I hope the Mars portion is just as successful. It will be good to provide Space X with some completion. Even better that the competition is another domestic company.
And better still that it's two giant egos behind the companies that already clash sometimes. That type of rivalry can sometimes create innovations that wouldn't naturally occur without that form of competition.
He thinks the market is going to shit the bed and sooner rather than later. Rather than risk his and his clients money on short bets and try to time it, since metrics he'd normally use to figure such things out don't seem to be matching up with reality, he's cashing out for now. Presumably he'll be back in once he thinks the market finds a new support level.
That's my take as well. The guy understood markets that held some tenuous relationship to reality. What this move really signals is that the market is so detached from reality that even someone with absolutely impeccable instincts on such markets is skittish. Something big is gonna happen, but he can't trust that he can read the tea leaves precisely enough to time it right. If he misses by far enough before the crash, his clients will be pissed that they could have earned more. If he misses the other direction, he'll be flayed alive by them. Best bet is to bow out.
Even the best poker player sometimes folds.
a rando gig worker who now knows they have more than $100,000 in assets
Not saying this is a good idea, but I don't think the gig worker would know if you're paying $6.99 or $2.99 for the delivery, which is what would tell them if you have more than $100k in assets.
Android github app is not allowed to save files into pyDriod3 data directory.
Android file manager app is not allowed to copy files to/from ibochs android app data directory.
In general data owned by app A is not readable/writable by app B. This is a pretty important security feature. There are ways for apps to choose to share data, but by default every app's data is private to that app.
I can see how that might inconvenience you, but I think it's Really Good Idea.
I can see you've not used one. The Quest 3 has graphics virtually indistinguishable from a wired headset.
Any ideas why the Internet is full of complaints about shit compression and lag? Is everyone using their Quest 3 incorrectly? I distinctly remember the complaints rolling in at the time. Checking back they are still there.
Just because it's standalone doesn't mean you can't use your GPU power on your PC. Actually I lie since the last wired headset I used had a lower resolution, so going wireless streaming with the Quest 3 was a leap up in graphics quality compared to my previously wired headset.
The problem with wireless HMD displays is latency and bandwidth. Years ago I remember seeing perhaps even for a short time shipping add-on product for Vive I think it was that were at least using high bandwidth 60ghz radios. Now people are going with standard WiFi and bolting on crappy compression schemes to make it work. This is a step backwards.
The Quest 3 on launch was smaller than any other HMD on the market, it was the opposite of bulky. You want bulk, buy a tethered headset. Untethered ones aren't bulky in the slightest. Literally every other headset released is setting small size records.
So an HMD with an internal battery and onboard CPU/GPU weighs less than a similar HMD without any of those things? Yea sure that makes.... total.... sense.
Generates more heat than the sun? You have that backwards. Heat is generated by my GPU. The headset doesn't generate any perceptible heat on my face and the front of it is lightly warm when I take it off after a gaming session. On the other hand when I use PCVR I walk back into the room with the PC, that room is very warm.
Do you ever read what you write before clicking submit? You can't have it both ways. Either the work is done within the onboard GPU generates heat or the same work is done somewhere else generating heat.
If you are wireless streaming and complaining about heat from the PC then as a basis of comparison you need to do the same shit on the onboard GPU and compare the results.
If you run something that makes the whole room heat up on a remote GPU running that same thing on the HMD if possible which it is not would result in the HMD melting into a pile of goo.
If you run the same workload of the HMDs GPU on a remote GPU then it isn't going to heat up the room because the remote GPU will be sitting mostly idle.
Yeah I'm one of those people, and I will not buy a tethered headset. There's just so many upsides to non-tethered and you've yet to point out a downside that wasn't simply the result of your ignorance.
Quest 3 is a shit experience compared to tethered HMDs for seated SIMs.
Run any PC game directly on the headset, with emulation if only an x86 version is available
This is literally an advertised feature, one that was part of their hands-on demonstrations. Valve has contributed heavily to FEX (a user mode emulator, so you're not emulating system libraries), which is integrated with Proton in SteamOS. Yes, it's subject to any potential compatibility limitations, but I don't see how it's "absurd nonsense".
I have no doubt it can run PC games. Yet the device is not physically capable of running any PC game directly on the headset. I could not for example run MFS, DCS or Elite. My PC is barely capable of doing that and it has huge fans dissipating hundreds of watts. If the HMD could do that
I'm sure it runs smartphone VR games just fine yet this is all quite a far cry from running any PC game.
And you can play a Quest 3 as long as you like if you plug it in to your computer. You can play games on your computer through the Quest 3 so that it is reliant on the power of the machine connected to it and not the device itself - and you can do the same wirelessly. There are also some games that you can install on the Quest 3 itself and play natively without a computer. Honestly, it sounds like you aren't really up to speed on more modern VR offerings.
While I've never used Quest 3 there are no shortage of complaints about shit quality of the compression.
That's a strange way of saying will run any game on a store that 99.9% of PC gamers use and already have.
This statement doesn't make much sense. VR is an outlier 99.9% of PC gamers don't care about. 99.9% of software on the game store does not support *ANY* VR HMD. All the HMD can do at best is run low spec flat games in flat mode.
Even my dad has a Steam account and the last time he played any computer game Obama was still president. Your complaint is just not a thing that anyone gives an iota of a shit about
I have no tolerance for app store monopolies and dependencies, DRM..etc. If it isn't available on GOG, open source or directly from the vendors I'm not interested. If an HMD requires a fucking account on steam to work at all I'm not interested. This is just as stupid as saying to use a monitor I need to have a Samsung account or to use a mouse I need to have a Razer account. No fucking way, no thank you.
These issues have in fact drawn significant attention and criticism over the years. I reject the disgusting appeal to popularity you are trying to invoke to cover for fundamentally anti-consumer and indefensible behavior.
This desire to tell people "fuck you, I got mine" is incredibly damaging.
You and me are fine with current browsing. A lot of people would like AI assist. Let them have it. Freedom to choose what you want is good.
We're fast approaching a point where AI and good can no longer be associated with one another. We're having it crammed down our throats from every direction, and there are a lot of us that aren't at all interested, yet get forced to use it at jobs, or because we haven't managed to escape a bit of software that has decided to push AI on us at all costs. When that happens with other things, forcing someone to do something against their will, we call it hostility. Why is it *NOT* considered hostility when it's AI being forced on us from the tech companies?
I guess they read a few EU laws and came to the conclusion that they need to provide a bare minimum by themselves if they don't want the EU to decide what they are required to provide.
Nah, their previous plan already provided the bare minimum, since it didn't restrict sideloading of unverified apps via ADB. This is just an attempt to calm the complaints by offering an even easier sideloading option. Unfortunately, it will probably make the whole scheme pointless, since malware authors will just train users to click through the scary warnings.
Whatever method it is, it will probably defeat the purpose of ending unsigned side loading. Whatever the hoops are, users will be trained to jump through them.
This is sadly true. They're going to attempt to throw up a lot of warning dialogs to dissuade users, but we know from long experience that users will click through anything to get to cat videos.
This is actually not a change, really, since they were already going to leave sideloading via ADB open, so their plan already included an "advanced user option" which users could be trained to do. This new thing must presumably be easier than ADB. My guess is that it will feature more scary warnings than enabling ADB, but will allow sideloading without using a USB cable to connect to another computer so that on balance it will be approximately as hard.
During another discussion of this I posted a story that an Android OEM related to me when I worked on Android security, when they asked me when we were going to "close the USB vulnerability", i.e. disable ADB.
The problem is that alternate app stores would have had to verify all their apps with Google which defeats the purpose of being alternate.
How so? The developer verification does not require compliance with any of the Play store policies or anything at all other than the rule "don't distribute malware", since distributing malware would result in the developer account (and signing certificate) being revoked, which is the point of the whole thing, to enable Google to shut down malware authors. Or at least to slow them down, since they'd have to register for a new account, with a different government ID.
This does leave determination of "what is malware" up to Google, but they've been doing that for a long time and I've yet to see any case where people disagreed with their assessment. Note that I'm talking about designation of malware, not about removal from the Play store. Identified malware is removed from the Play store, but there are lots of other policy violations that can trigger Play store removal.
It's called ADB.
The point was that that was going to go away as a route for unsigned apps to be replaced with a requirement for signatures even when using ADB or other alternative installation methods
This is not correct. Per the information on Google's developer console sideloading of unverified apps via ADB was not going to be disallowed:
Q: If I want to modify an app and install it on my own device, or if I'm a power user, is there a way to turn this verification requirement off?
A: We understand that's an important use case for many developers and power users. While the verification requirement itself is a core OS feature to help protect the broader ecosystem from malware and can't be turned off, developers and power users can still use Android Debug Bridge (ADB) to continue to build, test, and install modified or unverified apps on their own devices.
(Emphasis mine)
This information has been up since shortly after the announcement.
There are actually several, including language translation layers in some models.
You are misinformed. There is ZERO NLP processing in LLMs. The only "translation" is a simple lookup table that translates integer tokens to word fragments.
"If anything can go wrong, it will." -- Edsel Murphy
