Even with altitude, you still need time. The Earth moves around the Sun, the Sun around the galactic center, all at hundreds of km/s. Without a timestamp, lat/long/alt just tells you where something was, not where it is. Time was never optional.
Hmm I thought that, but we don't really live in a 3D world (or use the altitude parameter in a very meaningful way in life) so I wondered whether there's something else I was missing.
I wonder what makes you belittle the altitude dimension? Buildings have storys, humans can sit and stand, birds can fly, your eyes can move up and down your monitor.
The history and geography change very little whether the bird was flying while I was sitting and looking above the monitor. This is what I meant with "meaningful way". Even though a building has storys, life on 12th floor isn't much different from life on second floor. In any case, much less different than life 20 years ago, or ten kilometers away from it. In a sci-fi story or movie we see lives really in 3D, planets wrapped in habitations and such. THERE you can and should count altitude as meaningful, but we're not really there yet.
If you lived in a high place (Denver), you will find it different from a flat lowland (Chicago).
Also in Rio, how high you live can be a marker depending on which part of town you are. Favelas are on hills, whereas wealthy people in Zona Sul live down the hill closer to the beaches.
It's not that altitude isn't important, it's that it's basically determined by latitude and longitude (and time). Cultures don't exist directly above or below each other, especially not at the national scale being discussed here, and even at the micro scale, differences within a single high rise are presumably minor compared to the same distance laterally.
A mathematically ideal sheet of paper in a 3d-dimensional space can be addressed with both 2D coordinates, within the reference frame of the paper, and 3D coordinates, within the 3D space, but given that the paper has a location in space, 2D coordinates are sufficient to specify a point in 3D space.
You can do the same with geography. It's why generally only specify geographical coordinates with latitude and longitude, altitude is a given from those two, since you're as unlikely to be hovering in the air as you are to be immersed in bedrock.
"Developers talk not just about how the AI output is often flawed, but that using AI to get the job done is often a more time consuming, harder, and more frustrating experience because they have to go through the output and fix its mistakes."
This has not been my experience. Sure it feels like more work to fix the AI code problems sometimes - it is a different skillset than writing code from scratch. But the speed that I can deliver software has significantly increased by using coding agents.
I must be doing something very different from the anti-AI people on here. It is ridiculously empowering.
Got an issue in production? Give your agent the knowledge of how to locate the logs, and where the codebase is, and ask it to diagnose, and off it goes. It almost always finds the issue, and while it has been doing that, I've been able to get on with more productive things.
In terms of coding, if you work on it, and give it the correct guidelines, guardrails and ability to check its own work, it produces very high quality results.
The worst part is in such a short space of time I just don't think I can ever back to normal coding. I don't mind that, but it sucks when I'm offline.
I honestly don't know what people are doing wrong, or what sort of code they're writing that they can't get an AI to work well for them.
There's just different types of programming, I also find that if I give a decent description of a bug an LLM will often find the problem, and that great in a system that's mostly legacy and hard to repro. LLMs are also good for quick small scale from-scratch projects.
But there's a middle ground where you're really have to build something out that's super complicated and performant. Or do refactors that have a high quality bar. Situations where code bloat really matters. LLMs tend to create crazy amounts of code, not really thinking through the broader system and taking system level in-variance constraints seriously.
Honestly, the effectiveness of LLMs in coding depends a lot on what you're working on. If you're dealing with a software package like Odoo that's been around for literal decades an LLMs output can be borderline useless. The problem is that in its training data it has examples from every version that's ever been released and each succeeding major version makes breaking changes to the previous one, so pretty much what happens is that the LLM can't accurately tell what in its training data belongs to which version before concocting a reply.
Yes there's no Opus at all on Pro. GPT 5.5 is also missing. Then again what would you expect, the economic reality is beginning to hit. Also I can't be too mad when the "base" models (GPT 5.4...) are still available and decent.
When I see how fast Codex max thinking GPT 5.5 eats our enterprise seat credits almost anything else seems cheap (until we switch our live systems from 5.4 api to 5.5 api I guess)... good thing I'm not the one paying for those credits and tokens (which is probably how most of the money is going to be made on AI going forward, borderline free chatbots for normies are done)
On my personal account, Copilot Pro+ still only gave me back Opus 4.7, whereas my work's Pro account still lets me use Opus 4.6.
