Memory is Short
Jul. 18th, 2021 10:54 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
walkitoutI went over to Ars to see if there was anything interesting. There was! (More than a million dollar auction sale of a super mario sous vide. So. Weird. Moving on.) TSMC has decided that they are going to build more old-skool lines, both to supply specific customers who are pissy (car makers, presumably) and also just because there’s a lot of demand out there. They used to not do this, because staying right out at the edge let them collect all the profit and there wasn’t that much residual demand that they couldn’t service, and other old-skool, once relevant fabs picked up the crumbs that fell from the table. No biggie.
Of course the pandemic changed all that. Some months ago, when I was debating whether to buy a friend a Portal so our kids could use Portal for their cross country play dates, I had a series of angry conversations with my husband and others (oh, the damage I can do to relationships when I get an idea in my head!) about how long the chip shortage might last. I bought some TSMC stock and then waited for the news to roll out to confirm my theory. I was feeling pretty sure of myself.
It did, and, true to form, _that_ didn’t repair any relationships (ok, there was never really that much damage to my marriage — my husband knew me for over a decade before we got married, so he had a pretty good idea of the nature of the pig and the poke, shall we say). Being right does not make anyone love one. But highly emotional moments really sear themselves into my memory and also, I blog, so I have a number of ways of keeping track of What Happened When.
If only other people did that, too.
https://arstechnica.com/information-technology/2021/07/tsmc-signals-global-chip-crunch-may-be-easing/
There are problems with the article. Let’s start here:
“In the first six months of 2021, TSMC increased its output of micro-controlling units, an important component used for car electronics, by 30 per cent compared with the same period last year,“
I don’t doubt that this is an accurate representation of the call, however, it would have been nice if the journalist or someone, really, anyone had instead asked for the comparison to the same period in 2019. After all, the first six months of 2020 included a lot of pandemic related disruption, here, there, and everywhere. 30% better than a total disaster is … what, compared to pre-disaster?
Also:
“MCU production is expected to be 60 percent higher for the full year than in 2020, it added.”
Likewise — how does this compare to 2019? Also, how does this compare to what they projected for 2020 and 2021 pre-pandemic? I mean, if, in 2019, you were humping it to increase capacity by 30%, and instead you halved … well, you can see how all this is pretty Unhelpful.
https://www.freep.com/story/money/cars/2021/06/15/car-chip-shortage-2021/7688773002/
“The crisis began after carmakers pulled chip orders last fall, leaving them without supplies when demand suddenly surged weeks later.” (This is from the Ars article. The Free Press article gives a more accurate summary of the timeline.)
I think that this is a false statement. The orders may have been for fall delivery — I don’t really know — but the orders were canceled back in the spring or early summer of 2020. Not. Fall. Demand did not surge weeks later; it surged months later.
Altsuperego — and at the time I write this, it was 35/2 on votes, so positively received — said: “I just don't understand how the bean counters at every major auto OEM thought it was worth canceling their orders to save, what, a few million bucks?”
Well, on one level, the Ars article addresses this — they’d seen recessions before, and gotten hammered when they produced at pre-recession levels going into a recession. They’d figured out to taper early, but then this recession wasn’t like other recessions. Also, what altsuperego apparently has already forgotten is that back when those orders were canceled, vaccines were just being created and hadn’t been tested _at all yet_, and the broad expectation was that we’d be lucky to get something 50% effective, and the timeline on getting a vaccine in the past had been over a decade. Other commenters make these and other observations about the errors embedded, but mostly, I’m interested in the failures of memory necessary to make these mistakes in the first place. If you are old enough to be writing for FT (where the article at Ars originated) or commenting on it at Ars, you _probably_ are old enough to remember what it was like in March and April of 2020.
These are _not_ the only people to forget these very recent, indelibly printed in my memory, events!
https://www.theatlantic.com/science/archive/2021/07/unsolicited-seeds-china-brushing/619417/
That article is ludicrously long; you should just skip the first 3/4s of the article and start reading at the paragraph that begins, “I started with the market leader, Amazon.” This lets you actually read about the real journalism and the real conclusions, without wasting a lot of time on the Nonsense. What actually happened with the Mystery Seeds was that people ordered seeds in March and April of 2020 from 3rd party sellers on Amazon, because all the regular sellers were Sold Out. At that point, Amazon had a lot of new customers who weren’t yet up to speed on shipping speed variations, FBA, Amazon as a seller vs. Amazon as the market, etc., and so people ordered from 3rd party sellers in China and didn’t realize they wouldn’t get their seeds for a couple months. Because seeds are regulated across borders, the Chinese sellers marked them as “jewelry” or “stud earrings” or whatever for customs purposes, so they would pay a high enough amount at customs that customs wouldn’t look too hard and notice they were evading the regulatory aspects of seed shipping. (The article does not explain that part in any detail, alas.) Also, earrings are about the right form factor. Anyway. The seeds arrived at the recipient’s home, who either forgot they’d ordered it, or who was receiving it as a gift and didn’t know or had forgotten or whatever.
