Hacker Newsnew | past | comments | ask | show | jobs | submit | frei's commentslogin

The first song I tried mapped incorrectly. Even though there was a green verified checkmark!

> https://open.spotify.com/track/1Pfc1Qpj0s9vQumI0JvpBp

mapped to

> https://music.youtube.com/watch?v=ogoeWS6CDbI

but it should have mapped to

> https://music.youtube.com/watch?v=dfgKYWrRfoc


Maybe it was a UX bug that hampers the experience to the point that where the website is a better experience, and they felt that a bad app would hurt the brand more than no app.


Or, maybe there wasn't but they didn't care enough to test so they disabled it 'just to be on the safe side'.


Unfortunately the bot is not very good. Just by reading the rules I was able to beat it every time on multiple rulesets.

The matchmaking is also difficult to find a partner since the rules are so customizable, and there's no way to tell what rules potential partners could be searching with. Maybe a list of games that are looking for an opponent would be good.


I just read the first paragraph and then decided I wanted to play and I still beat the bot first try. At least that means the game rules are fairly discoverable :)


I didn't understand the rules until most of the way through the game and I still won against the bot. I'm not clear if its really doing anything better than playing randomly based on that.


It's a weighted random choice, so a bit better than randomly because of the weights, but not much better.


Yeah, the bot is good for learning, but after a few games it becomes easy to just go after it directly, kill the Os which connect the walls and win easily.



I've done some translating on YouTube. What it means is if your translation is in draft mode (you're still working on it) on 9/28 it will disappear. The video creator shouldn't need to do anything here.


Ah, okay, thanks for explaining!


The prescriptions in "If you’re a security conscious user..." make sense to me. If you use a unique password, sms/totp adds very little benefit.

However the section for "If you’re a security conscious vendor..." doesn't make sense. Credential stuffing is so common, and sms/totp is a great tool against it. You could prevent users from setting their own passwords, but that seems a little "too different" from existing sites that it could harm usability.


All the arguments in the post against SMS 2FA are also arguments against TOTP


There is no help desk that can give away your TOTP secret. That’s the big weakness with SMS, someone buys a phone and asks the clerk to give them your phone number.


It is a publicized weakness. In practice, phishing is scalable and social engineering isn’t. And humans are shit at detecting phishing attacks.


>The J.D. Power US Initial Quality Study, which gauges problems encountered in the first three months of ownership.

A very questionable definition of "reliability".


This video does a good job summarizing. https://www.youtube.com/watch?v=CkkiK4nplW0

JD Powers get maligned a lot, but they're an industry tool used by manufacturers to measure themselves against others. The survey has incentives to be accurate.

The Initial Quality Survey is basically a measure of how happy a customer is with their new car. I think a luxury brand like Tesla does especially badly because customers are already aware of (and sensitive to) potential issues with paint, body panel alignment & scuffing whereas mainstream brands' customers may not be scrutinizing their cars as much


It's not a good measure of reliability, but it's a good measure of quality control, where Tesla seems to be sorely lacking. I've had two Model 3s, and I love the car. I had one last year that I had to return when I changed jobs, but I loved it so much I immediately ordered another. With that context, the QC is really poor. The first one I got last year had lots of cosmetic issues that I spotted when I picked it up, mostly paintwork. I missed the fact that the cloth on one of the speakers was damaged until it was too late. The Model 3 that I got this year was delivered in March, just before lockdown, so I didn't get much chances to use it at first. I just collected it today from the service center where it had been to have a scary fault fixed. I was driving along when suddenly a loud chime, and regen braking was disabled and the handling changed completely, and the car no longer decelerated when lifting off the accelerator pedal. An alignment issue that was a quick fix (aside from the four week wait for an appointment and the two hour round trips to drop it off, and then collect it). I separately now need to wait for a mobile repair next month to replace a parking sensor. These sort of things should have been caught before the cars left the factory.


Perhaps it should be redefined as build quality, but it is still a decent metric worth analyzing.


Most product failures happen very early on, or very late in the design lifespan: https://en.wikipedia.org/wiki/Bathtub_curve


Also manufacturing problems are usually highly correlated.


AFAIK they lump everything from "I don't like the color of this AC control knob" to "sudden and unpredictable catastrophic engine failure" into this category.


Which may be problematic, but it's not like (as some here imply) that Tesla is being treated any differently to any other manufacturer.


If the headline said "Tesla vehicles have most complaints" it would be more accurate, reliability is a strange choice of words. The study is called "Initial Quality Study" and the axis label on the graph is "Problems per 100 vehicles."


To be clear, there, though - the word reliability is used by The Sunday Times, not by the survey.

Though you could argue that "I can rely on my vehicle to have a consistently high initial quality, coming from Manufacturer X" is consistent.


I don't doubt that Tesla's are less reliable by other definitions, but I agree that this isn't a great measure.

Tesla's have lots of new and untested technology. That's their appeal. If toyota wanted to be less conservative and start using blazing new technologies in their cars, I'm sure we'd see the same issues there.

Teslas are notorious for software bugs that eventually get fixed. What i'm more interested in is whether Teslas end up lasting a long time. Electric motors are a lot simpler and so in a sense, I suspect they have the potential to last for more years with fewer issues.


> Tesla's have lots of new and untested technology. That's their appeal.

This saw gets trotted out whenever Tesla gets dinged for reliability. So let’s dig into it. What untested technologies are Tesla deploying and how does the affect reliability?

