37 comments

[ 3.6 ms ] story [ 60.1 ms ] thread
All the bots that upvoted the scam replies... Its puzzling that Twitter does not do anything against them. I wonder how hard it would be.
Why would they? The bots inflate metrics and don’t hurt revenue. They’re a feature, not a bug.
The bots do hurt Twitter's reputation though.
There is no metric for that.
Wall street analysts are going to start writing articles about how this is a only a "burst week" and "not sustainable", and that the "rational" way of quantitatively analyzing the Tesla says the stock is worth zero.
It was a burst week and does not seem to be sustainable in the near-term at least given what they had to do to get that rate for one week. Not worth zero but certainly a difficult company to value.
That's before you mention if 7000 cars went through production. You could easily have stored 7000 cars half way through production and finished them over a week.

Heck, they could have 8000 cars parked waiting to go through quality control, discarded 1000 that didn't pass and rolled out 7000 cars.

The number is largely meaningless without further qualification.

Yes -- I would love a clear definition of what "factory gated" means and then an accounting of which cars were already close to finished. But I doubt we'll get that!
Do you know what the headline was 2 months ago? it was whether Tesla could make 2500 cars a week: https://www.theguardian.com/technology/2018/apr/17/tesla-hal...

(and they did miss that deadline)

They're now making in the territory of 5000 cars a week, almost double in 2 months. Regardless of how you interpret it as burst week or not, the production rate has ramped up significantly.

We are now seeing the goalpost being moved from whether Tesla can achieve this goal, to Tesla being lax with quality. Perhaps we'll hear soon 10K/week is unattainable. I guess that means they are making progress.

They aren't making, they are pushing out of the factory.

Each one of those cars could have taken 2 months to assemble, and they've just been waiting to finish and push it out.

Without more data, the 5000 cars out of the gate number is irrelevant.

Check Little's Law.

then was all the wall street articles 2 months ago, about missing the 2500 production goal, also irrelevant?

if meeting the 5000 production goal is irrelevant, then missing the old production goals would be just as irrelevant, right?

Meeting the goal isn't necessarily irrelevant, but without further data, it is irrelevant.

Tell me, do you consider "I can quality check 5000 cars in a week" equivalent to "I can produce 5000 cars in a week"?

Yes or no?

Tesla's valuation certainly is insane. They're valued higher than GM and Ford, the former moves over 100x more vehicles and both are actually profitable.

If you value Tesla as battery company that happens to make cars, it's just as silly since it's such a small portion of their revenue.

Lots of normal peeps invest in TSLA for moral/ethical/stylistic reasons. This in turn drives up the price.
Instagram is valued at 100B, Netflix goes neck and neck with Walt Disney. Not to mention Amazon. Market value has never been in lock step with revenue. And I'm sure you know it.

It's entirely possible Tesla will go burst. We'll see. However market value isn't a reliable indicator.

At least Instagram prints money (at the moment anyway). Netflix is almost as insane, but they're the (almost uncontested) leader in global online content distribution at the moment. Tesla is dead last against like 20+ competent profitable industry incumbents.
If you think there's 50/50 odds they'll either go bankrupt or be twice as big as GM/Ford some day, it's reasonable for their valuation to be on par with those guys (I don't think that btw, just trying to make sense of it).
It's Tesla's first step.

Just as SpaceX first showed that they could launch a payload to orbit, then showed that they could do it repeatedly; just as they first showed that they could land a booster, then showed they could do so dependably; just as they showed they could reuse a booster, then showed that they could do this rapidly and reliably.

First Tesla showed that they could meet their 5000 cars per week goal, now they have to show that they can do so repeatedly. I have a hunch that they'll do it.

They haven't shown that yet. They've shown that they can push 5000 cars out of the factory gates.

Pushing a car out of the factory gates is not equivalent to assembling a car.

It would be fair if they would provide accurate weekly production numbers for the last 6 consecutive weeks. Maybe the pattern was: 3000, 2000, 1000, 1000, 3000, 7000... Which is quite different than pattern: 3000, 3000, 4000, 5000, 6000, 7000...
If you’ve ever used software built in a similar “death march” fashion, you’d be scared to buy this batch, too.
Doubly so after hearing about the "semi-open outdoors tent in a hot climate" manufacturing strategy.

Who's to say some of the components aren't half melted or contaminated with natural elements from outdoors?

if a component melts in a tent how would you expect it to fare on a car, that sits outside in the sun all day rather often? around parts that often generate a lot of heat.
Are you saying that things can be manufactured in the same environment where they'll be used?

Even if all the parts are OK one would think not the whole manufacturing is temperature-insensitive.

I wouldn't worry if I saw Tesla as a responsible company, but I see them as desperate to achieve goals and capable of misleading people (e.g. autopilot being marketed as much more than what it currently is [1]).

[1] https://www.tesla.com/en_GB/autopilot?redirect=no : watch and read the embedded video too

As far as I know they aren't doing 'the whole manufacturing' in the temporary building, just final 'general assembly' of completed components.
They're being assembled. It's not like they're doing injection moulding or wave soldering circuit boards in open air. They have preassembled components that they're putting together.

The only thing I would be concerned about is dust or moisture intrusion.

Who'd to say it all hasn't turned to pasta either? Or they are importing special parts from Elons moon base. Who says they aren't doing that either?
A bit! But having spent some time in a model 3 lately, seemed fine!
Do you have any proof of the lax in quality of 3s being built, or just FUD? what does your experience with building bad software have to do with Tesla cars?
I believe in this FUD. I don't trust rushed jobs no matter who.
I drove past the Tesla plant in Fremont yesterday. The South lot had more than 20 auto-trailer trucks, fully loaded, and lined up to roll, sitting in a sea of 3s. It actually looked like they had produced so many that they didn't have enough drivers to haul them away.
>It actually looked like they had produced so many that they didn't have enough drivers to haul them away.

There's actually quite a large driver shortage right now. On one route we use at work, the rate doubled in the last year. The report was that some of the drivers are just fine with the life they were living before they got a big bump in pay, and want to drive less and make the same money.

Tesla did not want to hit the 200k domestic sale number till the start of a quarter to maximize the number of people who get the full $7500 federal tax credit. They were close to the number, so they have probably been storing up model 3's to hit the limit soon after July 1.
"7000 cars, 7 days" makes such a sticky tweet. It almost seems too perfect for someone not to assume some creative accounting/rounding here.