Author here. AMA.
mad respect to you for being open. thanks for sharing.
Ironically, today I have a part-time consulting gig at a company designing network switches, and I see them wrestling with the exact same problems we were tackling 30 years ago. So FlowNet would be a very close second.
Link?
I'm less convinced. Although to be fair, I'm really not the target market.
I think the root fail of this kind of business is that it's a Venn diagram of two circles, with very tiny overlap.
Firstly, flying commercial is "simple". You want to go from A to B at some point in the future. You want to buy a ticket, show up, and fly.
Sure, I get the TSA crap and so on, but it's fundamentally reliable. Not to mention cheap.
Private travel bypasses TSA, and sure there's glamor, and a bigger seat. But unreliable is built in also.
Or put another way, the jet owner goes wherever and whenever they like. The pay a lot for perfect convenience. But it makes that "empty return leg" unreliable. And of course, the further ahead you plan, the less reliable it is.
So there's one circle of people who can afford their own (whole or partial) jet. There's one circle who just want reliable A to B.
What you're left with is people who have enough to make it a business, but not enough to charter a plane themselves.
So your market are price insensitive enough to be preparedto pay more, but also don't really plan their travel, or don't need to travel reliably. Which just seems like a vanishingly small group.
In other words, what you are selling is worse than commercial (planing and reliability) coupled with worse than private (no control over where or when).
Which leaves such a tiny sliver of a market, where brokers practically outnumber customers.
You'd think that, but it turns out not to be true. There are a lot of constraints even on the owner: weather, mechanical failure, crew members getting sick. And it's even worse for them because when something goes wrong it can be a lot harder to recover. When you're flying commercial, or even chartering, the worst-case scenario is a delay.
The only real benefit to owning a plane is that you can keep some of your stuff on board.
> a tiny sliver of a market, where brokers practically outnumber customers.
It depends. If you measure the market size in terms of customers that's probably true. If you measure it in terms of dollars, it's not. It's a question of: all else being equal, would you rather pay less than more? Most people say yes no matter how fundamentally price-insensitive they are.
Not sure whether you're asking whether that sentiment exists or if it's correct. But either way, there's no easy answer to that. If you don't believe that's not helpful, but on the other hand, if your beliefs don't align with reality that's not helpful either. It's possible to succeed without strong convictions, and it's possible to succeed with convictions that don't align with reality, but if you have to choose one or the other, believing in yourself is a better than believing in reality if your goal is material success. (Look at Donald Trump.)
This is an incredibly common pitfall that people fall into time and again. Hell, even I am getting these vibes with all that recent hype about deep learning - despite I am no stranger to academia (in another area and some years back).
> recent hype about deep learning
Does anyone really look at contemporary AI as a Platonic quest for objective truth? It seems to me that there is pretty widespread clarity about the fact that this is mostly a commercial endeavor. If you want to talk about Platonic quests for objective truth we should talk about, say, mathematics or fundamental physics.
I realize I have a tendency to romanticize some of the papers I read even if I fully understand intellectually it is a wrong thing to do. Probably I should try concsiously train myself not to do it or something along these lines.
My experience tells me if I have a question or an issue it's very unlikely to be unique to myself. And (re-)reading an honest or even cynical account about something typically helps e.g. ribbonfarm on corporate hierarchies/politics.
https://blog.rongarret.info/2025/01/an-open-letter-to-democr...
This is not to say you can't make money at it. People make money on scams all the time. One might argue that it's a foundational element of the American economy. But when it comes to bitcoin, I understand the technology and its attendant risks too well for me to want to play that game.
I agree completely that BTC is basically a scam at this point, as are most of the top coins by market cap.
However, there are a few interesting cryptocurrency projects out there which are quietly getting better and better (ie, Nano). Do you have any interest in the potential of the field, or has BTC soured you on the whole thing?
That's a very hard question to answer because it depends on what you consider "that bathwater." I think using digital signatures is a win. But mining is a BIG lose IMHO. It is vastly better in every possible respect to rely on trusted third-parties. Even with bitcoin most people rely on TTPs because managing your own keys is fraught with all manner of peril.
> However, there are a few interesting cryptocurrency projects out there which are quietly getting better and better (ie, Nano). Do you have any interest in the potential of the field, or has BTC soured you on the whole thing?
I don't know anything about Nano. But the fact of the matter is that BTC is pretty much synonymous with crypto nowadays, and everything else (AFAICT) is riding on its coattails. No one would be taking any of it seriously if not for the fact that 1 BTC fetches USD100k or whatever it is nowadays.
About that:
https://blog.rongarret.info/2023/04/bitcoins-value-propositi...
(I'm the author, BTW.)
> and the fact that you can't easily reduce the amount of energy needed to mine is a feature, not a bug
We'll have to agree to disagree about that. Most people consider gross inefficiency a bug, especially when that inefficiency is a necessary part of the design.