It's just the next rung on the enshittification ladder. So many steps in our "progress" to enlightenment as a society, as a technology community, is just abstracting away work with a "good enough" solution that is around an 80% solution
That's fine for the first iteration or two, because you think "oh man this is going to make me so productive, I'll be able to use this new productivity to wring 40% of progress out of that 20% gap"
But instead we just move on to the next thing, bring that 20% shittified gap along with us, and the next thing that gets built or paved over has a 20% gap, and eventually we're bankrupt from rolling over all that negative equity
The counter argument for this is the comparison for traditional compilers. AI is "the new compiler", just for natural language. The optimization happens over time! But I am not so sure about that.
Except that the most glaring difference is that compilers are deterministic, while LLMs aren't.
Given the same input, compilers will always return the same output, while for LLMs. They won't, given the same input, they will return different output.
>If I move to IPv6 then my "internal" network address space is at the whim of my ISP.
This is a major problem to me before I'd go wholesale IPv6 at home as the primary way I address and connect to hosts
I have IPv6 enabled, but it's just all defaults. My traffic is going out over the internet on IPv6, my home automation stuff in the house using Matter is on IPv6, but for the few server-types that I have in the house they are still identifiable by me by their IPv4, and my addressing to get into my network from outside is via my ISP's IPv4 address
There really needs to be a universal way to bring IPv6 addresses to your ISP, so they're portable like a phone number. Both so that I can take them with me if I switch providers and so that my ISP can't arbitrarily change them from underneath me
With IPv6, it's common to have multiple addresses on an interface.
So on options is to assign yourself an [RFC 4193](https://datatracker.ietf.org/doc/html/rfc4193) fc00::/7 random prefix that you use for local routing that is stable, while the ISP prefix can be used for global routing.
Then you don't need to renumber your local network regardless of what your ISP does.
That's why I mentioned multiple addresses. The public addresses (assigned using SLAAC or DHCPv6) are for global reachability, while you use the local prefix for stable addresses within your network.
If you want stable global addresses, you should request an AS number and prefix, and choose a provider that allows you to announce it with BGP.
> There really needs to be a universal way to bring IPv6 addresses to your ISP...
There is. It's "Provider-Independent" address space.
It's used sparingly because widespread use of it would explode the size of routing tables.
I think you could also "simply" [0] become your own AS/LIR/whatever and negotiate with your ISP to route your prefix/subnet/whatever to your site (or some box in a colo somewhere that you attach to your site with some sort of tunnel).
[0] It is my understanding that it is often not at all simple to do this.
Why? You could easily block their range and it'd be blocked no matter where they went
IPv6 is already a nightmare for dealing with scammers and spammers. It's very often I get weirdly blocked because someone has abused my ISP's (AT&T) IPv6 block that I'm on and Wikipedia or whoever has blocked an entire /48 or something and it's virtually impossible to get a delegation outside of that range
Cisco, Juniper, and Arista make carrier hardware like cell phone radios and controllers and traditional telephone network switches?
While there's probably a little overlap in all of their product lines with Nokia (I mean Nokia makes simple ethernet switches so that carriers can buy all their gear from one vendor), most of those companies don't really compete in the same markets as Nokia
Cisco isn't selling into T-Mobile and AT&T's customer networks. Nokia isn't selling into JPMorgan's or Walmart's IP networks
That kind of stuff is the closest that they would come to compete with the others cited. They're all trying to get into datacenter gear, but Cisco specifically has gotten out of various levels of service provider network gear (they sold off all their cable network stuff, for example) which is where Nokia, Ericsson, etc all make their bread and butter
Cisco is still in the SP networking space, but they’ve been pushing heavily into datacenter and core routers generally (vs. edge which are more common in SP networks).
Granted, I only worked as a lowly dev in the Cisco SP routing team, and I haven’t been keeping up to speed with their work.
it's extremely unlikely that the plane wifi would be configured that way. trying to use a host file to make github.com respond on acwifi.com was definitely a red herring. It led to figuring out 53 was open, but was definitely not how the filtering was working
Also, this doesn't resolved the non sequitur in the OP. It claims there might be DNS blocking, not deep packet inspection. Also cloudflare has encrypted client hello turned on by default and the only domain that shows up for https connections in this case is cloudflare-ech.com, so if any Cloudflare website is whitelisted it would have to allow this domain, and consequently any other Cloudflare website.
zip codes are not geographic areas, they are a collection of mail delivery points. Sometimes those points are not geographically contiguous or may overlap
This is a fact of ZIP Codes that a lot of people stumble one. I've worked on GIS/mapping projects in the past where stakeholders wanted or assumed ZIP Codes to be polygons.
Another complexity that surprises folks is you can't guarantee a one-to-many state-to-ZIP Code relationship. There are several (I forgot offhand how many, I used to have them memorized) that span across state boundaries.