Well let’s review. The Model 3’s initial problems were metal panels not being installed correctly. A technology that has been successfully deployed on automobiles for over a century.

Now gaps are fixed, but they shouldn’t have been a problem at all. From what I understand the problem was caused by trying to have robots do final assembly and just lax quality controls, problems “legacy” manufacturers had solved decades ago.

Tesla’s iconic touch screens are delaminating [0] because Elon ordered screens to be installed that aren’t up to automotive grade standards. This is another unforced error.

And we haven’t even touched on the problems with the actual advanced technology, autopilot, which is once again oversold and underdelivers.

[0] https://www.thedrive.com/tech/27989/teslas-screen-saga-shows...


> Now gaps are fixed

Not really. They've improved on the 3, but people still routinely get some really wild tolerances on current Model 3s. And the Y has had an absolutely awful beginning of production, facing many of the same build quality problems that early Model 3s had.

Tesla is relying too much on buyer enthusiasm. As they expand into the mainstream, they need to avoid building a strong reputation of poor quality that will be really hard to shake.


I wouldn’t buy a Tesla. I don’t like the interiors, and like you said they've burned their reputation with me.

I was skeptical of how wide spread the the panel problems were with the 3’s initial rollout, until I examined the 3s in the parking lot at work. Every car was bad, including one that was a shockigly horrible.


> If toyota wanted to be less conservative and start using blazing new technologies in their cars, I'm sure we'd see the same issues there.

The Toyota Prius was the first mass produced hybrid vehicle and is notoriously reliable.


It's also essentially a regular car with a 144v battery and larger starter motor. Otherwise, very conservative.


By the same metric, isn’t Tesla just like a regular car but with a bigger battery, no engine and a really big starter motor?

I mean cars before that had all the adaptive cruise control stuff anyway which is the same but just a bit better right?


No, the interior and it's controls are completely different. The touch screen interface is a huge difference.


Like the touch screen a Prius has, but without the buttons?

I’m just saying if you can trivialise the huge R&D effort that went into the Prius, you can trivialise the Tesla in the same way. I think anyone who appreciates the nuance in engineering knows that it wasn’t just putting in a slightly bigger starter motor, they invented a whole category.

But the iPad is just a big iPhone, and the iPhone was just an N80 but it had a touch screen and you couldn’t install apps.


> Tesla's have lots of new and untested technology.

Tell us about some of these, beyond Autopilot. Because this is a common refrain from Tesla owners, some of whom believe that only Tesla has adaptive blind spot sensors, or that only Tesla will keep you in your lane actively (and not just do the pinball bumper drift sensor), and so on.

What other new and untested technologies is Tesla deploying today?


Seats. Other car makers don't use Tesla seats.


Your comment reminded me of an article I think I read here on HN: https://qz.com/1737145/the-economics-of-driving-seven-teslas...


This survey isn't really a "reliability" survey per-se. It's more a measure of how happy people are with their new cars


How is "problems in the first three months of ownership" not a good judge of "Initial Quality"?


It's a great judge of "Initial Quality", but a poor judge of "Reliability". When someone says their car is reliable, they mean it hasn't needed service for a decade.


It's "INITIAL QUALITY STUDY"... reading comprehension is going down even here..


I've tried Dadi and Sppare.me in the past and had good results on the front-end (I haven't retrieved any samples yet). I'm not sure the service could have been improved. What are you doing differently than existing mail-in sperm analysis/freezing companies?


Hey there - great question, and happy to hear that you are already being proactive + thoughtful about your reproductive health. A few things our team is doing differently than others in the space:

1. Clinic-level analysis: Most other solutions offer a basic analysis (count, concentration, volume + some motility). Our analysis is a comprehensive assessment that evaluates count, conc, volume, motility + morphology, which gives patients and their doctors the information they need to make decisions around treatment options

2. Personalized lifestyle recommendations - sperm health can be improved since sperm is regenerative. We provide all of our clients with data-driven personalized lifestyle recommendations for sperm health improvement.

3. Multi-geography storage - for every sample we receive, we split it into 3-6 vials and then distribute those vials to two different storage locations as an added measure of security. We get a lot of really good feedback about this from our clients.

4. Telehealth appointments - all clients have the option to speak with a specialist after they receive their results in order to learn more about their sperm health.

5. Client experience - we pride ourselves on providing a superb client experience. Each client receives a dedicated client service advisor to guide them through the process and answer any questions they have.


One more thing I'll add to Sarah's answer: in our case, the company was founded by - and run by - healthcare and fertility specialists. We are a healthcare company first, and a consumer company second, not the other way round.

This sounds minor but actually affects a lot of the way we do our work e.g. we ensure rigorous analysis, we take extra de-risking steps around cryostorage, we call clients to share bad results because it's more humane to do, we connect them to our urologist on staff, etc. etc.


Curious about this too. I recently looked into sperm preservation. Something like $1,250 to setup and $250/year after the first year. Seemed reasonable to me.


These rates aren't unreasonable! I've seen storage get as expensive as $1200 per year! We try to make our pricing as affordable as possible ($145/year)


That's exceptionally good. I'm paying Reprotech in Minnesota $200/year ($1000 upfront for five years of storage).

Wish I had heard about you in the spring, would've picked your org solely because of your georedundancy option mentioned in a sibling comment.


U2F doesn't protect you from an untrusted machine, it protects you from untrusted websites.

https://security.stackexchange.com/questions/157756/mitm-att...

The security model relies on the browser validating the origin.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: