There are probably many reasons for this, but here’s my take: the GA community is largely self-policing. There are dozens of regs one could break — whether it’s 8 hours bottle to throttle or busting VFR cloud mins — however the pilot community adheres to them almost religiously. To have someone so blatantly and publicly violate regs with a complete disregard for safety and property would be setting a terrible precedent for the FAA.
Additionally, this is in the best interests of the GA community as a whole, given that it always has been and will continue to be under scrutiny from the general public. No one wants there to be the perception that among them there is a 100LL cowboy who’s gonna bust through a bravo, slam a red bull, and ditch their plane over a neighborhood.
The hammer needs to come down, and it needs to come down hard.
Woof. At the rate packages get updated these days, and the amount of dependencies between them, that just isn't sustainable for any reasonably-sized project in server and -- especially -- frontend land.
It is implemented pretty well in a few languages. For ruby for example it's almost trivial to maintain a `vendor` directory that matches the current `Gemfile` and `Gemfile.lock`. The size changes without LFS mean that's a bad idea, but... you can do it.
I wonder what monetary loss in productivity was due to this change. We noticed this issue a bit before noon, tracked it down to GH, sent out company-wide comms notifying others of the problem, filed tickets with GH, had to modify numerous repos across multiple teams, and now it's 3pm and I'm here reading about it.
It's crazy how such a seemingly innocuous change, like this, could lead to such widespread loss in productivity across the globe.
Our conda-forge package builds broke. We had someone declare to us that tag downloads were never stable, just releases. This seems to be the opposite of the known truth about the previous status quo - but does go some way to demonstrating how little the state of the actual guarantees for this system were understood.
How would anyone (outside of GH) have known this? The checksums have been stable for years, and this issue resulted from an internal update to the version of Git being used. It also was not publicized, until this ex post facto blog post
This is a weird take, and I highly doubt retention is the primary reason they choose to limit the use of their web interface. In fact, I’m generally suspect of the hand-wavy speculation that some less savvy business unit made a decision in isolation. The two drivers that I would expect are: reducing friction and native compute capabilities.
People generally gravitate towards the path of least resistance, which is something you already see companies exploiting with things like unfavorable privacy settings defaults. Having an app is another instance of that. There is much less mental friction in pushing a button on a home screen compared to opening a browser, punching in a URL, possibly having to log in, etc.
Given that this is rideshare, this is important, because the services offered by other companies are effectively identical. If someone has Uber installed, they would be much more likely to go the path of least resistance — using the installed app — as opposed to going out of their way to use an equivalent offering.
Additionally, there are problems in mapping that would be very difficult, if not impossible, to solve in a browser. For example: localization (precise positioning) would be incredibly difficult without a native library, you wouldn’t be able to do any offline re-routing, like gmaps does, among other limitations.
Just this afternoon I was working on reversing a closed source library that wasn’t working on M1 under Rosetta, using Ghidra. If you get the chance, you should do a post on how you actually modified the program to get it to do what you want (as long as the fix isn’t trivial, like changing a constant).
My exercise today made me realize just how much more difficult the modification of the binary is than simply understanding it, as well as how much I hate the x86 architecture (and CISC in general).
There’s actually a pretty interesting approach in which: voters can see that their vote counted, everyone can tally anonymous votes, and it is resistant to DoS https://people.csail.mit.edu/rivest/pubs/RR14b.pdf
Of course, the real problem would be explaining to the general population that this works. Great in theory, but will never pan out
Big Ron can ask you to take a picture of the paper ballot as well, so regardless of mechanism you’re in a pretty sticky situation.
It’s not a perfect system, but I would personally find it comforting knowing that I could verify that my vote counted and that I could independently confirm the election results
Edit: haven’t read the paper in a while, but IIRC you can see that the vote counted without revealing details of who you voted for. They definitely would have covered this case
The complexity is not the depth of the problem, but rather the scope. Matchmaking — while complicated — is a very small subset of the domain. Pricing, demand forecasting, routing, payments, etc. are all equally important on the pax side. Similarly, there are many teams needed to support the overall ecosystem for things like driver onboarding, compliance, and support.
Also comparing a 1% return to a 30% charge isn’t exactly fair. Fixed costs like background checks aside, chargebacks, fraud, insurance, support claims, incentives among others warrant the markup
I could be completely clueless here, but it looks to me like a lot of the Lyft/Uber/etc cut is going to:
1) Overpaid, wasteful software engineering (no offense to us). Like someone here said, there are a lot of cheap software solutions out there to handle this stuff
2) R&D into crazy moonshot ideas
3) Advertising
4) Making a completely opaque techno-dystopian game out of people's livelihoods in order to manipulate drivers into working harder for less money and riders into paying less up front and as much as possible in the long run
If I'm right, I'd ideally like to completely destroy this market for good. Fragment it into a separate crappy home-made app for each city that you find out about via a poster on the wall in the airport when you land. I'm hoping that with enough transparency and flexibility, riders and drivers would keep each other in check and come to a happy and sustainable middle ground without making any VCs rich
(Or maybe Uber/Lyft are operating on a level of efficiency that would make my half-baked idea look like a joke. Ain't nobody disrupting Amazon or Apple from their basement any time soon)
Self-driving cars is not such a moonshot at this point, and indeed may be the only way they reach profitability (still a while off, though, and Lidars need to come down in price)
I think that ultimately a “second brain” system belongs at the OS level — as opposed to application — as it can greatly reduce friction from an interaction standpoint (e.g. double click iPhone power button to access your “brain” as opposed to pulling up a NFC credit card). I even went as far as getting the system concept patented while at Microsoft some years back: https://patents.google.com/patent/US20180349497A1/en
Additionally, this is in the best interests of the GA community as a whole, given that it always has been and will continue to be under scrutiny from the general public. No one wants there to be the perception that among them there is a 100LL cowboy who’s gonna bust through a bravo, slam a red bull, and ditch their plane over a neighborhood.
The hammer needs to come down, and it needs to come down hard.