…And on ones that do. Looks very useful for weeding out actual quality from the muddle. Would love to see someone doing this kind of work with other types of cables, too.
I could see this as a browser plugin - professional reviews of popular products where there are many, many brands of wildly differing quality such as cables and chargers. The problem is it seems that it'd need to be a browser extension, and most shopping-related browser extensions are pretty shady, so reputation would be an issue.
I suppose another option would be to just set up a site with reviews and make money with Amazon referral links.
I end up buying cables on Monoprice when I can just because the stuff they sell is usually good quality.
How will any company, especially a VC funded company, avoid problems and dilemmas seen by other online-submitted-reviews companies like Yelp that basically turn the company into either a fake review haven or a protection racket?
By focusing on what makes this reviewer notable: he's a verified buyer, systematically reviewing products, and with a resume to show he's worth listening to. We don't want reviews from any & all yahoos who sign up, we want reviews from vetted people who have a professional interest in checking numerous products with appropriate knowledge, and can write meaningful descriptions suitable for general audiences.
Personally, I'd very much like to see such a thing applied particularly to high-knockoff topics like USB cables (see lead post), camera batteries, printer ink, and other items where a lot of companies are providing a lot of low-price alternatives (some quality, some dangerously bad) to high-dollar brand-name accessories & consumables. For some there's WAY too many options, with HUGE variation of price & quality, to do anything other than either "buy cheap & popular and hope" or "pay thru the nose but it will work".
FWIW: if you sign up as an advertising associate https://affiliate-program.amazon.com with Amazon you can get up to 10% from every link you provide that turns into an actual sale. Build a website featuring appropriately curated reviews linking to Amazon and you could make a good buck - IF you can get enough readers. Over the years I've made a couple hundred dollars doing so, just by posting links when relevant to blog comments. Focused as a business? I'd expect good returns if done right, plenty to hire good unbiased reviewers.
A central review clearinghouse will never happen with cheap accessories like this. If model "1234-X" gets bad reviews, the same product will just be renumbered to model "9863-V2" and even the company name may change. So there's no way to keep reviews associated with the same product -- it's the same reason why there are no good review sites for mattresses across all retailers -- the manufacturers use a variety of names and product numbers for mattresses sold to different retailers, specifically to obfuscate them.
What's worse in the cheap accessories market is it is highly likely that groups of companies are buying and repackaging the exact same cables from the same manufacturer and slapping their own names and part numbers on them. So even if two companies are legitimately different, the product may not be.
companies doding bad reviews means that companies with good reviews will stand out, making selection easier.
as a consumer, i don't care about crappy products unless i end up purchasing them. i can avoid crappy products by only going for products with explicit positive reviews that i can trust. sure a grey area will continue to exist, but whitelisting is still effective for reviews, even if we can't blacklist.
Yeah, you'd need to cooperate with the manufacturer to "certify" only one exact product, and make sure that the certification is revoked when anything is changed.
The mattress issue actually sounds like a solvable problem, if you're willing to put in the effort. It's just not really worth it when you can also buy mattresses from IKEA or various online retailers which don't engage in those tactics.
That can be part of the reviews. "This product may look good, but Company X which makes it has a history of ninja-changes to their products."
A central (well, really federated would be better) is one of the most needed tech innovations today, because it will make the market for so many other products function more efficiently.
This is a really big thing. If you look at, say, a 10-port USB3 hub on Amazon, you will see the exact same thing, down to the boards (to say nothing of the casing), being sold under three or four different brands.
The thing is for a lot of low level electronics, you can buy "identical" products from a range of manufacturers, so while I know what you're describing happens too, it's tricky because there are also plenty of designs that "make the rounds" (e.g. lots of products are just tiny variations over reference designs from a chipset designer - you see this a lot with low end Chinese tablets, which are often almost-but-not-quite identical variations over what appears to be AllWinner and MTK reference designs).
Definitely. In this case, these are literally identical boxes I'm talking about (I have three from the same manufacturer, they put the same company name on the inside of the case).
I've used ebuyer (UK computer retailer) quite a bit and they moved away from their own review system to use Reevoo (the ebuyer one was better IMO however). Still Reevoo seems to have useful reviews in the computer/tech space.
Reading the headline, it sounded like astroturfing. Instead, he's being helpful and providing some very useful information from his experience as an engineer from google itself. What a thoughful fellow!
From a engineering point-of-view, he's saving himself a lot of product support headaches down the road when people try using bogus cables and then contacting Google when they fail to operate (or, worse, destroy the hardware under warranty)
It's probably true, since the Play store only sells one variety of cable, and Google has probably actually vetted it. In which case I can only see that suggestion as a helpful bit of advice for avoiding problems.
For sure. I'd hope they'll be self-aware enough to realize that it would be crappy to yank the reviews or punish the reviewer for recommending another store where the products aren't broken when Amazon is full of broken products. But I wouldn't be very confident of that.
I don't think Amazon worries about bad reviews, as long as they don't see only bad reviews. Bad reviews create trust in the reviews: Seeing bad reviews makes me reasonably certain that products with problems will get called out most of the time.
And often I end up buying products with bad reviews in preference to those with perfect reviews because bad reviews gives me faith a product has been reviewed "enough" by real people, and often the issues brought up are not important to me.
I'd say that for Amazon, these reviews are fantastic: Next time I want a type C cable, I'll check his reviews and buy one of the ones he gives good reviews - probably at Amazon. This is like popping into that store where you trust the staff because they talk shit about "their" own products when they're no good even if it means steering you to less profitable products.
For sure! This review[0] was particularly interesting. The vendor wrote back a nasty response about how their cables are tested at, and capable of delivering, 5A of juice--which is the point. The USB A device on the other end could get smoked.
Mr. Leung responds by painstakingly pointing out everything they have to do to fix their cables.
How much more are you willing to pay for that? If you're willing to pay a premium for tested and certified gear, why not just buy from your phone's manufacturer in the first place?
Because a lot of the Nexus 6P/5X accessories are sold out or have long delays before shipping. Also, I don't need another wall wart charger, just the cable.
Yeah, well it's easier to make things by the million when you don't care about testing, certification or product damage.
Part of the reason the cheap ones have so much availability is that they are cheap and easy to make and have little quality control -- even though they may not work.
Not really. I can buy super cheap micro-USB cables that work perfectly.
This seems to happen all the time when there's a new standard out. The disreputable companies whip together something shoddy and try to get those early sales before more reputable companies can move in. Meanwhile, companies like Amazon do nothing to stop this obvious fraud. Amazon should be stopping these sales.
Also, its disingenuous to pretend these are $2 cables. The only ones still available showing price are $14.99. These are much more than $2.
Arguably though, if a USB cable is $20 through Google, and $2 through an off-brand, I may want two of them, but buy six and consider any that don't work or stop working disposable. And still be saving money over buying from Google.
Consumers need to be aware of the risks they take when they buy bargain brand, but it doesn't negate the benefits of having bargain brand available.
That's a risk. But for the millions of people who successfully use $2 cables every single day, many of whom can't afford $20 cables, they're pretty glad $2 cables exist.
I might like to pay the premium for "tested and built correctly" but without also paying the premium for "the customer is too lazy to look elsewhere so let's gouge them."
I would like Amazon to provide better access to searching for and sorting commodity items. (and to remove from inventory defective ones on a limited basis)
For example, I'm searching for the least expensive bulk CR123A batteries. I'd like to sort by price per battery and buy the lowest with at least 3 stars. A fairly simple request.
But you can't. Amazon lacks (in most cases) per unit pricing information or the ability to sort by it. The batteries are cataloged in at least four different site departments meaning sorting is impossible over all of them. It's frustrating to have to try to deal with such a stupid problem.
You're probably thinking of AmazonBasics. They make a variety of products including USB cables, USB chargers, backpacks, monitor stands, and a bunch of other general stuff. I have several of their cables, chargers, and one of their backpacks. They are all fairly good.
really? How about working at a company that only sells products detrimental to your health? (read: McDonalds, Yum! Brands, Starbucks, Nestle, Kraft, etc... not to mention the entire tobacco industry, and plenty of other industries)
Why two star reviews instead of one for the cables that fail to meet the spec? Does Amazon / do people view 1-star reviews as bitter / exaggerated? Are they less visible?
They probably work for charging as well in most cases. The issue with most of them is they could allow too much current to flow, more than a micro-usb cable is rated for. Some consumers might appreciate this though, trading an miniscule fire hazard for their device charging faster.
It's not about what the cable can handle. Some of these cables can be plugged into an old device on the end that delivers power, while the cable (which isn't just wires) advertises the power delivery capabilities of a newer device.
Personally, I filter out one and five star reviews. To me, five stars means when you use the product, Jesus appears before you. One star means I ordered a cable and received a dead crab instead (pretty sure there's an XKCD that covers the last one).
