HN user

amiga386

6,885 karma
Posts1
Comments1,364
View on HN

I don't think a bunch of dodgy app vendors who put TVs in botnets is who you should be going to bat for.

Read it again: "LG Electronics is working with developers to remove the residential proxy option from their apps on the webOS platform"

It is working with developers it has a contractual relationship with, and has approved their apps for download from its "app store" -- this one: https://gb.lgappstv.com/main/tvapp

...where 42% seem to be ADDING THE TV TO A BOTNET.

LG TVs form a substantial botnet, and LG didn't appear to know or care.

This is not about a person using their own TV to add it to a botnet (WHY WOULD THEY DO THAT?), it's about "app developers" knowing that LG is lax in checking what "apps" do, and putting out trojan horse apps that, if the TV user installs them, adds the TV to a botnet, without the TV owner's knowledge.

This is entirely orthogonal to "can I run my own choice of software on my LG TV?", which is not addressed in the article at all.

That is not the case.

The article is quite clear that LG runs an "app store" where 42% of the "apps" likely contain residential proxies, and LG is going to make its "app store" developers stop doing that. From TFA:

“A residential proxy network is not an intended use for LG smart TVs, and LG Electronics is working with developers to remove the residential proxy option from their apps on the webOS platform,” Taylor said. “If this option is not removed, these apps will be suspended.”

This says nothing about your own software installed on the TV, should that be possible.

But even if you ran your own software on your TV, would you run a proxy service? Given you already have a residential ISP, you wouldn't need to hook your TV up to a botnet and buy residental ISP access for your scrapers via your own, or other unwitting suckers', TVs.

Your hypothetical of wanting to run a Tor exit node is silly. Most Tor exit node operators run them knowingly on dedicated hardware, usually in a datacentre, not their own home, and prepare themselves legally for when they inevitably get emails from law enforcement, or worse, raided. You'd almost certainly firewall the destinations it can reach, you wouldn't let it have access to your own network, and you wouldn't run it on your TV.

I tried it some time ago and found it was useless in that role.

Kodi is 99% a TV-like frontend to a local-disk media collection, its web interface is an afterthought, its performance for remotely served media is atrocious.

Jellyfin, Emby and Plex are media servers first and foremost.

The title here is burying the lede and should be updated to something like "Jellyfin founder, project leader and core contributer all leave the project".

It's not just Andrew [Rabert] leaving. This is an announcement by Joshua Boniface (the current project lead) and Anthony Lavado (a core contributor) announcing that they're leaving too.

Boniface's and Lavado's reasons are mostly akin to burnout, or the need for a family life, which I totally understand, especially as Boniface has been the face of the project for quite some time. I use Jellyfin daily, and I know exactly who Boniface is, I hadn't actually heard of Rabert, though I'm grateful for him founding the project. Boniface in particular has lead Jellyfin through huge levels of internal changes, bugfixes and improvements. It is throwing off the janky architecture of Emby, one part at a time.

Rabert's reasons are here: https://github.com/andrewrabert/jellium-desktop/issues/609

An obvious sore point is that Rabert has been AI-coding the Jellyfin desktop app, while the community (who bear the brunt of drive-by AI slop) have strongly opposed it. Even if he's a great coder and makes a wonderful desktop app with the aid of AI, that seems irreconcilable. Every wannabe contributor who shits up the project and wastes volunteers time with AI slop will say "you let that guy add his AI generated stuff, why not me?"

So he's leaving the Jellyfin project and he's going to rename his desktop app, to see if it stands on its own merits as a third-party Jellyfin client, rather than an official project. I wish the best of luck to him.

Even with that angle, it's true that the "Jellyfin community" (which includes all bug reporters, users on forums, etc.) can be very rude, directly to the contributors who may have inadvertantly added bugs. And Jellyfin is going through some growing pains, for example their 10.11.x series rewrote all the database handling to be a much better fit for the language (from raw SQL to .NET EF Core), which gives them a much more stable base to make changes upon, but it also had some performance regressions that rendered some third party Jellyfin apps unusable. (Those third parties' advice is not to upgrade). But at the same time, the project is getting there, supporting developers who are systematically fixing those regressions. The project has thousands of bug reports, but it also has thousands of fixes, and thousands of merged PRs. They are keeping up with the demand. I have every confidence Jellyfin 12.x will be great, and I'm looking forward to the new possibilities for plugins and enhancements it will open up.

