willseth

14 ms·

61.
by willseth 2y ago
Do you think you might like management if you knew it didn’t have to be permanent? Highly recommend Charity Majors’ blog about this if you haven’t already https://charity.wtf/2017/05/11/the-engineer-manager-pe
62.
by willseth 2y ago
If someone is looking for a job at 40 and they haven’t already built a big network, it’s not helpful to tell them they should build one. But if you’ve been working for about 20 years you probably do have at least some friends or colleagues
63.
by willseth 2y ago
I think you mostly only run the risk of pigeonholing yourself as a “startup person” if you jump on another startup. But startup people value startup experience, so as along as you enjoy startup work and are making solid contributions, I thi
64.
by willseth 2y ago
Have you used Pixelmator Pro? It’s far too complex to shoehorn into the Photos app. It also doesn’t make sense because like Photoshop Pixelmator images don’t necessarily originate from a photo. How do you create a blank document? Both pract
65.
by willseth 2y ago
Which things did you decide to move to duckdb?
66.
by willseth 2y ago
20k isn’t noise to most people in a 200k salary range.
67.
by willseth 2y ago
Your original comment as written claims that numpy cannot scale using threads because of the GIL. You admit that is wrong, but somehow can't read your comment back and understand that it says that. What you really meant was that combin
68.
by willseth 2y ago
This is your original comment, which as stated is simply incorrect > numpy isn't written in Python. However, there is a scalability issue: they can only drive so many threads (not 1, but not many) in a process due to GIL. Now you ha
69.
by willseth 2y ago
But what are you basing your personal judgment on? What would have been reasonable to you? My comment was about how under the EU's framework, Apple's behavior is totally rational. If Apple complies more than necessary, it could co
70.
by willseth 2y ago
Python has no threads or processes hard limit, and the pure Python code in between calls into C extensions is irrelevant because you would not apply multithreading to it. Even if you did, the optimal number of threads would vary based on wo
71.
by willseth 2y ago
This is misleading. Most of the compute intensive work in Numpy releases the GIL, and you can use traditional multithreading. That is the case for many other compute intensive compiled extensions as well.
72.
by willseth 2y ago
How are they not attempting good faith interpretation of the law?
73.
by willseth 2y ago
“just”... The DMA is intentionally vague about how to comply and the EU refuses to tell companies ahead of time whether their plans to comply are actually in compliance. When the amount of money at stake over the long term is far more than
74.
by willseth 2y ago
Intel typically calls their iGPU architecture "shared memory"
75.
by willseth 2y ago
I agree but some people love it. The rest of have a Logitech MX
76.
by willseth 2y ago
It seems you don't understand the lawsuit. Most of the claims are based on the actual mechanics of how algorithmic price fixing does violate existing laws.
77.
by willseth 2y ago
That is the entire basis for the RealPage lawsuit. The point is that if the effect on pricing is indistinguishable from price fixing, it doesn't matter if the act of colluding is abstracted into and laundered through a 3rd party with a
78.
by willseth 2y ago
I have a 14 Pro, and the main reasons I want to upgrade are for Apple Intelligence (mostly Siri-based) and for the better camera and video (as my family's default photog). If I didn't have kids I could wait a few more years on the
79.
by willseth 2y ago
The code he optimized sped up at least 1-2 orders of magnitude
80.
by willseth 2y ago
We use Doppler and it’s been great.
81.
by willseth 2y ago
What's premature? 50% of their runtime was spent in this one function call
82.
by willseth 2y ago
You're not wrong, but despite being complicated, this codegen is less than 200 lines of comment heavy code. This might have been a very light lift in comparison to whatever fundamental architectural changes would be necessary to solve
83.
by willseth 2y ago
It's pointless to argue what's "typical" when actual results are incredibly varied and specific to each operation. Also you didn't read the rest of my comment.
84.
by willseth 2y ago
Ime 2 orders of magnitude is closer to best case, assuming you have already made reasonable Python-based optimizations. But this problem has two components, function call overhead and conditionals. Moving the original logic into a compiled
85.
by willseth 2y ago
Yeah, and you could run a warmup script on deploy so no actual users pay for it. It's not a rare pattern.
86.
by willseth 2y ago
The problem isn't simply that repeated function calls are expensive, it's that the comparator logic is very branchy in order to support the different comparator options for their queries, even though any given query will only perf
87.
by willseth 2y ago
The improvement was 2x across the entire application, i.e. this hotspot was 50% of the overall runtime, and his bytecode optimization reduced that to an insignificant %. The improvement to the code he actually optimized was probably at leas
88.
by willseth 2y ago
Schiller reports to Cook. You are literally making things up to suit a definition of "organized under marketing" that no one uses.
89.
by willseth 2y ago
Now you're not only moving the goalposts, you're fabricating your own! App Store is not organized under marketing at Apple. Simple as that. You think there are too many "marketing types" running the App Store. Fine. That
90.
by willseth 2y ago
None of this supports your claim that these roles are "within the marketing department". Phil Schiller is a _former_ marketing VP, a position he left 4 years ago. Carson Oliver has been business management roles his entire tenure

More ›