But others don't agree with my rating system, so "ordered blue, but it was robin egg blue. Werst company evar!!!11" means one star. Five stars means "I work for the company that makes these, or I'm otherwise being paid to write this, or I simply didn't give a lot of thought to how rating systems work."
In this case, the cable fulfills most of the requirements, misses a some of them, but might work for your use case, so two stars.
Really? If a product meets your needs, is a fair price, and seems to be working fine after 30 days, you would not give it 5 stars? It would need to perform miracles to get 5 stars?
5 stars is "contained a 100$ coupon for amazon" or something that really makes you love a product.
The product quality of LEGO bricks – survives a century and so on – is worth 5 stars, as it’s far above expected quality for such things.
For a USB cable it would need to be amazingly resistant to wear – survives being folded and survives a dog chewing on it and so on – and it would need to be perfectly manufactured to get 5 stars.
In Hardware, usually very few high quality products – Miele washing machines, LEGO bricks, IBM ThinkPads, etc – reach something that could be called "5 stars".
If a product meets your needs, is a fair price, and seems to be working fine after 30 days, you would not give it 5 stars?
And then along comes the cable which, when I plug it in, causes a religious miracle to occur. How do I rate that one? Five stars is already taken. (Or maybe it should be two stars. I want to get my work done without being distracted by a prominent religious figure appearing next to my desk.)
To put it another way, when I purchase something I expect that it will work fine after 30 days and that it meets my needs. That doesn't warrant five stars. That's three stars: does what I expected, doesn't suck but doesn't stand out from the rest all that much. If it were half the price of cables of equal quality, then I'd give it five stars. But IMO the product has to be outstanding in some way to warrant five stars, otherwise why have a rating system at all?
OTOH, Amazon doesn't help with this. Click on "critical reviews" on Amazon, and three star reviews are listed. That's just broken. Three stars is middle of the road, not outstanding in anyway but not deficient, either.
So, no, given your parameters, I would not rate the item five stars. I realize the rest of the world doesn't agree with me, and I'm loathe to participate in such a skewed system (everyone gets a trophy). My concession is I'll give four stars to something I'd normally give three stars, just so it doesn't show up in the "critical review" section, broken as that might be.
I'm more inclined to reserve 5 stars for really good films and books (even if they're not in the truly great category). For a cable though? If it does it's job, appears well-made, and is a good price, I'm not sure why I'd hit it in the ratings. I suppose I can leave lower numbers of stars in a quixotic quest against review inflation but doesn't seem really fair to the products involved.
If it does it's job, appears well-made, and is a good price, I'm not sure why I'd hit it in the ratings.
And hence my quandary; I don't consider three stars to be a "hit...in the ratings". It does the job, doesn't stand out in any other way, three stars. But, as you point out, it's not fair in the inflated system that we have available. So which wins out, an honest review system that benefits consumers, or review inflation that benefits product sellers? I haven't decided, nor figured out any means of compromise.
As I once heard about bug priority, if everything's a Priority 1, then nothing is a Pri 1. And if everything that isn't complete shite is five stars, then really nothing is five stars. Five stars just means it doesn't suck. Then why don't we have just one star or no stars from which to choose? In other words, the precision of the rating system is not what the five choices would imply.
Unlike bugs, you are not triaging all products on Amazon together. If I search for a certain type of USB cable, I expect to see 5 star products. I'm not disturbed that Spirited Away is also rated 5 stars. I'm not confused and decide to buy a great film instead of a cable.
> "Engineers do not give out 5s. If you tell an engineer that the possible rating scale is 1 to 5 and they like you, they give you a 4. Because tomorrow they might love you and then they would give you a 5. They probably won't, but they might, so they give you a 4 so that they have room. We all are totally okay with this. But the world is also full of this other class of people who if they like you give you a 5. They're quite consistent about it. 'I like you. I see nothing wrong with that. I'm giving you a 5.'"
"We all are totally okay with this." Just be aware when you cross cultural boundaries. In one real-world story, a woman was asked, on a 10 point scale, how much pain she was in. My take of the story was that she interpreted it linearly, so said something like "2", where most people would have said "9" or "10". She was told to come back when the pain got worse, when it was already excruciating. My memory might be off - it's from Bliss's essay at http://www.radiolab.org/story/233143-pain-scale/ .
eBay really dug themselves into that hole when they introduced the star rating system. Basically they mapped everybody who wasn't actually known to be bad onto 5 stars. Which is why you get messages begging you to rate them 5 star.
Personally 5 stars would be when the box it came in had water soluble ink and it made a mess on your clothes, but the company offered to pay the dry cleaner costs. Even though there was no expectation for them to know that the shipping company would leave package in the rain / a puddle for so long.
Anything above "meh, it didn't crash and burn immediately, I guess it's okay" is 5 stars on eBay.
Uber is the worst, as I understand it from others, having never used it. It sounds to me like Uber has a binary system with other ratings just to confuse you: "ding this driver and end his contract if he gets anymore ratings like this, or...don't."
This is such a great service to everyone! I wish more companies let their engineers do that. Too few knowledgeable people call out the crap we're being sold.
Google doesn't have a strong policy against things like this. It mostly boils down to using your best judgement. "Would you like to read this on the front page of the New York Times?" is the usual question to ask yourself. I also like to ask myself, "is this factually correct, and am I providing proof?"
On HN, I tend to limit my responses because I usually end up being attacked for something I have nothing to do with, by virtue of some grudge the commenter has with Google. If you want to call me out for my own work, go ahead, I value feedback greatly in areas that I can control. If you have useful commentary, I'll try my hardest to make sure that people with understanding in that area see your comments. But if you just want to yell at me by virtue of who pays my salary, I find that all very unenjoyable and would prefer to not be involved.
I also notice that lurkers with blogs tend to like to take quotes from HN and convey them as the official word of Google out of context, which I also don't like being involved in. Learned that lesson my first week at Google :)
Finally, one also ends up attracting the crazies that email me threats against other Google employees. All those get forwarded directly to Security Operations and don't get a reply from me.
HN is a pretty big community these days and isn't quite the bastion of sanity that it once was. This limits how open many people will be with the community more than any policy.
Is there a simple test that someone could do with a multimeter to test these cables themselves? Or is it more than just having a pullup resistor on the right pins?
The average consumer has no idea if one of these cables will actually do what they proclaim. The only way to really know is when other customers review the product, which we know is barely useful most of the time due to fraudulent reviews. Great job on the part of this Google engineer for writing such clear and technical reviews of the products.
Now that an expert has called some of these cables out as not adhering to the spec shouldn't they be taken off of Amazon? Should they be reprimanded for false advertisement?
Amazon users do rely heavily on reviews, generally. I think with the Google engineers' reviews promoted to the top by the number of people who have marked them as helpful, market forces will finish the job. Few people will buy them, and those that do will have done so against solid advice. Amazon, of course, can delist things or not at their pleasure.
Except when sellers send their stock to Amazon, to be distributed from their warehouses (fulfilment by Amazon), I recall reading about their stock being comingled, such that one seller was being legally sought after for selling pirated DVDs that another seller had sent to Amazon.
Sorry I can't find a better source at short notice.
And in a nutshell, that's why people buy stuff from apple. imo, their brand is "here's a good thing, and we promise it will work". Amazon's is "you're on your own, but it's really fucking cheap!" That's not to say amazon doesn't have good customer service, but the whole thing is still shit. I just wanted a [cable|dongle|hdmi adapter|battery|plant spray|charger] that was what it claimed to be and worked correctly the first fucking time. The fact that you have a good return policy doesn't fix the fact that I still don't have X.
Funny story. I wanted to buy a cheap iPad for my son. I could have got one from Amazon but after hearing so many stories about scams or refurb devices or devices with someone else's state on it I decided not to deal with that headache and ordered direct from the Apple website.
Upon turning on the device for the first time it showed a screen saying it was controlled by the LDS Church and wouldn't proceed through guided setup until I acknowledged that. After some time on the phone with customer service, we determined there was no way to remove this configuration. I ended up having to return it to an Apple store. The rep there said "yeah that happens sometimes" (someone pulled from the wrong pile in fulfillment most likely). So sometimes you're damned if you do, and damned if you don't.
If anything Amazon's customer service is better - to resolve an issue like this for items they ship I don't have to drive anywhere, just put the box back on my front doorstep. I wish they would make it clearer when you're buying things from them, and would do a better job of associating compatible products. This seems like data they should be able to have, and UI they should be able to build.
Annoyingly in places it looks like data they have and turns out but to be. I once bought a desk lamp which below it included the "people buy these together" box, which showed the llano and a bulb.
Frustratingly it wasn't the right bulb, but I just fed a further data point into Amazon's model, and reinforced their belief that its useful to list those two items as a pair.
