And please don't give me "that was on consumer models only". I really couldn't care less. A company pulling off such dirty shit winds up on my eternal shitlist. No matter how great their "professional" products are.
I'm sure there are countless other examples.
Interesting that the founders of Superfish are both Israeli, I wonder if they both served with the Unit 8200 that was posted on here recently?
Even if Superfish was GPL licensed, the issue would've persisted. As article mentions even laptops coming with Linux pre-installed hardly come with the Kernel and userspace tools source code and you need to download them separately. Should Lenovo have preloaded Superfish as Linux kernel module and shipped it to its consumer laptops, United States Department of Homeland Security would have still advised removing it.
>In both cases higher authorities hired by people to protect them, protected them. Not FSF,
The FSF did a better job of protecting them with their unwavering stance. Neither would ever have caused an issue if you followed their advice to never use non-free software.
Pretty much the only things on my phone that aren't adware or spyware are free software apps.
Still, I think it used to be much worse or at least more obvious in the early 2000s. At that time Shareware was still a thing and it was normal for installers to add completely unrelated 3rdparty software. Probably that's also half the explanation why Windows used to be so unreliable, half of the ecosystem around was just random stuff.
JFC. I'm sorry, but this "article" can be dismissed on that alone. Software vendors already are inserting spyware into their "wares." Did the author do any research?
> The Free Software coterie is fond of insisting that words mean what they say they mean, and that is a profound misunderstanding of the nature of language. Such linguistic naïveté is not an asset in pursuing political goals.
Uh, take your snooty diaeresis and shove it where l'accent aigu don't shine?
Defining terms is par for the course in the general intellectual sphere. Academic papers, data sheets for microchips and ISO standards for programming languages all define terms and insist on using them in the defined way.
If that's not enough, there is outright misinformed claptrap:
> With all that said, the intent of the adherents to the term Free Software is to seek to promote certain freedoms for the users of software, by depriving the creators of software (at least in the United States) of the rights afforded them by Congress under Article I, Section VIII, Clause VIII.
What? Creators of software are deprived of freedoms by the Free Software Foundation? Not simply encouraged to copyleft their software, while retaining all rights to license it in other ways at the same time?
Crazy.
More abhorrent than that is the position that free software should be done in such a way that nobody claims credit, justified by a glib quote from Harry Truman. Harry Truman was a politician and, for a time, president of the United States. Presidents rely on other people to execute plans, and implicitly get credit for everything good that happens during their reign anyway, as well as blame for the bad. So it's easy to make seemingly magnanimous statements of that sort. I would rather say that the absence of a blame environment is more conductive to getting things done, rather than the absence of credit.
Software is a creative work made by specific authors, who undeniably deserve credit. Credit transcends even copyright. To falsely claim authorship of a work in the public domain isn't a copyright violation, yet it is plagiarism. It will forever be plagiarism, even when the work is millennia old. Falsely claiming to have produced a cave painting that is 30,000 years old is plagiarism. Some unknown prehistoric person is credited with that.
The western intellectual tradition is heavily steeped in credit. Theories, equations, industrial processes, objects and phenomena in nature, and other entities connected to ideas and discovery, are often named after individuals. Maxwell's equations, Einstein's relativity, Rayleigh scattering, Bell's palsy, Fermat's Last Theorem, Pascal's Triangle, Haber-Bosch process for ammonia production, Early effect in transistors ... credit, credit everywhere in STEM, philosophy, the humanities.
I find it obnoxiously odious for someone to dare suggest that free software developers should disrobe themselves of claims to credit. In free software, credit is your only reward, more often than not.
Oh, he's one of that kind of cunts. Someone with an axe to grind against FLOSS because he obviously hasn't studied history of copyright in addition to his misunderstanding of FLOSS.
Look at the giant tantrum basically every software vendor is throwing over the quite reasonable restrictions that the EU placed on spyware. I have to tell users about the spyware I'm inflicting on them and ask for consent first? End of the world.
But hey, if this guy's really a software architect then he might've cured my Impostor Syndrome, so I guess he has my thanks.
