|
The technical and scientific illiteracy of the average Citizer is only surpassed by Mr Robert Space himself. 'Last I heard, Mr Space was working on quantum entanglement to help solve the latency issues with server meshing!', said an idiot youtuber now on my 'do not recommend' list. 
|
#
?
Apr 11, 2024 09:30
|
|
|
|
| # ? May 21, 2024 12:10 |
|
|
Can someone break down for me how they think this solves... anything? Like in terms of big space battles in EVE the problem was always that the more entities in one place, the number of checks and calculations that need doing goes up as well, exponentially (and time dilation was brought in to allow the server to keep up). So now CIG have taken the big number of calculations and instead of having one "big" server doing them, they have e.g. 10 meshed servers all doing them individually, increasing the work by 10? for what gain?
|
|
#
?
Apr 11, 2024 12:05
|
|
|
NtotheTC posted:Can someone break down for me how they think this solves... anything? Like in terms of big space battles in EVE the problem was always that the more entities in one place, the number of checks and calculations that need doing goes up as well, exponentially (and time dilation was brought in to allow the server to keep up). So now CIG have taken the big number of calculations and instead of having one "big" server doing them, they have e.g. 10 meshed servers all doing them individually, increasing the work by 10? for what gain? My limited understanding of this mess is also that , there is always going to be one master server or main central server to control the info going back and forth between all these supposed endless mesh servers right? Call it a controller server for the purposes of my brain Now sure you can scale up endless amounts of “mesh” servers to support millions of players , just spin another 1 up instantly (because nothing in the internet world is instant with zero delay ever) But as you scale up the amount of servers it scales up the data transmission back to the one controller server and eventually that server will not be able to handle the load And we are smack back into 1993 era servers all over again But I suppose Robert space has some genius idea to fix this and we are just waiting for server port loving tech v5 to come online and fix it forever
|
|
#
?
Apr 11, 2024 12:17
|
|
|
Ngl, the vid managed to fool me. I think there is some innovation being discovered here. Is it worthy of praise? Time will tell, and we are only one decade in, lol. What I want next is for actual real game devs who make actual games to steal this tech and make something actually practical with it - instead we got: 700 million spent on buggy doors that can't even kill you 100% of the time. Coming soon! - witness your friend sliding off the face of earth in real time when you press the elevator button! Watch as their mesh-bones take on individual trajectories after being hit by a space-bullet - now without lag. New collision-less adventures await!
|
|
#
?
Apr 11, 2024 12:20
|
|
|
i think they honestly innovated a worse way to do things they backed themselves into a corner using the wrong engine then had to cobble together that system to accomplish what other multiplayer games have been doing for decades
|
|
#
?
Apr 11, 2024 13:39
|
|
|
I mean my vague recollection of The Early Years was stories about croberts micromanaging with a demand of It Had To Be Real, no game designer shortcuts like "designating livable areas based on geometry, there has to be real gas simulation" so the best engine for the job was never going to save them anyways
|
|
#
?
Apr 11, 2024 13:49
|
|
|
The result has always been that no matter what you do, you cannot sidestep the fundamental fact that the data of what's going on must be replicated to all relevant clients. Determining the relevant clients is a long-standing problem that I've usually seen referred to as "interest management". There has been piles of research into this concept to minimize data transfers as much as possible. The fact remains that if a large battle is happening pretty much everything that is happening is relevant to every client in the area and you run into the exact same problems that Eve solved with TiDi. It would really be worse for SC though because of its real-time nature with physics etc. They couldn't reasonably use TiDi. I don't see how server meshing has anything to do with this or really how it's relevant to players of the game at all. It's beneficial to the developer in that they could theoretically spin up or down the size of servers dynamically to handle the load but there could also be negatives in that the more density there is in an area the more server subdivisions there are. This would help the actual compute performance but adds additional overhead in data transfer, which again is the main issue in the first place. It's an interesting technology in a "why would you do that" kind of way while providing extremely fringe benefits. It's a huge cost in both money and added complexity and I foresee very little concrete difference in the end result with significant potential for being worse than more standard implementations.
|
|
#
?
Apr 11, 2024 13:52
|
|
|
Blah, blah... Extraordinary claims require extraordinary proof. Star Citizens has spent a huge budget delivering very little and presented almost nothing in their demo. I'm strongly on the side that either they are faking it, voluntarily overlooking the serious limitations of what has been done (look for the next Jesus tech to be announced in a few months?) or have merely replicated something that's commonplace in other MMOs but slapped a new name on it so their ignorant fans will slurp up the juice.
|
|
#
?
Apr 11, 2024 14:09
|
|
|
Oh one of these servers is lagging for some reason, all your shots never synched. Oh it crashed, everyone inside this capital ship also crashed to desktop One of these clients is playing from rural Australia over Starlink with 600ms
|
|
#
?
Apr 11, 2024 14:11
|
|
|
Ursine Catastrophe posted:I mean my vague recollection of The Early Years was stories about croberts micromanaging with a demand of It Had To Be Real, no game designer shortcuts like "designating livable areas based on geometry, there has to be real gas simulation" so the best engine for the job was never going to save them anyways A whole bunch of good physics systems put in place rather than "cheating" can be good, but like only if affecting the gameplay. Like immersive sims are great making sure there's a bunch of interacting sytems in place a player can mess around with for resutls. Roberts though seems to just want to put that stuff in just becuase. Complex systems are only as good as they are fun to mess about with, otherwise you really are just wasting everyones time and cpu/gpu cycles. Also I'm guessing a lot of the stuff they've been working on comes out of the box on the latest unreal engine.
|
|
#
?
Apr 11, 2024 15:18
|
|
|
I'll preface this by saying that I picked up SC about a year ago and I've been playing off and on since and actually enjoy it. Feel free to disregard my opinion accordingly if you like. 1) I think the server meshing demo is real. Obviously it worked as well as it did in the demo because they had extremely limited loads on each server so it was nice and smooth. The "real world" testing of it shown at the end of the video was in a much rougher state and will probably continue to be so for at least a year as they attempt to scale it and put it live. 2) I don't know if this is actually in any way revolutionary. I would be surprised if something like this hasn't been done before in other games, though perhaps not on the scale they intend for SC. I say intend because there's a big difference between making it work in a tech demo and making it work across dozens of servers hosting hundreds (or thousands?) of players. 3) I don't think this will, or is intended to, solve the EVE online problem of hundreds of people in the same location. Even if it enables breaking up a visible range region of space into different chunks to be handled by different servers everything in visible range still needs to be replicated on all of the servers that can "see" it. Notice how in the demo the server transition points are blocked from each other by a corner so server 1 can't see server 3. Server 1 is still covering all of the areas of server 2 that it can see, as is server 3. To me this implies that putting 200 ships in one area will still break poo poo because server meshing won't help. If anything the duplicated areas will mean more processing power dedicated to the same task. The advantage would seem to be allowing thousands of players in the same instance of the world, while dynamically adjusting how much space is assigned to each server depending on how many players are present, but limited by view distances and line of sight.
|
|
#
?
Apr 11, 2024 15:28
|
|
|
I can enjoy this game too when I play with someone else or only play for like 30 minutes. The game works and looks just fine for me and I rarely have any glitches; certainly not the ones that get shown in videos. I guess I am one of the lucky ones (I know those glitches are real though). The trouble is that the game has a very, very limited gameplay loop and the limitations of size and scale of actual content on planets or zones make even Stanton very very small in comparison to many games despite its physical in game size. I am all in on making fun of SC for what it is, the management of it, etc. But I DO think some of the things in it are super cool (and I hope the next big patch comes through as the UI update and inventory management is a MUCH needed improvement). In the end though it is hilarious how much it has cost to develop versus what it offers and it is important to recognize the grift despite any good things you find on the surface. Do I hope it all works out? Yes, I do hope so for the sake of the genre, the devs who have been probably put in a career black hole, and the $40 pledge I did; not for CIG’s or Croberts’ sake though. Am I gonna make fun of it and promote making fun of it? Yes.
|
|
#
?
Apr 11, 2024 15:49
|
|
|
Agreed, the game is a tech demo. It doesn't have the required content and progression to be a game by modern standards. I just happen to like the tech they're demoing and hope (however unreasonably) that they will eventually turn it into a game. I also think either their development process is ludicrously inefficient, or squadron 42 had better be the second coming for the amount of money they must have pumped into it, because I sure don't see close to a billion dollars in the PU. Disgruntled Bovine fucked around with this message at 15:55 on Apr 11, 2024 |
|
#
?
Apr 11, 2024 15:52
|
|
|
 There's an insignificant gain to be had from having each server process the inputs from and state of “its” clients, but ultimately, as mentioned, the problem is the exponential transaction costs between all the parties involved. The distributed load doesn't help that in any way, and it has been done before. It's called client-side processing — the same idea, except the server isn't involved at all aside from traffic arbitrage. Of course, client-side opens the door for all kinds of cheating, so you may not want to go that way, but from a technical standpoint, that's actually all “server meshing” is. And it still doesn't address the transaction bottleneck. If anything, it makes it worse. It's “the cloud” all over again, with people once again forgetting that “the cloud” (or in this case “the server”) is just someone else's computer. From the standpoint of each connected client, it makes no difference if “someone else's computer” belongs to a different player or to CI¬G. Tippis fucked around with this message at 17:34 on Apr 11, 2024 |
|
#
?
Apr 11, 2024 17:30
|
|
|
Bottleneck was the word I forgot to use Yes there will always be a bottleneck and I think CIG is just smoke and mirrors glazing over that aspect for now with server meshing wow look it’s great ! And they use the term spin up instantly a lot. Like I said before nothing is instant and there will be even maybe a millisecond delay in info passing from A to B while their magical servers spin up to take over the increased load And what does a millisecond delay translate to when in the real world with real people internet connections ? That’s right you shoot mister space ship and miss by 5 parsecs because he’s already in the next solar system But I’m sure all the other hundred games that could have solved this problem years ago are simply too dumb to do it
|
|
#
?
Apr 11, 2024 17:53
|
|
|
I was doing some calculations for a how much it would cost to send say 100 bytes of data for each connected client every so often (say 100ms), and multiplied that by the number of clients in an area (say 100). It works out to around 10MB/s for the server to send data about every client, to every client. That number isn’t horrible until you start looking at real AWS prices when leaving their network onto the general internet. 10MB/s works out to about 25TB of egress which is going to cost the server operator like $2K/mo. Unless they’ve brokered a deal with Amazon, I wouldn’t surprised if they’re paying 10s of thousands on network egress alone every month, depending on how many people are concurrently playing SC and how unoptimized their network is. No amount of server meshing is going to fix that and unless ship sales keep up, they’re going to be lighting cash on fire to keep any sort of live service thing up once they go “1.0”. cmdrk fucked around with this message at 18:31 on Apr 11, 2024 |
|
#
?
Apr 11, 2024 18:26
|
|
|
Shazback posted:Blah, blah... Extraordinary claims require extraordinary proof. Star Citizens has spent a huge budget delivering very little and presented almost nothing in their demo. I'm strongly on the side that either they are faking it, voluntarily overlooking the serious limitations of what has been done (look for the next Jesus tech to be announced in a few months?) or have merely replicated something that's commonplace in other MMOs but slapped a new name on it so their ignorant fans will slurp up the juice. Games like Destiny use a similar process, you have a "world" server that tracks the basic state of the world, event triggers, basic sanity checks on PC and NPC interactions and sends a regular tick update to all clients. Then you have an actual host server that all clients send their data to and performs a real time stream of player actions to all the connected players. And in destiny's case it's an extremely simple example. Separation of concerns across multiple servers for a single instance of an area is not a new thing. At the end of the day you still need to commit the worldstate to all players and all the different servers need to communicate with each other so you are not saving on bandwidth at all. Guild wars 2 has separate state servers, heck even the comparitively ancient world of Warcraft has different hosts the client talks to to perform different actions. Server meshing is probably a big technical achievement in the sense of CIG getting it to work with their engine, but if they would have had a proper project scope from the start they would have started with server / client interactions instead of putting themselves into a corner.
|
|
#
?
Apr 11, 2024 18:26
|
|
|
Disgruntled Bovine posted:3) I don't think this will, or is intended to, solve the EVE online problem of hundreds of people in the same location. Even if it enables breaking up a visible range region of space into different chunks to be handled by different servers everything in visible range still needs to be replicated on all of the servers that can "see" it. Notice how in the demo the server transition points are blocked from each other by a corner so server 1 can't see server 3. Server 1 is still covering all of the areas of server 2 that it can see, as is server 3. To me this implies that putting 200 ships in one area will still break poo poo because server meshing won't help. If anything the duplicated areas will mean more processing power dedicated to the same task. This really gets at the crux of it though. How is this any different than server instances that MMO's have been using for 20+ years? It's just renamed to "server meshing" but it doesn't actually enable anything new. Star Citizen at one point promised hundreds/thousands of players all in the same place at the same time fighting epic space battles. That still won't be possible with server meshing yet we are being told that this is a revolutionary technology but nobody can actually explain why.
|
|
#
?