So, my gut says, it's entirely possible that Pro+ will continue to have more segregation on model availability...
FTA
> Last week, we also rolled out temporary changes to Copilot Individual plans, including Free, Pro, Pro+, and Student, and paused self-serve Copilot Business plan purchases. These were reliability and performance measures as we prepare for the broader transition to usage-based billing. We will loosen usage limits once usage-based billing is in effect.
There's enough weasel wording here that I would expect only certain models get re-enabled on Pro.
e.x. lots of people seem to get good enough results from Opus 4.6, personally I prefer it over 4.7 in GH Copilot... locking that down to Pro+ would be, given this salvo of enshittification, a 'logical' move on their part.
A "few seconds" to pay in bitcoin? So the captain is supposed to be watching for a response via email with his finger hovering over the pay button? Is the recipient address static? Surely they would use unique payment addresses if they have any hope of obfuscating payments.
This all sounds more like a TV show script than an actual thought-out plan to me.
The payments are going to a government. They're not using bitcoin to hide the payments, they're using it because receiving USD or Euros or whatever would mean that a hostile government could seize the funds from the bank.
The tracking is unique though. I don't know who had the $20 in my wallet before me or what series of payments it was a part of, but crypto has the curious property that over time, essentially all crypto money will have at some point passed through wallets associated with controversial entities or transactions.
Transaction fees are based on the complexity of the inputs/outputs, not the value transacted. You are literally paying for the minimum amount of data necessary to prove you own the funded sending-address, paying to write those hashes and amounts into the blockchain. The institution handling this offchaing lightning branch can implement fees in whatever structure you agree to transact, including percentage based.
Lightning is just an off-chain out-branch, which will eventually be re-integrated onto the main blockchain (based on its original funding/terms). The benefit of this is that single entities can branch off the main blockchain, which is limited in its total blocksize/capacity.
The only limits are those by the handling lightning institution. This differs from bitcoin's main public blockchain, which rewards/creates approximately six blocks /hour, each with a limit of just a couple megabytes.
Lightning is a protocol and there can theoretically be many disjoint networks. The biggest network is usually what's considered to be 'the' lightning network. Double spending, which would require 'settling' a superseded lightning transaction, is prevented by a penalty mechanism that makes it so the malicious party loses all the funds in their 'channel' when caught (which will be, at smallest, the amount the original payment was made for).
I'm still thrown. I don't see how an after the fact penalty can work.
Let's say I have 1 BTC. I buy something for 1 BTC on lightning network A. Simultaneously (within nanoseconds) I buy something for 1 BTC on lightning network B. I never plan to use BTC again (or if I do, I will use a different wallet, etc.) Do I just get two purchases? Is there a meta-network clearing house, and if so, why are there many disjoint networks.
Or do I need to have moved my BTC into the lightning network A or B before I spend it?
You have to commit the money upfront into not just the network, but the link on the network.
Basically a Lightning connection (or channel) is two parties locking up some money (in any amount and any split they want) and then repeatedly re-agreeing on what the current split is. At any time they can close the channel which unlocks the money according to the latest agreed split. It's cleverly set up so that if either one cheats (by trying to finalize an earlier split), the other one gets to overrule it and keep 100%.
The most money that can be transferred is when the split is 100% to one party. Then you need to finalize it and create a new one.
It's not as magical as its proponents think. It is better than the base protocol in some cases - if you have connections and money. The sender having to lock up actual money in advance, in the maximum amount they can foresee sending, is a real buzz killer for the sender. And if the sender is some central relay - if you want to receive via a central relay and not peer to peer - you'll have to convince them why they should do that. Usually by actually prepaying, in real bitcoins, a few percent of what you want them to lock. Which is more than the transaction fee for a nontrivial base layer transaction.
To date, my own full verification node rejects all lightning/segwit blocks, until a concensus level of 6+ is reached. I think both were BIP errors, but rejoined the main concensus network ~2018.
Also participated in the O.G. bitcoin hardfork, back in August 2017, supporting the larger_blocks camp (but now mainchain, only).
It also takes a week to get your money when finalizing the channel. This is because the other party has a week to post on the blockchain a proof that you cheated. Only if no cheating proof is posted can you actually finalize the channel.
On the main chain, bitcoin transactions can have "scripts" that describes who and under what condition this money can be spent.
You have to lock your bitcoin on the main chain in a script that shares the bitcoin between you, and another lightning network user (typically a hub)
The trick is that a lightning transaction happen by signing transactions to the other party that changes the way the bitcoins are split, without broadcasting it to the main chain. You only broadcast to the main chain when you want to unlock the bitcoin.
Broadcasting an earlier transaction will result of you losing the found because subsequent transaction contains secrets that allow the other party to take them.
So I allocate my one BTC to lightning network A, so I can spend it there. I try to spend it on lightning network B and it rejects the spend because it's allocated to A in the main blockchain, even if I never spent it on A. There exists some mechanism for me to switch my bitcoin to B (by broadcasting on the main chain)... maybe? But then A needs to sign off that I don't have pending transactions?
Normally, network A and network B would be connected by a path, so your bitcoins can be routed through the destination by moving balances of each intermediaries. If there is no route to the destination, you can't send them as is. You'd need to take the bitcoin out of network A and open a channel with B. And that's operations on the main chain so it may take some time.
Once you've allocated BTC, you cannot doublespend it elsewhere (well, you can try, but verification nodes will reject it so the blockchain will never accept).
> Transaction fees are based on the complexity of the inputs/outputs, not the value transacted
Not on the lightning network. Fees are used to incentivize or disincentivize routes across channels.
> The institution handling this offchaing lightning branch can implement fees in whatever structure you agree to transact, including percentage based.
No institution is needed. Even if one is used as an intermediary, when using lightning non-custodially, the economics of lightning are such that fees are determined by the nodes in the payer's desired route. If it's a custodial transfer from one user to another, no routes are needed.
No way they're boarding boats. They can get an accurate enough cargo weight within seconds visually (maybe minutes, depending on how computerized it is)
Why would you want to obfuscate payments if you can track how many ships entered the gulf using transponders? Regarding money laundering, you use Tornado Cash or Monero
I use YNAB. I thought about building my own now that AI coding make this feasible. But the moat that I can't cross is the integration with my bank accounts. Plaid and the like are too expensive and don't cater to one-off users like me.
Has anyone been able to find a personal financial data provider that has a reasonable price?
As a few others have said Plaid is actually rather cheap if you only have a handful of accounts. I created my own personal finance tracker when Intuit Mint shut down and Plaid costs me $1.80 per month for all my linked accounts which feels very reasonable to me
Check out Lunch Flow, that's the exact reason I built this :) we Aggregate multiple providers behind a simple api for global coverage, and with a pricing friendly to individuals not businesses.
Plaid has a pay-as-you-go option that's only about $2/month for this use case. (I believe the current rack rate PAYG pricing is 30 cents per month per connected bank login).
I think GoCardless stopped accepting new accounts recently, which makes it a bit harder to rely on now.
Feels like the options for EU/SEPA are still pretty limited. I’ve been looking into alternatives as well, curious if you’ve found anything that works well also?
Otherwise, with the scraping approach Woob https://woob.tech/ (FLOSS) works well enough on some banks... It's damn absurd that banks and even supermarkets do not offer authenticated feeds for data export but that's is unfortunately...
Me personally I think banks do their best to push people toward cryptos only because of their crappy services... Even the worst CEx offer better data access the banks...
While I applaud this effort for flagging AI slop generators, I would caution users to not always ignore your "foes".
Reading opposing viewpoints is so important and is becoming a lost art in our society. I encourage you to understand and empathize with people you don't agree with. It will help all of us in the long run.
The second half of that argument was not in this article. The author was just relating his experience.
For what it is worth, I have also gone from a "this looks interesting" to "this is a regular part of my daily workflow" in the same 6 month time period.
AWS Amplify was supposed to solve this. It is built on the AWS CDK. But I found it too clunky and slow and gave up on it.
These days I use SST (https://sst.dev) which is built on top of Pulumi. I find this to be a manageable infrastructure-as-code solution for AWS deployments.
Sybil attacks are very difficult to mitigate while maintaining privacy and decentralization. I see a nod to anti-gaming in this proposal, but it does fully address this.
Why did you choose Bitcoin for this? Other blockchains can have significantly lower fees and allow the distribution of application-specific reward or governance tokens.
"Geography is three dimensional" doesn't correctly communicate the time dimension.