Hacker Newsnew | past | comments | ask | show | jobs | submit | more ronsor's commentslogin

I think the fact that regular stores are now stocking high-seas set top boxes is more proof that streaming is too overpriced now and media companies are too greedy.

I don't think they're stocking these boxes. A lot of retailers let anyone list products on their website - just as Amazon allows third party sellers to list products. The one I found on BestBuy's website says "Sold & shipped by Evolution Blazed Inc"

Article seems to indicate at least one model can (or, could... maybe Censys has notified them and they were pulled) be bought off the shelf in store at Best Buy

> In a recent video interview, Ashley showed off several Superbox models that Censys was studying in the malware lab — including one purchased off the shelf at BestBuy.


Back in the heyday of torrents and burnable optical disks, retail DVD players could usually play random video files procured from the high seas.

They sure did, but what they didn't have is a built-in mail-order bootleg DVD catalog!

> text obfuscation against LLM scrapers

Nice! But we already filter this stuff before pretraining.


Including RTL-LTR flips, character substitutions etc? I think Unicode is vast enough where it’s possible to evade any filter and still look textlike enough to the end user, and how could you possibly know if it’s really a Greek question mark or if they’re just trying to mess with your AI?

Afaik most LLM datasets use FastText or something similar to detect the language of the data and if it's spam, and some additional small language models to detect if text is "educational" or desirable in some other way. Often text is filtered in instead of filtered out, so anything unusual like this probably won't pass the filter, you don't need to detect it explicitly.

Ultimately the AI will just learn those tokens are basically the same thing. You'll just be reducing the learning rate by some (probably tiny) amount.

I assume that anyone trying to "filter" the text could just render it and then OCR it.

This works for ASCII, and you could just “smush” these special Unicode chars into ASCII lookalikes but then your AI won’t be usable by people who actually use these chars as part of their language.

> and how could you possibly know if it’s really a Greek question mark or if they’re just trying to mess with your AI?\

I mean how could YOU possibly know if it's really a Greek question mark... context. LLM's are a bit more clever than you're giving them credit for.


I think the bigger problem is that if the dataset was sufficiently poisoned, LLMs could start producing Greek question marks in their output. Like if you could tie it to some rare trigger words you could then use those words to cause generated code not to compile (despite passing visual inspection).

It's the Internet protocol. End users are not supposed to interact with it directly.

What exactly would replace IPv6? It's just an implementation detail, but an important one if you want to make the rest of the stack suck less.


The Windows kernel has certainly improved—BSODs are now rare—but the userspace has only gotten worse. The end result is a decline in usability and dozens of new ways for your OS to enter a permanently unusable state without a BSOD.

There are gains and losses in UX, I agree. I avoid the ads stuff via Pro (though not completely the telemetry, that said it's for games and a separate Enterprise device for Windows-only PoSh). I think the big spot for 'Recent' on the start menu, which I disable recent, is a waste of space.

But I don't use the start menu in the way of Windows' past; it's always Win+type what I want.

We did gain with things like tabbed Explorer or a right-click menu not infested by COM extensions taking ages to load.

I'm not aware of any of these 'dozens of new ways' to make Windows unusable in the way I use it, then again Windows doesn't really force any one happy path, there are often five different ways to do one thing.


That’s just your opinion, man.

Like, what do you mean by “permanently unusable state?”

This is all just vague nonsense.

Windows bad because Microsoft, or something.

I’m gonna guess you’ll come back and say something dated or exaggerated like “OneDrive nags you all the time” (nope, it can be fully uninstalled in windows 11 just like any app).


> That’s just your opinion, man. > Like, what do you mean by “permanently unusable state?”

I mean ways for the OS to lock up in ways that require a reboot. This is an objective criteria.


Sounds like you don’t know what the word “permanent” means. If you can reboot and everything works again that doesn’t sound all that permanent.

I use all three major OSes regularly and none of them lock up in ways that require a reboot with any level of regularity, never mind entering a “permanently unusable state.”

In my experience I find that at present in 2025, rebooting Apple systems seems to fix occasional little wonky problems [1], my Linux laptop needs reboots or hard restarts for occasional sleep issues, and my Windows 11 desktop’s most frequent problem tends to be graphics driver crashes while playing games (and that’s partially my fault for choosing AMD instead of Nvidia). That is the kind of problem that used to cause full system lock-ups but Windows 11 actually manages that failure relatively gracefully.

But the point is that the only operating system I interact with that never has any issues on the OS level are my Linux servers, but that’s really an entirely different use case with much less complexity and risk than a desktop workstation. And even then it’s common practice to manage Linux servers as cattle rather than pets and just destroy/replace them when there’s an issue.

[1] Out of all the desktop operating systems, I think Apple has the highest quantity of hastily added features that ship with rarely-fixed minor bugs, while the Windows team doesn’t even attempt to add features at anything close to Apple’s pace. macOS has this struggle to keep feature parity with iOS and the iPhone which itself has a economic mandate to iterate quickly. For example, since iOS 26 I’ve been having random issues with Guided Access and the screenshot tool on iOS that only resolve when I restart the phone. I’ve also needed to cycle Bluetooth since iOS 26 on occasion to get my AirPods to connect successfully.


The task bar not being movable is a huge (and also completely silly) example of stuff going worse.

I have some things that simply don't work anymore that kinda didn't bother me with easy workarounds on Win 10.

I would have to physically alter my desk setup for certain things with this new anti-feature. Not even sure how this could be argued as a win. (FWIW, I have 3 monitors and the bar used to be on the right one on top - so if I have a laptop half in front of it (no space next to the monitorr) I could do everything. open programs, look at the damn clock, etc.pp - now it's at the bottom and because of this annoying constraint called gravity I can't affix my laptop to be out of the way on top of the screen)


I would disagree with feature additions; Windows brings new features often every month as detailed in their patch Tuesday relnotes. Some aren't enabled right away, they do staged rollouts now, but they're much faster than Apple's (generally) once-per-year feature update release.

> gdscript is not a serious option imo (stupid whitespace language..They need to drop it and focus on C# like Unity did years back with UnityScript.. Seriously XNA and then Unity have cemented C# as the other language of game devs alongside C++. Even KSA's custom new engine is C# for example)

I'm going to break this down:

> gdscript is not a serious option imo

Subjective, but fair enough. If this is a general argument against "scripting languages" versus more "advanced" languages, I should remind you that Lua and JS are still very popular languages in game development.

> stupid whitespace language

The only other big "whitespace language," Python, is immensely popular and widespread in server, client, AI, etc. applications. This is a question of taste, and has nothing to do with the objective quality of GDScript.

> They need to drop it and focus on C# like Unity did years back with UnityScript

The irony is that JavaScript (real JS, not a JS-like language) is almost certainly very viable in a game engine nowadays.

> Seriously XNA and then Unity have cemented C# as the other language of game devs alongside C++. Even KSA's custom new engine is C# for example

C# is popular for a variety of reasons (fast, mature ecosystem, etc.); however, "just use C# because others are using it" is by itself not a compelling argument.

Syntax and language aren't that big of a deal for professional developers.


I disagree, going from C# with full visual studio (having a compiler with a lot of information about your code), and all of .net available, to gdscript feels like moving to a kids toy language like Scratch.

I could do everything in gdscript, but it would be slower, more error prone and less maintainable.

.. Ive done some js recently, I did a web app. and the difference is shocking..

* you can actually make spelling mistakes

* there no find all references

* You cant rename things

* without types autocomplete does very little

* without compiler autocomplete suggest random things based on spelling, not actually viable fields

I could go on.. All these things add up to make the work go slower, and produce more error prone and less maintainable code. I thought the freedom of no types and less boilerplate would make js really fast to write but I found it the opposite in practice.

I can write 1000's of line of game code in C#. Complicated stuff like procedural level generation and I dont even need to run it, usually its perfect 1st time. I couldn't do that in js, the IDE experience is not the same.


You probably violated a Nintendo NDA by even mentioning that the exporter exists on the forums.

Wrong documentation is perhaps worse than no documentation. Although Apple provides little, at least it is usually accurate, and what's left you know you must reverse engineer.