Apr 11, 2024 18:45
|
|
|
I keep coming back to how the original pitch was, what? A Wing Commander-esque space combat game. And then they started promising the moon.
|
|
#
?
Apr 11, 2024 18:46
|
|
|
TuxedoOrca posted:I keep coming back to how the original pitch was, what? A Wing Commander-esque space combat game. And then they started promising the moon. Not just the moon. Every moon in each one of the 100 systems at launch.
|
|
#
?
Apr 11, 2024 19:02
|
|
|
I'm getting strong "my MY perpetual motion machine is the real deal" vibes from CIG. Incredible amounts of wasted time as the industry's least capable minds twist into pretzels trying to circumvent math and physics with "one neat trick" that nobody ever thought of before.
|
|
#
?
Apr 12, 2024 00:07
|
|
|
You guys are all forgetting quantum entanglement! Mr Space has personally assured me it works perfectly, and you can shoot through the window off an enemy ship that is on another server entirely, getting a clean head shot as they pass at hundreds of metres a second. With next to no latency! (Yes, that was a scenario they posed)
|
|
#
?
Apr 12, 2024 02:46
|
|
|
Popete posted:This really gets at the crux of it though. How is this any different than server instances that MMO's have been using for 20+ years? It's just renamed to "server meshing" but it doesn't actually enable anything new. Am I mis-remembering or didn't they recently have an event where they managed to jam ~800 people into their one existing solar system for their Static Server Meshing PoC? So they are (kinda) starting to reach those numbers  (Dynamic) Server Meshing is also kinda going to be required to manage compute spend in cloud-land. cmdrk posted:I was doing some calculations for a how much it would cost to send say 100 bytes of data for each connected client every so often (say 100ms), and multiplied that by the number of clients in an area (say 100). It works out to around 10MB/s for the server to send data about every client, to every client. That number isn’t horrible until you start looking at real AWS prices when leaving their network onto the general internet. 10MB/s works out to about 25TB of egress which is going to cost the server operator like $2K/mo. This is small potaties when it comes to AWS. I bet their bandwidth costs are probably less than 10% of their cloud bill, with compute services(EC2, EKS), database (RDS), and message queing (SQS) costs being leaps and bounds beyond what they spend on egress bandwidth in AWS. And also, yes, AFAIK they did ink a deal when they rebranded their engine as StarEngine (was it 8 years ago? xD) and started hosting with AWS. I'd be willing to bet that deal has expired though.
|
|
#
?
Apr 12, 2024 03:44
|
|
|
I think people are conflating dynamic server instancing (which is common already for decades), with CIGs patent pending magical thinking server meshing. Dynamic server instancing is just regular interest based bubbles around each _client_, and as you leave the edge of the _server_ bubble you pass dynamically into another server bubble. Server meshing was _supposed_ to be that PLUS you can see everything in neighbouring bubbles and interact with them. As in shoot from one bubble to another seamlessly. Using magic.
|
|
#
?
Apr 12, 2024 04:49
|
|
|
Wicaeed posted:Am I mis-remembering or didn't they recently have an event where they managed to jam ~800 people into their one existing solar system for their Static Server Meshing PoC? So they are (kinda) starting to reach those numbers I believe they attempted this recently yes and from the sounds of it it did not go well. They had to back off that number to get it at all stable. I Also have no idea what the test was. Was it just getting 800 people on a server? Where they testing combat in this situation (seems unlikely)? The issue isn't getting thousands of player entities into a single server that I believe can be done, it's having even remotely acceptable performance for an FPS game that supposedly simulates ever single bullets physics.
|
|
#
?
Apr 12, 2024 04:53
|
|
|
Experimental Skin posted:I think people are conflating dynamic server instancing (which is common already for decades), with CIGs patent pending magical thinking server meshing. Right but what from a networking/software perspective is the difference? I would say none. Just because they are claiming entities will flawlessly transition between server boundaries (load in/out) without any noticeable sync loss is fantasy once you start reaching an appreciable number of clients/entities that need to be tracked. They are going to run into the same issues that have always been the blocker to a truly massive realtime FPS MMO game.
|
|
#
?