While I'd agree that the 10.11.x releases have terrible performance on some APIs to the extent that I rolled back to 10.10.x and marked it as held, until they release 12.x...

Jellyfin is a daily driver for me and works amazingly. I have thousands of films, TV shows, books, music, and it lets me navigate and use them perfectly well.

I have a wonderful little Linux media server.

What sort of bugs affect you?

desolate parks

Criminals hide out in the dark parts of urban parks, mere minutes away from well-lit, more populated areas of the city. They're not going to get bored. Victims walk through the parks and get jumped, occasionally dragged off into the darker parts.

To give a example: The Meadows in Edinburgh: https://www.openstreetmap.org/way/233760765

A park between the university and student accomodation. Very popular with students going home at night. Apart from a few lit paths, entirely dark. Most paths are lined with trees, like so: https://www.edinburghnews.scotsman.com/webimg/b25lY21zOjAxYz...

Here's a small sample of incidents. There are genuinely criminals who hang about in the dark parts of this park, late at night, in order to attack victims. They don't have to wear ghille suits, they need only stand behind the trees.

- 2011: https://www.bbc.co.uk/news/uk-scotland-edinburgh-east-fife-1... "Two men who grabbed a young woman before one sexually assaulted her in an Edinburgh park are being sought by police. The 19-year-old woman was walking through the Meadows at 03:30 on Tuesday when she was attacked at a spot north of Melville Drive."

- 2011: https://www.scotsman.com/news/charges-over-sex-assaults-in-m... "Lothian and Borders Police issued an open letter to students and residents in response to the incidents, advising them not to walk alone in the Meadows at night"

- 2012: https://www.bbc.co.uk/news/uk-scotland-edinburgh-east-fife-2... "The 30-year-old was walking through Middle Meadow Walk at about 22:30 when two men approached him. One of the men punched the victim in the face, knocked him to the ground and continued assaulting him before taking his personal belongings."

- 2014: https://www.bbc.co.uk/news/uk-scotland-edinburgh-east-fife-2... "The 25-year-old was walking near the tennis courts at the east end of the park at about 04:30 on Saturday when he was approached by two men. After asking the man for money, the two attackers punched him several times, causing minor facial injuries, and also tried to steal his watch."

- 2014: https://www.bbc.co.uk/news/uk-scotland-edinburgh-east-fife-3... "Steven Davidson, 24, attacked the 19-year old as she made her way home through the Meadows in November 2014. [...] Davidson ran up and grabbed her from behind. [...] The woman had been making her way home just before 07:00 [...] she passed out and came round to find him lying on top of her. He then raped her."

- 2016: https://archive.news.stv.tv/east-central/1374940-three-assau... "Police are investigating after three assaults were carried out on the Meadows in Edinburgh within 15 minutes of each other. In each of the incidents, which took place between 3.10am and 3.25am on Sunday, the attackers robbed or attempted to rob the victims of cash or possessions."

- 2020: https://www.bbc.co.uk/news/uk-scotland-edinburgh-east-fife-5... "The attack on the 37-year-old took place on The Meadows at about 23:15 on Saturday"

- 2023: https://www.edinburghlive.co.uk/news/edinburgh-news/edinburg... "The incident took place at around 4am on Sunday, January 29."

I'm not saying we should floodlight the park. What has helped most is regular police patrols of the park. But I'd be lying if I said the criminals don't hang out in the dark parts of this park, waiting for victims. They do.

TFA doesn't even claim that. TFA links to an advocacy website trying to convince authorities to put out less light pollution. It claims improved lighting (presumably brighter) doesn't reduce crime, and it claims a reduction (but not elimination) of lighting doesn't increase crime.

I would say that going from no lighting at all, to some lighting, on urban paths, improves safety (not tripping over unexpected obstacles) and reduces crime (presuming there was any crime while it was unlit). The article doesn't refute that. I would suggest that if a major city decided to eliminate all street lighting, it would be absolute chaos.

The problem is not just with people who want to walk places, but people who want to mug/murder/rape those that do.

As an example, in the UK, councils are expected to light commonly-traveled paths, both so path users can see where they're going, and to reduce crime. Unlit paths in city parks are still traversed, and are often a haven for crime.

"Dead of night" isn't the right image. 5PM in winter with clouds (as is typical in the UK) has roughly the same level of natural light as 2AM. People are still coming home from work in conditions where natural light is low enough to lead to accidents and crime.

