I've been pondering something similar as a modern approach to fat binaries, basically around a table like
CREATE TABLE functions (name TEXT, arch TEXT, body BLOB);
The advantage would be that binaries could be partially fattened, i.e. every function would have at least one implementation in some cross-platform bytecode (like WASM), and then some functions would get compiled to machine code as necessary, and then the really-performance-dependent functions would have extra rows for different combinations of CPU extensions or compiler optimization levels or whatever — and you could store all of these in the same executable instead of having a bunch of executables for each target.
As a bonus, it'd be possible to embed functions' source code into the executable directly this way, whether for development purposes (kinda like how things are sometimes done in the old-school Smalltalk and Lisp worlds) or for debugging purposes (e.g. when printing stack traces).
> There's always the lazy approach of storing JSON blobs in TEXT fields, but I personally shy away from that because you lose out on a huge part of the benefits of using a SQL DB in the first place, most importantly migrations and querying/indexing.
SQLite at least provides functions to make the “querying” part of that straightforward: https://sqlite.org/json1.html
More critically: just because a company is worth $1 trillion doesn't mean it has anywhere close to $1 billion in cash at all, let alone able to be earmarked to a given project (at the presumed exclusion of other projects).
Granted, a project like this probably doesn't strictly need all $1 billion all at once, but I'd argue it's better to get whatever necessary funding upfront instead of risking having sunk a partial investment without being able to obtain the rest should the company's financial situation change.
Microsoft had over 94 billion in cash and cash equivalent as of June 2025.
My assumption is that the real reason this is a loan from the government and not paid directly by Microsoft has to do with other factors, like Microsoft not waiting to be in the hook for the billion dollars if the partner company folds, or the potential for loan forgiveness, or other incentives that make the effectively loan cheaper than cash.
RCS has been a royal pain for me on Android, too. Partially my fault since I'm using non-default ROMs (LineageOS on my Fairphone 4, which I then replaced with GrapheneOS on my Pixel 9a), but also mostly Google's fault for taking as janky of an approach as possible when it comes to its Messages app (which seems to be the only actively-maintained Android SMS app with RCS support, because of course it is).
The Graphene folks have at least been making progress on getting it working (my understanding is that Messages expects special permissions from Android and Play Services that GrapheneOS has to specifically whitelist without blowing massive holes in the Google Play sandbox, and without those permissions it fails to verify the phone number for certain carriers — T-Mobile included, in my case). Hopefully whatever fix they come up with works for the long haul; it was really annoying to have RCS working fine for all of two weeks only for it to immediately start failing again when the required RCS endpoint switched from Google's Jibe instance to whatever T-Mobile is allegedly maintaining themselves.
The rest of the world isn't on WhatsApp. What a bizarre claim. Vietnam uses Zalo. Japan uses Line. Korea uses Kakaotalk. China uses WeChat. Iran is Telegram.
And in the US more people are using iMessage than SMS thanks to iPhone's 58% market share.
I don't know about you, but I personally talk with Iranians more on Whatsapp than telegram. I know the Iranian government did ban whatsapp for a while, but its still popular. I remember reading an article on here about a whatsapp leak, and it mentioned that there are over 60 million whatsapp users in Iran. Considering that Iran has a population of around 91 million, that's a huge majority of the country.
Can confirm, my family back in Iran doesn't use Telegram and haven't for quite some time. They're all on WhatsApp. Telegram seemed to be popular in Iran during the Whatsapp ban and it switched back to Whatsapp being dominant it seems. Which is very annoying to me because I loathe Meta and don't use any of their products.
I think Germany has a high amount of users on Signal, it's quite interesting seeing the stats about messaging apps in different countries, it's very fractured internationally while being very consistent inside borders.
I for one fucking hate that most of Sweden uses FB Messenger, it's the clunkiest of them all, and since I don't like using it all I constantly miss important messages from friends from not having the app installed and checking Facebook once in a blue moon :/
This is probably a lot more work than you're willing to put in, but the Facebook Messenger bridge for Matrix has actually been reliable, set'n'forget, and headache-free for me, whenever I have to use it for Marketplace.
I wouldn't otherwise mention it, but this is one of the few sites where "Stand up your own messaging server" isn't a completely insane suggestion.
>it's very fractured internationally while being very consistent inside borders
I think it's caused by the network effect [1].
>I for one fucking hate that most of Sweden uses FB Messenger
I agree. Denmark is the same, everybody uses FB Messenger or, even worse, Snapchat.
And don't even get me started on payment systems: Sweden has Swish, Denmark has MobilePay, Italy has Satispay, etc. It's completely fractured and it's so annyoing when travelling across the EU.
At least there's a new European system called Wero [2], I wonder if it's going to help fixing this situation.
> The rest of the world is on WhatsApp and doesn't even know what RCS messaging is.
Absolutely _not_ the case here (France), the overwhelming default is SMS (and now RCS). Sure people use WhatsApp but also Telegram just as much these days, but in both cases it's _not the default_.
Maybe because it's been, I don't know, one to two decades that SMS have been unlimited in even the most basic plans.
Also RCS Just Works here, I've seen my non-Apple contacts move to RCS over time as they got OS or phone upgrades.
I'd blame NA carriers, which, from afar, seem to have a habit of screwing up in so many ways.
Early adopter syndrome strikes again. None of my friends or family have Whatsapp, Whatsapp doesn't (currently) work with other services, and all of us have had SMS for nearly as long as we have had cell phones.
Slow cable Internet and 120v residential electricity are two more examples. I fortunately have fiber now, but I'll be stuck dreaming of 240v outlets and appliances for the rest of my life.