Am I correct in assuming this is some kind of enterprise managed provisioning? I am surprised that they both ship it that way, and that it can't be removed. I guess Apple doesn't care about killing resale value (in the case of it being unremovable), but you'd think these kind of screw-ups would be a reason not to ship it this way. Do the restrictions extend deep into the hardware, or are they just saving the enterprise customer like 3 minutes per device pushing a profile over USB?
It's the Apple Device Enrollment Program (http://www.apple.com/business/dep/) and it allows enterprises to exercise "mobile device management" at the time of device activation. Apple says that once devices are enrolled they can't be removed (because, presumably, rows cannot be deleted from a database after being added... >sigh<).
It's not that easy. DEP is meant to also protect private property, so it piggy backs on activation lock to bind a device to a certain (company) Apple ID. Obviously, there must also be no way for any person to social engineer an Apple Care representative over the phone to get DEP lifted from a device, and this is why Apple Care can't do that, period. The only possible way to remove that is going through the private key bound to the MDM for that decide, the same MDM that requested DEP while buying the device in the first place.
(Obviously, much like for activation lock, it is indeed possible that someone with "root access" - so to speak - to Apple systems would be able to manually disable DEP; but the point is that this possibility isn't exposed on any user-level user interface)
I was not aware that there was any mechanism, even for the rightful owner, to have DEP membership removed from a device once it's enrolled. I'll have to look into that.
>Obviously, there must also be no way for any person to social engineer an Apple Care representative over the phone to get DEP lifted from a device, and this is why Apple Care can't do that, period.
That's not obvious to me. Those same people can be social engineered into giving someone access to your data, which is an order of magnitude worse than a device wrongfully moving accounts.
But once there are 500 reviews on a product it basically gives you that confidence.
This isn't a troll comment. Ben Thompson makes the argument that the reason why EBay, AirBnB, and Uber work is because you basically are replacing the trust you had in a brand with the trust you have in community reviews. He calls it 'Aggregation Theory'. I find it rather compelling to explain why these types of marketplaces work.
Problem is another supplier can claim to sell the same item and benefit from the positive reviews. I've seen a lot of this with unbranded items like USB chargers for cars.
Bingo. You always need to check the most recent reviews because of this: a solid history of 5000 5-star reviews is meaningless if the past 200 are 2-star reviews saying, "Not what I expected".
Exactly. And then you look up and you're 20+ minutes into skimming reviews in order to buy a $3 care remote battery or cable and wondering why the hell you didn't just drive to target / walk over the apple store because you'd be done with the errand by now.
I ran into exactly this earlier today. I needed to get a new Macbook charger and just buying one directly from the Apple store nearby was faster than (as I first attempted) trying to find one on Amazon without "this is actually a counterfeit imitation" warnings in the reviews.
At this point I just flatly don't trust Amazon anymore for small electrical devices, because there's too much of a chance (even for things labeled with brand names) that it's actually a cheap knockoff that will short-circuit and set something on fire eventually.
Between 2008 and 2013 when I had an iPhone, their USB cables would regularly break down on me. The problem was the white casing around the wire would break down and start to thread. So I think the rule of Apple promising high quality is true a lot of the time, it baffles me that they sold that low quality cable for so long. Ever since I switched to decives that support Micro-USB, cheap cables have yet once to break.
Yeah, Apple uses rubber instead of plastics to wrap their cables. And it appears to have two problems. The common one with rubbers that they deteriorate by moisture. And it seems like the thickness of the layer is about as thin as the rubber's consistency allows. Meaning if you chip it in any way, the whole cable wrapper will slowly start to unravel.
I've now bought a couple of Nohon braided cables from Aliexpress and they seem to hold together nicely. But of course them being cheap Chinese cables they are finicky with iPhone 6+, but work fine with my 5s.
Every time I read something like this, I wonder "what the hell are you people doing to your Apple cables?" Now, I'm not saying you abuse your cables or anything. Maybe it was bad luck, maybe it's me just being extra careful with cables (doubt it, they're disposable in my book). But I've got a whole box of white cables. I finally threw an old 30-pin away that was so old it had the latches on the side that you had to squeeze to get the cable out. 30-pin connector was cracked, but it still worked. Threw it away only because we only have one 30-pin device left in the house, and the cable was sort of broken.
My wife did finally kill the Lightning cable in the car that was a few years old. One cable failure in probably ten years. That and $4 will get you a coffee at Starbucks, so take it FWIW. I just wonder why some have better experience with Apple cables than others.
>> The average consumer has no idea if one of these cables will actually do what they proclaim.
Isn't that what standards compliance is supposed to be for? Shouldn't the product have to be tested against the standard in order to get labeled USB X.Y ?? Or doesn't USB have such a requirement?
The government is the one that should apply the laws to protect consumers and fair competence. Once something fishy has been detected they have to act.
Does Underwriters Laboratories do that kind of testing? I'm honestly curious-- I don't think I've ever seen their logo on anything as cheap as a single USB cable.
If so, you can look for their logo and have... some confidence at least.
It's the same for Android. It's open-source, but you cannot say that it's derived from Android unless you pass Google's tests (and according to Chinese independent manufacturors, you need to know people from the inside to advance your application).
And to put in terms of real-life - in Shenzhen, China I've seen people selling whole spools of "QC PASS" stickers, as well as any logo or guarantee label you want.
To be fair, QC pass stickers are going to be sold like that just because you'll need a lot of them if your only way of identifying pass/fail is via a sticker. I don't see why you wouldn't buy those en masse, and use them on a real assembly line -- especially in a country where most plastic goods are produced.
Having worked at a company shipping a USB cable with our product, I know we explicitly couldn't use the USB logo on the cable without additional (expensive) testing. So our cable just didn't have the logo.
Theoretically, people who purchased complain to their state attorney general or report the manufacturers to the FTC. There might be some recompense under consumer fraud laws. If enough people had bought them, there might be enough people to form a class action (doubtful though).
The problem is that amount of $ involved here is just not something anyone is likely to get up in arms about. It's $5. Most people will just write it off and never think about it again. No lawyer would bother either unless there were 10,000s of consumers willing to come forward and complain.
Just one of those shitty situations. Caveat emptor.
It's $5 for the cable, but you could potentially damage the USB C port on a laptop, and some rather expensive fruit branded laptops now charge solely via the USB C port, which you could argue destroyed the value of the laptop. That sounds like it might make a small-claims-court case sensible to me.
I was thinking the same thing. And some of these guys would not be able to capably defend very many small claims suits at one time. I always wondered what would happen in that case.
Very likely they're not even going to be around under that name by the time you file your suit, but instead operating under a different name. Good luck finding them in Shenzhen in order to collect.
I don't want Amazon being the product police, in the same way I don't want them refusing to sell products that they think compete with them (Chromecast). Amazon should be neutral about the products it sells, instead giving people the tools they need to make buying decisions, like reviews and ratings.
We've done a similar analysis of USB 3.0 host controllers and the state of their drivers cross-platform. It's amazing how bad the state of USB 3.0 is, even 5 years after devices started shipping. Even the latest Intel host controllers (8 & 9 series) initially shipped with bad drivers. I look forward to getting our results posted.
Absolutely. Data corruption from VIA based host controllers was probably the worst issue we've seen, and the USB 3.0 ASMedia host controllers are probably the worst now that VIA was able to solve their issue with a driver update.
Most hardware on the market today was released before the xhci spec 1.0 was finished, and I agree that ALL hardware on the market does not even properly implement the spec that was available during its design. Fortunately, most issues can be worked around using quirks.
The hardware is the root of the problem, but I would also add that the xhci_hcd kernel module, even in the latest kernels, is well behind Microsoft in terms of handing these device-specific quirks. The Microsoft USB 3.0 stack introduced in Widows 8 solves just about every host-specific issue we've ever seen. ASMedia is the only host controller vendor that bothers to provide their own Windows 8/10 driver (both for their USB 3 Gen1 and Gen2 chips). No one else provides an alternative to the Microsoft provided usbxhci.sys.
That said, all of our tests have been limited to bulk transfers, and we're only using existing user space libraries (WinUSB, LibUSB, IOKit). We do push throughput, we have tight latency requirements, and we keep a large number of buffers queued at all times, which does push the host controllers further than your typical storage device. (We make streaming data recording equipment for electrical engineers and embedded developers)
I don't want to bash xhci_hcd though; their progress has been fantastic, despite the minefield of problems created by all these host controller vendors. I would like to say a special thanks to @sarahsharp, for all of the hard work put into the xhci host controller module. Without Sarah, I don't know who will answer when the xhci_hcd doorbell rings.
If only there were more of this stuff! The kind of information in his reviews is technically detailed but clear enough for the layperson, and it answers the real question I have in mind when I'm looking at the 1-3 star reviews: does it actually work for what I want it to do?
The strange part of these reviews though is that counterintuitively many of these cables would be considered to "work for what you want it to do" when a spec-compliant cable would not.
