"Ah I will just quickly make a change to that old hobby project. Let me boot up the VM which has the only working development environment for this 2012-era Rails app."
*VM fails to boot on M1 Macbook*
"Oh no."
That was an hour or two ago, and I've been untangling the dependency tree to find some combination of software which is young enough to run on this processor but doesn't have total incompatibility with e.g. Rails 3.2.21.
I had finally gotten through fiddling with a bunch of gem versions, bundle installed successfully, DB re-created successfully, tried to install the seed data and then:
Now it has been a long, long time since I've done Ruby development on a daily basis, but I seem to remember that disagreements about the contract for Integer bubbling up from your web framework mean you're about to have a very, very bad day.
Currently wondering if I should just bite the bullet and buy a new computer to be able to run the old VM and save myself pain here.
Is this thread going anywhere? No, not really. This is one of those days I hate everything about computers. Save yourself, there is still time.
Hmm since I still have a working server on a VPS I wonder if I can clone the VPS, drop the DB, switch to development mode, rehydrate the DB, and then use the new VPS as my new VM...
thisisfine.gif
The current strategy involves a) upgrading to a newer version of VMWare to b) use their ovftool (the only one running on Mac required a new Fusion license) to c) convert the VM to an OVF to d) upload to DigitalOcean and make a VPS which I will e) SSH into to install Microsoft...
Remote Desktop from the command line then f) Remote Desktop from Microsoft's Mac app into the Linux VPS to access the dev environment that I know works, effectively turning the cloud server into a somewhat slower-than-locally-hosted VM.
...
I LOVE TECHNOLOGY.
Holy #*%(#%*( this actually worked.
Now I'm just connected via a relatively slow pipe to Singapore to do the thing I used to do locally.
Hmm should I force upgrade ubuntu precise to ubuntu bionic.
"How disruptive could that possibly be..."
(Yay for snapshots.)
Sorry, my lovingly artisianlly maintained "pet" devbox from 2006 through 2015ish: you are now officially "cattle."
After approximately 6 cumulative hours of DevOps (not counting upload time for about 100gb of images) I have successfully made a trivial edit to my old-as-the-hills side project!
I feel irrationally accomplished.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
My other favorite in this extremely specific genre is the Dagger and the Coin series, which as a banking/dragon crossover suffers from too little banking and too few dragons, but is otherwise one of my favorite fantasy series ever.
Back to DB: I want to spoil all the ways it is both a good and frustrating account of banking for a dragon but they'd be spoilers, and part of the joy of the book is "Hah nice idea but he hasn't thought of HOLY #%*#(* HE HAS."
On one hand, great demonstration of microecon 101. I was contributing to the problem; I'd just rent a new one on spec every morning regardless of whether I'd need it because I go to my local convenience store daily anyway and might need walking-around power at some point.
So I guess my advice is Charge More so people can charge more?
OK, obligatory joke out of the way, here's the reason why they'd give this service away for a month: 7/11 is fairly clearly paying them to, to buy foot traffic. Each rental/return cycle causes two trips in.
The $850 million was covered with an emergency equity injection (organized among large Tether customers), but I think industry observers have not yet cottoned onto the practical equivalency of that seizure and a fine for money laundering.
Tether has held out hope, including when raising the equity (the LEO token), that the money seized from Crypto Capital Corp's accounts will be restored to them after they demonstrate that they are the beneficial owners. Many people seem to believe this is the natural outcome.
I think the natural outcome is closer to "Oh if you are indicted by multiple governments for money laundering including for Columbian cartels then whichever jurisdiction the money was seized in will probably keep it and maybe, maybe, maybe we investigating governments divvy it."
It always struck me as absolute madness back in my software days a) how much working on the checkout drove sales relative to any product thing I could do for equivalent work and b) why I had to do this, because shouldn’t it be a solved problem.
The fundamental thesis of Stripe Checkout is that if you have less than a thousand engineers working for you there is no way to have a sufficient number of them working on your payments UX.
We centralize this engineering effort to keep you and your team free to do actual work.
This also lets us get returns to specialization, chase marginal opportunities which make no sense for most individual customers given effort involved but which are great probabalistically across network (“+1 bps for all Japanese customers transacting in dollars? AMAZING”), etc.
Note the non-disparagement clause in the order that Tether will, with absolute night follows day certainty, break, and probably within the next 48 hours.
A Japan anecdote on the hospitality industry: after being shown our seat at a hotel’s (very nice) buffet it was explained:
“While of course your children are welcome to anything at the buffet if they are hungry and you don’t want to be rushed we could give them prepared plates.”
“Oh sure.”
“Would you like a plate built around [fish / meat / curry / other mains]?”
“Two of the noodles.”
*They arrive in under a minute.*
“Please take your time in browsing the buffet.”
And of course because Japan they’re nutritionally balanced and presented well. (Someone cared about making the colors pop.)