I can't argue with the specific usecase presented in that GitHub issue, I'm going to keep telemetry enabled. But if you argue that vendors "usually won't be inserting spyware", in the face of evidence like free software products seeing value in phoning home with telemetry data, I think I have the moral right to reject your argument in full.
Your software is either trustworthy, part of that being that it doesn't perform surveillance on me, another part being that it installs through well-known automation (apt, nix, etc). Or it will never become part of my trusted computing base (yes, that term is another blast from the past).
If I am forced to use it, it will end up in some isolated VM or throwaway tablet, with the minimum of access required for the functionality I need. I will consider it a second class citizen and generally deprecate it as much as possible (eg for communication apps, work to move the conversation to a different medium).
I can't think of anyone I know who has a car that needs to phone home. But that's a very limited sample size, so you know. Also, I'm most likely in a different market to you, we've never had anything like OnStar make inroads here into domestic vehicles - some commercial operators are using telemetry on their trucks etc.
But rest assured, if all our cars were phoning home, I'd be making a massive fuss.
For example, an insurance company in my country has recently launched an app that will "measure" your driving and offer lower premiums if your driving is "safe" according to their algorithms. It's obviously opt-in, but at some point, the difference between a discount for opting in, and a penalty for opting out, becomes hard to differentiate.
You don't have any rights to review their algorithms if you feel that they got it wrong, it's a combination of Hail Corporate and Hail AI, and context is lost because it's impossible to capture that. E.g., does heavy braking indicate you were driving poorly, or did you encounter a situation where heavy braking was necessary, such as the damn cat down the road that thinks it's invincible deciding to make a sprint for it in front of you? Is acceleration in excess of their defined limit unsafe? Or were you accelerating more than you normally would, because someone gave you space to turn into the road and you didn't want to needlessly hold them up, given their courtesy?
And given what I've seen of the FAANG algorithms, I don't want algorithms from companies nowhere near FAANG level making decisions about me. A personal favourite of mine was FB removing a comment of mine, because my sister said she'd totally marry my wife, on account of how, well, pretty damn awesome my wife is, and I'd replied "Haha, I'll fight you" - and FB had flagged that as "hate speech/incitement to violence".
Anyway, thank you for coming to my TED rant.
You don’t know anyone with a Tesla? https://www.tesla.com/support/connectivity
Or a Honda? https://hondalink.honda.com/#/
Or a Bmw? https://connecteddrive.bmwusa.com/app/index.html#/portal
Or a Toyota? https://www.supraconnect.com/app/index.html#/portal
You don’t know anyone with a Chrysler, Dodge, Fiat, Jeep or Ram brand vehicle? https://www.driveuconnect.com/
You didn’t hear about the remote control vulnerability 6 years ago? Chrysler recalled their entire fleet to fix it. https://www.wired.com/2015/07/hackers-remotely-kill-jeep-hig...
Definitely no Chryslers, Dodges, Fiats, or Rams. They are very limited in market reach indeed in NZ, unless we're talking vintage cars, then there's a few more.
I also had the, ahem, "privilege" of owning a GM built "Toyota" previously[1]. It was... shall we say, a cacophony of interesting and bemusing engineering choices - the boot/trunk lid was incredibly heavy, yet the latch mechanism was made entirely of plastic, and to turn on the interior lights, you had to turn the dashboard brightness dial all the way up, and then a bit more, and then the interior lights would turn on. Only took me 3 months to figure that out. Admittedly, was handy for turning on the interior lights without looking away while driving, but was not at all intuitive.
Oh, I do know someone with a couple of Jeeps though, the poor bastards.
I guess you could rephrase my statement as "I don't know anyone with a car built after 2016", although as far as I can tell in NZ, Toyotas, at least, ship with telematics as an option for fleet management, rather than a default.
[1]: https://en.wikipedia.org/wiki/Chevrolet_Cavalier#Toyota_Cava...
A lot of consumers (I would guess most) prefer to have default-on connectivity to provide services like theft protection, automatic emergency support in case of accidents, and up-to-date navigation maps, among many other modern conveniences.
