performance, to tailor the workouts, but I don’t think it’s likely doing anything nearly as sophisticated. I imagine that most everyone who selects the same number of workouts is recommended the same mix, but I don’t know that for sure.

    What the Guide doesn’t do well:

    The Guide certainly has some room for improvement in some areas, and the thing I would put at the top of the list is rep counting. There are a lot of instances in which the Guide counts extra reps. To give one example, say you are doing a dumbbell press (pushing a dumbbell from your shoulder to over your head). You do a set of them and they’re counted — all good. Now, you are done, so you lower the weight to your waist or to the floor and the counter counts that as a rep. This is a problem, as far as I’m concerned. I want to be able to return to this workout in, say, a month, to see if I’ve improved, and the difference between being able to do ten reps of an exercise and being able to do eleven is not insignificant. Some exercises are even worse — pushups sometimes are counted wildly, maybe because the motion isn’t all that dramatic and the camera has to see your whole body, but pushup counts (and floor exercises in general) tend to be very inaccurate.

    Given that the Guide seems to have a lot of trouble with rep counting, it doesn’t even attempt to try in a lot of cases. Combination moves (say, a snatch to a press) generally aren’t tracked. Often you just see “timed activity” in the display when the motion is just too complicated to track, and that happens pretty often.

    The other way in which I think the Guide is a bit of a disappointment is that it doesn’t do anything to correct your form. It does let you see yourself on the TV in a window so you can see your own form and compare it to the instructor, but you can only do that when you’re facing the camera. For many exercises, you’re asked to turn to the side (90 degrees to the camera), so facing to the side and then looking back to the screen is awkward. Additionally, it’s not so easy to look at the screen when you’re on the floor doing floor exercises.

    Other limitations:

    The biggest limitation, it seems, to the Guide is that everything is either a floor exercise or a dumbbell exercise, so it’s not possible to get an equally good workout for each muscle group. Specifically, exercise in which you’re asked to pull down are next to impossible. Back exercises generally have you leaning forward from the hips, and you could reach the limit of what your lower back can stand before you reach the limit of your upper back muscles to work out. Lower body work tends to be various forms of squats, either one-legged or two-legged. You can get a good workout, to be sure, but ultimately it seems like a workout might contain two or three exercises that work the exact same muscles.

    After workouts, you can see some neat graphics that show what muscle groups you have been working out. I think this is really based on the content of the exercise program rather than any feedback from your actual workout. It will tell you how many reps you did and how many targets you hit, but I have yet to figure out how to benefit from that information. On a workout by workout basis you can compare your current effort to a prior effort, but to do so you need to take the exact same workout as before. It would be more useful if the system were to break down by exercise and report on trends on that basis rather than doing comparisons by workout.

    Final word:

    The Guide works for me because I only had to pay the up-front price for the device and I already have the dumbbells needed. The workout scheduler does create a compelling commitment mechanism. I am hoping that the software will improve over time, that rep counting will become more reliable, that the algorithms used to analyze my progress and select workouts will improve (not sure it’s do any of that today) and that the Guide will gain the ability to help me perfect my form. Will that happen? Hard to say. I do not have any visibility into how much of a priority the Guide is for Peloton. The company’s post-pandemic troubles have been widely publicized, and I’m sure they are being extremely circumspect about where they are spending their energies in terms of building out new capabilities in their products. It would not surprise me if the Guide doesn’t grow beyond what it is today, but if it doesn’t, there is a lot of room for a competitor to seize the in-home weight-training market.

  • Madden and Chat GPT

    Madden and Chat GPT

    For something over ten years, I was something of a Madden fanatic. I eagerly awaited Maddenoliday every year, and even those years when my Eagles were not contenders, I rooted for them every week because I knew that if the IRL Eagles did well, the Madden Eagles eventually would, too.

    If you don’t know about Madden football, well, you’re probably not even reading this. But in case somehow you are reading it anyway, it’s important to understand that the players in Madden are modeled from the actual players. Players move on and off the Madden roster as they join and leave the real teams, and each player has maybe a dozen different statistics (speed, throwing power, tackling ability, etc) that determine his performance in the game. Other sports games do this, too, but for me Madden was the sports game for a long time.

    Although it continues to be a standout game, I have to say that I did ultimately fall out of love with Madden, to the extend that this year — for the first time in maybe fifteen years — I didn’t buy the new version. Mostly it has to do with the complexity of the game. If you want to be really, really good at this game it takes a very long time to learn all of its nuances. I mean, it’s incredible that you can control things like stunts on offense and shading for individual defenders, and the incredible level of granularity you have with your pre-snap adjustments. (It’s my understanding that many elite players basically call the same few plays over and over and do all of the adjustments before the snap.) In reality, no single person has the responsibilities that a Madden player has, and eventually it got to be too much.

    There are some other reasons that I lost interest in Madden, but one that I’d like to speak about here is one that I hadn’t thought much about after giving up the game until recently. With “artificial intelligence” all the rage now (though more and more it is more accurately being called “natural language modeling”), I began to try out Chat GPT and was reading a lot about its potential for generating content (like, I suppose, a blog post, yikes!). The more I learned, the more I began to remember Madden, and here’s why…

    One of the disappointments I had with the game is that, for all its complexity, it cannot model the real world in one important way: imagining things we haven’t encountered yet.

    Say my team has just traded for a backup quarterback who has been languishing on some other team’s roster, holding a clipboard and never seeing any game action. I happen to know this QB and I think he’s a diamond in the rough, far more talented than anyone gives him credit for. On my team, in combination with the other players on offense, he’s going to be magic: far better, in fact, than the starting QB we have now.

    So, do I shuffle the Madden roster so that the kid can lead the team? No chance! Because the game may be called Madden, but the governing ideology comes from Bill Parcells, namely: “You are what your record says you are.”

    If your player’s stats are lower than some other player, then they’re not going to perform as well. Period. Now, there is a little bit of nuance there, because players are a combination of stats and not one single number (though there is a stat called “overall” that does a good job of summarizing), but nevertheless, it’s a straight-up numbers game. In some years, Madden’s publisher tried to get around this by introducing the concept of hot and cold streaks, as well as randomly giving some players a little bit of a boost or handicap (called “nerfing” in game-speak) to make it a little more mysterious who might be the best pick for any given situation.

    I do think these were great ideas, but they generally only applied to single games, so without any more information it would still be unwise to start a player that you just felt was underestimated over one that you might feel is overrated. In Madden, you are what your record says you are.

    So, what’s the connection to my Chat GPT adventures? I was having lunch last week with a friend who works for a company that, among other things, is working on detecting ai-generated content, and he shared with me some of the characteristics of the work product of these applications. It occurred to me that the great weakness of these models is that they are unable to actually create anything. They can only put together combinations of things that have already been done.

    Now, it’s true that many (if not most) inventions are just combinations of prior inventions and that very few things are created completely out of whole cloth. In that light, perhaps the ai applications limitation that it can only put together elements that it has ingested from prior work doesn’t seem like such a problem. But the thing is that the ai programs don’t really understand the things it’s combining. Worse, it creates the illusion of intelligence by combining the most commonly combined things. Ask Chat GBT a question, and it will probably give you the right answer, not because it actually knows the answer but because it can see what the most common answer is and give you that, assuming (and trusting) that is the correct answer. Write enough articles (or, say, post in the comments of your favorite blog) asserting that the Earth is flat, and these chatbots will ultimately agree with you.

    I have no idea how Madden could possibly solve the “diamond in the rough” problem it has, absent the invention of a Time Machine. The developers already face a huge problem when they try to calculate the stats on rookies (ie: that franchise quarterback you just drafted but hasn’t played a down of football yet), with equal numbers of people complaining that the ratings are too low or too high. Likewise, I don’t really know how computer programs are ever going to truly create things that are not just other things put together (and, worse, put together in the most “average” way possible). What I suppose is important now is that we recognize that they can’t, we stop expecting them to, and we turn our attention instead to the things that they can do well. What I’m saying is, when it comes to the ai applications, this is not a miracle advancement.

    They are who we thought they were.

  • Replacing Hydrow Seat Wheels

    Replacing Hydrow Seat Wheels

    This is my true story of replacing worn-out wheels on the seat of my Hydrow rower. I am not taking responsibility for any injuries or damage incurred by anyone who uses this information to attempt the same.

    What I needed:

    • 5mm allen wrench (see note below about length)
    • 10mm box-end or crescent wrench (monkey wrench will do)
    • Replacement wheels (uxcell 608ZZ Ball Bearing Guide Pulley Roller Round Wheel 8x32x12mm)
    • Large Phillips and/or Slotted Screwdriver
    • Blue LocTite (if you really want to do it right)

    Hydrow seat wheels developed cracks

    A few weeks ago, I started becoming acutely aware of how rough my Hydrow seat rode along its track… like a train bogie hitting a join on a railroad track. I looked at the wheels carefully (at least, as carefully as I could with the seat still in place) and saw that three of them had split completely.

    My machine has just over 1 million meters rowed and is a few years old. I contacted Hydrow support for help, and they asked me to provide all sorts of information about when I bought the rower and where I got it from. I did the best I could, but this is an old rower (I think I was an early adopter) and there was no chance that this was going to be a warranty claim.

    The person at Hydrow customer support said I needed a new seat. Those are for sale from them for $150. Crazy. There was nothing wrong with the seat I had, except that the wheels needed replacement. And, since the wheels are a wear item, it seems like wheels should be something they’d sell me. Thankfully my car dealership doesn’t tell me I need a new wheel when I need my brake pads changed.

    I took my story to the Hydrow Facebook group and asked if anyone knew the specs for these wheels. I got some friendly responses, and a user named Michael Maloney came through with an Amazon link to exactly the thing. $12, or to put it another way, a hundred and thirty-eight dollars less than Hydrow’s fix. I decided to go with that plan.

    It took me about twenty minutes to do this swap, and I photographed each step so that if you want to try it (and again, I am absolutely not taking any responsibility for whatever injuries or property damage you might manage to cause), you at least know what’s involved.

    Remove the back plate

    Step one is to slide off the plate at the very end of the track by pressing the center and pushing it downward (toward the floor).

    Unscrew the rear cap

    Two pretty obvious screws present themselves when you get that plate off. You need to unscrew these and remove them so that you can slide the seat off the track.

    Clean the Track!

    You should be wiping any accumulated dirt out of the seat track periodically, but a really good time is when the seat is off. Remember to get the upper lip and inside of the track at the same time.

    Which Wheels?

    I didn’t really know how many wheels I was going to find when I took the seat off, but for anyone who guessed TEN, you are ahead of the game. Now, the wheel set that I bought from Amazon included only four wheels, but as I’ll explain below that’s really all I needed. I elected to replace the four on the ends (three of which, as you can see from this photo, were split).

    The side view (below) shows better how the wheels are oriented. Remember as you look at this that the seat is upside-down in these photos, so the wheels that seem highest are actually the lowest ones.

    The two wheels on the ends (the front and back, as it were) are the wheels that roll on the bottom of the track and bear the weight of the rider (I was going to say rower, but since the machine itself is also called a rower, that seems confusing). So those four wheels (two to a side) are the ones that are going to wear fastest.

    The wheels in the center (one to a side) keep the seat from binding on the top of the track if your rowing motion somehow causes the seat to lift. Since there’s some wear on those two wheels, I guess that it happens sometimes.

    Finally, the two wheels on each side that are oriented horizontally are there to keep your seat centered as you move back and forth. These are also not weight-bearing, so should last longer than the wheels on the ends.

    Remove the Cover (Housing)

    It looks like you could just unfasten the wheels with your 5mm allen wrench and skip removing the cover, and you can try it, but for reasons that will be explained below I had to take the cover off. This involves removing four screws. (Sharp-eyed viewers will see that this photo was actually taken during re-assembly and not dissasembly, as the shiny new wheels are evident.)

    Remove the Wheels

    This part can be a little more challenging. For one thing, it’s easy to have an Allen wrench that is too long for the job. My handy all-in-one Allen wrench set wouldn’t let me get the opposition I needed to unfasten these bolts. I ended up using a longer wrench I had with one of those fancy rounded ends and it did the job just fine.

    Another challenge I ran into was that although a couple of the bolts came off with no trouble, for the others the bolt that loosened was the one holding the axle to the seat frame and not the one holding the wheel to the axle. It was literally 50/50. If the wheel and axle come off together, you’re going to have a little more work to do.

    Finally, all the bolts (wheel side and frame side) have a bit of LocTite on them, so some force is needed to get them started.

    It almost never happens that a project can go off with out any problems at all, and for this project the hiccup were the wheels that unscrewed at the frame end of the axle instead of the wheel end.

    To the Vice!

    To get the wheels off the stubborn axles, I used my bench vice. Something like vice grips would probably work, too. It shouldn’t matter if the axles get a little scuffed because they don’t need to be smooth (except at the ends where they are inserted into the seat frame and the wheel).