Alas, my workshop didn't come with 240 already run, so that was an added expense to get my welder set up.
An electric tea kettle that didn't take an hour to warm up would be very nice.
My well pump runs on 120v, and when the motor kicks in the whole house knows.
240v has lower voltage drop over distances, puts off less heat due to lower amperage for the same wattage, and since we're dreaming, we could switch over to a sane plug design like Type F or G instead of A and B.
Running the same wattage device at 240V instead of 120V would decrease the amperage, assuming the device was designed to handle either voltage.
My desktop PC uses about 600W running at full tilt. It can take 120V or 240V. At 120V, it will pull 5A to run its 600W load. At 240V, it'll only use 2.5A. This means for the same gauge of wire, it'll experience less resistive losses and thus be cooler and less prone to overheating.
You wouldn't change the outlet to a higher amperage outlet, you'd just change to the 240V equivalent of that same amperage rating. For the US, it looks pretty much the same as a regular wall outlet but has the blades horizontal instead of vertical. Something like this:
> Running the same wattage device at 240V instead of 120V would decrease the amperage, assuming the device was designed to handle either voltage.
Well yes, but usually the whole point of switching to 240V is to get more power than what 120V can supply. The people complaining about electric kettles being “slow” in the US compared to the EU would still be complaining if those kettles always pulled the same number of Watts on both 120V and 240V, because it's the Watts that determine how fast the water heats up. The amperage is therefore probably going to be at minimum approximately the same in that case — and probably higher if you're doing something more intensive (and therefore requiring more current) with that new 240V outlet than just running an electric kettle (like running a stove or a clothes dryer or an air conditioner or an electric car charger or a rack of 10+ of those 600W-PSU-laden computers — hence those usually getting beefier 20A+ circuits while everything else in a house might be 15A).
I'm not saying it's 100% that way, but a large chunk works like that.
Videoconference, chat, collaborative document editing are pretty much centralized in the hands of private companies, even if open source solutions do exist.
SMS also has crazy weird limitations with messaging across countries due to ISP pricing, even though the messaging apps such as whatsapp have no problem with this.
I don't know- I'm in England whastapp is the default and it makes me sad.
I was hoping when I first learnt about RCS that it could be an alternative to Meta owning everyone's comminications channels, but I've given up that hope a fair while ago.
I remember installing WhatsApp and it proceeded to delete all contacts from my phone. Haven't ever installed WhatsApp ever since. Have told people to either contact me through e-mail, google chat, LINE, discord or irc after that incident.
That's not true at all. Random data point. Estonia. I have a _single_ contact that uses WhatsApp. Everybody else is reachable via FB Messenger/Discord/SMS/Signal/Google Chat/Instagram.
One issue with Google's RCS implementation is that they've added root detection, something mandatory if you follow the RCS payments spec. Google will probably eventually want to mirror Apple's "send money*" feature to their messenger which precludes GrapheneOS and other non-official software (including Google's GSI images).
*: unless someone does a chargeback after, which makes the money disappear from your account, a major source of "oops I accidentally sent (too much) money (to the wrong person)" scams
Yeah, that root detection is the bane of my existence, beyond just RCS. Even entirely ignoring my phone having much stronger security than with the stock OS (and therefore rendering the whole “security” excuse to be complete BS), if I want to take on the risk of using an “insecure” device for payments or whatever then that's my choice to make and mine alone.
Your credit card probably has a policy where they take on the liability for fraud. At least in that case, you're not the one primarily taking on the risk for using an insecure device
The risk they'd allegedly take on by letting me use an “insecure” device is far lower than the risk already inherent in, say, the card having an RFID chip in it that anyone can silently scan from a distance unless I happen to have the foresight to buy and use a fancy RF-blocking wallet (that actually does block RF signals), or the card having all of the authenticating info¹ printed directly on it such that anyone who has access to it for the whole three seconds it takes to snap a photo of both sides can then use it to make purchases on quite literally every website that accepts credit cards.
Needless to say: letting me use an “insecure” device for tap-to-pay would considerably lower their risk compared to me not using a device at all and instead using a physical card — even, again, ignoring that my device is in all likelihood considerably more secure (and therefore exposing them to even less risk) than it was in its stock configuration.
----
¹ except for my ZIP code, which is easily guessable if you know roughly where I live — which I don't exactly keep particularly secret!
I use a rooted android phone with a custom ROM and I'm on your side. I was just pointing out that you wouldn't be taking on all the risk if your credit card provider assumes some liability for fraud.
I get the same impression w.r.t. RISC-V v. MIPS similarities, just from my (limited) exposure to Nintendo 64 homebrew development. Pretty striking how often I was thinking to myself “huh, that looks exactly like what I was fiddling with in Ares+Godbolt, just without the delay slots”.
No, it's perfectly apt. One comment is stating that using LLMs allows them to gloss over the details. The responding comment is saying that glossing over details is not a great idea, actually. I think that statement holds up very well on both sides of the analogy. You can get away with glossing over certain details when building a little shed or a throwaway python script. If you're building a skyscraper or a full-fledged application being used in the real world by thousands or millions of people, those details being glossed over are the foundation of your entire architecture, will influence every other part of the decision-making process, and will cause everything to crumble if handled carelessly.
As a bonus, it'd be possible to embed functions' source code into the executable directly this way, whether for development purposes (kinda like how things are sometimes done in the old-school Smalltalk and Lisp worlds) or for debugging purposes (e.g. when printing stack traces).