Cosine Thoughts

Why do companies keep trying to use AI/token usage as a performance metric? Because they've lost (or never had) the ability to quantify success. Put another way, it is the exact same reason that vibe coders exist and succeed until they fail. AI is not currently reliable and the people who tend to rely on it most, also let it replace their own judgement and faculties. And they let it get that way because of a few (sometimes big) early wins.  Tokenmaxxing is an obvious result of an obviously flawed system.  How is the system "obviously flawed?" I've written up a number of examples of my struggles with AI at work. But it boils down to this; AI isn't actually "smart" in the human sense and the results are non-deterministic. This means that the 2 developers using identical prompts can wind up with WILDLY different tokens spent for the same problem. AND more than likely, the developer that spent more tokens ends up with the inferior solution.  Identical prompt...

I've already made my arguments against AI on a financial level and on a technical level. Now they've started suggesting that you combine these catastrophes into one big, uber-catastrophe: don't even write prompts any more, just write loops! The hubris! AI is not ready. Period. I'm NOT saying that AI will NEVER be ready. I'm saying that, at the current moment, we're not there. Not even close. If you thought the crazy AWS bills of the past were something, wait until the first engineering teams start taking this advice. I guarantee you that it will be epic. A little part of me actually wants companies to just take this advice and let Claude run without any token caps. Why? Not because I want to see those companies fail, but because I want to see how fast Anthropic can destroy itself when these companies suddenly can't afford the bills which they need them to pay to keep the lights on. It would be short term pain for long term gain in the software industry. A me...

A recent development in the past year or so at our dojo has been an "internal shinsa" to test for whether or not the dojo leadership would allow us to challenge for the next shinsa. I love the idea. My own feedback? I wish I had done worse. I have a problem. I don't like positive feedback. It doesn't tell me much. By no means was the performance perfect, and there were critiques. They were simply stifled by the positives. It deflates the sense of urgency and need for me. I suppose that there is an up side. It forces me to be the negative one. We didn't get to see the recordings, so I can only focus on what I was told and what I normally don't do well. And that too is a blessing in disguise. That performance was singular. Both in the literal sense and in the sense that I was feeling REALLY GOOD that day. I've never been able to perform so "freely". So, in some sense, focusing too much on that performance might not be as helpful. The critiques I di...

No. At least, not yet. Then again, we also should not have scrapped the carbon tax. There are a few key pillars to my reasoning: The government is operating a deficit. And a big one. This helps. The world is currently in an energy crisis, and anything which promotes reduced consumption is a good thing. Prices in Canada are currently much more favorable than most other places. There are other things that the government can do to help. Some of it with that money. The first point is pretty obvious. Higher gas prices means that the government makes more in tax revenue from oil and gas. At a time of economic uncertainty, when the government is already trying to spend its way out an economic crisis this is actually a bit of good luck. And we shouldn't turn around and piss that money away. The second point is also kind of obvious. The straight of Hormuz being closed means that there is less oil and gas to go around. We shouldn't be cutting prices to help us pretend that this isn't...

Sometimes things make a lot more sense in retrospect. The difficulty jumping from a 10kg yumi to a 14kg yumi was one of those things for me. For those whose jaw dropped reading that, I will preface this by saying that the draw weight I chose based on a recommendation by my instructor and that I actually went with a lighter draw weight than was recommended. I can't remember the exact recommendation but it was 15-16kg. I also did use a 15kg once before the purchase and was able to draw it. But I was already tired and couldn't tell how much of the difficulty was due to fatigue vs. the bow itself. It has been over 2 months now. It is still somewhat difficult getting into DAISAN.  It isn't a completely alien feeling any more and it is much more manageable. I can also keep shooting through at least 2 dozen arrows now. Which is to say; progress. I do miss how much easier my 10kg dojo bow felt. I'm actually considering getting a second bow; either a 12kg or a 16kg. The idea is ...

This will be more of a normal journal entry. I think it is worth stating these things, but it doesn't feel very formal or informational. Here are some things I've encountered so far: Humility Not knowing all of the terms for the associated equipment Equipment mistakes (poor tsuruwa construction) Concerns Storage and blind-spots in care/maintenance  Excitement New "toy" Learning new things and overcoming some of the above Knowing I still have even more to learn So, yeah "humility".  I will admit now that I was not properly prepared for this moment. I think that, for next year, I will recommend more equipment maintenance lectures. There is definitely more for me to learn with regards to how to maintain my equipment and even what all of the individual items are. The first mistake was confusing the uchifukuro with a yumibukuro. To be fair, we only use the uchifukuro in the dojo and they both have the same basic form. In fact, both even have VERY similar dimensio...