(Not wholly descriptive of things I’ve seen in my career but really important to understand broad strokes of how large software-oriented orgs operate compared to small shops and large non-software orgs.)
(And while there are lots of differences between and within companies it is, unsurprisingly, trivially true that e.g. eng team randomly picked at Google is more similar to one at Amazon than a randomly selected writer of code at Washington University or the IRS.)
(For one, astoundingly higher likelihood that an engineer randomly picked at Google has worked for Amazon than that they worked for IRS!)
It's a trueism in tech / finance that "Your most valuable asset leaves the building every night" (though I suppose these days might need to update that language) but underappreciated that things which resemble IP move around and are remixed as people move around and do remixing.
Aside: one of the reasons that hiring is sort of incestuous among a circle of tech firms is a belief that there is a very large body of industry practice that you need to know to be effective and that if you have 15 years in your career and 0 in big tech you're effectively green.
I always believed that to be disastrously untrue with respect to technical skills specifically but am coming around slightly with regards to the technical skill that is "knowing how to operate at AppAmaGooBookSoft", which includes e.g. dependency management by org chart not PR.
There's an obvious followup question "To what degree does this sort of thing describe development at Stripe in 2021?" and I think it deserves more considered writing by someone closer to it day-to-day; hopefully that happens sometime.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
That’s probably not “Rails create” but very well could be “Here’s how we deploy on Rails and why we made these choices.”
Or in banking not “Here’s what net interest is” but “Let’s talk about funding sources available to banks and the relative costs and limitations of them.”
I remain surprised by how much better my life got by the really simple expedient of making batteries a pervasively available utility, even though I'm not often away from an electrical outlet during coronavirus compared to more typical life pattern.
The dominant use case for me is "Drop daughter off at school, would really like to go to a cafe and enjoy breakfast but cannot justify it if cell phone is off, see I only have 4% and think about going home... pop into konbini, rent battery for ~$3, get to enjoy coffee."
And it's such a good business for all parties that the local konbini moved the station, which both handles payments (with the app on your phone) and fulfilment of the battery, away from the register because frequent transactions were blocking foot traffic to the checkout.
Also there is culturally more willingness to grind a bit on marketing and sales where a lot of us back in the day reaaaaaally wanted to make a software vending machines, put it on Internet, and never have to talk to anybody.
An interesting thing from the crystal ball: a lot of what has historically made this form of software entrepreneurship attractive is that most founders of it live somewhere where they are the best game in town for a software dev.
One of many articulations of the thought, though I wouldn't use the word "bubble" these days; in core tech (ex-crypto) degree of investment seems extremely rationale to me.
Software takes money to build, and that money has historically been raised by professional money people.
Software makes software people money.
It is very not obvious that professional money people add value other than the money, and to the extent they don't, why have them.
A very useful thing to know, if you sell things that could be used by a company that would have need for a hundred pretty nice workstation GPUs, is how this calculation gets made in practice:
Technologist: Residual value is $4k contingent on sale.
Manager: Do we have a buyer on the approved vendor list who will do The Usual with less than one phone call required?
Technologist: No.
Manager: Trash them.
Technologist: Why?!?
Manager: We’d only spend hundreds of dollars of staff time, I.e. yours, if this was a simple matter of boxing them up and giving them to vendor. That might or might not be worth it given typical ROI on your time given current workload/priorities. However.
So Bitfinex/Tether apparently lost about $23 million to an operational error in setting Ethereum fees, which would be terrifying due to low equity cushion if regulated.
OTOH, most crypto folks believe that Tether effectively has backup up to full equity value of Bitfinex/Tether.
In the more formal financial world there is the so-called “source of strength” mandate where if you own a bank your holding company/etc is required to recapitalize a failing regulated subsidiary even if, considered from its own perspective, it would strongly prefer not to.
The argument in crypto is “I mean sure they’re not required to recapitalize Tether but everyone knows they would; it would torch their rep if they didn’t.”
To which one could make observations about the bail-in that happened.