I'd bet dollars to donuts you didn't make your format an ISO standard.

What matters to me, if I'm communicating with my government (or vice-versa), is that they only use recognised standards.

I will need a web browser, email client, document editor... but not a specific web browser, email client, document editor.

This takes effort from governments, because the vendors of all the products they consume want to try and leverage their government connection to get lock-in and thus make their products mandatory for everyone in the country.

For example, even if government developers only use Chrome browser and neglect to test on Safari, you can end up with millions of Safari or iPhone users unable to perform legally essential tasks, and have no workaround. The same is true for using proprietary office document formats, and that includes Adobe when there's a difference between PDF-as-specified versus PDF-as-implemented-in-Adobe's-proprietary-software.

The importance of standards is that they tell every vendor exactly how to conform and not-conform to the standard. Hence the standard itself has to be made be robust. Hence why a standards committee has to own it, not a proprietary vendor. If it's a proprietary format undergoing standardisation, the standardisation's job is to file off the rough edges and make it interoperable. It doesn't take some fragile code that only works in one way, and say "that's the standard". It throws out the fragile bullshit like Microsoft's broken date handling that they don't even bother to fully document in OOXML.

For comparison, a better standard and standards body (even though they have their own problems e.g. being dominated by Google) would be HTML5 and WHATWG. Everything is carefully specified so that all browser vendors can implement a standard, identically. That even includes things like word-wrapping (a small subset of document layout!), and there are test suites to catch deviations from the standard.

Not so much TDF as Italo Vignoli

I noticed this HN comment from four months ago and boy does it have his number:

https://news.ycombinator.com/item?id=47630474

I'm confident the person who most wants to sabotage LibreOffice's success is Italo Vignoli. [...] Most of his blogs are about how awful OOXML (Microsoft Office's open standard) formats are [...] This is a self-sabotaging marketing approach. [...]

And now I look at who authored this article. "2026-07-17 Italo Vignoli"

While that's true, it glosses over the battle to make DVDs convenient. Hollywood did not want them to be convenient, they wanted them region-locked and unrippable, and spent a fortune prosecuting anyone who thought otherwise:

https://en.wikipedia.org/wiki/DeCSS

We have Derek Fawcus, "mdx", "DVD Jon" et al to thank for making DVDs worthwhile.

Not even "Rent". Rentals are priced by the time you rent for. If you want to rent something for 30 years, you can, and you'll keep paying for 30 years.

This is a one-time cost and you just don't know when they're going to snatch it back from you. They won't tell you. They won't even give you a notice period. They don't know themselves. They only find out when the licensor they're sublicensing from demands "too much" for ongoing licensing and they just give up and pretend they didn't sell you that and take your money.

The button would have to be "Licence, subject to unilateral revocation at any time."

It's a magical superpower.

It kept https://en.wikipedia.org/wiki/Gary_McKinnon from being extradicted to the USA. Apparently the courts don't accept "your opsec is shit and I got in with default passwords", but they do accept "I have autism"

Let's try it in action:

- "Mr Wallace, we have several credible reports that you harrassed TV production staff by going around with no underpants on, and finding excuses to take your trousers down. What do you say to that?"

- "Did I mention I have autism?"

( https://www.bbc.co.uk/news/articles/cx24lxl85wyo )

Count Binface 10 days ago

In case you're wondering why there's so much interest in the Count, the reason is his opponent.

Nigel Farage is a populist gobshite. He's in trouble because he refused to declare a £5 million donation from a crypto billionaire, Christopher Harborne. MPs are expected to declare any "gifts" that might sway them.

https://en.wikipedia.org/wiki/Reform_UK_financial_allegation...

After Farage took that "gift", he became an MP and lobbied the Bank of England to adopt a stablecoin that Christopher Harborne was heavily invested in.

Recently, the press found years and years of undeclared dodgy donations to Farage and his party, many from convicted fraudster "Posh" George Cottrell, author of book How to Launder Money.

Sleazy MPs who take bribes are investigated, can be reprimanded, can be recalled by their constituents.

Farage is trying to get ahead of the curve by resigning now, then competing in the election for his successor. He's hoping to pause the investigation into his finances, and wants to pull the populist move of "The People support me!"

Every major party sees this trap and have boycotted Nigel's self-election. They'll fight him in the next election, after he's recalled for sleaze.

