I was a hurdler in high school and mastering stride length was almost the entire point of practicing. It's equally weird to me to see someone claiming to be fluent in English who has never heard the word. Maybe a reminder that we're not as knowledgeable as we think we are and what we choose to consume on YouTube is a tiny smittance of human experience. Running is a fairly universal and important thing for nearly any land animal, hardly a niche thing to talk about, but if you had ever talked to or listened to runners speaking English, you'd have definitely heard them talking about their strides.
I don't now if you consider it "reasonable" but the Gigabye Aorus TRX boards even from 6 years ago came with a free PCIE expansion card that held 8 M2 sticks, up to 32 TB on a consumer board. It's eATX, of course, so quite a bit bigger than an appliance NAS, and the socket is for a threadripper, more suitable for a hypervisor than a NAS, but if you're willing to blow five to ten grand and be severely overprovisioned, you can build a hell of a rig.
Are you sure? I've seen plenty of motherboards bundle a PCIe riser to passively bifurcate the PCIe slot to support four M.2 drives in an x16 slot or two in an x8 slot, but doing eight M.2 drives in one PCIe slot would either require a PCIe switch that would be too expensive for a free bundled card, or require PCIe bifurcation down to two lanes per link, which I don't think any workstation CPUs have ever supported. And 32TB is possible with just four M.2 SSDs.
I don't think it's a mistake. Non-subscription without clear gamification and engagement tricks seems like Rosetta Stone basically and that's the longest-running, most effective language learning software I'm aware of. When I joined the Army 18 years ago, while waiting for school assignment, I had the opportunity to train provincial reconstruction teams as a role player and had to cram learn some reasonable level of conversational Pashtun and that was what we used, before smart phones were widely adopted. The only real innovation I've seen since then is conversational LLMs that allow you free-form practice without a human partner, but even back then, scripted conversational practice was pretty good as long as there was a wide enough diversity of scripts.
Problem is more along the lines of "solo developer" here. Hacker News seems to have a real thing about this niche for whatever reason, but when doing something like this that I think requires real expertise in a wide variety of subjects that aren't software development, I think you need help. There's no way something like Rosetta Stone was developed without the input of experts in language learning and teaching, for instance. Knowing the platforms, programming languages, frameworks, and app store onboarding and delivery processes is already a lot for one person, but expecting to know the target domain on top of that is expecting an awful lot from yourself. I don't think it's a great sign trying to crowdsource business strategy from a free web discussion board, as a single example. This is the kind of conversation you want to have with your private team of people you know for sure have the experience they claim to have, not anonymous comments.
This is going to sound like a gripe and I swear it isn't, but is there a plan for a reasonably full suite of minimal userspace tools? I don't expect util-linux itself in Rust, but something like:
- coreutils
- findutils
- libmagic and file
- tar and some compression libs
- grep, awk, sed
- the shell and all of its builtins
- something functionally equivalent to openssl or GnuTLS
- some ssh client and server
- curl
- a terminal-based editor
- man-db and texinfo
- some init system and bootloader
- pick a package manager, any package manager, and rewrite it in Rust
Barring all of that, maybe just busybox but written in Rust. That should give you roughly what you need for a non-graphical system. coreutils isn't nothing, but it's a pretty small part of the system, with much of it ending up implemented by the shell in most distros.
systemd sort of did this to certain parts and it's the worst thing that happened to Linux. Standardizing on Rust political rewrites of tar and libmagic is going to be an epic disaster and a decade of never-ending fun for Rust evangelicals.
also worth mentioning that the ones that are user facing are much lower priority than machine facing one. no one cares about CVEs in ed, but a CVE in cp would be really bad
It also seems wrong? libsodium explains the logic in its name right on its about page. It's a fork of NaCL (the chemical formula for sodium salt), which itself is a plain acronym for "networking and cryptography library." Google doesn't seem like a good example, either. Wasn't that meant to be an allusion to the very large number googolplex, as in Google exists to tame the unfathomably large amount of information on the web? The author may or may not like those names, but they have a logic just like grep and awk do.