Usually what people want is to "make my device charge fast" - in which case the cables not implementing the spec correctly work while compliant cables would not.
Of course, there is a reason that compliance exists! Chances are you are plugging this cable into a wall-wart made in the past few years with higher tolerances than needed, and things will work out fine. Until they don't. Then you may burn out a PC USB port, or potentially I suppose catch a cheap chinese wall wart on fire.
At least that is my current understanding having spent the past few days attempting to decipher how this all works. The correct way to go about this transition to USB-C is to toss out all your USB-A chargers, buy USB-C chargers, and if you need to adapt; adapt in the USB-C -> USB-micro/mini direction vs. the USB-A -> USB-C direction.
The tldr on this is basically if you plug into a USB-A charger, you should not be able to "rapid charge" at 3A. If you are, you are out of spec and could potentially damage something you plug that cable into. Just give up on the dream of using your old chargers and expect to replace them with chargers that implement USB-C there. If you need to charge USB micro devices, invest in some USB-C -> Micro cables (or adapters) instead.
Just slapping the "the other end is 3A capable" resistor onto the cable isn't good enough, because a cable has no idea what's on the other end.
You can still rapid charge with a compliant type-C/A cable, but it needs to be negotiated through one of the other specified mechanisms (that directly talk with the other end and thus can figure out what the type-A charging side can actually provide).
The only physical USB-C device and cable I have seen are the ones on the new Apple TV. Apple even helpfully included the cable with the developer devices.
What I find interesting here is that Benson is essentially using his employer as a reputation guarantor.
I guess depending on the company's HR that could fall either way of
a) This is great, we're getting nice PR
b) Mr Benson, we need to have a quiet word about what it is you do here and what Google is not in the business of.
Really interesting.
edit: I assumed this is part of his job. I just find the publishing of the result along with Google's name interesting.
If/when I evaluate software for a solution as part of my job, I wouldnt dream of publishing my findings and referencing my employer - but I like what Benson has done.
My assumption is that he's working on some sort of project where he's buying many cables on Google's dime to test them, then reporting his results on Amazon.
My worry would be the cables that get good reviews being switched out for crappy cables down the line. I've had problems with this for cables and adapters before - I'll re-order from Amazon and get a slightly different product that is not 100% compatible with the old one.
Isn't there some kind of SKU/ISBN-like identifier that you could use to prevent that? Enforcing consistency would be tricky, but there should be certain quantifiable properties, especially when proper specifications are involved, to rate against.
Unfortunately, there is not. Amazon will often group similar unbranded products together under the same ASIN.
(This was a particularly serious issue for Arduino a few years ago, when Amazon started listing genuine Arduino products under the same ASIN as third-party clones: https://blog.arduino.cc/2013/07/10/send-in-the-clones/)
Even large, US-based companies sometimes change stuff without updating the SKU. This was a huge problem back in the day with getting WiFi adapters that worked in Linux - I've seen chipsets changed out without updating the SKU, even as far as chipset brand. If you were lucky, there'd be a hardware revision listed on the product near the serial number, but sometimes not even that. I think this has happened with a lot of the RTL-SDR adapters as well.
As a more dramatic example of companies changing the product without changing the SKU/whatever, the faulty GM ignition switches were changed mid-production (from the bad design) without an update to the product serial number. Confused some of the engineers investigating it at first.
Might be doing this officially for Google, given the official-sounding tone and the amount of time and money it must take to buy and analyze every adapter/cable on the market.
Yet, if we go to a conference and Benson The Google Engineer is there and we happen to find out he's been testing cables and we ask which cables he finds to be the best, it's OK.
This is very common among Googlers, and it bugs me because so many buy into it without further question, but in every case here, the reviewer also includes a detailed analysis describing how he came to his conclusion, rather than merely implying authority, and his reference to Google seems only to be in his capacity as working on a product team that might explicitly care about quality of accessories in a first generation technology (if USB PD can still be considered in its first-gen phase).
Would not be surprised at all if he expensed every single one of these purchases. He's doing God's work, we shouldn't be complaining here. :)
I'm not complaining at all. I just find it interesting that at one organisation this 'reporting-out' is accepted if not proactively encouraged. A more old school organisation might come down hard on an employee.
This might actually be part of his job. Why would HR have anything to say about it? Engineers do tech conferences as 'employee of XYZ,' how are product reviews any different, assuming they're in context?
I would think it's probably sponsored in that they want their customers who buy cables for their pixel and Google stuff from Amazon to have a good experience and actually get working stuff.
> If/when I evaluate software for a solution as part of my job, I wouldnt dream of publishing my findings and referencing my employer - but I like what Benson has done.
I think that's the whole point - for Google to chime in on cable quality so that customers know what to buy for their Google devices.
I'm guessing that Google saw that a large percentage of their customer support issues were caused by poor cable quality, and that most of their customers bought their cable from Amazon.
This seems like a way for Google to pre-empt that customer support cost by making sure their customers are buying cables that they know are good.
Most of the time other companies hand-wave their way around this problem by offering 1st-party cables (usually at a fairly high price).
It kind of makes sense: he worked on products that use USB-C (Pixel laptops), but if the market is littered with bad cables USB-C will have a bad image which will be bad for the products he worked on.
So what he's doing is in the interest in his employer, and for the benefit of the products he worked on. He's not doing completely unrelated stuff, like reviewing routers while claiming "I work for Google so I know how shit works!"
Additionally he's preventing Pixel owners (Google customers) from buying bad cables on Amazon then blaming Google because the laptop doesn't charge. So it really looks like he's doing his job.
People frequently buy multiple cables for various reasons. For example, I have three microUSB cables for my phone, which only came with one - for charging at my home computer, work computer, and in my laptop bag.
I do the exact same thing. Keeping one at home, one at work, and one in the bag means one less thing to worry about leaving behind/not having when you need it.
I don't see why shipping a good cable with the product makes a difference. My house is full of USB chargers, because I want to be able to charge multiple devices several places in the living room as well as our bedrooms. Then there's the portable battery packs etc.
One cable per device is insufficient for a lot of us.
Then you create a list of recommended cables. You don't bury that data in amazon reviews - if that's the intended purpose of releasing this information
They do have a list of recommended cables -- the ones that they sell. Regardless, that doesn't mean that the Google Store is the first place people go to get a "standard" cable, especially as it isn't necessarily the cheapest.
To many, a cable is a cable is a cable, and preventing someone from buying a commodity part that might damage their phone is a good objective, no matter how it's done.
Posting Amazon reviews might not be the optimal way to get me to buy the "right" cables, but it's a pretty effective way at preventing me from buying the wrong ones.
If only there were some way for google to get high quality usb-c cables to customers! Like not charging $13 plus shipping for a 1m cable (a ridiculous profit margin), and keeping them in stock. Or hell, throwing 5 of them in with everything they sell that uses usb-c. I mean, that would have to cost dollars. Whole dollars, on a $500 phone!
The fact that the market is full of low-cost USB-C cables that aren't reliable may suggest that the cables cost $13 because it's actually a $13 problem to get it right. ;)
Oh they are produced on a massive scale, and they are cheap - in Shenzhen, it's easier to find them now than the ordinary ones. Every other vendor has them stacked up to the roof.
The ones on Amazon are all $15-$20 and even monoprice is $10
Since Google isn't the manufacturer they are not making a ridiculous profit margin.
Whole dollars on a $500 phone is more likely to be Google's entire profit margin, though. The profit margins on phones is razor thin unless you're Apple.
Just because monoprice sells them for $10 one at a time, I don't think that tells you much about the cost to put more than one in phone package you're already handling when you buy millions at a time.
Monoprice also has volume pricing and it's also $10/ea even when you're buying 50+. The simple fact is nobody that makes cables is selling them for cheap currently. The volume isn't there yet, and they still need to recoup costs from spinning up a new production line and a new SKU and all that.
And LOL @ the idea that Google sells millions of phones.
Manufacturers got to lengths to squeeze pennies out of the Bill of Materials (BOM) in products like these. I remember listening to some conversations and being blown away at the tiny savings here and there, and our product was over $100.
But so what? Why is google selling phones -- to promote android and google as a brand, or to make money? If it's to make money, fine, then eek every penny out of that BOM. But they claim it's the former, and cheap cables are apparently a brand image problem, so they should solve it by not cheaping out on the large investment they've already made.
Google promotes Android by selling the device near cost. Even if they're not making money, they aren't going to sell devices at a loss (-$5 times a hundred million devices is real money), which means the only way to increase the manufacturing cost is to increase the sale price.
When it's only a dollar or so you might expect the customer to want to pay the extra dollar in exchange for the benefit, but in practice the device is going to be priced at a threshold like $499 and the psychological difference between $499 and $500 would measurably drive down sales.
But the bigger problem is that this is hardly the only facet of a device you could improve by making it cost a tiny bit more. Add them up and it would add $100 or more to the price of the device. So you can't have them all, but decide against any one in particular and everyone wants to know why you're cheaping out over a buck.
Some USB C cables are actively marked. I.e. they have a chip in them that says "I support power delivery" and so on. It can also be used to create Apple-style authenticated cables which is a bit shit.
Anyway if these cables contain such chips $13 seems reasonable.
At least for me, the Google name adds zero credibility (nobody can tell if you are a dog, or unemployed, on the Internet). The detail of his reviews, and their high rating, is what sells it for me. That there is a plausible explanation for why he's doing it is just icing on the cake.
I actually had the same reaction. He has a vested interest in pushing specific vendors that work well with his employers products.
This means because of his reputation (Google engineer) it's easy for him to dismiss one cable manufacturer and support another while getting some form of a kickback either directly or indirectly from said manufacturers.
Sure, he's doing his job, but he's also in the position to pick and choose winners. As such, I would discount his advice based on the possible bias towards products that could possibly line his own pockets.
>This means because of his reputation (Google engineer) it's easy for him to dismiss one cable manufacturer and support another while getting some form of a kickback either directly or indirectly from said manufacturers.
That's always a risk, but I think it is more of a risk having used his real profile and Google affiliation. On the face of it, I want to think that Google pays Benson well enough that he needn't go tell lies on Amazon as a side job.
>As such, I would discount his advice based on the possible bias towards products that could possibly line his own pockets.
This isn't like for example "audiophile" cable reviews where people just make crap up in ambiguous terms hoping to sell a thousand-plus dollar pair of speaker cables knowing full well that no one can argue sensibly about the warmness/depth/color/scent of the sound.
He pretty clearly explains his major complaint being the lack of a proper 56k pullup so that the cable identifies itself correctly. This is testable by almost anyone. It makes very little sense to publicly tell a lie that can be verified so trivially.
The review is extremely negative, explicitly telling people not to buy Nest products. Those are (and were at the time of that video) a Google product, as Google acquired Nest in 2014.
Brad still works at Google. I'm sure Benson will be fine.
I used to test Xbox 360 network connectivity, during which we'd regularly buy consumer routers, hubs, and switches and ensure compatibility with Xbox 360's network stack. (There were different types of NAT which could cause problems, and some had problems with low MTUs.)
It never occurred to me, or anybody else working in the lab as far as I know, to post the results of our testing on Amazon reviews. We just did the testing.
Seems to me having stuff like this is a good thing... I think a lot of headaches could have been averted had you been able to, or advised to do likewise in posting reviews in common sales channels (Amazon, Newegg, etc).
I find the willingness of Google to put their name of this very interesting. This does not seem to be a "my views are mine alone and not my employer's" thing. I wonder about the ethical and legal aspects of this. What stops Google from bashing Microsoft products this way? What happens if he makes a mistake in testing, and the loss in reputation causes a small company to go out of business?
I would think this is appropriate disclosure, and probably part of the job... to say "this is great" or "this sucks" without disclosing that you were paid to review the product and/or the context it's unethical.
This is entirely ethical, and the context is stated clearly... I think it's great.
Other companies have employees just like that. Sadly, if those folk pull a stunt like that, someone will be having a word with them.
I say "sadly", but I've seen the flip-side of this, too. A properly-formed query on usenet will turn up some back-in-the-day posts from Microsoft employees (posting with a microsoft.com reply-to) who were real assholes. (And not to pick on Microsoft, though they certainly have it coming; just the first counter-example that came to mind.)
So I guess I'm conflicted. Yeah, Benson's being really helpful and it looks good for Google, IMO. But with a different employee, it could turn really ugly if someone's back is turned.
I've definitely seem some embarrassingly bad attitudes from Googlers on social media. (One recently which basically said every serialization protocol that wasn't Google's was inherently harmful, because Google didn't make them.)
But obviously, there's many people like Benson who provide incredibly sagely advice that do a service not just to Google's own customers, but competitors' like Amazon's customers as well.
Once a company reaches a certain size, they have way too many people to police all of them across the Internet, or ensure all of their hires contribute meaningfully online.
It would be good if he did resistance tests as well. I killed my phone's battery by absentmindedly using a shitty cable that was basically tinsel wire.
(details: I was full-time tethering while plugged in. I assume the charge controller thought the battery was charging but it was actually slowly discharging)
That info deserves being tied to the model of the phone, at least. Not because that phone's thus inherently bad, but because that's a buggy design. I understand you're saying that the cable fooled the charge controller into letting the battery fall completely flat? Wow.
Getting said info to the right place so next version's charge controller is fixed would be nigh impossible, but dropping the info in lots of high-exposure places might be a start.
It was an HTC EVO 4G (non-"LTE"). Quite old at this point.
I can only surmise what actually happened. I was full-time wifi tethering for a month or two and the symptoms along the way seemed to point to this cause. The phone eventually stopped taking power at all, until I plugged it into a USB port instead of the OEM 1A charger. At which point I realized the higher current was the problem, and investigated the cable.
It actually felt like a really nice cable - very robust and strain-reliefy. But apparently that's because the whole thing was mostly plastic and essentially headphone wires. Lesson learned!
If my theory is correct , it's a pretty stupid design. But I feel like there's probably many weird corner cases in consumer electronics that don't matter because 99% of people won't find them, and the ones that do won't realize it's a bug, and by the time anybody could put it together that specific model has been obsoleted anyway.
I can't even remember the last time I used a USB cable on my phone for data, the only time I ever plug it in is for charging, so I don't care too much about signal integrity as long as it will charge my phone without damaging it.
When I do care about signal integrity (i.e. connecting a USB hard drive to my computer, I use name-brand cables)
Amazon reviews become useless as Amazon allows third-party vendors from China and HK to flood the original listing with clones.
So people buy based on the review and then they find out the vendor shipped a clone, so then they give a bad review but it's not for the original listing.
Amazon made a huge mess with that, they should never allow third parties to sell on the original listing.
Another fun thing I saw on Amazon a few months back was items for sale with reviews for another product. Shopping for a small FM transmitter, I found a listing where every review more than a few months old was for a roof-mounted car antenna.
I suppose there is a Ship of Theseus problem when trying to detect tricks like that automatically - surely sellers should be able to reword description, update product image, and change the product name slightly, but where do you draw the line and how do you handle that? Heh.
I've had a similar thing, where a vendor had clearly kept the same listing for many years. They were the same type of product (tablets) but the specs referenced for the oldest listings were quite hilarious. For a long time this seemed to have been fine, sort of, as the shipped products consistently were better than those described in older listings.
Until they weren't. We bought one that was a substantial step down (screen resolution in particular, which was exactly why I'd picked it).. Got a refund, but Amazon simply didn't seem to care that the reviews had no relevance to the current model, which was a let down (though a lesson to read the reviews closely and pay attention to their age...)
I'm not sure why you're being downvoted (perhaps for singling out China/HK?), I've had problems like this on Amazon as well when buying chargers/cables.
I'm confused--if the person purchased from Listing A, but the vendor shipped Product B, are you saying the user does not have the option of leaving a review for Listing A?
I wish someone would do this for micro-usb2 cables. I've bought so many different ones and found that they don't carry enough current from the charger to properly charge my Nexus 6.
Yeah I'm still not sure how that works. My phone will randomly charge really quickly on the wall adapter for weeks, and then at some point it decides it's going to charge slowly for the next few weeks. Same adapter, same cable.. it seems to be wacky. It's all original from Samsung.
The micro-USB connection is also really loose with this particular cable<->phone combination (cable is fine with other phones; phone is fine with other cables). Used my old Nokia micro-USB for this reason and charging was always slow, then connected the Samsung one and it was a lot faster.. for a while.
I never knew the cable could do something like this, unless it's something like being way too long. From your comment I take it that you do need a certain amount of copper in those wires to work? So more expensive cables might actually be better? (I always laughed at "gold plated extra quality HDMI cables of 30cm for 30 euros because HDMI is a digital signal.)
One part is the wire gauge. You need something like 24awg (its the American system, don't ask me why) on the power pins (higher number, thinner wire, iirc). Most use 28 on both power and data pairs.
Then the charger has to actually follow the USB spec, not the very similar Apple one. This means either a microchip that can handle the USB handshake, or a resistor across the data pins at the charger end to signal that this is a charger (Apple has one resistor on each pin, with a slightly different resistance).
Now on top of that you have Qualcomm pushing their fast charge system that involves a chip at both ends, and higher voltage than the 5V that USB normally uses. The higher voltage allows for more watts without upping the ampere (V*A=W, iirc).
Note that the Qualcomm system is similar (but i don't think they are interchangeable) with the USB3 system (the latter can go all the way to 20V, but that requires beefy cables).
Frankly the more i think about it the more i expect to hear about someone getting electrocuted because they mistook their data cable for their dumb laptop charger.