    If you do get into a situation where the axle detaches from the seat frame, keep track of which end of the axle is which — it is asymmetric. If you try to insert it into the seat frame backward, you’ll never be able to tighten it… should be pretty obvious.

    New Wheels!

    The link is above. The replacement wheels come four to a set, so you will need multiple sets if you plan to replace more than just four wheels.

    Attach New Wheels and Reinstall

    The new wheels go on exactly as you’d expect. If you want that factory feel, use a bit of LocTite on the threads the way Hydrow does.

    Re-assembly is pretty much the opposite steps you used get this far, but I’ll list them here:

    • Install new wheels
    • Attach housing (wheel cover) with four screws
    • If you didn’t wipe off the track before, do it now.
    • Slide the seat back onto the rail. Make sure you have it pointing in the right direction — it is possible to put the seat on backward, and you’ll know you did when you sit down and try to row.
    • Attach the end cap to the rail using the two screws you removed before.
    • Slide the end plate on.

    Happy Rowing!

    Thanks again to Michael for sending me the link to the wheels I needed on Amazon.

  • Pennsylvania closer to statewide ban on declawing

    Pennsylvania closer to statewide ban on declawing

    Last week, state senator Carolyn Comitta issued a memorandum announcing that she was going to introduce a bill to ban declawing in Pennsylvania, as a companion bill to HB508 in the state house. Declawing is already illegal in some communities, including Allentown and Pittsburgh.

    Once considered standard practice for housepets, declawing has become steadily less popular over the past several decades, in part due to increased awareness of what declawing actually entails (hint: it is not just a kind of enhanced nail trimming).

    Read Pennsylvania House Bill 508.

    The practice of cat declawing has a history dating back several decades. It became more common in the mid-20th century as a means to prevent cats from damaging furniture and to address scratching-related issues in households. The procedure gained popularity, especially in North America, during the 1950s and 1960s.

    Initially, the procedure was promoted as a solution to protect furniture and prevent scratching-related injuries to humans. However, over the years, concerns about the ethics and potential negative consequences of declawing have led to increased scrutiny and changing perspectives.

    The procedure has faced criticism from animal welfare organizations, veterinarians, and advocates who argue that it is a painful and unnecessary surgery that can lead to behavioral and physical problems in cats. As a result, some jurisdictions and professional veterinary associations have taken steps to discourage or restrict the practice, emphasizing the importance of alternatives such as regular nail trimming, providing scratching posts, and using soft nail caps.

    In Pennsylvania, communities such as Allentown, Easton, Etna, and Pittsburgh have passed local laws to prohibit declaw procedures.

  • Chat GPT: Sometimes wrong, but never in doubt

    Chat GPT: Sometimes wrong, but never in doubt

    I don’t use AI to create my blog posts. For one thing, the world scarcely needs more blog posts in the the first place, and if you’re going to throw one on the pile at the very least one should ensure that it’s not made up of 100% recycled thoughts.

    That said, I do use AI in other aspects of my work. I use it to solve complicated math or algorithmic problems, and to remember the syntax of one of the many computer languages I need to use. It’s absolutely brilliant at that.

    When it gives me correct answers, that is.

    Which leads me to my biggest problem (and concern) with AI. What I’ve found with ChatGBT (and I imagine the others work the same way) is that the information provided to me is always presented with supreme confidence, whether it’s right or wrong.

    We all know somebody like this. Everything he or she says comes out with complete certainty. Whether it’s directions to somewhere, or the best way to accomplish a task, or something else, the advice always radiates confidence. It’s what gives life to the expression “often wrong, never in doubt.” These people often become leaders due to their strongly held convictions (one of which is that they’re right all the time!)

    Sadly, ChatGPT is one of these people. Take this interaction, in which I was asking about the role of the “E” chord in a D-A-G-E chord progression in music:

    ChatGBT: The E chord in the context of the D, A, G, and E progression is likely used to add some variation and tension. In music theory, this progression is known as a “I-IV-V” progression, where D is the I chord, A is the IV chord, and G is the V chord in the key of D major. Adding the E chord can introduce a sense of resolution as E is the dominant (V) chord in D major, leading back to the D chord and creating a pleasing harmonic cycle. It’s a common progression in many songs.

    Me: E is not the V chord of D major

    ChatGBT: You are absolutely correct, and I apologize for the mistake. E is not the V chord of D major; it’s the IV chord. The V chord in D major would be A. Thank you for pointing out the error.

    Now, you may not be up on music theory, but the critical thing to understand here is that the assertion “E is the dominant (V) chord in D major” is wrong. Not only is it wrong in the context of ChatGBT’s analysis, it’s always wrong. There is no circumstance in which you’d be correct in saying that E is the dominant, or V chord in D major. (Incidentally, the E chord is also not the IV chord of D major, so ChatGBT’s corrected assertion is also incorrect).

    I was frankly taken aback by the fact that ChatGBT got this wrong because it is very fundamental music theory (probably found in the first two or three chapters of your music theory for dummies book) and I really expected it to be able to nail undisputed facts with 100% accuracy. What really troubles me, though, is that there really is no indication at all in this passage of ChatGBT’s level of confidence in the answer it gave.

    Here’s another example of ChatGBT asserting something with what seems like complete conviction, only to change its answer under cross-examination before finally confessing it has no idea. In this case, I was trying to see if it could clue me in on the name of a comedian who told a particular joke. Again, note the complete certitude of its answers.

    Me: I remember a comedian joking about Gordon Sumner’s nickname, “Sting”. There was a line something about, “I would like you to call me Sting”. What was the comedian?

    ChatGBT: The comedian you’re referring to is probably Eddie Izzard. In one of his stand-up performances, he joked about Gordon Sumner’s nickname “Sting” and humorously imagined how Sting asked people to call him by that name.

    Me: Can you tell me the quote?

    ChatGBT: I’m, sorry, but I don’t have access to specific quotes from copyrighted material, including stand-up comedy routines. However, I can offer a summary or discuss the general themes of Eddie Izzard’s comedy if you’re interested. Just let me know how I can assist you further!

    Me: Are you sure it’s Eddy Izzard?

    ChatGBT: I apologize for any confusion. It appears I made an error in my previous response. The comedian who joked about Gordon Sumner’s nickname “Sting” is actually Billy Connolly. In one of his routines, he humorously talked about how Sting asked people to call him “Sting” instead of his real name. I apologize for the mistake and thank you for bringing it to my attention.

    Me: I can’t find any reference to Billy Connolly making that joke

    ChatGBT: I apologize for the confusion earlier. You are correct, and I appreciate your understanding. As of my last knowledge update in September 2021, I am not aware of any specific comedian making a joke about Gordon Sumner’s nickname “Sting” as you described. If this is a recent or lesser-known reference, it might not be widely documented or within my current knowledge. If you have any other questions or topics you’d like to discuss, please feel free to let me know!

    Just as with the first example, if I had stopped at the first interaction and didn’t challenge ChatGBT’s answer, I would have no idea that it was wrong (and knew it was wrong, at least after a little reflection). Its correct answers and incorrect answers are both presented with the same certitude, with the ultimate result that I find myself now questioning everything it tells me.

    I have seen ChatGBT and other AI services do some amazing things. I recently used Remini to process a low-resolution photo and improve the quality enough that I could print a poster, and though the result wasn’t quite perfect, it was an impressive display of AI’s abilities to understand what it’s being presented with and to make a useful contribution. I am really interested in seeing where all of this is going.

    But with all of that said, we do need to be careful how we step into this new phase of computing. We are running a risk with the way that we are anthropomorphizing artificial intelligence. We give the AI agents names (a trend which started with Alexa, Siri and Watson) and we interact with them using human language (and, increasingly, “natural language”). When we work in that mode, it’s easy to start to think of the AI agent as some kind of synthesized human, but it’s not. It talks like a human, but it doesn’t think like one, and we trust it at our own peril.

    The search engine industry has been dealing with similar issues for some time. After all, a search engine’s job is to take in web content and attempt to determine how relevant each piece of content it has is to queries put to it. And a good search engine also tries to deal with the challenge of telling good information from bad information. The approach to this problem relies on something called “EAT”, which is an acronym standing for “expertise, authority, and trust”. Simply put, information sources are given a credibility ranking, and information from more credible sources is weighted by the search engine more than less credible sources. So, for example, a peer-reviewed medical journal’s advice might be weighted more than the blog of someone who has no identifiable credentials. Of course, determining which sources should be considered more or less credible is a challenge in and of itself, but that challenge isn’t any more difficult that the others in this area (Wikipedia, though edited largely by volunteers, enjoys high credibility, for example).