As others point out, it's almost impossible to get a new car these days that isn't connected. Anything in the last five years in particular. It's pretty disgusting and one of the reasons I'm in no hurry to own anything new. I have a '97, '02 and '10 and they all work quite well for me and I intend to run them into the ground. And if I'm diligent they should last me until I am no longer fit to drive.
In NZ we tend to buy used Japanese imports because they're so damn cheap (IIRC, Japan has very strict rules about the age of a car for pollution control purposes, so they get shipped off to Australia and us after hitting that age limit, and we don't charge tariffs, unlike the Aussies), most people here would be lucky to buy a new car once or twice in their life.
Unless you go the usual route and start a building company, tick up a new Ford Ranger on the company's credit, transfer ownership to your spouse/partner, and then go into liquidation leaving behind devastated people who were trying to build their first home - and subcontractors who really need the $12k you owe them.
Cars Have Your Location. This Spy Firm Wants to Sell It to the U.S. Military:
https://news.ycombinator.com/item?id=26492322
One company wants to sell the feds location data from every car on Earth:
https://news.ycombinator.com/item?id=26511649
Military Unit Conducting Drone Strikes Bought Location Data from Ordinary Apps:
Nissans do, my Leaf does. They connect to a mobile network or WiFi and upload data.
https://www.nissan.co.uk/ownership/nissan-infotainment-syste...
I've been driving EVs for the past 10 years (LEAF -> Model X -> I-PACE) during which time I haven't used a gas pump. After reading your comment I had to go searching for this thing about ads playing while you're filling up. I found this Reddit post about being forced to watch ads before being allowed to even start the pump!
https://www.reddit.com/r/assholedesign/comments/819tcf/forci...
Oh. My. God. I'm disappointed. Not surprised. But disappointed.
> playing inside of the car
Yeah, that's definitely going at the top of my list of car "misfeatures" that would make me run screaming to another brand.
Goddamn, I don't want this post to be a screenshot for someone to point out that we can see the future.
It's definitely the way the industry wants to go - I mean, free data, why not? Bit like the FAANGs, dress up the data collection with some features people want.
Might just be a case that no-one I know owns a late model car :D We're big on older Japanese imports in NZ, god bless our lack of tariffs. Although RIP our local car manufacturing industry, god bless neoliberalism.
That is very much not true, on both counts.
There is a lot of nuance in this area.
Monitoring how well your own systems are working and how they're being used is one thing. It's obviously reasonable and necessary for a variety of practical reasons.
Monitoring how someone else's systems are being used, even if they happen to be running some of your software or incorporate some equipment you made, is something else. If you're no longer responsible for those systems and ownership has been handed over, including remote access or phone-home functionality means crossing some lines that maybe shouldn't be crossed, particularly not without the full knowledge and genuine consent of the person whose system you are communicating with.
This is like saying nobody cared about mass surveillance before Snowden. The problem is apparent, but the realization isn't evenly distributed.
The only difference between this Lua server and other Software Augmented with Additional Surveillance (SaaS) is your trust. You apparently trust them to not sell the data trove to a surveillance company (or to a VC who eventually will), but I see no reason to. Heck, I've been answering no to popcon for over a decade now, even though Debian is outstandingly trustworthy.
Even Backblaze, a company whose core product is securely storing your data, just recently suffered from an in house attack - apparently their security team didn't foresee the need to protect against javascript injection by their own marketing stooges. When data is there for the taking, most people cannot restrain themselves - the problem is endemic. The only solution is to assure the privacy of data, through means such as Free software, E2E encryption, and not collecting it in the first place.
I sure as hell am making a fuss about it, which is exactly why I drive a car that's old enough to vote (and pretty soon will be old enough to drink, smoke, and/or buy a handgun in California).
There is no need to believe. Just have a look at F-Droid and GNU/Linux repositories.
Where do you draw the line between spyware and legitimate feature?
Where do you draw the line between malware and unintentional faulty code?
Review time is a scarce resource, and a bad actor will always find a way to bypass automated checks.
We could charge access to Debian repositories for example. For example one repo could have bugfixes, and the broader audience could have a 1-year-old repo. Sure your can build from source, get a zip from a torrent, or find someone who hosts a free mirror of the paid repo, but as a company it’s much easier to get it from the official source, for security compliance. (Btw I’ve always said I’d be happy to spend $200 per year per employee for a desktop OS of macOS quality, so probably much less for our servers).
This would offer price discrimination: - Free for individuals who want to spin off a small website, - Paid when you need to prove compliance to security regulations, - Developments still go to the pool of mankind’s marvels, instead of being locked with a commercial license and dying with the company, - Customer can still fork or fix a bug themselves by building from source.
The line that separates legitimate features and spyware is blurry, especially when a bad default ends up leaking information of the user.
With the apparent lack of both viruses and telemetry and the like in libre software even with the laughable default security of linux distros (your home containing basically everything free to read/modify for any user app), I would say that FOSS has quite a good track record in this metric. Nonetheless, a stronger security should be developed/strived for because it still boils down to a trust based system, but it makes me happy that the default of most people is to not cause harm.
FLOSS does not prevent malware from existing, but it makes it highly unprobable. You can always find bad examples, but how frequent are they?
Android is built on Linux. Android-based TVs and phones are sold to consumers every day with spyware preinstalled.
So we have broad adoption of GPL software. Spyware is still happening. What’s the next step?
I can pretty much guarantee that none of those "smart" TVs are running GPLv3 code.
Hehe, and then there's the Affero licence... https://opensource.google/docs/using/agpl-policy/
The hypothetical scenario you describe is the one where you build Linux from scratch and you find a spyware. And this, AFAIK, is not happening right now, and one of the reasons is because it's open source and you can build it from the source code.
And a typical android distribution - eg the one installed on a TV - includes a bunch of other software too.
And the hypothetical scenario I want is where I buy a TV and it doesn’t contain spyware. It seems that the existence of Linux has, if anything, made that less possible, given that any old hardware company can grab a Linux kernel and shove whatever spyware they want on top of it before selling it to me.
This has nothing to do with "So we have broad adoption of GPL software". Android (AOSP) is Apache. We don't have "broad adoption of GPL software". This scenario doesn't exist.
If the software of the TV is GPLv2, you must have access to the source code (even with Apache this can happen to some extent). The TV manufacturer must give you the source code of the software running on the TV (with GPLv3 they should also guarantee that you can re-flash your devices). This of course is an ideal world, and it's the freedom FSF searches for.
This way, you (or someone else) can audit the source code and check for spywares. You might be able to rebuild and reflash the software.
Unfortunately not everything is open source. On Linux you have (for example) NVIDIA drivers that are closed source. You can put a spyware in there, even if Linux is GPLv2.
EDIT: btw, you first asked "What’s the next step?". I believe the next step is to raise awareness of the rights one has over GPLvX software and to be able to truly enforce those rights. Like if you ask Samsung for the source code of your TV and they don't give it to you, you can (like) go to the nearest police station and press charges.
I haven't quite grokked why Linus decided GPLv3 was bad, but he strikes me as a pragmatist vs. the FSF idealists.
Linux just wants anyone who changes the code to contribute back. He doesn't care how they exploit the software he gives them.
Also, you seriously argue that availability of a great kernel is somehow a bad thing??
Android is built on GPLv2 Linux. GPLv2 does not protect the user against locked down hardware and firmware. In contrast, GPLv3+ does protect the user. The user does not have the freedom to modify their software on their hardware, so there is much less freedom.
You’re really just supporting the FSF’s arguments against non-free software.
But it's kinda obvious that the lack ob a big and serious actor is the main reason why nothing seriously bad ever happend in linux-land so far. Everyone switching to Open Source would change this and the watchmen would likely be unable to protect their world.
I mean just look at the mildly bad things happening on mainstream-systems, which everyone knows about and still ignores them, because "that's just how it is". At the end people willingly use bad software & services, because their subjective gain is higher than the small price they pay.
This is a terrible document that is riddled with so many inaccuracies that it doesn't convince anyone other than the uninformed. If anything, the last few years have highlighted the need for Free Software and proved many of the alarmist warnings of advocates were prescient actually and not so alarmist.