I bought the TechMatte 2x USB-C to Micro-USB item that he gave two stars for use with my 5X. At least one of the items I got was actually defective. The other worked but charged very slowly. I wish I had seen his post before buying!
Now I am nervous because I subsequently bought a cable that he hasn't reviewed yet...
This has me wondering if there is some way to incentivize good reviewers. There is Consumer Reports and some more tech-specific sites but there is still a long tail of products that you have to buy without good reviews.
It's probably more effective to try to weed out the reasonable brands.
For instance, I don't think Amazon is going to sell broken crap using their AmazonBasics brand, and they still charge quite a bit less than Belkin (I don't think they sell USB C yet, but that's besides the point).
Didn't burn down Google campus or shoot my dog, +1 star.
Joking aside, I'm somewhat more likely to read a two star review vs a one star review because a lot of the one star reviews are whiny rants and not actually helpful.
I have one of these, and it happens to work fine (seemingly anyway). Unfortunately he hasn't reviewed anything similar that is actually in spec, so I don't really have a choice here. (other than buying new cables/charger/car chargers, which I deem wasteful)
I just bought an HDMI "hub"/duplicator, and just now realized that now all HDMI cables are the same - some support 1.3, others 1.4 (or were these the numbers and such).
Maybe it's okay, so far the Wii/PS4 and ChromeCast work fine through it... I'm just posting this here, as no one informed me about possibly complications at BestBuy, but the great folks at Staples actually forwarded me to BestBuy to look for such device.
My father told me a story about service in Turkey, basically if you go to a store and they don't have right away what you need, they may go out find/buy it for you, just to please you. Or buy you a coffee, while waiting in s shoe shop.
Whether he exaggerates a bit or not, I like this approach. Yes you might lose customers, but you'll always leave a smile in good customers. They might come back, just for that little appreciation...
"USB" and the various USB logos are trademarks of the USB-IF. I was under the impression that non-compliant products can be sued for trademark infringement by the USB-IF. Sure, they could be sold without the logo, but not being able to put the word "USB" in their product description would make them unsearchable and kill their online sales...
I miss a simple page with information om how I can perform tests, like on the USB-C cables, myself. I find reading specs to take to much time to be able know how to perform them.
Example is the 57MB zip-file for the USB-C-spec: http://www.usb.org/developers/docs/
397 comments
[ 4.1 ms ] story [ 447 ms ] threadI suppose another option would be to just set up a site with reviews and make money with Amazon referral links.
I end up buying cables on Monoprice when I can just because the stuff they sell is usually good quality.
Personally, I'd very much like to see such a thing applied particularly to high-knockoff topics like USB cables (see lead post), camera batteries, printer ink, and other items where a lot of companies are providing a lot of low-price alternatives (some quality, some dangerously bad) to high-dollar brand-name accessories & consumables. For some there's WAY too many options, with HUGE variation of price & quality, to do anything other than either "buy cheap & popular and hope" or "pay thru the nose but it will work".
FWIW: if you sign up as an advertising associate https://affiliate-program.amazon.com with Amazon you can get up to 10% from every link you provide that turns into an actual sale. Build a website featuring appropriately curated reviews linking to Amazon and you could make a good buck - IF you can get enough readers. Over the years I've made a couple hundred dollars doing so, just by posting links when relevant to blog comments. Focused as a business? I'd expect good returns if done right, plenty to hire good unbiased reviewers.
as a consumer, i don't care about crappy products unless i end up purchasing them. i can avoid crappy products by only going for products with explicit positive reviews that i can trust. sure a grey area will continue to exist, but whitelisting is still effective for reviews, even if we can't blacklist.
The mattress issue actually sounds like a solvable problem, if you're willing to put in the effort. It's just not really worth it when you can also buy mattresses from IKEA or various online retailers which don't engage in those tactics.
IKEA manufactures wherever possible for the lowest price. The difference is that IKEA employs quality control measures.
A central (well, really federated would be better) is one of the most needed tech innovations today, because it will make the market for so many other products function more efficiently.
Example USB-2 cable: http://reevoo.com/p/startech-com-usb-2-0-active-extension-ca....
I've used ebuyer (UK computer retailer) quite a bit and they moved away from their own review system to use Reevoo (the ebuyer one was better IMO however). Still Reevoo seems to have useful reviews in the computer/tech space.
And often I end up buying products with bad reviews in preference to those with perfect reviews because bad reviews gives me faith a product has been reviewed "enough" by real people, and often the issues brought up are not important to me.
I'd say that for Amazon, these reviews are fantastic: Next time I want a type C cable, I'll check his reviews and buy one of the ones he gives good reviews - probably at Amazon. This is like popping into that store where you trust the staff because they talk shit about "their" own products when they're no good even if it means steering you to less profitable products.
Mr. Leung responds by painstakingly pointing out everything they have to do to fix their cables.
[0] http://www.amazon.com/review/R2DQ7OH7PQG24F/ref=cm_aya_cmt?i...
http://www.righto.com/2012/10/a-dozen-usb-chargers-in-lab-ap...
I'd like to see amazon just develop in-house testing for certain types of stock. Cables, power supplies...
Part of the reason the cheap ones have so much availability is that they are cheap and easy to make and have little quality control -- even though they may not work.
This seems to happen all the time when there's a new standard out. The disreputable companies whip together something shoddy and try to get those early sales before more reputable companies can move in. Meanwhile, companies like Amazon do nothing to stop this obvious fraud. Amazon should be stopping these sales.
Also, its disingenuous to pretend these are $2 cables. The only ones still available showing price are $14.99. These are much more than $2.
Consumers need to be aware of the risks they take when they buy bargain brand, but it doesn't negate the benefits of having bargain brand available.
This isn't just a few bad cables or a bad batch that somehow got loose. This is purposely producing things that don't meet spec.
I would like Amazon to provide better access to searching for and sorting commodity items. (and to remove from inventory defective ones on a limited basis)
For example, I'm searching for the least expensive bulk CR123A batteries. I'd like to sort by price per battery and buy the lowest with at least 3 stars. A fairly simple request.
But you can't. Amazon lacks (in most cases) per unit pricing information or the ability to sort by it. The batteries are cataloged in at least four different site departments meaning sorting is impossible over all of them. It's frustrating to have to try to deal with such a stupid problem.
But others don't agree with my rating system, so "ordered blue, but it was robin egg blue. Werst company evar!!!11" means one star. Five stars means "I work for the company that makes these, or I'm otherwise being paid to write this, or I simply didn't give a lot of thought to how rating systems work."
In this case, the cable fulfills most of the requirements, misses a some of them, but might work for your use case, so two stars.
4 stars is "Does its job better than expected"
5 stars is "contained a 100$ coupon for amazon" or something that really makes you love a product.
The product quality of LEGO bricks – survives a century and so on – is worth 5 stars, as it’s far above expected quality for such things.
For a USB cable it would need to be amazingly resistant to wear – survives being folded and survives a dog chewing on it and so on – and it would need to be perfectly manufactured to get 5 stars.
In Hardware, usually very few high quality products – Miele washing machines, LEGO bricks, IBM ThinkPads, etc – reach something that could be called "5 stars".
And then along comes the cable which, when I plug it in, causes a religious miracle to occur. How do I rate that one? Five stars is already taken. (Or maybe it should be two stars. I want to get my work done without being distracted by a prominent religious figure appearing next to my desk.)
To put it another way, when I purchase something I expect that it will work fine after 30 days and that it meets my needs. That doesn't warrant five stars. That's three stars: does what I expected, doesn't suck but doesn't stand out from the rest all that much. If it were half the price of cables of equal quality, then I'd give it five stars. But IMO the product has to be outstanding in some way to warrant five stars, otherwise why have a rating system at all?
OTOH, Amazon doesn't help with this. Click on "critical reviews" on Amazon, and three star reviews are listed. That's just broken. Three stars is middle of the road, not outstanding in anyway but not deficient, either.
So, no, given your parameters, I would not rate the item five stars. I realize the rest of the world doesn't agree with me, and I'm loathe to participate in such a skewed system (everyone gets a trophy). My concession is I'll give four stars to something I'd normally give three stars, just so it doesn't show up in the "critical review" section, broken as that might be.
And hence my quandary; I don't consider three stars to be a "hit...in the ratings". It does the job, doesn't stand out in any other way, three stars. But, as you point out, it's not fair in the inflated system that we have available. So which wins out, an honest review system that benefits consumers, or review inflation that benefits product sellers? I haven't decided, nor figured out any means of compromise.
As I once heard about bug priority, if everything's a Priority 1, then nothing is a Pri 1. And if everything that isn't complete shite is five stars, then really nothing is five stars. Five stars just means it doesn't suck. Then why don't we have just one star or no stars from which to choose? In other words, the precision of the rating system is not what the five choices would imply.