I don't really build software any more and have moved into other parts of the business. But I'm still a huge user of software and I'd just echo all the other comments asking if it's so easy to get all these great tools built and shipped, where are they? I can see that YouTube is flooded with auto-generated content. I can see that blogspam has skyrocketed beyond belief. I can see that the number of phishing texts and voicemails I get every day has gone through the roof. I don't see any flood of new CNCF incubating projects. I don't see that holy grail entire OS comparable to Linux but written in Rust. I don't see the other holy grail new web browser that can compete with Firefox, Chrome, and Safari. It's possible people are shipping more of the stripped down Jira clones designed for a team of ten that gets 60 customers and stops receiving updates after 2 years but that's not the kind of software that would be visible to me.
If you're replacing spreadsheets with a single-purpose web UI with proper access control and concurrent editing that doesn't need Sharepoint or Google Workspaces, fine, but if you're telling me that's going to revolutionize the entire industry and economy and justify trillions of dollars in new data centers, I don't think so. I think you need to actually compete with Sharepoint and Google Workspaces. Supposedly, Google and Microsoft claim to be using LLMs internally more than ever, but they're publicly traded companies. If it's having some huge impact, surely we'll see their margins skyrocket when they have no more labor costs, right?
It's always amusing to see what Python has become thanks to NumPy and Django and how widely deployed it became when its intended purpose was always to be easy to understand, not performant. But just like JavaScript, people are going to hack it to death to make it fast anyway.
As much as people complain, maybe if I was still 22 and dirt broke, I'd do something like that, but more likely I just wouldn't watch TV. I didnt own a TV back then and it was fine. Now, sure, I don't exactly like being nickle and dimed from a pure intellectual perspective, but these streaming services are what? Like $15 a month a pop? That's 1/40 the cost of groceries. It's annoying but makes no difference and isn't anywhere near worth the hassle of starting and stopping. If it was a $120 a month gym subscription or the old cable bundles I used to pay $200 for, then it's getting to the point that it's worth caring about.
The stickiness is probably just that. Even as they raise prices, it's still less than we're paying for pretty much anything else. Gas, electricity, food, housing. Cut Netlix and well great, I just reduced my monthly spend from $5000 to $4980. Really making a dent there. I can retire comfortably now. It's almost as patronizing as the old avocado toast thing. Avocado toast might be overpriced and nowhere near worth it, but it isn't the reason anyone is broke.
I do keep a vague eye on subscriptions/credit cards/etc. that I'm really not getting value out of over the course of months.
But, yes, if you're either poor or optimizing points on an airline or whatever is sort of a hobby, then sure. But otherwise, it's just not very interesting to many of us and involves mental overhead we can just live without.
RL is a technique for finding an optimal policy for Markov decision processes. If you can define state spaces and action spaces for a sequential decision problem with uncertain outcomes, then reinforcement learning is typically a pretty good way of finding a function mapping states to actions, assuming it isn't a sufficiently small problem that an exact solution exists.
I don't really see why you would want to use it for binary classification or continuous predictive modeling. It's why it excels in game play and operational control. You need to make decisions now that constrain possible decision in the future, but you cannot know the outcome until that future comes and you cannot attribute causality to the outcome even when you learn what it is. This isn't "hot dog/not a hot dog" that generally has an unambiguously correct answer and the classification itself is directly either correct or incorrect. In RL, a decision made early in a game probably leads causally to a particular outcome somewhere down the line, but the exact extent to which any single action contributes is unknown and probably unknowable in many cases.
It's great that opportunities like this exist. Doing a project like this at all is such valuable experience. You must have learned a ton and can take that with you for all future projects. The only real quibble is the experimental setup is not really scientifically valid. UV light on its own kills algae, so you're going to detect a monotonic effect roughly equivalent to the altitude increase assuming a reasonably constant rate of altitude increase just from the cumulative exposure. That's not the same thing as detecting a change purely because of altitude.
Who cares, though? Scientists train for many years to learn the details of experimental methods in their specific domain. The engineering and hacking experience on its own is what really matters here.
reply