Effectively he is standing unopposed... except for the joke candidates. And who better a joke candidate than a quick-witted comedian in a ridiculous outfit? £500 to rip the piss out of Farage for a month is a bargain.

The thing populists hate most is being ridiculed and not taken seriously. It really pricks their ego. Will Nigel spend summer arguing with a bin? Will Nigel lose to a bin? That's for the people of Clacton to decide.

all else being equal, most C software would indeed benefit from being written in Rust instead

I don't even think this is true.

You can certainly say "the security of computer systems would be a little bit better if most software was written in memory safe languages" (even more so if you went a bit further and said "with automatic memory management").

However, a lot of software exists and is useful, and the only reason it exists is because the author(s) found a way they liked to express that software.

So, for example, there are millions of companies out there, especially small ones, running all their finances on Excel. Not even accounting software. Excel. I am totally OK with this. I do not need all these companies to rewrite their Excel workflows into frontend applications backed by a relational database, even though that would be "better" in a lot of ways (more robust, easier to backup, easier to bring someone else onto...). Those little business owners understand Excel and build models and count numbers and they're happy with that. If some kind of edict compelled them to use something "proper" instead, they might not even go into business, and the world would lose whatever it is their business does.

The same thing goes for software and languages. Each language, whether it's F#, Haskell, Common Lisp, PHP, ... brings with it its own kinds of expressiveness and usefulness, and ecosystems of programmers and libraries/modules form around it. Some languages are a better "fit", sometimes for the problem domain, sometimes for the programmer's mind, sometimes for community building. It's difficult to compare any two languages because of this, and if you were to say "language X should not exist, all software written in X should've been written in language Y", you have to accept in your thought experiment that were that the case, their is likely a huge amount of software which would not exist just out of the people who made it not being happy about language Y and, if it were the only choice, would choose not to create the software they created in language X at all.

This is somewhat deconstructive.

So you need the solo developer not to contract out or buy in existing music, graphics, 3D assets, animations, marketing, or you won't call them a solo developer.

Right, so do you also need them to create the 3D engine or are they allowed that off the shelf? Oh, they need to make it themselves. You're strict!

Ok, so they're allowed to write for a platform? Oh, no they're not, that's relying on other people's code.

And writing in an existing language? Tsk tsk tsk. Got to invent the programming language yourself, otherwise you need to list the entire GCC/LLVM team as your collaborators on the game.

They have to create their own silicon too, it's cheating to rely other people's chips, how can you call yourself "solo"?

Are they allowed to sell it on Steam or do they need to build their own store and payment networks? Heck, should they get themselves accredited as a payment network. Oh, and as a bank.

And presumably, if the game needs to be translated to any language other than the developer's own, they have to do that translation themselves, right? Not rely on experts in that language. Can't really be a "solo" dev that way, can you?

And so on.

Building a game involves effort, and millions of decisions. Is the gameplay right? Is the story right? Are the graphics right? Design the characters, the levels, the world. Make the game run. Make the game available?

I can accept that solo developers will sometimes make the graphics/music/"assets" themselves, sometimes buy off the shelf, sometimes pay others. But unless they hire that person full time to collaborate on the game... they're still the solo developer.

They will definitely lean on existing 3D engines, libraries, plugins, font engines... and that reminds me, I've almost never seen a game developer design their own fonts. These reusable components can be used in games, and some are even intended to (e.g. engine plugins). But do they define the game experience? Generally, no. That's on the game developer.

Here is Jonathan Blow's placeholder art for Braid: https://upload.wikimedia.org/wikipedia/en/2/2f/Braid-art-1.j...

Here is how good David Hellman, the artist he hired, made it: https://upload.wikimedia.org/wikipedia/en/2/2e/Braid-art-2.j...

David Hellman is credited as the game's artist, but it's still effectively Jonathan Blow's game from top to bottom.

Unfortunately this sputtered out during the SSD transition

I advise people to backup their data (emails, writings, pictures, photos, home videos, phone contents, etc.) and worry about their music/TV/film/game collection later.

Most people's personal data collection doesn't run into dozens of terabytes, while most people's media collection does.

You can get portable 5TB HDDs, one or two of them is usually more than enough for most people's own files. Plug them in and backup regularly, e.g. monthly, they last a lot longer that way than if you backup once and throw them in the cupboard.