Yep, this fact eluded me earlier in the year. I was supposed to map out all ZIP codes in the US and color their boundaries based on certain stats we had. We were surprised to find many areas in the US were empty because they didn't have ZIP codes. I did a quick search and found out ZIP codes are driven by mail routes, and that instantly made sense to me, but the product stakeholders were very surprised to learn it.
One thing I just recalled is that if you maintain a small exceptions lookup table (i.e. the ones that span state boundaries), you can use ZIP Codes as a way to uniquely look up a county name.
For example, health care plans in the US are county-specific with regard to premiums, co-pays, etc. (based on demographics). Allowing someone to type in their ZIP Code to get started can be a better user experience than having them pick their county.
Yeah it makes it easier. And I’m appreciative of the idea of making address entry easier for users but if they have to disambiguate did you actually make it easier?
I think Zillow does it best. You just type your address in a box and it looks up the normalized full address.
My ZIP code happens to be shared by two separate cities and there are a few websites (Github, I'm looking at you) that will fail the payment, registration, etc. attempt if you don't enter the municipality that it THINKS is correct.
Exact opposite problem at my former rental: Two different properties with different ZIP codes... but they had the same address on the same road, just a mile apart (different jurisdictions).
I lived in a house; the other location was a nail spa. Strangers sometimes visited thinking they were at the right address (they weren't) to get their nails'did (they didn't).
A lot of this is on the people doing the platting in the first place. I would note, for example, that where I live in Oak Park, when they replatted the suburb after splitting from Austin Township with a different grid than is used in Chicago and the neighboring suburbs that
1. They kept the Chicago grid on the edge streets of the village so that, e.g., 110 North Austin would be across the street from 111 North Austin and
2. If they had kept the usual new 100 at each block system, the north-south streets on the south end would have been 1200–1249 which would have been identical to the numbers of the next block south in Berwyn and Cicero so the last block on the north-south streets is instead 1150–1199.
Contrast the borders of Los Angeles which in some areas are almost fractal in their complexity (there are buildings which straddle the boundary between L.A. and its neighbors and many blocks where adjacent buildings are in different cities). For whatever reason, the powers that be decided that the incompatible address numbering between adjacent cities should be retained so you will have weird discontinuities in building numbering along a block depending on what city the building lies in. I remember my wife having a doctor’s appointment in a building which was one of those which crossed the border so it had two different addresses assigned to it, one for Los Angeles and one for Beverly Hills.
A lot of P. O. Boxes have their own zip+4 address. You can spot these because the P.O. Box number and the +4 are usually the same (or occasionally the P.O. Box number is the last two digits of the five-digit zip code combined with the +4 so that a piece of mail with a box number and no zip code can be delivered to the correct post office in a large city).
There’s a less-known 11-digit zip code which is unique for every delivery point (so down to the individual residence). I’m not sure if multiple apartments in the same building have distinct 11-digit zip codes, but this does imply that a zip+4 cannot have more than 100 delivery points within its bounds.
Apartments will commonly have multiple Zip + 4 for the building so I imagine each apartment gets unique Zip 11.
Open Google Maps, go to Central Park in NYC, search for apartments and randomly pick one. Then go USPS Zip Code lookup (https://tools.usps.com/zip-code-lookup.htm) and punch in the address leaving off any apartment number so it will show all available addresses. I used 225 E 63rd St New York, NY. Appears they have 8 Zip + 4 assigned to the complex.
I tried again with building in Philly and same story. Each floor of 16 apartments got its own Zip + 4.
And census tools like Geocorr can translate between zip/zcta and other geographies, weighted by population or land area, without doing any spatial calculations.
The military postal code system is seperate from the US post office's zip code system. Often (and easily!) conflated but actually operating in parallel. Made even more confusing because many U.S. based entities can be reached in both systems.
zip codes are not geographic areas, they are a collection of mail delivery points.
Indeed. My zip is from a neighboring state but has a -xxxx to route specifically to my mail distribution center. Even without the last 4 digits they manage to figure it out, just slower.
Yeah, they're cops. Cops aren't going to arrest other cops. Their superpower is being the people who are supposed to enforce the law, if they decide to break it who is going to stop them?
Yeah, this weirdly splits the Atlanta metro area in half between two regions based on the counties, and while north Atlanta and south Atlanta metro have decidedly differing cultures (along mostly but not entirely racial lines) the split is completely arbitrary on county lines with Fulton County, GA jutting upwards as if the 10 miles across that county don't represent anything on either side of it
Fulton County is a weird shape for historical reasons - it absorbed the counties to its north and south during the Depression - and historically the northern part of Fulton County (everything north of the Chattahoochee River) was Milton County. If Milton County still existed it would probably end up in Woodard's "Greater Appalachia" over "Deep South".
We can ignore current settlement patterns because Woodard does. In a recent paper (https://www.sciencedirect.com/science/article/abs/pii/S00330...) he does explain the methodology, although I don't have access - but from the snippets I can see it appears that he's essentially trying to work out who the first European settlers in each area were. So it doesn't matter that north Fulton County is full of carpetbaggers from up North and immigrants. (I write this as I sit in an office in north Fulton County; I am a carpetbagger from up North and many of my co-workers are immigrants.)
It makes sense for the split to be along county lines just because a lot of data will be available at the county level, but it occasionally produces absurd results. I occasionally have mocked these splits as "I drive to Appalachia for ramen", because I used to live in DeKalb County about a mile from the DeKalb-Gwinnett county line - according to Woodard's map, DeKalb is "Deep South" and Gwinnett is "Appalachia" - and I liked a ramen place just over the county line. (Since then both I and the ramen place have moved.)
Maybe I misinterpreted what this is supposed to show. Is this based on data from like 150 years ago or is it based on how things are today?
This is similar to something I saw on reddit over the weekend which was a similar map but based on local cuisine. I live in North Fulton County now, but I'm originally from central Alabama and the dividing line for the cuisine was between "soul food" and whatever other term they had come up for deep fried food
Basically it was white people southern food vs. black people southern food (which, at the end of the day is actually not that different)
curious if this Appalachia vs. "Deep South" thing is really just a racial divide in the data with "Deep South" being African American descendants of slaves across the Black Belt and Appalachia being the more white population
The map in this post is historically based, I think, but they don't say that very loudly.
And definitely some of what we're seeing in this data is a racial divide - but the racial divide in the South goes back to where slave-based agriculture was and was not viable.
A major reason nuclear plants are super expensive is because we do it so rarely
Every reactor and every plant is bespoke, even if they are based on a common "design" each instance is different enough that every project has to be managed from the ground up as a new thing, you get certified only on a single plant, operators can't move from plant to plant without recertification, etc
Part of that is because they are so big and massive, and take a long time to build. If we'd build smaller, modular reactors that are literally exactly the same every single time you would begin to get economies of scale, you'd be able to get by without having to build a complete replica for training every time, and by being smaller you'd get to value delivery much quicker reducing the finance costs, which would then let you plow the profits from Reactor A into Reactor B's construction
> A major reason nuclear plants are super expensive is because we do it so rarely
Once you have your supply chain running, and PM/labour experience, things can run fairly quickly. In the 1980s and '90s Japan was starting a new nuclear plant every 1-2 years, and finishing them in 5:
Exactly. What is needed is a SpaceX-like enterprise, where the engineering effort is concentrated in building economies of scale. To me it's clear that nuclear energy's pros largely outweigh the cons, and that it is a perfect complement to solar and wind power generation.
I'm not sure. They have more injuries per worker than their competition [1]. Space should already not be "let's work too fast at safety's cost", nuclear really can't.
Injury rate is 6x other space vehicle manufacturers. If you were to slow them down by 6x they would pretty close to the 20 years it’s already taken to get SLS/constellation to do a test launch.
Nuclear submarine power plants are not in any way a technology useful for utility scale power generation.
To start with they use fuel enriched to weapons grade.
They aren't cost effective vs the amount of power produced, and the designs don't scale up to utility scale power.
Submarine plants are not some sort of miracle SMR we can just roll out.
The Navy is willing to page cost premiums a utility company cannot, because for the Navy it's about having a necessary capability. There's no economic break even to consider.
I thought I'd mention that ship supplied short power has been a thing for ages. USS Daniel Webster even trained for this for new years eve apocalypse nothingburger. And its almost always been used for only powering something critical. Today's subs are <10MW. Nothing for utility scale. I can't imagine the economics are ever good. More of a: we've already got this boat.
Secrecy isn't the obstacle here. Naval reactors are optimized for combat performance, costs be damned. They aren't economically efficient for commercial power generation.
At least Russia is doing fine with SMRs, thought the fuel enrichment level is around 20%. They are building new reactors all the time and they seem pretty efficient. E.g. they have even floating nuclear plant: https://en.wikipedia.org/wiki/Akademik_Lomonosov
I'd be fine with us just having the USA navy operate them we build them for carriers and subs just double or triple the order and plug em into the grid.
Submarine reactors run on super high enriched fuel instantly one could instantly repurpose into a bomb. Lots of gen 4 and 5 reactor designs that combine low cost, compact footprint, and running on less expensive and carefully controlled fuel.
> If we'd build smaller, modular reactors that are literally exactly the same every single time you would begin to get economies of scale
You can also build standardized, modular LARGE nuclear power reactors. The French and the Japanese did it and managed to builds lots of large reactors with relatively short build times
Yes, as a Portuguese with some patriotic feelings (but also self critical when needed), I was impressed myself with the quality of this report given the time frame.
That's fine for the first iteration or two, because you think "oh man this is going to make me so productive, I'll be able to use this new productivity to wring 40% of progress out of that 20% gap"
But instead we just move on to the next thing, bring that 20% shittified gap along with us, and the next thing that gets built or paved over has a 20% gap, and eventually we're bankrupt from rolling over all that negative equity
reply