I like Godot, but you're not getting it running on a PS2.

gdscript does not do any fancy JIT or AOT compiling. I don't see why you can't get gdscript running fairly easily.

gdscript itself, sure, why not. Porting the entire Godot engine would be a real feat though, especially because of the PS2’s technical limitations.

Someone did compile Godot (v2) for PS2 and did run it but seems like they gave up trying to run/export any games with it[1][2]. Somehow, the PSP port[3] (also v2) is more successful[4].

[1]: https://github.com/ivandortulov/godot-ps2 [2]: https://itch.io/t/3658957/compiling-godot-for-the-playstatio... [3]: https://github.com/technicaljicama/godot-psp [4]: https://www.gamebrew.org/wiki/3D-Luck_PSP


Farming points for later shilling.

Community forums/support from big companies like Microsoft and Adobe tend to be completely useless. In most cases, all threads follow the same flow:

* Question with reasonable amount of detail.

* A reply from some "Community Helper" (Rank: Gold): "did you try reading the help files?"

* Another person with a "Staff" badge: "this isn't our department"

[Thread closed.]


Or

* Helper: This is a great suggestion which I'll flag for the team to add support (5 years ago)


For what it’s worth the people who made that sort of post are probably vaguely annoyed at the lack of progress on this change, or on other ones on their own particular list of requests that have been moldering for half a decade while everyone spends three dev cycles adding half-assed AI bullshit features.

At least it's not Qualcomm support forums.

"Talk to the sales about this functionality. [Thread closed]"


I have some respect for the Oracle's honesty in putting stuff like "this bug can't be solved in the cheapest version of the software, buy the upgrade package X if you need it fixed" right on the forum.

There’s absolutely nothing wrong with this. Every enterprise OSS company operates like that. People paying for support and funding the project get to make requests. Anyone else can submit a PR or be happy with the free software. It’s a pretty good deal if you ask me.

Granted, Oracle charges a lot just to even use the software, but I still don’t think it’s unreasonable to limit certain types of requests for higher paying customers. Pay base price and you get to use the software, get updates and call tech support. Pay a premium and they prioritize bug fixes and features for you.


The "no guarantee of fitness for a purpose" people put on the terms of software they sell is bullshit. There is something wrong with selling software with some functionality and then requiring customers to buy other pieces of software to make that functionality work.

That said, yes, they still handle that bit better than most large companies.


You could ask the company to remove that clause for you, but it may come with two or three extra zeroes at the end of the price tag because of the legal and support ramifications that come with it.

You could make such a clause illegal, but then all software would have to come with those two or three extra zeroes.


Hah, this gave me a good laugh. There have been countless times where I have ran into this exact kind of situation, and it's not just limited to Microsoft and Adobe.

This is true, I chose to pick on MS and Adobe because the article closes with the admission that the author has backup Windows machines to run Adobe Creative Cloud in the 'inevitable' event that Linux has a problem.

For myself, those issues have been largely evitable; I think my longest current uptime on a running linux install is approaching 5 years..


Many OpenSource forums and software are like this. None of the help is there to help you use the system. It’s there for you to gain some deep knowledge that you don’t care about.

But I’ve said it before and I’ll say it again. some Linux distro needs to adopt some hardware line and partner with them to release a known good line of computers and polish the hell out of it. Like System 76 but nicer.


Almost all community help forums (for commercial and open source software) suffer from what I like to call "HaveYouTrieditis". You post a question, and without any root cause analysis or even a description of why it might work, people start posting "Have you tried X?" and "Have you tried Y?" and "You should try Z." These kinds of responses are almost always unhelpful.

I'm asking for help because I don't want to just try random things.


Or "Did you try rebooting?"

The Microsoft Way (tm)

"Hello, thank you for posting. I understand you are having a problem with X. I assure you that you have my deepest sympathy in this time of trouble, and I will stop at nothing until this problem is fixed.

Have you tried running sfc /scannow?

...

Restart in Safe Mode?

...

Backup and Reinstall Windows. You can use OneDrive to backup if you have not already purchased a subscription."


Nailed it.

Lmao true.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: