SOLID often comes up against YAGNI (you ain’t gonna need it).

What makes software so great to develop (as opposed to hardware) is that you can (on the small scale) do design after implementation (i.e. refactoring). That lets you decide after seeing how your new bit fits in whether you need an abstraction or not.

I never beat Hollow Knight (yeah yeah, I know), so I’m working my way through that. I think I’m about half way? Obviously hard to tell.

TL;DR: Yes, you subsidize, because the amount each seat pays is proportional to their ticket price, but it’s a small amount per flight. But you also need to think about what you consider “fair” in this instance.

Sorry, been offline for a few days. I’m curious, so let’s check the math:

Their example is a flight from Atlanta to Orlando. The most common flight on that route is a Delta B757-2001. SeatGuru can provide us with the most common seating layout: 24 First class, 21 Comfort+, and 135 Economy.

The tax that the NYT Editorial Board is looking at is mostly likely the 7.5% Passenger Ticket Tax, which is about 1/3 of total fees on a two-leg itinerary2.

Looking at Delta’s fares for about two months out on randomly chosen dates in October, I see economy fares of about $270, economy plus at about $350, and first at about $570 (I tried to take a median, but it’s very approximate). Those fares are round-trip, so let’s cut them in half for $135/175/285. At 7.5%, the tax comes out to ~$10.13/13.13/21.38 per seat. That tallies up to $2156.40 when we multiply out by the number of seats, pretty close to the $2300 value the video claims, so I’m comfortable saying my numbers are “right enough” for the example.

Depending on how one structures the ownership of their private jet, the equivalent tax for them is zero dollars (they are exempt).

There are then taxes that both flights would pay on a per flight basis or based on fuel consumed. A private jet would pay proportionally more of those than a commercial airliner (due to the lower % passenger weight of total weight), but those are a small part of the total fees, especially for the commercial flight. (I’m seeing about $300 per plane, so I think NYT was counting some of those fees but not all, as they said that the private jet would be paying about $60 in fees).

The problem with the whole “subsidizing” conversation is that it depends on what perspective you take. If you look at it on a person-by-person basis, then sure, each passenger on the commercial flight probably pays less than the passengers on the private jet (assuming 2 ppl or something).

But FAA resources aren’t provisioned on a “per-passenger” basis, they’re provisioned on a “per-flight” basis, with some modifiers based on:

1. Origin and destination airspaces: busier ATC spaces require more resources, and the gain for an additional resource is not linear because of handoffs;
2. Size of plane: bigger planes do take up “more room” in the sky, and therefore tax the ATC more than smaller planes, plus they compete for larger runways, where smaller planes can use smaller runways at airports that have them;
3. Distance flown: Planes that fly farther use more ATC resources en-route.

Let’s be generous and say that our B757-200 takes 4x the ATC resources that the private jet does (I would bet the real factor is closer to 1.5-2.5x). So for a total of 5 units of ATC resource, 4 are used by the commercial jet, and 1 is used by the private jet.

The commercial flight therefore pays $(2156.40+60)/4=$554.10 per ATC resource, and the private jet pays $60 per ATC resource. Equal distribution would be $(2156.40 + 60 + 60)/5=$455.28. So the private jet is receiving a “subsidy” of $455.28-60=$395.28 per flight.

If we divide that subsidy over all of the passengers on the flight by fare, then we get about $1.81/2.34/3.81 based on seat class. That isn’t much of a subsidy per passenger, about 1.4% of your ticket price.

But let’s think about the other side of the equation: Chartering a plane from Atlanta to Orlando costs about $12,5003. Taking on an extra $395.28 would be an additional 3.2% per flight, which is admittedly more than the 1.4% of the fare for the commercial passenger.

And hold on… we are talking about passenger transport when we’re talking about both flights… so let’s look at how efficiently those FAA resources are used. Keeping that 4x factor for a flight of the same distance, we have 4 FAA resources spread over a flight with a capacity of 180, let’s assume 80% full for 144 passengers, using about 0.03 FAA resources/person. Now let’s look at the private jet, which recall uses 1 FAA resource for the same flight plan. Let’s be generous and assume 6 passengers. That’s 0.17 FAA resources/person. The commercial jet is more than 5x more efficient in its use of FAA resources.

These are people who literally create complex corporate structures for their private jets just to avoid that 7.5% excise tax, AND they tend to have much more disposable income. I think they can pay the extra $400 for their inefficient use of FAA resources.

If I were making the rules (which is absurd because not only am I not an expert but I am also Canadian), I would make the FAA fees per-itinerary filed with the FAA and incorporate three factors:

1. Category of departing and landing airports4
2. Takeoff weight of aircraft
3. Itinerary flight path length

The FAA has a whole section on their website about airport planning, so I would use that to figure out how to apportion these factors to best approximate the factors required for FAA resource allocation. I’m sure there are planners at the FAA that have this all broken down already.

So yes, unless you get a super-discount fare, you are subsidizing private jets assuming that the fair apportionment of costs is based on how FAA resource capacities are planned. It’s not much per passenger, but it adds up across all of society, and is another way that the US economy moves wealth from the lower classes to the upper class.

While that’s true, and so First Class and Business Class subsidize private jets more than Economy Class does, that doesn’t change the fact that Economy also subsidizes private jets.

As a percentage of the total weight of a plane, passengers and their luggage constitute a much larger percentage of a commercial flight than a private one. So they are “more utilized” than a private jet, and can spread that cost over all their passengers.

Also, larger planes that fly longer distances cross more ATC zones, using up more ATC resources. They also take up more “room” in the sky, as e.g. ATC needs to leave more room for jet wash behind a heavy. So it makes sense from multiple perspectives that bigger planes pay more.

You also have to consider hobby pilots. Charging them the same amount as a 747 would be insane.

So it’s a tradeoff: the Canadian system makes smaller planes pay more, proportionally, than a per-ticket model; but not so much more that it harms the smallest personal planes.

It’s also just simpler. Personal plane? Private jet? Commercial passenger flight? Cargo plane? Same calculation for all of them.

(Yes, you could try to make it “only for flights with paid passengers”, but then pilots of private jets would all of a sudden have a lot of very rich friends with whom they do a lot of personal flying. It’s just so much easier if there’s nothing subjective about it.)

Oh thank goodness. Coffee Stain is by far my current favourite publisher, and I was worried with them getting caught up in all this Embracer stuff.

I don’t frequent that world much these days, but I personally preferred the agent/pull model when I did. I can’t really articulate why, I think I feel comfortable knowing that the agent will run with the last known config on the machine, potentially correcting any misconfiguration even if the central host is down.

The big debate back in the day was Puppet vs. Chef (before Ansible/SaltStack). Puppet was more declarative, Chef more imperative.

I also admit, I don’t like YAML, other than for simple, mostly flat config and serializing.

I further admit that Ansible just has a bigger community these days, and that’s worth something. When I need to do a bit of CM these days, I use Ansible.

So wait… Did Puppet have a license change as well recently? Is this just preemptive because it looks like Perforce is starting to change things?

The Free Software Foundation requires “CLAs” as well. I have no fear that they’re going to rug-pull. I don’t think we can use that as the indicator. IMO, it’s even a good idea to have a CLA so that’s no conflict that the project owns the code.

The warning for me is if the project is run by a company, especially a VC-backed company. Joplin isn’t, so I would be comfortable using it (although I don’t).

JIRA Data Center: What am I? Chopped liver‽

https://www.atlassian.com/enterprise/data-center/jira

Agreed that JIRA is… not the greatest tool.

Ahh, the comment I was looking for

I would have also accepted: “Haskell did it first.”

X2go is the successor to NX and works well IMO, though I’ve never tried Rustdesk to compare.

This is interesting because I’ve been thinking about switching from Debian to Arch. I’m already running Nix inside of my Debian installation to get more recent apps (I don’t like how snap interacts with the rest of the system, so I avoid it if I can).

Is there anything else on a more base OS level (like apt v pacman) that you’ve noticed is different, if you’re willing to share?

I tried to switch to Tidal, but I found their app not as good, their integration with Sonos lacking, and no parental controls, which is important to me. Music selection was pretty good. A lot of niche stuff isn’t there, sadly. For example I sometimes listen to college acapella groups, and there just isn’t as much there. All the popular music is there though.

I mean, you got my upvote already, but one big reason is that Robertson wanted to control all the manufacturing of the screws and the bits. Phillips licensed his patent out and let anyone make them just taking a tiny licensing fee. Made a fortune on volume. Robertson: good engineer, bad businessman.

I don’t fully agree with these, but these are the cases I’ve heard of:

• Deeper integration with webcams
• USB authentication devices like Yubikeys

I think these are better served with extensions or specific browser protocols that communicate with native apps in order to keep the crazy web world more isolated from the high-value computer world, but what do I know? My guess is that someone at Google went “You know, we’re creating a lot of these specific protocols to communicate with webcams, printers, and now we want to do authentication dongles. You know what? They all use USB? Why don’t we just create a general way to access USB?”

In the immortal words of Dr. Ian Malcolm:

There’s the topic of this conversation, WebUSB. I happen to believe that a missing feature here for Firefox is a good thing, mind you…