> "Engineers do not give out 5s. If you tell an engineer that the possible rating scale is 1 to 5 and they like you, they give you a 4. Because tomorrow they might love you and then they would give you a 5. They probably won't, but they might, so they give you a 4 so that they have room. We all are totally okay with this. But the world is also full of this other class of people who if they like you give you a 5. They're quite consistent about it. 'I like you. I see nothing wrong with that. I'm giving you a 5.'"
Personally 5 stars would be when the box it came in had water soluble ink and it made a mess on your clothes, but the company offered to pay the dry cleaner costs. Even though there was no expectation for them to know that the shipping company would leave package in the rain / a puddle for so long.
Anything above "meh, it didn't crash and burn immediately, I guess it's okay" is 5 stars on eBay.
Sound alike the whole reason is to make their device owners able to make better choices.
On HN, I tend to limit my responses because I usually end up being attacked for something I have nothing to do with, by virtue of some grudge the commenter has with Google. If you want to call me out for my own work, go ahead, I value feedback greatly in areas that I can control. If you have useful commentary, I'll try my hardest to make sure that people with understanding in that area see your comments. But if you just want to yell at me by virtue of who pays my salary, I find that all very unenjoyable and would prefer to not be involved.
I also notice that lurkers with blogs tend to like to take quotes from HN and convey them as the official word of Google out of context, which I also don't like being involved in. Learned that lesson my first week at Google :)
Finally, one also ends up attracting the crazies that email me threats against other Google employees. All those get forwarded directly to Security Operations and don't get a reply from me.
HN is a pretty big community these days and isn't quite the bastion of sanity that it once was. This limits how open many people will be with the community more than any policy.
Now that an expert has called some of these cables out as not adhering to the spec shouldn't they be taken off of Amazon? Should they be reprimanded for false advertisement?
Sorry I can't find a better source at short notice.
http://www.ecommercebytes.com/cab/abn/y13/m02/i25/s01
Upon turning on the device for the first time it showed a screen saying it was controlled by the LDS Church and wouldn't proceed through guided setup until I acknowledged that. After some time on the phone with customer service, we determined there was no way to remove this configuration. I ended up having to return it to an Apple store. The rep there said "yeah that happens sometimes" (someone pulled from the wrong pile in fulfillment most likely). So sometimes you're damned if you do, and damned if you don't.
If anything Amazon's customer service is better - to resolve an issue like this for items they ship I don't have to drive anywhere, just put the box back on my front doorstep. I wish they would make it clearer when you're buying things from them, and would do a better job of associating compatible products. This seems like data they should be able to have, and UI they should be able to build.
Frustratingly it wasn't the right bulb, but I just fed a further data point into Amazon's model, and reinforced their belief that its useful to list those two items as a pair.
(Obviously, much like for activation lock, it is indeed possible that someone with "root access" - so to speak - to Apple systems would be able to manually disable DEP; but the point is that this possibility isn't exposed on any user-level user interface)
That's not obvious to me. Those same people can be social engineered into giving someone access to your data, which is an order of magnitude worse than a device wrongfully moving accounts.
This isn't a troll comment. Ben Thompson makes the argument that the reason why EBay, AirBnB, and Uber work is because you basically are replacing the trust you had in a brand with the trust you have in community reviews. He calls it 'Aggregation Theory'. I find it rather compelling to explain why these types of marketplaces work.
And still, no guarantees.
At this point I just flatly don't trust Amazon anymore for small electrical devices, because there's too much of a chance (even for things labeled with brand names) that it's actually a cheap knockoff that will short-circuit and set something on fire eventually.
I've now bought a couple of Nohon braided cables from Aliexpress and they seem to hold together nicely. But of course them being cheap Chinese cables they are finicky with iPhone 6+, but work fine with my 5s.
My wife did finally kill the Lightning cable in the car that was a few years old. One cable failure in probably ten years. That and $4 will get you a coffee at Starbucks, so take it FWIW. I just wonder why some have better experience with Apple cables than others.
Isn't that what standards compliance is supposed to be for? Shouldn't the product have to be tested against the standard in order to get labeled USB X.Y ?? Or doesn't USB have such a requirement?
https://www.ftc.gov/about-ftc/bureaus-offices/bureau-consume...
If so, you can look for their logo and have... some confidence at least.
It's the same for Android. It's open-source, but you cannot say that it's derived from Android unless you pass Google's tests (and according to Chinese independent manufacturors, you need to know people from the inside to advance your application).
Theoretically, people who purchased complain to their state attorney general or report the manufacturers to the FTC. There might be some recompense under consumer fraud laws. If enough people had bought them, there might be enough people to form a class action (doubtful though).
The problem is that amount of $ involved here is just not something anyone is likely to get up in arms about. It's $5. Most people will just write it off and never think about it again. No lawyer would bother either unless there were 10,000s of consumers willing to come forward and complain.
Just one of those shitty situations. Caveat emptor.
(I am not a lawyer, this is not legal advice.)
So do some rather expensive shiny-metal-branded laptops.
https://www.spinics.net/lists/linux-usb/msg131265.html
http://www.spinics.net/lists/linux-usb/msg122354.html
http://marc.info/?l=linux-usb&m=134861220631257&w=2
http://www.spinics.net/lists/linux-usb/msg63980.html
...shit's so broken I could keep doing this all day long.
Most hardware on the market today was released before the xhci spec 1.0 was finished, and I agree that ALL hardware on the market does not even properly implement the spec that was available during its design. Fortunately, most issues can be worked around using quirks.
The hardware is the root of the problem, but I would also add that the xhci_hcd kernel module, even in the latest kernels, is well behind Microsoft in terms of handing these device-specific quirks. The Microsoft USB 3.0 stack introduced in Widows 8 solves just about every host-specific issue we've ever seen. ASMedia is the only host controller vendor that bothers to provide their own Windows 8/10 driver (both for their USB 3 Gen1 and Gen2 chips). No one else provides an alternative to the Microsoft provided usbxhci.sys. That said, all of our tests have been limited to bulk transfers, and we're only using existing user space libraries (WinUSB, LibUSB, IOKit). We do push throughput, we have tight latency requirements, and we keep a large number of buffers queued at all times, which does push the host controllers further than your typical storage device. (We make streaming data recording equipment for electrical engineers and embedded developers)
I don't want to bash xhci_hcd though; their progress has been fantastic, despite the minefield of problems created by all these host controller vendors. I would like to say a special thanks to @sarahsharp, for all of the hard work put into the xhci host controller module. Without Sarah, I don't know who will answer when the xhci_hcd doorbell rings.
Usually what people want is to "make my device charge fast" - in which case the cables not implementing the spec correctly work while compliant cables would not.
Of course, there is a reason that compliance exists! Chances are you are plugging this cable into a wall-wart made in the past few years with higher tolerances than needed, and things will work out fine. Until they don't. Then you may burn out a PC USB port, or potentially I suppose catch a cheap chinese wall wart on fire.
At least that is my current understanding having spent the past few days attempting to decipher how this all works. The correct way to go about this transition to USB-C is to toss out all your USB-A chargers, buy USB-C chargers, and if you need to adapt; adapt in the USB-C -> USB-micro/mini direction vs. the USB-A -> USB-C direction.
The tldr on this is basically if you plug into a USB-A charger, you should not be able to "rapid charge" at 3A. If you are, you are out of spec and could potentially damage something you plug that cable into. Just give up on the dream of using your old chargers and expect to replace them with chargers that implement USB-C there. If you need to charge USB micro devices, invest in some USB-C -> Micro cables (or adapters) instead.
You can still rapid charge with a compliant type-C/A cable, but it needs to be negotiated through one of the other specified mechanisms (that directly talk with the other end and thus can figure out what the type-A charging side can actually provide).
https://thunderbolttechnology.net/blog/thunderbolt-3-usb-c-d...
I've seen a couple Dell laptops but not much else. It's a much better form factor. Not to mention the elimination of USB superposition: :-)
http://9gag.com/gag/6756457/usb-superposition
I guess depending on the company's HR that could fall either way of
a) This is great, we're getting nice PR
b) Mr Benson, we need to have a quiet word about what it is you do here and what Google is not in the business of.
Really interesting.
edit: I assumed this is part of his job. I just find the publishing of the result along with Google's name interesting.
If/when I evaluate software for a solution as part of my job, I wouldnt dream of publishing my findings and referencing my employer - but I like what Benson has done.
My worry would be the cables that get good reviews being switched out for crappy cables down the line. I've had problems with this for cables and adapters before - I'll re-order from Amazon and get a slightly different product that is not 100% compatible with the old one.
(This was a particularly serious issue for Arduino a few years ago, when Amazon started listing genuine Arduino products under the same ASIN as third-party clones: https://blog.arduino.cc/2013/07/10/send-in-the-clones/)
Most recently a "good ol' American made" can opener turned out to be a cheap China counterfeit: https://www.amazon.com/review/R3MJPDPNRM9HXR
Would not be surprised at all if he expensed every single one of these purchases. He's doing God's work, we shouldn't be complaining here. :)
Depends on how busy the HR dept is I guess.
