FTDI have been anti-consumer for years - their last several drivers have introduced intentional instability and Code 10 errors for suspected counterfeit devices.
I think this is totally crappy. I see what they're trying to do (create market incentive for consumers to insist on real FTDI chips) but the reality is that it's just screwing over innocent consumers who buy a device.
It's similar to black market goods. A consumer wants the lowest possible price. It turns out the goods he bought are stolen property. He didn't know, but he is not waived of responsibility. If he was truly unwitting he will not be prosecuted, but the goods will still be repossessed, etc.
Basically, it is the consumer's drive for the lowest price that creates the market for these illicit goods, so they are not blameless. Additionally, illicit supply chains are hard to attack and often times the consumer "really should have known better", so one of the ways to attack that supply chain is by slapping the hand of the consumer who patronized it, whether or not they actually meant to buy stolen goods.
It's hard to deal with the "Well it was cheap and he was shady but I just didn't ask too many questions" purchases; responsibility is very diffuse, with everyone doing their best to avoid responsibility.
No, that's not the case here. Fake FTDI chips are going to be used in absolutely everything, products from Alibaba to name brand professional stuff. It's pretty much guaranteed not everybody has complete control of their semiconductor supply chain, and even if they do there's an incentive for all companies to cut costs where they can. A consumer buying a product doesn't make an informed decision about the type of serial interface in their devices, much less whether it's genuine or not. There's an expectation that the product will work and that falls on the manufacturer, who might not even be responsible either.
It shares the critical element of diffuse responsibility. Everyone can half-reasonably shrug their shoulders and say, "Well it isn't MY fault".
If a Sony product happens to have a fake FTDI chip in it, this is FTDI's way of incentivizing Sony (via angry customers) to manage their supply chain, because as you say there's an incentive even for Sony to cut costs where they can- perhaps by turning a blind eye when they get some suspicious batches of chips for a great price, and claiming ignorance later...
Everyone, in demanding the lowest price no matter what (all the way up the supply chain) bears part of the blame.
We are of course witnessing a visceral response on the part of HN voters reacting to my comment, who are exemplifying why this is a hard problem to tackle. It isn't MY responsibility!
> Everyone, in demanding the lowest price no matter what (all the way up the supply chain) bears part of the blame.
Your use of the word "blame" implies that you think there's some wrongdoing on the part of someone other than FTDI. As far as I can see, there's absolutely nothing wrong with the production and use of these clone chips except that they are being labeled with FTDI's trademark. They're piggybacking on FTDI's software work, but that's nothing that the government has an interest in stopping. If FTDI doesn't like people using their software without buying their hardware, they can resort to more traditional means like not giving out their software so freely or including DRM.
Your use of the word "blame" implies that you think
there's some wrongdoing on the part of someone other than
FTDI.
I don't see it as particularly controversial to say that, when someone selling an item claims it's a certain brand, I expect that to be the truth.
For example, if I buy an apple iphone I expect to get an apple iphone and if the supplier instead sends me a fake I regard that as wrongdoing on the part of the supplier.
Likewise, if a designer has specified an FTDI part and someone in the supply chain has substituted a fake, I'd regard that as wrongdoing.
Right. That would be the trademark infringement I mentioned. But aside from that, the fakes get the job done. Aside from who ends up getting the revenue, it's basically no different than if FTDI started producing a new revision of the product that had a different internal layout. Accidental second-sourcing doesn't really hurt anyone other than the first source. Everyone downstream of whoever bought the counterfeits is innocent, and even the company that procured the counterfeits has probably only made forgivable mistakes given that the counterfeits are near-perfect substitutes. The supplier of the counterfeits is guilty of trademark infringement, but is otherwise fulfilling all their obligations to provide the required component.
Accidental second-sourcing doesn't
really hurt anyone other than the
first source.
It hurts the entire electronics industry industry if I can't trust that a part is what it's labelled as, or if I can't trust a supplier not to deliver fake parts.
If your suppliers can substitute a fake FTDI part, why not label 10% precision resistors as 1% precision, or label 1,000-operating-hour capacitors as 30,000-operating-hour, or label parts that failed temperature range binning as having passed temperature range binning?
And the people who really lose out from this aren't the Apples and Samsungs of this world, who do enough business that the promise of future work can keep the suppliers honest - it's the small manufacturers and kickstarter projects that aren't big enough to have the leverage to keep their suppliers in line.
None of that corner-cutting is being alleged here. Nobody but FTDI has been complaining about the counterfeits. This has every indication of being more like a big pharmaceutical company complaining about generic drugs. If these clones are actually deficient in some way, then they're a much bigger problem, but that doesn't seem to be the case here.
As a consumer I had no way of knowing the motorcycle I bought was stolen. It came with the keys, the title, there was no evidence of thievery (most commonly broken parts around the ignition). The police said, "Wow we wouldn't have known this was stolen without running it in our database either".
I didn't know. But it still gets repossessed, and I'm still out $1k.
(True story, and my first brush with the laws around black markets)
Except in this analogy, someone sold you a Ducati with Pirelli tires that happened to be counterfeit. And suddenly Pirelli shows up and torches your bike because someone used a tire that had their logo.
These FTDI fakes aren't stolen, though. There is a massive difference between stolen, counterfeit, and clone. If the chips are only sold as "FTDI-compatible", then I would even say they should be completely legit, like Dalvik vs. Java.
If the person from whom the motorcycle was stolen discovered one night that it was at your house and just took it back, they would be committing a crime. Even though the motorcycle was rightfully theirs, there is a procedure to go through to get it back. Simply showing up and taking it back is a form of vigilantism. What FTDI is doing is also a form of vigilantism.
The point of restoring stolen goods to their original owner is to undo the damage to that owner. It's not to punish someone who innocently and unknowingly came into possession of the goods at a later stage. The latter person losing out is necessary in that case only because justice can't be done entirely to both parties at the same time (unless the intermediary responsible for the theft is found and forced to compensate everyone who lost out).
In this case, screwing the innocent purchaser of a chip that is a knock-off is entirely unwarranted. It's hard to see how it wouldn't be a breach of the Computer Misuse Act 1990 in my country (England), which would seem to make impairing the operation of the component a criminal offence here.
Edit 1: Clarify wording/citation.
Edit 2: If I'm looking at the right company web site, these guys are actually headquartered in Scotland. I'm not sure whether the exact same law applies in their jurisdiction.
You clearly don't know how the semiconducter supply chain works. All of this stuff is made in asia and changes hands 3 or 4 times before it makes it's way into a distribution channel. Unless you are buying straight from ftdi, there's a chance these can end up in your products. Same goes with most commodity ICs.
Of course it applies to me. I am not supremely confident I own no counterfeit hardware. Especially when it comes to some of the cheap cables and adapters I own.
The FTDI OSX drivers will still randomly cause kernel panics when the device is unplugged and a program still holds an open file descriptor for the serial port. While working with a company which was developing and selling products with FTDI chips in them, we made FTDI aware of this issue and attached many panic logs.
It's been about 5 years and they've still yet to actually fix it.
Oops, was dealing with a similar issue with the Prolific 2303, which was dealt with by using a newer driver, I haven't run into the issue yet with FTDI (which I don't as much).
The funny thing is that we reported this sometime in 2009 (might have been 2010), they "fixed it" and released a new version, but the new version only crashed somewhat less often.
OS X ships with an FTDI driver since 10.9. I suspect this one is made by Apple. Maybe because of this kind of issues. Do you still have crashes on 10.9+?
It's not a great situation, but where does FTDI responsibility for supporting counterfeit devices start and end?
If they let it continue more devices may hijack their USB VID, perhaps with bugs and glitches. Should they patch their driver to protect their reputation?
Your 'bugs and glitches' strawman is humorous because FTDIintentionally introduced bugs and glitches with the counterfeit devices, so they're certainly not trying to protect their reputation, just their profits.
FTDI's responsibility for supporting counterfeit devices ends a long way before tampering with them.
I think the moral high road to take in this situation would be to ignore the clones, but failing that, the approach which Prolific have taken (make the driver refuse to start on a known counterfeit device) seems a lot better than intentionally introducing bugs and not completely ludicrous like damaging the hardware.
Not to mention that FTDI's chips are not exactly known for their bug-free nature. If I had a dollar for every time I ran into a bug in their RS485 implementation, I would be a friggin' millionaire.
I think this is totally crappy. I see what they're trying to do (create market incentive for consumers to insist on real FTDI chips) but the reality is that it's just screwing over innocent consumers who buy a device.