Apr 12, 2024 04:56
|
|
|
Popete posted:Right but what from a networking/software perspective is the difference? I would say none. this is why theyre gonna use quantum entanglement, u see
|
|
#
?
Apr 12, 2024 05:10
|
|
|
Popete posted:I believe they attempted this recently yes and from the sounds of it it did not go well. They had to back off that number to get it at all stable. I Also have no idea what the test was. Was it just getting 800 people on a server? Where they testing combat in this situation (seems unlikely)? The issue isn't getting thousands of player entities into a single server that I believe can be done, it's having even remotely acceptable performance for an FPS game that supposedly simulates ever single bullets physics. I believe it was a stress test, which if they had to back off the player count to discover where they became stable/playable, would the goal of a stress test.
|
|
#
?
Apr 12, 2024 05:33
|
|
|
Wicaeed posted:I believe it was a stress test, which if they had to back off the player count to discover where they became stable/playable, would the goal of a stress test. Sure but if they wanted to know the player count that would make the game a buggy slideshow mess they could have just looked at their current player limit cause the game runs like crap.
|
|
#
?
Apr 12, 2024 05:35
|
|
|
TuxedoOrca posted:I keep coming back to how the original pitch was, what? A Wing Commander-esque space combat game. And then they started promising the moon. I think it was something like "Hey I'm Chris Roberts, remember Wing Commander? I did that...imagine that, but also imagine what we could do with modern tech!" And the Backers went: "Could I fly from one moon to another moon in a separate system, with my own ship that is crewed by AI & my friends?" And CR went: "Absolutely! (Write that down!!)" And the Backers went: "$$$$$$$" Popete posted:Sure but if they wanted to know the player count that would make the game a buggy slideshow mess they could have just looked at their current player limit cause the game runs like crap. Server stress test, not a client stress test. Pretty sure they are quite aware of the limitations of their client side at this point.
|
|
#
?
Apr 12, 2024 05:44
|
|
|
Popete posted:Just because they are claiming entities will flawlessly transition between server boundaries (load in/out) without any noticeable sync loss is fantasy once you start reaching an appreciable number of clients/entities that need to be tracked. They are going to run into the same issues that have always been the blocker to a truly massive realtime FPS MMO game. You're not missing anything. It's the same line as always - CIG says magic will happen and the reason no one else does it is because no one has had the vision of Robert Space to make it happen before. The fact that everyone else "fakes it" (read: found out it's laughably not feasible) is just proof of the genius. That's it, that's the entire scam.
|
|
#
?
Apr 12, 2024 06:05
|
|
|
Mailer posted:You're not missing anything. It's the same line as always - CIG says magic will happen and the reason no one else does it is because no one has had the vision of Robert Space to make it happen before. The fact that everyone else "fakes it" (read: found out it's laughably not feasible) is just proof of the genius. That's it, that's the entire scam. anything more advanced would have confused the "its not illegal to run slaves if you just kill them and then only have corpses" crew
|
|
#
?
Apr 12, 2024 11:57
|
|
|
Time_pants posted:Is the thinking here that the 1.0 release is going to happen this year? Nah, there is no way. SQ42 may have some chances to be announced this year for some kind of partial very limited release perhaps in 2025. Something like maybe a handful of initial missions and the rest of the campaign to be released later based on reception. Or some such. But the mmo, SC? No frikking way that is in the cards for 2025. Unless. Unless they just package whatever they have now as 1.0 and be done with it with all kind of promises for the future etc. But the current crap is so bad it would probably be review bombed hard and I don’t think Crobbert would want that. He probably much rather stick with Early Access indefinitely. MedicineHut fucked around with this message at 13:30 on Apr 12, 2024 |
|
#
?
Apr 12, 2024 13:27
|
|
|
TuxedoOrca posted:I keep coming back to how the original pitch was, what? A Wing Commander-esque space combat game. And then they started promising the moon. IIRC it was more a prettier and open version of Freelancer as the basis of the original pitch. But by the last few days of the Kickstarter they were essentially claiming they'd deliver a AAA quality space sim for under $20 million....at a time when AAA had at least double that budget. Like the promises they made just for an extra 100k were ridiculous. The original pitch was fairly reasonable...but after the first couple of stretch-goals promised it was clear they had no actual plan for the game and that it'd never deliver. Gerrund_ing fucked around with this message at 13:35 on Apr 12, 2024 |
|
#
?
Apr 12, 2024 13:32
|
|
|
Disgruntled Bovine posted:3) I don't think this will, or is intended to, solve the EVE online problem of hundreds of people in the same location. Even if it enables breaking up a visible range region of space into different chunks to be handled by different servers everything in visible range still needs to be replicated on all of the servers that can "see" it. Notice how in the demo the server transition points are blocked from each other by a corner so server 1 can't see server 3. Server 1 is still covering all of the areas of server 2 that it can see, as is server 3. To me this implies that putting 200 ships in one area will still break poo poo because server meshing won't help. If anything the duplicated areas will mean more processing power dedicated to the same task. This is all well and good but really isn't ground-breaking like it has been portrayed as if that's all it really is. Like I said, this sounds more like a feature built to save the developer money on AWS costs (which is nice). The actual stuff that they're talking about is things like shooting across server boundaries, seeing across boundaries, etc. As far as data is concerned this is exactly the same thing as if you were on the same server with interest management with a little bit of extra latency involved. The dynamic sizing and whatnot sounds reasonable, I don't foresee the "shooting across servers" concept working out but ideally that is not a situation that should occur very often. The vibe I get from the community is that they do think that server meshing will allow larger groups of people in the same location. https://starcitizen.tools/Server_meshing The 4th paragraph makes a lot of claims that don't really follow from the description of server meshing. My main issue is that CIG does not provide a clear description anywhere that I can find of what the actual concrete result of server meshing is or will be. https://robertsspaceindustries.com/comm-link/transmission/18397-Server-Meshing-And-Persistent-Streaming-Q-A is the best to be found as far as I can tell. There seems to be a bit where he says that there could be multiple servers running across a single Javelin if it had hundreds of players in it. It does also bring up that game clients still have to render everything you can see even if the server can handle it which might end up being the bottleneck. This page even still talks about shards and replicating planetary bases across shards. In general it seems to be a lot more complex and limited in scope than what a lot of people imagine. Because there's no definitive explanation of it it's left people free to dream up their own reality of what server meshing will enable.
|
|
#
?
Apr 12, 2024 14:19
|
|
|
Arguably, the main point of server meshing is to get commandos to keep throwing money at CIG by making them believe that it'll eventually make all their dreams come true. With this in mind, I'd call server meshing a resounding success.
|
|
#
?
Apr 12, 2024 14:28
|
|
|
I really don't think that's the case. I know this thread is all in on SC being a scam but I think Crobberts is genuinely trying to make what he thinks is the greatest space game ever. Is he skimming more off pledges for boats and mansions than he should be? Probably, but I don't think his intention is to run a grift. I also don't think he has anyone who is willing to tell him he's wrong, so he has the same problem as George Lucas with the prequels or Elon Musk where his bad decisions don't get push back.
|
|
#
?
Apr 12, 2024 14:58
|
|
|
Meanwhile back in the wankpod... https://clips.twitch.tv/CrazyCleverDiamondSoBayed-8yzcDJ1vLeJYezyl
|
|
#
?
Apr 12, 2024 15:06
|
|
|
Disgruntled Bovine posted:I really don't think that's the case. I know this thread is all in on SC being a scam but I think Crobberts is genuinely trying to make what he thinks is the greatest space game ever. Is he skimming more off pledges for boats and mansions than he should be? Probably, but I don't think his intention is to run a grift. I also don't think he has anyone who is willing to tell him he's wrong, so he has the same problem as George Lucas with the prequels or Elon Musk where his bad decisions don't get push back. A lot of fomer emplyees have said pretty much exactly this. Works being done, but the project is just extremly mis-managed with roberts wanting to give sign off's on everything, but than taking ages to actually give sign offs stuff and changing his mind on descions on stuff that has already had a lot of work already done on it and what not.
|
|
#
?
Apr 12, 2024 15:20
|
|
|
|
| # ? May 21, 2024 12:10 |
|
|
We'll probably never know intentions, but we can see the results. Or lack of them. It comes to the same thing. Whether they're grifting in earnest at the top, just pouring money down a bottomless hole while chasing the impossible, or a cheeky little melange of both. Doesn't really matter.
|
|
#
?
Apr 12, 2024 15:30
|
|