necubi

8 ms·

61.
by necubi 2y ago
This appears to be for low-power embedded devices, not anything that would have a GPU
62.
by necubi 2y ago
Automatic reference counting is a form of GC, and optimized versions of it end up very similar to tracing GC: https://web.eecs.umich.edu/~weimerw/2012-4610/reading/bacon-... (except that ARC can't handle
63.
by necubi 2y ago
Thanks for pointing that out! I've updated the table to use speedups rather than (incorrectly computed) percentages.
64.
by necubi 2y ago
The schema is only known at runtime, which prevents ahead-of-time code generation. JITing would be probably be an improvement though (at least in our use case, stream processing, where we're basically always willing to pay higher upfro
65.
by necubi 2y ago
I'm not very familiar with the internals of serde_json, but it's largely solving a different problem. Serde_json is deserializing individual records (rows), and typically using compile-time code generation. (Alternatively, it can
66.
by necubi 2y ago
Interesting article! I climb at Berkeley Ironworks which is the successor to City Rock, but didn't know all of the history. The story ends in 2000 when Ironworks "represented the next generation of climbing gyms", but the tre
67.
Fast columnar JSON decoding with arrow-rs(arroyo.dev)
56 points by necubi 2y ago|7 comments
68.
by necubi 2y ago
No worries! We definitely rely heavily on DF (it’s an incredible project!). Part of what makes it so great is its modularity—it’s a toolkit for building sql systems, which is extremely cool.
69.
by necubi 2y ago
This isn't quite correct (I'm the creator of Arroyo). We use DataFusion to implement parts of our SQL support (in particular the planner and the expression interpreter) but we have our own dataflow and operators. By contrast Synna
70.
by necubi 2y ago
I don't follow your first point—the compiler is pointing out exactly what the problem is (the argument has the incorrect type) and then telling you what you likely wanted to do (borrow the String). What would you see as a more helpful
71.
by necubi 2y ago
In reality, this practice long predates modern corporations ( https://en.wikipedia.org/wiki/Shanghaiing )
72.
by necubi 2y ago
Funny enough, exactly the same at Lyft. Redis everywhere. The original version of the dynamic pricing system was a series of cron jobs reading and writing to redis, before it was replaced with a Flink pipeline (which still wrote to redis fo
73.
by necubi 2y ago
Please don't post LLM generated comments on HN.
74.
by necubi 2y ago
That's not correct. It's for people who think they might be targets of surveillance operations, like journalists and activists. > Lockdown Mode is an optional, extreme protection that’s designed for the very few individuals who
75.
by necubi 2y ago
I think it's probably worth learning a little bit about economic history before talking about the benefits of the gold standard. Before the fed was created, the US economy was constantly booming and busting. There were recessions ever
76.
by necubi 2y ago
The article is about how bad land and energy policies (essentially central planning)—dating from the post-war period—are holding back the British economy. Is that neoliberalism?
77.
by necubi 2y ago
DuckDB is an OLAP (analytical) query engine, sqlite is an OLTP (transactional) database. Modern OLAP engines store and represent data in columnar formats, which makes them very fast at queries that touch many rows (particularly if only a fe
78.
by necubi 2y ago
The parent is right, investors would almost certainly have 1x preference, which means they get back the first $230M in an acquisition. The valuation doesn't matter if you exit for less than money in. Founders and employees would get no
79.
by necubi 2y ago
> Device Timeline: Your Ai Pin will continue to function normally until 12pm PST on February 28, 2025. After this date, it will no longer connect to Humane’s servers, and .Center access will be fully retired. > Device Features: Your A
80.
by necubi 2y ago
Can you name a single ICO that produced something of value before the SEC cracked down on what are (obviously) unregistered security offerings?
81.
Autopsy: No foul play in OpenAI whistleblower's suicide(sfstandard.com)
22 points by necubi 2y ago|6 comments
82.
by necubi 2y ago
"In Ascension" is really wonderful sci-fi novel set partially on Ascension Island. Recommended if you like Ted Chiang or Jeff VanderMeer style weird, meditative sci-fi.
83.
by necubi 2y ago
He’s talking about local governments in places that have little or no fire risk. Should people living in dense cities be paying to underground power lines to rural communities hundreds of miles away?
84.
by necubi 2y ago
Mexico vaccinates its chickens, the US does not ( https://www.unmc.edu/healthsecurity/transmission/2023/02/17/... )
85.
by necubi 2y ago
I assume they're talking about the longer history of the distributed systems lab at Berkeley, which was AMP before RISE. (It's actually now Sky Lab[0], each of the labs live for 5 years). AMP notably is the origin of Spark, Mesos,
86.
by necubi 2y ago
In 2022 Lightbend relicensed Akka from Apache 2.0 to BSL, which was a huge problem for all of the other opensource projects (like Flink) that used it as part of their coordination layer. At this point most or all of them have moved to Pekko
87.
by necubi 2y ago
I was a bit unclear there—I was referring to templates, not C++ macros. I believe that concepts in C++20 have improved the situation (although I haven't personally worked with a C++20 codebase and don't know how widely adopted it
88.
by necubi 2y ago
I'm not necessarily going to defend proc macros, which have a lot of problems but are also better than the alternative in many languages (like Go's over-reliance on codegen). Macros-by-example are pretty nice though, even with the
89.
by necubi 2y ago
I think they may be too quickly assuming that there's never a need for long-lived branches. For example, for any company (like mine) that does on-prem deployments, we need to support old versions for an extended period of time by backp
90.
by necubi 2y ago
What vast number of layoffs? The national layoff rate is currently 1.1%, which is lower than the the pre-pandemic average [0]. The NYS layoff rate is 0.5% which is one of the lowest in the country. [0] https://fred.stlouisfed.org

More ›