Galanwe

58 ms·

391.
by Galanwe 2y ago
> The vast majority of hedge fund clients are high net worth individuals, with very few pension clients for the typical hedge fund. That is very, very wrong. HNW have to small tickets to be interesting for most hedge funds, where the min
392.
by Galanwe 2y ago
Very close to my experience, the most challenging course I had was "Semantics" during the 2nd year of Magistère. To this day I'm not sure I fully understand what I was taught during that course.
393.
by Galanwe 2y ago
The underlying theory is primary school math if you stay in a decimal fixed point format. All you have to do is choose the number of decimals you want. Say, half the number of digits of the biggest number that fits your desired size. Half t
394.
by Galanwe 2y ago
I think this is but a detail in the context of the article. Yes, you can implement your state machine on SIMD sized states, but there are also millions of other optimizations you can do like that, aligning your reads, prefetching the file,
395.
by Galanwe 2y ago
Looks nice, though in 2024 I don't have much files / photos / etc on my local computer anymore. My photos are now 100% taken with my phone, and they are searched / stored / synced through e.g. Google photo
396.
by Galanwe 2y ago
Last I used NuGet, circa 2013,it allowed packages to run arbitrary code during install, which resulted in dozens of shareware like pop-ups for trials.
397.
by Galanwe 2y ago
The downloaded packages themselves are shared in pip's cache. Once installed though they cannot be shared, for a number of reasons: - Compiled python code is (can be) generated and stored alongside the installed source files, sharing t
398.
by Galanwe 2y ago
From my experience, the issue is not the means to query the data, it's the actual idiosyncrasies of the data. The key metric is thus not "users seem to be more independent as we are receiving less inquiries for help", because
399.
by Galanwe 2y ago
So the customer data is actually stored on Snowflakes AWS accounts? What difference does it make what underlying storage / provider it uses then? Also does that mean every data query to snowflake goes out/in to/from internet
400.
by Galanwe 2y ago
The salesforce.com servers are temporarily unable to respond to your request. We apologize for the inconvenience. Thank you for your patience, and please try again in a few moments.
401.
by Galanwe 2y ago
I don't quite understand how Snowflake works. My understanding was that you had to grant storage access (e.g. S3) and compute access (e.g. EC2) from your account to Snowflake, which would then use said resources to perform queries that
402.
by Galanwe 2y ago
That extra 25% being unbearable self promotion makes it 100% more annoying.
403.
by Galanwe 2y ago
Or you need to have enough experience with any IaC tech that setting up a bucket with proper permisioning is a breeze. Once you get your own little pulumi/terraform library, these things really stop to be a pain.
404.
by Galanwe 2y ago
Would be nice to have it for Github Enterprise for sure!
405.
by Galanwe 2y ago
The article misses the point of JWT: it's dead simple to implement. I don't implement JWT because I fancy big data terascale technologies. I do it because it's mostly stateless, meaning it's very easy to mock locally, an
406.
by Galanwe 2y ago
Huh, I mean yeah gcc, clang and msvc, for desktop, servers and GP embedded hardware. That's like 99.99% of the compiling that most likely happen in the wild. Far from tunnel vision.
407.
by Galanwe 2y ago
Yeah, well, that's the kind of thing you say once you have so much money that effectively "giving most of it away" won't quite change your quality of life.
408.
by Galanwe 2y ago
I think it has to be more of a compiler driven movement. GCC and clang have to be the ones that accept to implement the standard of an alternative committee. Now I get this alternative committee doesn't exist yet, and there is no agree
409.
by Galanwe 2y ago
After 10m of using act I gave up, because of the millions of "oh so slightly different" environment than github actions.
410.
by Galanwe 2y ago
I don't know of anyone happy with the C++ standard committee. Some parts of the new std features were copied from Python, which is good, a semantic and naming a lot of people already know. Yet some weirdo decided to half copy the namin
411.
by Galanwe 2y ago
Employees are constantly saving cost or adding value, that's what they are paid for.
412.
by Galanwe 2y ago
When you think about it, the "winner takes all" martingale is already forbidden in most states of law. Two companies are not allowed to merge if their total market share would past 30%, because that would allow a monopoly and thus
413.
by Galanwe 2y ago
I think this boils down to the existence of megacorps. IMHO a state should just not allow corporations to grow beyond a certain marketcap level. I see no real world benefit to allow such companies to exist, it creates "too big to fail&
414.
by Galanwe 2y ago
> you can partition your dataset and process each partition on separate core, which will produce some massive XX or even XXX GB/s? Yes, but as I mentioned: > multicore decompression is not really possible unless you manually twea
415.
by Galanwe 2y ago
Last I checked you can't get much better than 1.5GB/s per core with LZ4 (from RAM), up to a maximum ratio < 3:1, and multicore decompression is not really possible unless you manually tweak the compression. The benchmarks above
416.
by Galanwe 2y ago
Well, see, that's the point of the conversation where I just don't follow the Rust logic. "Safe" and "unsafe" are scoped properties. They apply to the block of code that they encompass, by very definition of th
417.
by Galanwe 2y ago
The subtle differences (or lack of) between threads and processes don't really matter IMHO. A data race happen when two concurrent accesses on the same memory location happen, one of these accesses is a write, and there is no (correct)
418.
by Galanwe 2y ago
I don't think that's unrealistic, it all depends on which sphere of development you're navigating into. In highly concurrent/long lived systems, you often end up using a multiprocess approach between your producers and c
419.
by Galanwe 2y ago
You can have both data races and race conditions in a perfectly safe Rust program. Rust can only offer its safety for the lifetimes it tracks, and this tracking is limited to a single instance of your process. In other words, if you perform
420.
by Galanwe 2y ago
Rust safety has very little to do with bugs. Rust has its own definition of what "safe" means, which is arguably a subset of memory safety. You can have perfectly safe Rust code with data races, deadlocks, memory exhaustion, etc.

More ›