I made the switch from a Pixel 2XL to a 7 last year, and dang, the 2XL was better in most every respect. Fingerprint access with the 2 worked 99% of the time on the first try vs. about 10% (probably being generous) on the 7.
And voice-to-text is just as broken. I did side-by-side tests, and if you’re in perfect conditions, facing the phone directly, with no background noise (i.e., never), then the 7 performed okay, but it would fail horrible in any other case. The 2XL performed quite well, relatively. I even replaced the 7 thinking there was an issue, but the new one did the same thing.
Unfortunately, lack of updates means the 2 is no longer usable for many purposes.
It's a separate video on his channel which is probably a better listen than this one. This one is a bit of conjecture, perhaps true. The other has been validated by multiple sources, AFAICT.
Short summary - Someone had tied in most of their home automation to Alexa and found their account cancelled suddenly one day. They went through the automated recovery systems and were told to contact support, which they did. Support ended up transferring them to an Amazon exec (let's assume "manager") who told them their account was disabled because an Amazon delivery driver reported that someone said something racist to them over their video doorbell (which wasn't a Ring, ironically).
Upon investigating, checking cameras, logs, etc,, the owner determined that (a) no one was home at the time of the delivery, (b) the driver was wearing headphones, (c) the doorbell had done an automated, "Hello, how can I help you?" response to the driver as they were walking away (presumably ring-and-dash or drop-and-dash delivery, as usual).
The driver had apparently, with the headphones on, completely misunderstood.
It took over a week to get Amazon to review all the evidence and reactivate the account. No apology at that point (although I believe I saw they subsequently have).
That's a bad look for Amazon, and the Youtuber makes a valid point that it's a bad idea to trust control of your home to a company that will make such boneheaded decisions.
IMHO, the only correct response for Amazon here is firing at least two people involved in the debacle, apologizing publicly, and promising to review and adapt their policies in response to the incident. Any halfway decent PR department at anything other than a mega-monopoly would be scurrying to do exactly that, but not Amazon apparently.
> firing at least two people involved in the debacle
The idea that after a mistake companies should fire people leads to company cultures that are overall worse for everyone, including customers/users. Demand that they apologize, even that they give compensation, but not firing.
During my tike there, Amazon's failure culture, as in yes, mistakes and failures happen, and as long as don't happen twice and lessons are learned nobody gets blame or the axe, was one of the things I liked best.
This whole asking for punishment is what actually drives a culture in which these kinds of things do happen more frequently, because everyone involved just wants to cover their asses.
Okay, "firing" might be too strong, but that policy of "tolerating mistakes" (as long as it doesn't happen twice and lessons are learned) seems to have created a corporate culture where (if this story is true, and it seems to be confirmed):
- A customer can be mistakenly called a racist
- Their home automation systems they bought and paid for disabled
- Any digital content (Kindle books, Amazon Prime Video purchases, Audible books, etc.) they bought (sorry, "licensed") revoked.
- It takes more than a week to resolve after being provided with clear evidence of the company's mistake. I mean, good grief, at least the manager/executive should have reactivated the customer's account during the review process, but they opted for "guilty until we've taken our sweet time reviewing the evidence and make sure they're innocent".
- After all that, the customer isn't even offered an apology, much less compensation.
This isn't just a "mistakes and failures happen" situation. Failures and mistakes occurred at multiple points in this process and along the decision chain, and apparently no one involved had the common sense to break out of the resulting insanity-loop.
Yes and I find it similarly extreme that a company would disable someone's doorbell because one person claimed something racist was said through it (?!?!). Are they going to set up a little Amazon Ring court to adjudicate every claim??
If Amazon hadn't taken that extreme step in the first place, the stakes wouldn't be so high, and there would be less reason to discipline the employee (for the record I don't think they should be fired in any case).
Ironically, the doorbell was the thing that wasn't an Amazon product and wasn't disabled. ;-)
And yes, I recanted on the 'firing" part, but I still feel that Amazon's "resolution" here was weak-sauce compared to the "extreme" action taken in the first place. At this point, I'm guessing they wish they'd offered the customer some minor token (say 2 years of free Prime at a minimum) compensation in return for an NDA on the topic. 20/20 hindsight ;-)
Amazon is a large entity with complex processes. Firing one of the people involved in implementing those processes does not make things better for users or affect how much power Amazon has in the situation.
Someone implemented this process, as well as the policies that drove it. They clearly needed an incentive to think through the PR consequences (if nothing else) before imposing their incompetence on a paying customer.
Firing people for negligence in similar situations would likely have had just such an effect.
"Code working" isn't necessarily black-and-white. For a new user (the one asking the question), the code may appear to solve the problem, but may have corner-cases or even security risks. That's entirely possible with user-generated code as well, of course, but GPT/AI allows it to be produced at a much higher rate, with the person who posted the answer often not being capable of (or not caring to) validate or correct it.
Sure, but if you try it out, you pretty quickly realize it's a hallucination. Unfortunately the type of GPT content we're now getting on Stack Overflow and its sibling sites is mostly unvalidated GPT hallucinations.
Agree with the middle part - At the moment, the policy implemented by corporate is "Don't ask; don't tell". If someone says they used GPT or other AI for their answer, it's disallowed. If they try to hide the fact, there's not much the community can do to get it removed.
And while I'm not a moderator, as just a user I've flagged over 1,200 answers on Stack Overflow (and several of the smaller communities like Ask Ubuntu) that were subsequently removed. Automatic detection was never the sole criteria that was used to determine if it was AI - It's entirely possible to spot GPT content using multiple methods. I don't publicly talk about most of these, since we do have a group of users (sometimes spammers) who attempt to hide their use and make it more difficult to detect. See some of my additional notes on the topic on https://meta.stackexchange.com/a/389674/902710
And there's the problem on SO. Previously, we could do exactly that - Flag the content for a Mod to review. Now Mods are pretty much prevented from taking any action when we (the community members) and they believe it is a bot.
I saw one user yesterday post 10 lengthy, detailed answers in an hour, in 3 different programming languages. But the Mods aren't allowed by SE to consider that (or pretty much anything) to be an indicator that it's AI-generated.
Yeah, that may shoot down my "good intentions" theory.
According to the event's sponsorship page (linked elsewhere in the comments here), the speaking slot was 1500€ on top of the 990-4500€ "base". We can assume that they didn't get the "prepayment" discount since if they had already paid, this wouldn't be an issue.
What's still unclear is whether they are pulling out of the speaker slot only, or the entire sponsorship. Regardless, Couchbase shouldn't be withholding those funds just because their employee is unwell. They are essentially saying that they were planning on paying for that, but now they won't since they can't attend. That falls on them and is not the fault of T3chfest. They should still be expected to pay up.
And maybe they plan to? Maybe this is just a misunderstanding between parties where Couchbase told T3chfest that they wouldn't be able to send the speaker, but in the confusion someone took that to mean they wouldn't be paying for it either, when that wasn't stated?
Again, don't know, so I'm not going to jump to assumptions other than to say that, based on the response, I would expect (and hope) that Couchbase would still pay the sponsorship fees that they (clearly) committed to.
People seem to be attributing "bad intent" here that may not have occurred. In my experience, situations like this occur quite often for various "good intention" reasons. Usually they are resolved before there's an issue, but sometimes they don't work out.
For example, the sponsorship form might not have been signed by someone with signing authority. Usually, it's a lower-level marketing (or even engineering) person that is working on getting everything squared away. The peon who is "negotiating" was told by their management that they want to do it. The peon assures the person who they are working with at the event that it's going to happen (because they've been told that).
The time comes to print the materials, and the form still hasn't been signed. The event has two options - Forgo the possibility of the $5k sponsorship by telling Couchbase that they haven't signed the commitment, so they can't print their logo. Or hope that things comes through.
Again, 99% of the time, things work out. Printing the materials even though they didn't have the agreement (or money) in hand was the right call for the organization.
But sometimes it doesn't work out ...
Maybe (pure hypotheticals following, because we don't yet know) the person who was working on the sponsorship at the Couchbase end left the company. Maybe their boss, who had given the "okay" even left the company. Now there's no one at Couchbase who knew where this agreement was in the process - They only know that there's no sponsorship commitment that has been made. And they question why there doing it in the first place.
That's just one example.
Maybe the person who signed the sponsorship form wasn't a director and didn't have signing authority, but thought they were authorized to sign something as "simple" as a sponsorship form. Oops! The employee is fired for signing a $5k agreement, the agreement isn't technically valid, and ...
---
Sometimes things are moving along beautifully and someone in management (rightfully) asks to have legal do a "quick review" of it, and there turns out to be something that the attorney has issue with (often an indemnification clause, inserted by an attorney on the other end). They end up not being able to resolve it and get the sponsorship agreement signed at all. Deadline passed, logos printed already ... uh oh.
---
If there was a sponsorship agreement in place, then let's be realistic - It's doubtful that Couchbase would be thinking they could just get away with "pulling out" four days before the conference.
Something else is going on here. I've been through this before myself -- Usually we get the agreement finalized in time, for the good of all parties.
But is T3chFest really "out" this money? Would they have found another sponsor in time? Or did they go ahead and print the materials because they really had no other choice, and no time to line up another sponsor before the deadline since Couchbase hadn't signed the forms yet?
