The author mentioned Unity in the README, but every other game engine I know requires NativeAOT support. I hope it doesn't run into the limitations of NativeAOT otherwise it's just Unity-only.
Even Unity games compile AOT via IL2CPP. It's optional on PC/macOS but required for most other platforms. The only real reason games may not build with it enabled is if they want to allow modding via IL decompilation.
I have teams with 1-2 permanent members and 8 more that may or may not want to check like... maybe once a week at most. Seat limits really mess with the "compliance officer needs to do something every once in a while but do we really need to pay for a separate seat?" issue with per-seat pricing.
A heavy user and a one-time-monthly user are different costs to the product but charge me the same. ;_;
If having copyright were a prerequisite of training data this would be true.
But in the US this hasn't been tested in the courts yet, and there's reason to think from precedent this legal argument might not hold (https://www.youtube.com/watch?v=G08hY8dSrUY - sorry don't have a written version of this).
I would imagine if we use a very strict interpretation of copyright, then things like satire or fan-fiction and fan-art would be in jeopardy.
As well as learning, as a whole.
Unless there is literally a substantial copy of some particular piece of copyrighted material, it seems to be a massive hurdle to prove that analyzing something is copyright infringement.
Most people in the fanfiction community recognize that it's probably not strictly allowed under copyright. However, the community response has generally been to do it anyway and try to respect the wishes of the author. Hence why you won't find Interview with a Vampire fanfiction on the major sites.
If anything, I think that severely hinders the pro-AI argument if fanfiction made by human authors are also bound by copyright.
ETA: I just tested it out and you can totally create Interview with a Vampire fanfiction with Bing Compose. That presumably is subject to at least as strong copyright as human authors and is thus a copyright violation.
> Copyright protection is available to the creators of a range of works including literary, musical, dramatic and artistic works. Recognition of fictional characters as works eligible for copyright protection has come about with the understanding that characters can be separated from the original works they were embodied in and acquire a new life by featuring in subsequent works.
Creating a work using Harry Potter or Darth Vader or Tarzan ("As of 2023, the first ten books, through Tarzan and the Ant Men, are in the public domain worldwide. The later works are still under copyright in the United States.") is a copyright infringement.
Creating Interview with a Vampire fan fiction with Bing - Bing didn't have any agency. The question of copyright infringement (I believe) should be only applied to entities with agency to (or not) ask for copyright infringing works.
> if we use a very strict interpretation of copyright, then things like satire ... would be in jeopardy.
Satire, criticism, reviews and journalism are explicitly permitted under fair use.
If I wish to publicly express my disdain or praise for your art, it is necessary that I can show samples / pictures/ photos when I express whatever my deal is.
The difference is when writing satire its not strictly necessary to possess the work to do so. You can merely hear of something and make a joke or a fake story. Training data on the other hand uses the actual material not some derivative you gleamed from a thousand overheard conversations.
> So if you delete your image the entire trained data set is invalid because they no longer have license to the copyright?
The portion of the training set might. The actual trained result -- the outcome of a use under the license -- would, at least arguably, not.
Of course, that's also before the whole "training is fair use and doesn't require a license" issue is considered, which if it is correct renders the entire issue moot -- in that case, using anything you have access to for training, irrespective of license, is fine.
Let's say you post an image, and I learn something by viewing it, then you delete the image. Is my memory of your now deleted image wiped along with everything I learned from viewing it?
Unfortunately computer memory, unlike your memory, is so easily wiped. Having the infrastructure in place to make sure it happens on the other hand, seems more like human memory.
How derived data is handled after copyright is revoked is a question thats hard to answer.
I suspect that the data will be deleted from the dataset, and any new models will not contain derivatives from that image.
How legal that is, is expensive to find out. I suspect you'd need to prove that your image had been used, and that it's use contradicts the license that was granted. It would take a lot of lawyer and court time to find out. (I'm not a lawyer, so there might already be case history here. I'm just a systadmin who's looking after datasets. )
postscript: something something GDPR. There are rules about processed data, but I can't remember the specifics. There are caveats about "reasonable"
The fees are higher everywhere. Their pitch is "our fees are higher, but we have higher-end customers who spend more", which roughly tracks with reality. But a hot dog stand isn't gonna have whales but an upper-end store might, so it's not surprising to see less support it.
High-end businesses accept Amex, because it's worth it. Major chains can probably negotiate good deals. Tiny/seasonal businesses often use middlemen like Zettle that charge high fees and accept almost every payment method imaginable. Those in the middle who use traditional payment terminals and pay list prices may still avoid Amex due to the high fees.
Good! I'm looking forward to having to uninstall a crypto miner on my mom's phone because it told her the only way to play a slots app is to allow third-party apps.
Computers overflowing with naively self-installed adware/malware is a recurring and persistent problem with some relatives of mine. Nothing you can say stops them from inadvertently doing it again, they simply lack the sophistication to understand, and then complaining that their computer is slow or doesn't work anymore. Putting these people on iOS is a godsend, these kinds of issues don't happen there.
Am I misunderstanding something or couldn't this just be made an optional setting?
So users can lockdown the ability to install stuff if they want without overcoming various hurdles (maybe allow users to add customised message so if a user tries to do it a message will pop-up saying "Your Son/Daughter/whoever has said you should never disable this! Call them before doing this if someone has asked you to!").
I feel like there is so many options/info hidden from consumers about their devices that really shouldn't be. And preventing it or hiding it only really serves the companies themselves, not the end user.
Far too defeatist an attitude and a poor comparison that plays directly in companies that want walled-gardens hands.
Fortnite is an immensely popular game people search for to buy, whereas malware etc are almost by definition not something you think "gee whiz, might go buy that".
For the type of user being discussed it seems very simple to just say/have a setting of "only allow downloads from official app store" combined with the above. Although tbh this may all be rendered moot by AI LLM security style tools that can actively monitor and prevent users from doing stuff like this.
Nope, my parents had 20 things installed on their Android devices because they clicked yes to something.
I insist on the fact that you should respect the end user’s choice of platform and not try to change the platform to earn your 30% more. If you really hate it - don’t develop for Apple. I mean at the end of the day you can tell your customers buy an Android phone and see what they prefer - your app or the Apple eco system.
You need to first go to the settings to allow installations of Third Party Apps, then you get a warning of Google Play Protect that you have to expand, and then you have to confirm that you know the App is a security risk and explicitly go forward with the installation.
Your parents did this on purpose
We don't get by just fine - there are lots of viruses. But it is much easier to install an iPhone app than a program on your computer, so it is definitely more likely.
Nokia’s Threat Intelligence Report of 2021[0] shows that Windows made up over 23% of all malware infections, in 2020[1] that was almost 39%.
They seem to have skipped 2022 and 2023 doesn’t seem ready yet.
More interesting however is looking at Android since Google has made efforts to match iOS in sandboxing the last few years, as well as the context provided with the statistics.
Where 2020 “only” saw Android come in at 26.64% with iOS coming in at 1.72%, in 2021 Android accounted for a whopping 50.31% of the infections while iOS didn’t even register on the charts.
Let me repeat that again: over half of all infections in 2021 were on Android devices.
Were these super sophisticated attacks? Let’s see, because Nokia, understandably so, dedicated significant sections of their reports to Android.
In 2020 they stated (emphasis mine):
> In the smartphone sector, the main venue for distributing malware is represented by Trojanized applications. The user is tricked by phishing, advertising or other social engineering into downloading and installing the application. The security of official app stores, such as Google Play Store, has increased continuously. However, the fact that Android applications can be downloaded from just about anywhere still represents a huge problem, as users are free to download apps from third-party app stores, where many of the applications, while functional, are Trojanized. iPhones applications, on the other hand, are for the most part limited to one source, the Apple Store.
In 2021 they stated (emphasis again mine):
> Among smartphones, Android devices remain the most targeted by malware due to the open environment and availability of third-party app stores.
> […]
> The number of Trojans targeting banking information through Android mobile devices has skyrocketed, putting millions of users around the world at financial risk.
> […]
> Banking Trojans can arrive on smartphones in a variety of ways, often disguised as common and useful apps. When run, they request a variety of permissions needed to perform their desired behavior, then often remove their icon from the application pane, effectively disappearing from the device. In many cases, the apps never provide the promised functionality that enticed the phone's owner to install them and are forgotten quickly after disappearing. However, they remain installed and continue to run as background tasks, using a variety of tricks to collect user information. These may include capturing keystrokes, superimposing their own transparent overlays onto bank login screens, taking screenshots and even accessing Google Authenticator codes.
So it looks like in most cases users are being tricked to install malware and grant permissions.
This all also explains why the whole “muh sandbox” argument carries little weight.
Not only is the sandbox but a single layer of a bigger Swiss cheese model, the sandbox isn’t gonna help your mom if she’s tricked into granting permissions.
So I ask you again to define “just fine”, because from where I’m standing Windows making up more than 20% of all malware infections is far from “just fine”, let alone Android’s more than half.
And I know you said x86, but the two and a half Linux users don’t really make a significant dent in statistics, nor is x86 the relevant platform for this discussion.
On top of that you can bet your ass that iOS users will be prime targets, certainly more desirable targets than random Android and Windows users, because of potential ill gotten gains.
This sounds like a false dichotomy. It's possible to opt-in for a "dev" mode which gives the user more choice, but not implemented because of the potential profit loss.
Imagine arguing against press freedom because a relative might fall for some disinformation.
Of all the arguments in support of Apple's prohibition against third party browsers, your mother's gambling addiction is the saddest, most bizarre, and by far the least convincing.
The film was shot almost entirely in Unity, then the final render used some extra stuff and other tools. But the scenes, shots, etc were done in Unity. A quick google shows you this.
That article I seems to imply Unity only used to generate "camerawork" data (camera positions and movements), and the actual rendering is done on other system using that camerawork data. IIRC on 2019 (Lion King release year) Unity didn't even have ray tracing support yet, so I think the final render might not use Unity at all, but Unity editor itself used extensively during production.
In contrast, Unreal Engine rendered scenes seems to be used in actual footages on some recent shows. Was Unity acquisitions of CGI firms meant to pursue this direction, or boost Unity editor usage during firm production like in Lion King?
You are correct, they used it for production, building the scenes, getting camera data. I then assume they exported the scenes and camera data to whatever they used for the final render. So Unity was used for 70-80% of the work.
It is possible but I don't know. Just was trying to point out that it's more than just some demo that it was used for.
While this should be especially true for destinations, such as hotels or even apartments, there still needs to be fast-charging infrastructure for road trips. And right now, the not-Tesla charging infrastructure is hot garbage.
Half the plugs don’t work.
The other half is at 1/4 speed, or less.
And because it’s all mostly older people, they don’t understand that it takes like 30 minutes to go from like 0 to 50% but 45-60 minutes to go from 50% to 100%. Which is annoying because:
With so few chargers, and the ones that are working at slow speed, it is VICIOUS out there. People in Porches yelling at other people to get spots.
I don’t know what it’s like for Tesla people, since its twice as much to charge than at the other chargers, but right now the road trip experience for EVs is awful just because of so few working chargers.
This is before you get to the 20-30 minute charge times.
I have had to wait at a Tesla Supercharger once in 4 years, and that was early on.
They are building new stations in the US faster than one per day (over 400 last year, tracking to over 500 this year). Each stations contains at least 8 chargers, some have 12, 24, 40, or even more.
Charging stops are usually 10-20min since it's best to use the fast part of the charging curve and move on if you're trying to maximize trip speed.
Definitely agreed about fast chargers needing to be reliable and smart. Part of Tesla's magic is a charge network and in-car navigation that talk to each other so you are routed to working chargers at the right part of your trip. And congested chargers can be automatically routed around.
Hopefully the new agreements in the US allowing other cars to access Tesla's network will promote more competition all around between charging providers to provide a better experience.
Tesla owner in Europe. My experience is pretty good even in non Tesla stations. Especially Ionity ones.
However I don't understand your comment about Tesla's costing twice as much. In Italy Tesla's for Tesla owners il less than half compared to 3rd parties (90ish cents kWh vs 44ish)
>While this should be especially true for destinations, such as hotels or even apartments
It's really not though? I've never come across an L2 charger that doesn't just work. The DC fast charging is where the problems always arise. At least with an L2 charger, the hotel/apartment can easily track the power consumption being utilized, and also lock it down to only allow charging as they see fit via mobile app, NFC card, or pin. With a 14-50 it's a free-for-all.
Sure, physically broken or turned off chargers aren't going to be magically fixed by making it a 14-50 receptacle though. If people don't care to maintain the charging infrastructure it doesn't matter what it's based on.
My point is: I've never plugged in a working J1772 and had it fail to communicate or fail to charge if it's an otherwise working charger. I've definitely seen that with DC chargers and there are plenty of known issues between certain models of cars and certain brands of DC fast chargers.
In Canada the supercharger experience is great. In 3 years I’ve never had to wait for a spot. The superchargers are usually located in convenient locations in mall marking lots or next to coffee shops or restaurants. I’d love to switch try a different EV but the lack of a supercharger network prevents that.
My experience is in northern California. Probably no one can speak for the experience across the entire US. FWIW, this guy in Minnesota seems to have had a very similar experience to what I’ve seen:
Is this a station with a wait list? Did it fail to charge? No, there were multiple dispensers open without anyone there and it worked on the first try. On his 1,200mi road trip they encountered one dispenser not working but there was another one unused right next to it.
So no, he's not constantly encountering long lines with belligerent people at multiple broken fast chargers, he's pretty much always rolling up to charging locations that are mostly empty and worked on the first try. Technology Connections is not the person to point to trying to showcase how bad charging is in the US. If you're going to paint his experiences as the norm, then I guess most CCS chargers do work fine and there aren't usually lines.
I didn't watch the video you linked, but I accept your summary of his experience then (1 year ago). The video I link is from 2 weeks ago. A good half of the video is describing many frictions he has encountered with non-Tesla charging networks. Perhaps his experience/evaluation has changed in the intervening interval.
I personally have experienced those frictions myself, in my limited EV experience. I am glad that you have had a better experience - I absolutely want non-Tesla charging networks to be high reliability and easy to use! For everyone, but also for myself, since I don't want to purchase a Tesla.
In the final analysis, I don't know how Technology Connections feels about the state of non-Tesla charging networks. In fairness, I'm not sure you do either. It sounds like his experience was more positive a year ago. But from his most recent video, it doesn't sound like he feels like non-Tesla charging networks work very well or reliably.
To be clear on one point: I haven't encountered lines or interpersonal conflict at charging points. I have encountered plugs not working, payment not being accepted, and rates lower than advertised capability. Before I had those experiences, I had heard people on HN complain about CCS charging network reliability. After, I saw Technology Connections complain about it. All I can say is that my experience has corresponded with those reports.
I don't know where you're getting "a good half of the video", it barely talks at all about his own personal experiences of CCS. It mostly talks about the history of the plugs, discusses Tesla's process of opening up NACS (and his previous reservations on it), describes how automakers are switching to it, talks about the slow roll out of CCS, talks about how Tesla chargers and 800V cars might not play nice for a while, and then talks about V2G. Could you give me some specific timestamps where he's spending half the video talking about these negative experiences? Which few sections are mostly his experiences? I've seen the video a few times before and just re-scrubbed/watched at 1.75x and I didn't see much of his personal experiences at all.
He does briefly touch on charging networks should work on reliability and just having a new connector won't necessarily improve reliability, but I'm not disagreeing with that. I've got no doubts those who do have bad experiences legit had bad experiences, I'm just suggesting it's often regional and not like every non-Tesla station is plagued with non stop problems. And given his own video publications it seems like his area and the places he traveled had reliable chargers. Maybe most of the issues are on the West coast. Either way that's not half the video nor is it really him talking personal experiences.
Also, kind of funny you're suggesting I watch a video when you refuse to watch one that rebuts your own points. You might want to re-watch both videos if you want to understand his experiences with EV charging. He's got a few of them, and they're generally not filled with routinely bad experiences.
Also, go check out Out of Spec Reviews. They've got a number of videos rightfully critical of the charging networks and several videos showing lots of them failing all at once. But notice how in most of his videos with CCS cars, the experience is almost always rolling up to a station with empty ready to use dispensers, plugging in, and getting a charge. And it doesn't get brought up like some "woah, its actually working this time!" No, instead he'll mention " these are some great Delta units" or "these ABB chargers are top notch" if anything gets mentioned at all. So obviously his experiences, which this is a guy that seems like he's practically on the road every day in multiple different EVs, seems like it's mostly that they work but with some big examples of them not.
"So now, why are automakers suddenly itching to make this switch? Well, there's a pretty simple answer: the CCS charging networks available here in the US all kinda suck."
Now, is that judgement based on his "personal experience?" I don't know. But that's what he says. He does say about Electrify America that while he's never "been stranded" by it, (18:37), he describes having problems with using the app to use the chargers. He describes more than half the stations having broken NFC readers. He says "It wasn't great and that experience is happening to far too many owners of new EVs." "Meanwhile, the Tesla supercharger network largely just works." He implies that redundancy at EA stations is insufficient (20:42). At 21:07 says that activating the charger is harder than it needs to be...
At 21:30: "The main problems with CCS networks are poor maintenance, horrible up-time, largely terrible apps which are often the only way to start charging, ..." He then says: "Let me go through those one-by-one." And the he does.
Through 26:56 (this section is labeled "The Many Reasons We're Switching"), he describes issues with non-Tesla charging networks. (You're right that he doesn't blame CCS for the problems, and neither do I.)
I didn't watch your video because I conceded your characterization of it. You don't agree with my characterization of this video, but I think you missed this section.
I don't think it's reasonable to characterize his take on non-Tesla CCS charging networks in this video as anything other than quite negative.
Other than him experiencing broken NFC readers and poor app experience it doesn't sound like a lot of personal experiences being shared. Making general statements of uptime isn't necessarily personal; I also agree they should work on uptime because clearly other people do experience problems and I've seen dispensers down I've just used the open one next to them whenever that's the issue. That whole "Many Reasons We're Switching" is largely him talking generally, but a few specific personal experiences are mixed in as short quips (like the charging latches). Him talking about the networks going with more expensive cables and them being slower to replace the cables when failed isn't necessarily a personal experience, just explaining why they struggle with reliability.
As for the app experience, I wouldn't know, I don't use apps to activate the chargers. And that sounds like that was easily half his personal frustrations.
Do you have a CCS car? You mention a limited experience with charging networks. How many years have you used them? How many times have you tried charging at a non-Tesla charger?
You're moving the goalposts. He expresses his judgements about the reliability of CCS charging networks in the video. I suppose that you don't have to credit Technology Connections' assessment if you don't think it's based on his personal experience. But I think you now agree that his assessment (right or wrong) about the state of non-Tesla charging is not positive, and can be (and is) summarized as: "it kinda sucks". And that is what we have been disagreeing about.
I don't accept or trust your pivot to personalize the discussion by focusing on my personal experience. I never said I had vast experience with EV charging (I said it was limited). But for the benefit of anyone reading this thread, I will tell you my experience, and then I'm going to disengage with you.
I have owned a CCS vehicle (Toyota Rav4 Prime) for about 6mo. I live in San Francisco. I have tried to charge it about a dozen times at non-Tesla chargers. I succeeded one or two times. I have also observed friends charge their Tesla vehicles at Tesla chargers about a half dozen times.
In my personal experience, I have had connectivity problems (unable to pay because NFC didn't work, and EA app had no signal in a garage), and chargers labeled as up in the EVgo app were not functional. On the Plug Share app I have seen a non-functional station (https://www.plugshare.com/location/37345), where it has been labeled as up for over 12mo, during which time it has never worked. It happens to be the exact station I have most wanted to use.
Those are some of the exact problems Technology Connections mentions, which is why his video did resonate with me.
It sounds like you've had a good experience with non-Tesla charging, which is great. I wish my experience was as positive as yours, because I am unable to charge at home at my rental unit. I would love to have a great experience with non-Tesla charging, both for myself, and for wide adoption of non-Tesla EVs. It sounds like where you live non-Tesla charging infrastructure works more reliably than it has for me here.
Maybe you're going to tell me I'm a moron and don't know how to charge my car. But I haven't had a great experience at it.
I'm not saying you're a moron or that your experiences didn't happen. I've suggested several times I think a lot of the worst experiences are possibly regional as the vast majority of bad experiences I see here mention being on the West coast. But telling people asking if it's normal for over half the chargers to be practically non-functional, always long lines, etc. as the norm across the whole US is a stretch. It would be like me telling someone outside the US summers for the whole country is like 105+F every day.
I'm not suggesting the non-Tesla chargers are perfect, I'm not even saying they're particularly great. They do need to have faster turn around when even a single dispenser goes offline. They've been adopting regular credit card terminals over needing the apps, which is how it should have been from the start. I'll definitely agree overall Tesla charging experience is better. They've been way more responsive to problems and made the better bet on cheaper but easier to replace components. I agree with practically everything he said in that video, but that doesn't mean it was personal experiences. A lot of my "they need to improve" agreement is looking at plugshare around the country, reading comments like yours, seeing videos on YouTube from places like Out of Spec Reviews, and others, but not much personal other than seeing like 1 of 4 chargers out at the places I've gone to.
I'm just suggesting it's not the way the OP said it is across the entire United States. The poster was essentially describing a charging hellscape, someone asked if it's like that across the whole US, and you said yes based on your experience in SF. Well, maybe SF isn't the whole US.
I am trying to suggest Alec's experience isn't as bad as failing to charge 83% of the time. It sounds like he had a road trip where the EA app was glitchy (which he'd prefer to use over a credit card as his car has free charging through the app), he's run into a few dispensers which should have worked save for a broken clip and used another one at the site, and he's seen some percentage of dispensers out of service when he's gone to charge but still managed to charge. But from most of the content he's posted over both of his channels and other channels he's participated in, it looks like his overall success rate is much higher than 17%. Can we agree on that?
Another poster here mentioned plenty of empty chargers in the Midwest. Another mentioned no problems in the PNW outside of holidays. Multiple people in California talk about charging hellscapes. Maybe it's not a US thing? Maybe we shouldn't tell people it is?
Yeah it can be terrible... on a trip I was recently waiting for a DC fast charger to become available, the guy using it was charging from 80 to get to 100%. I had a car full of kids that I couldn't get home because I was at 15%. I asked if I could at least charge for 15-20 minutes and he said "I was here first, not my problem."
I charge at home 99% of the time so it's rarely an issue, but I need to be more cautious on road trips than I hoped. Next time I'll plan ahead to charge earlier.
> . People in Porches yelling at other people to get spots.
Are you in Cali? In the midwest, Tesla chargers are still empty (like, in 45,000 miles of driving my tesla I have waited 1 time for ~5 minutes on a roadtrip). Even the non-Tesla chargers are cricket land still.
Obviously EV adoption is higher out west, but also I think more apartment living that need to public charge all the time. Out here - you would only ever charge on a roadtrip as 90% of charging is at home.
I think a lot of EV charging stories are region specific. I pretty much never encounter down chargers, there's always been one working open dispenser available any time I've wanted to charge at a CCS fast charger. I'm not saying people's horror stories aren't real, just that it doesn't seem to be the story everywhere.
West coast US; and it’s awful. The only times a spot is free is at like 6am, otherwise you’re waiting at least 30 minutes (actively, so no one can take your spot in line) to wait another 30 minutes to charge.
PNW here with an i4. The only time I’ve had problems are on holiday weekends. EA kind of sucks though, lots of broken chargers although most of the time it doesn’t matter. I only use chargers on road trips, so my experience is limited.
If an update goes really wrong (ie. battery died in middle of an update), the Mac may need to go into DFU mode and get a firmware write over USB.
I don't know if this is the actual format the firmware update uses, but the ports listed in the official documents match the "special" ports in the wiki there.
Honestly, with flash so cheap these days, why isn't the firmware structured as two "partitions", each with a timestamp and checksum, and a small stub that
(1) is never overwritten
(2) is capable of accepting firmware updates over serial connection / TFTP / etc.
(3) checksums the newest firmware "partition" runs it if the checksum checks out
(4) falls back to the older "partition" of firmware if the checksum indicates corruption and/or partial update
Having had to TFTP-update several (and in one case exacto-knife jump two pins on the flash) WRT-54G firmware updates before I learned that the WRT-54G firmware update had issues with FireFox's HTTP POST implementation, I really wish firmware updates were more robust. A firmware update also bricked an early FitBit of mine.
If a sufficiently robust integrity check is used, then an update failure is just a no-op, not a bricked (or even reduced capability) device, no?
Now, pushing a bad firmware version is another story, and impossible to prevent unless the firmware updater contains some impossibly complex static analysis tools to formally verify some set of correctness properties of the incoming update.
