• 3 Posts
  • 16 Comments
Joined 2 years ago
cake
Cake day: May 8th, 2023

help-circle


  • I think it is more a drip pricing scam to increase revenue for the airline, especially when it is for things that don’t have an incremental cost for the airline. Can’t compete with other airlines? No problem, advertise a lower price than your competitors, but then dream up things your competitor offers as included that almost ever customer wants (and perhaps even try to create problems for customers but charge to make them go away). Now you get customers in the door for the lower initial price, but almost all customers end up paying more than if they had just gone with the competitor.

    It is not beneficial to the customer because it reduces the efficiency of the market (and hence competition) by making it harder to quickly compare prices and get the best overall offer.

    Other industries do the same - insurers with exclusions, retailers trying to make warranties an optional extra (where regulations allow them to do it), ISPs trying to drip price extra charges.

    If a business has absolute upfront honesty about all extra charges, but they genuinely have a reason to charge extra for some customers doing things that cost them significantly more, then that is a different matter, and not necessarily bad for their customers. But the second they try to conceal part of the price and progressively reveal it, it really is a form of scam.


  • Lemmy instances are no different to any other website, in this regard. To ‘take over’ an instance would be to take over hosting of a website - which would mean either re-pointing the DNS somewhere else (and getting a copy of the database), or to take over the hosting of it (e.g. if it is hosted with a cloud provider, or you are physically taking possession of the hardware).

    Taking over the DNS in nearly any gTLD or ccTLD (short of some kind of compromise at least) generally requires one of the following: 1) a process initiated by the registrant, or 2) proving that you are the registrant, or 3) proving you owned the trademark in the domain before it was registered or 4) waiting until the domain expires, any grace period is up, and then being first to register the domain.

    If the admin is completely gone, and they are the individual owner of the instance, you could wait for (4) and try to drop-catch the domain. But domains are generally registered for a minimum of 1 year, and often up to 10 in advance, so it could be a long wait. And even then, you wouldn’t have a copy of the database. It is quite possible the actual hosting of the instance has not been pre-paid for anywhere near that long (or might fail in any number of ways, or fall out of date and get compromised, or need some kind of manual intervention following a problem), so it could go down and not come back up a long time before the domain name expires.

    If the admin has made same arrangements in advance for takeover of the instance in case they are unable to continue, the picture can be a lot better. They might, for example, if they created a legal structure (company, not-for-profit organisation) for the instance, and that is the registrant and owner of cloud resources etc…, then maybe other members of the organisation could call a Special General Meeting (or whatever similar procedure the org’s rules set up in advance say), appoint a new president, and then the new president can prove their authority to any providers involved to get access to the organisation’s resources (e.g. hosting server). Or maybe they could set up a dead-man’s switch system to automatically email credentials for all the resources if they don’t check in for a couple of weeks; or give a few trusted people the credentials (possibly encrypted with something like Shamir’s Secret Sharing to ensure n of m trusted friends - e.g. any 7 of 10 or something, need to combine their secrets to get the credentials) to take over the instance. But any of those things would require they have planned for that eventuality in advance.

    Otherwise, of course, all users of the instance could just chose another instance (possibly posting a last message with a public key to that instance, to establish the link to their new account).


  • The problem is that negotiation only means something if the parties trust each other to follow it (or an agreement will be enforced by external parties with credible ability to do so).

    Russia and Ukraine already have an agreement - the Budapest Memorandum on Security Assurances signed in 1994. It opens with this: “The Russian Federation, the United Kingdom of Great Britain and Northern Ireland and the United States of America reaffirm their commitment to Ukraine, in accordance with the principles of the Final Act of the Conference on Security and Cooperation in Europe, to respect the independence and sovereignty and the existing borders of Ukraine”. Note that the 1994 borders of Ukraine include Crimea and the entire Donbas region.

    If agreements only apply until Putin feels it is ready to take more of Ukraine, and then Putin can just take more by force, but not risk what it has ‘locked in’ through Ukraine upholding its previous agreements, then an agreement is meaningless for Ukraine - at best, all it does is lets Russia recover economically and rebuild reserves so it can prepare for a large push and take more of Ukraine next time.

    The only ways this could meaningfully be settled would be:

    • Ukraine leaves Russia in a materially worse position than before the war, and remains prepared to do it again for ongoing deterrence, so that game theoretically it would be a bad idea for Russia to attack Ukraine again ever.
    • Major political upheaval in Russia moves them to legitimately democratic, pluralistic and trustworthy leadership, and Ukraine and Russia then come to a new agreement that both sides will genuinely uphold.
    • An agreement includes other countries who will credibly enforce the agreement - e.g. Ukraine agrees to give up some of their territory, and the rest of NATO agrees that Ukraine immediately will become part of NATO and be protected by Article 5 - so that Russia cannot take any more territory from Ukraine ever without unthinkable consequences for Russia.











  • I think we need to try to get Firefox’s user base up fast (and the user base for other browsers that are ultimately controlled by non-profits) - if non-commercial browsers dominate or even have 30+% market share, if they say no to something bad for users and the open web, it doesn’t happen. While non-commercial browsers are a small minority, if they say no, services that work everywhere else follow Google / Apple and consider breaking Firefox acceptable collateral damage, and then Firefox etc… becomes an ever smaller minority, so they get forced into things like this.

    The trouble is FAANG get advantage by posing an insidious threat - they treat users well when they are trying to gain market share, and invest heavily and maybe briefly offer a superior user respecting product. But when they get the market share to give them the leverage, the switch part of bait-and-switch comes out, and we see them try to take down the open web to cement their position against the non-profits, and make their browsers inferior for users to bump up revenue (enshitification, to borrow a term from Cory Doctorow).


  • And in fact will save you CPU cycles. For a bit, Chrome had a slight performance edge over Firefox. But once Google got the market share, Firefox caught up and got ahead, and Chrome didn’t invest in keeping up, so Firefox is generally faster. The only exception is a few sites (especially Google ones) seem to be heavily optimised for Chrome, but not necessarily as much for Firefox. If you stay away from those sites, Firefox is generally faster.

    Plus Chromium is increasingly becoming more hostile to efficient ad blocking add-on implementations - so if you want to block ads (generally recommended due to ad networks doubling as paid malware distribution networks), Firefox or other Gecko-based browsers are generally the best bet.


  • “With the narrowest of majorities – five seats – in the House, Republican Speaker Kevin McCarthy needed near-unanimous support for the more than 1,200-page bill. That forced him to appease ultra-conservative members of the party, who pledged not back down in negotiations.”

    Is working with moderates of a different brand of party really worse than kowtowing to the demands of extremists in the same brand of party as you? Because it seems that would be the obvious way to get moderate bills passed if you were moderate and that is what you really wanted.