I relied upon Mathematica / Wolfram Language heavily for a couple of decades, was an active member of the community of sophisticated users, and had my own work reported on and reprinted by Wolfram themselves in their official blog. About two years ago I decided to shift to Python. It's a much less elegant language and I'm nowhere near as productive in it, but the shift may have nonetheless been the single most important decision I have made in my career. Within a year I had abandoned elegant code, swiftly written but useless to anybody but me, and was able to solve problems in my firm's existing codebase. The impact for me, and for my firm, has been significant and entirely positive.

I keep a Mathematica license but only to solve math problems. I may never write another line in Wolfram Language.

The video card has been upgraded. The USB ports have been upgraded. The optical drive was upgraded. The drives have been upgraded repeatedly. The RAM has been upgraded. The ROM was flashed to a (slightly) newer version. It's running 24/7/365 in an unheated garage and I figure I'll keep it in its current role until it finally dies, at which point it will probably be replaceable by a $50 Raspberry Pi.

I love the 3.4 / 5 rating for that.

From the article, "Most likely, the latest RockYou iteration contains information collected from over 4,000 databases over more than two decades."

No need to surmise — Apple states that they kept battery life constant in the move from M2 to M4, and early reviewers report that this appears to be true.

Part of the challenge in going electric is that electric motors don't thump and growl, and the feel and noise of the vibrating engine is part of the fun of Autotopia.

At my high school they had Apple IIs that ran UCSD Pascal (and had a special card in them enabling this, maybe?), and a single IBM PC (5150 I think but maybe an XT) that could run Turbo Pascal. Turbo Pascal compiled so much faster than USCD Pascal that it seemed almost like magic. Plus the computer had the amazing IBM buckling-spring keyboard. I remember writing a Hammurabi clone on that system (https://en.wikipedia.org/wiki/Hamurabi_(video_game)).

I think your understanding of Thrasio is a bit off. They bought the individual stores. They may have eventually shared a software platform with those companies but that wasn't part of the original pitch. Rather, if you established "Gilbert's bicycle wheels" as a vendor on Amazon and managed to turn a profit selling wheels in the marketplace, Thrasio might offer you a lump sum in exchange for your brand name (?) and future cashflows. There was earn-out provision, presumably, to keep the sellers working, but I don't remember the details if I ever even knew them.

The individual shopkeepers may have used Alibaba-and-operations-management software from Thrasio, or third parties, or developed it themselves, what Thrasio cared about was gross profit.

Upper90 pitched me repeatedly raising money for Thrasio. There were multiple rounds of financing, all very large. I could probably find the pitch decks if I dig deep enough. The basic idea was "people can make money as Amazon marketplace vendors. Some people can do this in a smart, automated way, figuring out what things to bulk-import from AliBaba, how much to spend promoting them, etc." Most of the financing was equity-style I think; you put money in now and hoped that at some point down the road they were able to sell Thrasio to somebody else (but to whom?).

Personally I never understood the business; my theory is that if it's profitable enough to be interesting to Upper90's limited partners, Amazon will inevitably squeeze them to capture the margin for themselves. That company doesn't leave a dime on the table.

But that may not be what happened — Thrasio management say the business will continue to operate through and after bankruptcy. If margins had dropped near zero, they wouldn't bother. We'll find out more during the bankruptcy process.

I've bought all my Apple hardware from their refurb store for years, with no regrets. That typically saves at least 10%, even if the thing you're buying is still being sold new by Apple. Stacking a military or .edu discount gives you another 10%. Use the Apple Card and you've saved another 3%. Compounded together, a haircut of at last 21% from the price of the a machine.

In the years since I signed up for Netflix Disc, I've moved several times. I've changed jobs several times. I've gotten married, had kids. Netflix has been part of my family longer than most things I own. There may not ever be another source for the potpourri of titles that Netflix Disc carried. Does any streaming service have Heißer Sommer, the East German beach musical? How about Dick Powell's 1937 musical "On the Avenue" or even 1985's "Prizzi’s Honor"? How many services do you need to subscribe to if you have a diversity of taste, or a family with people of different preferences? Netflix Disc was great and we'll all miss it.

"Bernie Madoff, home chef"?

"Jeff Skilling and Ken Lay, always ready to help a friend move furniture"?

"Jeffrey Epstein, kind to animals"?

I've been a member of Netflix since their initial promo offer back in 1998(?). They let each member's queue hold 500 films, and mine has been pegged at that number forever; as quickly as we watch a movie, we come up with others we want to watch. The selection on Netflix discs is *much* wider than on any streaming service. With Netflix disc you can decide to watch every Winona Ryder movie, or every Elliott Gould movie, or every episode of a ten-year-old French crime drama. How do you do that with streaming, unless you subscribe to fifteen services?