Anyway. Over the next months and years, I expect to see a _lot_ more of this kind of WTFery, where people who lived through the last year and a half nevertheless behave as if they were on another planet and missed the whole thing.
Of course the pandemic changed all that. Some months ago, when I was debating whether to buy a friend a Portal so our kids could use Portal for their cross country play dates, I had a series of angry conversations with my husband and others (oh, the damage I can do to relationships when I get an idea in my head!) about how long the chip shortage might last. I bought some TSMC stock and then waited for the news to roll out to confirm my theory. I was feeling pretty sure of myself.
It did, and, true to form, _that_ didn’t repair any relationships (ok, there was never really that much damage to my marriage — my husband knew me for over a decade before we got married, so he had a pretty good idea of the nature of the pig and the poke, shall we say). Being right does not make anyone love one. But highly emotional moments really sear themselves into my memory and also, I blog, so I have a number of ways of keeping track of What Happened When.
If only other people did that, too.
https://arstechnica.com/information-technology/2021/07/tsmc-signals-global-chip-crunch-may-be-easing/
There are problems with the article. Let’s start here:
“In the first six months of 2021, TSMC increased its output of micro-controlling units, an important component used for car electronics, by 30 per cent compared with the same period last year,“
I don’t doubt that this is an accurate representation of the call, however, it would have been nice if the journalist or someone, really, anyone had instead asked for the comparison to the same period in 2019. After all, the first six months of 2020 included a lot of pandemic related disruption, here, there, and everywhere. 30% better than a total disaster is … what, compared to pre-disaster?
Also:
“MCU production is expected to be 60 percent higher for the full year than in 2020, it added.”
Likewise — how does this compare to 2019? Also, how does this compare to what they projected for 2020 and 2021 pre-pandemic? I mean, if, in 2019, you were humping it to increase capacity by 30%, and instead you halved … well, you can see how all this is pretty Unhelpful.
https://www.freep.com/story/money/cars/2021/06/15/car-chip-shortage-2021/7688773002/
“The crisis began after carmakers pulled chip orders last fall, leaving them without supplies when demand suddenly surged weeks later.” (This is from the Ars article. The Free Press article gives a more accurate summary of the timeline.)
I think that this is a false statement. The orders may have been for fall delivery — I don’t really know — but the orders were canceled back in the spring or early summer of 2020. Not. Fall. Demand did not surge weeks later; it surged months later.
Altsuperego — and at the time I write this, it was 35/2 on votes, so positively received — said: “I just don't understand how the bean counters at every major auto OEM thought it was worth canceling their orders to save, what, a few million bucks?”
Well, on one level, the Ars article addresses this — they’d seen recessions before, and gotten hammered when they produced at pre-recession levels going into a recession. They’d figured out to taper early, but then this recession wasn’t like other recessions. Also, what altsuperego apparently has already forgotten is that back when those orders were canceled, vaccines were just being created and hadn’t been tested _at all yet_, and the broad expectation was that we’d be lucky to get something 50% effective, and the timeline on getting a vaccine in the past had been over a decade. Other commenters make these and other observations about the errors embedded, but mostly, I’m interested in the failures of memory necessary to make these mistakes in the first place. If you are old enough to be writing for FT (where the article at Ars originated) or commenting on it at Ars, you _probably_ are old enough to remember what it was like in March and April of 2020.
These are _not_ the only people to forget these very recent, indelibly printed in my memory, events!
https://www.theatlantic.com/science/archive/2021/07/unsolicited-seeds-china-brushing/619417/
That article is ludicrously long; you should just skip the first 3/4s of the article and start reading at the paragraph that begins, “I started with the market leader, Amazon.” This lets you actually read about the real journalism and the real conclusions, without wasting a lot of time on the Nonsense. What actually happened with the Mystery Seeds was that people ordered seeds in March and April of 2020 from 3rd party sellers on Amazon, because all the regular sellers were Sold Out. At that point, Amazon had a lot of new customers who weren’t yet up to speed on shipping speed variations, FBA, Amazon as a seller vs. Amazon as the market, etc., and so people ordered from 3rd party sellers in China and didn’t realize they wouldn’t get their seeds for a couple months. Because seeds are regulated across borders, the Chinese sellers marked them as “jewelry” or “stud earrings” or whatever for customs purposes, so they would pay a high enough amount at customs that customs wouldn’t look too hard and notice they were evading the regulatory aspects of seed shipping. (The article does not explain that part in any detail, alas.) Also, earrings are about the right form factor. Anyway. The seeds arrived at the recipient’s home, who either forgot they’d ordered it, or who was receiving it as a gift and didn’t know or had forgotten or whatever.
Anyway. Over the next months and years, I expect to see a _lot_ more of this kind of WTFery, where people who lived through the last year and a half nevertheless behave as if they were on another planet and missed the whole thing.