I think that's the whole point - for Google to chime in on cable quality so that customers know what to buy for their Google devices.
I'm guessing that Google saw that a large percentage of their customer support issues were caused by poor cable quality, and that most of their customers bought their cable from Amazon.
This seems like a way for Google to pre-empt that customer support cost by making sure their customers are buying cables that they know are good.
Most of the time other companies hand-wave their way around this problem by offering 1st-party cables (usually at a fairly high price).
It kind of makes sense: he worked on products that use USB-C (Pixel laptops), but if the market is littered with bad cables USB-C will have a bad image which will be bad for the products he worked on.
So what he's doing is in the interest in his employer, and for the benefit of the products he worked on. He's not doing completely unrelated stuff, like reviewing routers while claiming "I work for Google so I know how shit works!"
Additionally he's preventing Pixel owners (Google customers) from buying bad cables on Amazon then blaming Google because the laptop doesn't charge. So it really looks like he's doing his job.
"if the market is littered with bad cables USB-C will have a bad image which will be bad for the products he worked on."
I don't buy this tail wagging the dog argument. Sorry.
One cable per device is insufficient for a lot of us.
To many, a cable is a cable is a cable, and preventing someone from buying a commodity part that might damage their phone is a good objective, no matter how it's done.
Posting Amazon reviews might not be the optimal way to get me to buy the "right" cables, but it's a pretty effective way at preventing me from buying the wrong ones.
https://store.google.com/product/usb_type_c_to_usb_standard_...
Since Google isn't the manufacturer they are not making a ridiculous profit margin.
Whole dollars on a $500 phone is more likely to be Google's entire profit margin, though. The profit margins on phones is razor thin unless you're Apple.
And LOL @ the idea that Google sells millions of phones.
When it's only a dollar or so you might expect the customer to want to pay the extra dollar in exchange for the benefit, but in practice the device is going to be priced at a threshold like $499 and the psychological difference between $499 and $500 would measurably drive down sales.
But the bigger problem is that this is hardly the only facet of a device you could improve by making it cost a tiny bit more. Add them up and it would add $100 or more to the price of the device. So you can't have them all, but decide against any one in particular and everyone wants to know why you're cheaping out over a buck.
Anyway if these cables contain such chips $13 seems reasonable.
This means because of his reputation (Google engineer) it's easy for him to dismiss one cable manufacturer and support another while getting some form of a kickback either directly or indirectly from said manufacturers.
Sure, he's doing his job, but he's also in the position to pick and choose winners. As such, I would discount his advice based on the possible bias towards products that could possibly line his own pockets.
That's always a risk, but I think it is more of a risk having used his real profile and Google affiliation. On the face of it, I want to think that Google pays Benson well enough that he needn't go tell lies on Amazon as a side job.
>As such, I would discount his advice based on the possible bias towards products that could possibly line his own pockets.
This isn't like for example "audiophile" cable reviews where people just make crap up in ambiguous terms hoping to sell a thousand-plus dollar pair of speaker cables knowing full well that no one can argue sensibly about the warmness/depth/color/scent of the sound.
He pretty clearly explains his major complaint being the lack of a proper 56k pullup so that the cable identifies itself correctly. This is testable by almost anyone. It makes very little sense to publicly tell a lie that can be verified so trivially.
https://www.youtube.com/watch?v=BpsMkLaEiOY
The review is extremely negative, explicitly telling people not to buy Nest products. Those are (and were at the time of that video) a Google product, as Google acquired Nest in 2014.
Brad still works at Google. I'm sure Benson will be fine.
It never occurred to me, or anybody else working in the lab as far as I know, to post the results of our testing on Amazon reviews. We just did the testing.
Huh.
What about it? I'm sure Google makes minor changes to their search algorithm that drive small companies into bankruptcy all the time.
This is entirely ethical, and the context is stated clearly... I think it's great.
I say "sadly", but I've seen the flip-side of this, too. A properly-formed query on usenet will turn up some back-in-the-day posts from Microsoft employees (posting with a microsoft.com reply-to) who were real assholes. (And not to pick on Microsoft, though they certainly have it coming; just the first counter-example that came to mind.)
So I guess I'm conflicted. Yeah, Benson's being really helpful and it looks good for Google, IMO. But with a different employee, it could turn really ugly if someone's back is turned.
But obviously, there's many people like Benson who provide incredibly sagely advice that do a service not just to Google's own customers, but competitors' like Amazon's customers as well.
Once a company reaches a certain size, they have way too many people to police all of them across the Internet, or ensure all of their hires contribute meaningfully online.
(details: I was full-time tethering while plugged in. I assume the charge controller thought the battery was charging but it was actually slowly discharging)
Getting said info to the right place so next version's charge controller is fixed would be nigh impossible, but dropping the info in lots of high-exposure places might be a start.
I can only surmise what actually happened. I was full-time wifi tethering for a month or two and the symptoms along the way seemed to point to this cause. The phone eventually stopped taking power at all, until I plugged it into a USB port instead of the OEM 1A charger. At which point I realized the higher current was the problem, and investigated the cable.
It actually felt like a really nice cable - very robust and strain-reliefy. But apparently that's because the whole thing was mostly plastic and essentially headphone wires. Lesson learned!
If my theory is correct , it's a pretty stupid design. But I feel like there's probably many weird corner cases in consumer electronics that don't matter because 99% of people won't find them, and the ones that do won't realize it's a bug, and by the time anybody could put it together that specific model has been obsoleted anyway.
When I do care about signal integrity (i.e. connecting a USB hard drive to my computer, I use name-brand cables)
So people buy based on the review and then they find out the vendor shipped a clone, so then they give a bad review but it's not for the original listing.
Amazon made a huge mess with that, they should never allow third parties to sell on the original listing.
I suppose there is a Ship of Theseus problem when trying to detect tricks like that automatically - surely sellers should be able to reword description, update product image, and change the product name slightly, but where do you draw the line and how do you handle that? Heh.
Until they weren't. We bought one that was a substantial step down (screen resolution in particular, which was exactly why I'd picked it).. Got a refund, but Amazon simply didn't seem to care that the reviews had no relevance to the current model, which was a let down (though a lesson to read the reviews closely and pay attention to their age...)
The micro-USB connection is also really loose with this particular cable<->phone combination (cable is fine with other phones; phone is fine with other cables). Used my old Nokia micro-USB for this reason and charging was always slow, then connected the Samsung one and it was a lot faster.. for a while.
I never knew the cable could do something like this, unless it's something like being way too long. From your comment I take it that you do need a certain amount of copper in those wires to work? So more expensive cables might actually be better? (I always laughed at "gold plated extra quality HDMI cables of 30cm for 30 euros because HDMI is a digital signal.)
One part is the wire gauge. You need something like 24awg (its the American system, don't ask me why) on the power pins (higher number, thinner wire, iirc). Most use 28 on both power and data pairs.
Then the charger has to actually follow the USB spec, not the very similar Apple one. This means either a microchip that can handle the USB handshake, or a resistor across the data pins at the charger end to signal that this is a charger (Apple has one resistor on each pin, with a slightly different resistance).
Now on top of that you have Qualcomm pushing their fast charge system that involves a chip at both ends, and higher voltage than the 5V that USB normally uses. The higher voltage allows for more watts without upping the ampere (V*A=W, iirc).
Note that the Qualcomm system is similar (but i don't think they are interchangeable) with the USB3 system (the latter can go all the way to 20V, but that requires beefy cables).
Frankly the more i think about it the more i expect to hear about someone getting electrocuted because they mistook their data cable for their dumb laptop charger.
Now I am nervous because I subsequently bought a cable that he hasn't reviewed yet...
This has me wondering if there is some way to incentivize good reviewers. There is Consumer Reports and some more tech-specific sites but there is still a long tail of products that you have to buy without good reviews.
For instance, I don't think Amazon is going to sell broken crap using their AmazonBasics brand, and they still charge quite a bit less than Belkin (I don't think they sell USB C yet, but that's besides the point).
Joking aside, I'm somewhat more likely to read a two star review vs a one star review because a lot of the one star reviews are whiny rants and not actually helpful.
Maybe it's okay, so far the Wii/PS4 and ChromeCast work fine through it... I'm just posting this here, as no one informed me about possibly complications at BestBuy, but the great folks at Staples actually forwarded me to BestBuy to look for such device.
My father told me a story about service in Turkey, basically if you go to a store and they don't have right away what you need, they may go out find/buy it for you, just to please you. Or buy you a coffee, while waiting in s shoe shop.
Whether he exaggerates a bit or not, I like this approach. Yes you might lose customers, but you'll always leave a smile in good customers. They might come back, just for that little appreciation...