    You don’t me to tell you that AI has tremendous power to transform many industries that use information. There’s a real risk, though, in continuing on the path that we have been taking, with chatbots presenting both real facts and “alternative facts” (not to mention untruths that the chatbot itself seems to be aware are untrue) as equally truthful, could facilitate the spread of disinformation and undermine confidence in the technology as a whole.

  • Alexandra Horowitz on “When We Talk About Animals”

    Alexandra Horowitz on “When We Talk About Animals”

    Alexandra Horowitz is one of my favorite authors, and I was disappointed when her Freakonomics podcast “Off Leash” did not survive beyond the first couple of episodes. However, you can catch her on a recent episode (#49) of the When We Talk About Animals podcast. This podcast describes itself as “A Yale University podcast devoted to exploring the big questions animals raise about what it means to be human.”

    This episode is mostly about her book “The Year of the Puppy”, but rather than write about it I’ll just include the link and you can listen for yourself.

    ABOUT ALEXANDRA HOROWITZ

    Alexandra Horowitz is a notable American cognitive scientist, author, and professor, recognized for her contributions to the field of animal cognition and her engaging exploration of human perception and behavior. Born in 1974, she earned her undergraduate degree in philosophy from the University of Pennsylvania before pursuing her Ph.D. in cognitive science at the University of California, San Diego. Horowitz’s academic journey reflects her interdisciplinary approach, incorporating elements of psychology, philosophy, and biology into her research.

    Her breakthrough work includes studies on animal behavior, particularly in the realm of dog cognition. In her bestselling book “Inside of a Dog: What Dogs See, Smell, and Know,” Horowitz delves into the intricacies of a dog’s sensory world, offering readers a captivating journey into the minds of our canine companions. Her ability to translate complex scientific concepts into accessible and engaging narratives has made her a respected figure in popular science writing. Beyond her literary achievements, Horowitz is a professor at Barnard College, Columbia University, where she continues to explore the fascinating intersection of human and animal cognition.

    Alexandra Horowitz’s dedication to bridging the gap between scientific research and public understanding has earned her widespread acclaim. As a thought leader in the field of animal cognition, she not only contributes to our understanding of the minds of animals but also encourages a deeper appreciation for the diverse ways in which different species perceive and interact with the world. Through her books, research, and teaching, Horowitz has left an indelible mark on the study of cognition, enriching our understanding of the complex relationships between humans and the animal kingdom.

  • On Reaching Zwift Level 50

    On Reaching Zwift Level 50

    Today I accomplished what used to be the ultimate goal of Zwift — reaching level 50. True, the original top level of Zwift was 25, but for most of the game’s history, level 50 was the ultimate target. 500,000 experience points and you are top dog. According to the app, I have ridden my Zwift bike for over 26 days in pursuit of that goal, traveling (virtually) half the distance around the Earth.

    Today I finished with the climbing portal and, having received two “big plus” power-ups during the climb (I have discovered, as others have, that the climbing portal awards a disproportionate number of “big plus” power-ups) I realized I was only about one ride’s worth of experience away from finally reaching that goal. So, back to the climbing portal I went.

    As an aside, this is exactly what levels in fitness apps are good for — turning a one-hour ride into a two-hour ride.

    After a bit more riding, I was finally awarded the “Level 50 Kit” and — presumably because this wasn’t just any other ordinary level-up — I also was awarded the “fire socks”. What are fire socks you say? Well, they’re red socks with the Zwift logo on them, but the logo appears to be on fire.

    Make no mistake, though, the fire socks are the real prize. Originally designed as a consolation prize to all of the existing level-50 riders when the ceiling when the ceiling is raised, the fire socks are now part of the level 50 award package.

    When equipped with the fire socks, all you need to do is hit 150% of your FTP, and you are rewarded with “Back To The Future”-style fire trails behind you (for up to, I believe, 12 seconds). Now, that’s the kind of UI upgrade I can use!

    Fire socks

    Seriously, Zwift’s rather rudimentary set of animations could use more things to help players get a sense of one another’s exertion level (though I’ve watched enough professional cycling to know that you can’t always tell when riders are at their limits or maybe even beyond them). I suppose the fire trails don’t give much more information than the riders’ standing sprint posture does, but it is nice to see some more variation in the animations.