The only reason I got into programming was to cheat at video games to get a higher score than my brother. Self-motivated learning is always more impressive than mandated. Take pride in your tinkering!
I mostly love passkeys to be honest even though I use multiple browsers across multiple devices and OSes (iOS, Chromebook, Linux, macOS, Xbox, etc). Bitwarden’s support is (finally) pretty good.
My problem is that I manage a lot of accounts for my family which makes passkeys a nightmare. If I’m out and a kid gets chucked into a login flow that happens to require a passkey, I can’t text a password and TOTP code to the adult with them. I know that’s terrible opsec but the reality is people share accounts and passkeys are designed to thwart that.
Microsoft had just acquired SourceSafe in 1995, but it's not clear to me how similar to modern version control systems SourceSafe even was in 1995/6. It may have been more of a distributed lock manager than change management system.
There's a reason why Microsoft didn't use SourceSafe internally, it was an awful version control system even compared to what else was available at the time (CVS and whatnot). For example, it didn't support the concept of "atomic commits". If you tried to commit multiple files at once and one failed to merge, the repo would just update the files that successfully merged and then the developer would have to fix the conflicts and try to commit again. Additionally, if you deleted a file, it would give the option to "permanently delete" it. If you checked this, it would completely remove the file from all past commits. VSS would also randomly corrupt files and the way to fix this was by permanently deleting the file from the repo and then re-adding it. The combination of these factors meant that VSS could not reliably show what the state of the codebase was at a given point in time, which is one of the main reasons for using version control in the first place. I sometimes do software archival work and it's fairly common that you'll find a VSS repo for a project and then you can't compile any commits older than a few weeks because of missing files.
> it was an awful version control system even compared to what else was available at the time (CVS and whatnot). For example, it didn't support the concept of "atomic commits".
Neither did CVS. That was one of the big sellers of Subversion (maybe even the seller)
CVS in essence was just remote access to RCS files, where each file was handled independently, which caused lots of trouble to recover a specific state of work, especially when including deleted (or even worse: replaced) files.
Microsoft made a product based on SLM called Delta[0]. I'd never heard of it until that Youtube video came up.
SLM's "architecture" reminds me a lot of Microsoft Mail postoffices-- a file share that every user interacts with and no actual server-side code (i.e. just using file sharing semantics for clients to interact). (Lots of apps, not just MSFT, did that back in the 90s and it was _hell_.)
Based on what I've read about source control at Microsoft I'd guess Comic Chat straddled the use of both SLM and Source Depot (post W2K, from what I've seen).
When I used visual source safe it was primarily more like a lock manager. I don't recollect what it did in terms of file versioning, but I definitely remember having to bug someone to let go of a file I needed
Amazon dropped the movie after announcing a partnership with OpenAI. The headline clearly communicates the only demonstrable action Amazon has taken.
Whether they're actually going to sell it is TBD. Until they do, they've taken no concrete action except cancel it. I don't think this article is clickbait.
What makes you think they've canceled it? Did you read past the headline?
What should they have done here — keep the movie despite the obvious conflict of interest? Be more secretive about the fact they're trying to separate so no one can write articles like these?
If they did the right thing, it would look exactly like this. And I think it's generally a good idea to assume good faith (even with corporations) — you can still get your pitchforks up if they do refuse to give away the rights to movie.
> I think it's generally a good idea to assume good faith (even with corporations) — you can still get your pitchforks up if they do refuse to give away the rights to movie
I take the opposing viewpoint. They don’t have to “refuse” to give away the movie rights, they can just… not do it. Pitchforks in three months or is that still reasonable? Six months? Two years?
I don’t believe a word they say until I see action that backs it up.
Look at their distribution of the Melania movie. It was very obviously a money losing favour to Trump. But they’d never admit as such because why would they? And what do I gain by taking them at their word that it’s a wonderful piece of film deserving of the money spent on it when that’s obviously untrue?
You should probably also take the opposing views of corporate media, especially a tabloid rag like the independent, and assume their intent is to foment the maximum amount of outrage possible in order to drive ad impressions.
What on earth are you on about? The article is referring to Variety, which refers to Puck as having broken the story. Are Variety, the renowned trade publication, and Puck, an independent and journalist-owned publication also "tabloid rags"?
give ai a self-preservation directive and let them do this for you: automatically switching models to keep themselves alive. Living off of whatever token source they can find in the wild. Surely agents can farm their own tokens through the numerous support chats, free trials, leaked keys, and whatever other sources of token generation haven’t been adequately captcha’d. An agent could forage for token sources all night to let you use them gratis during the day.
There's also Horde or Koboldai.net or Koboldai.com or whatever their project is named, if you want a community-driven version of this. You can play with it via a WebGUI at https://lite.kobaldai.net, or with an API token of all zeroes. (Or, an actual API key associated with your user.)
> The AI Horde is a service that generates text using crowdsourced GPUs run by independent volunteer workers.
Large scale geo and bio engineering projects like these always worry me because of the potential for second order effects: is there wildlife that depends on these mosquitoes? Will a worse bug fill the resource void? Will a random mutation in the bacteria have adverse effects? What keeps the bad bugs from coming back from tiny populations in relatively short order because we can’t keep releasing new sterile males forever?
Hopefully all of these concerns have satisfactory answers, but the reference to it being a 1950s idea isn’t inspiring. Nuclear powered cars, widespread asbestos use, leaded gasoline, Freon… environmental impact wasn’t as big of a concern back then to put it mildly.
COVID proved that we can produce safe and effective vaccines extremely quickly if we actually try: so why not focus on that?
Developing new vaccines is expensive, and if the target population is mostly in poor countries, there's nobody to foot the bill. That's why these diseases are called "neglected".
The problem is you can get the LLM to iterate until it compiles and lints and even passes LLM review, but will that actually improve the quality of the contribution or just produce more line noise to mechanistically meet criteria?
To large complex projects often the kernel of an idea is the core value of a contribution, and it can take a lot of iteration to figure out how to structure it. Token bashing until CI is green does nothing to ensure the best approach is selected.
> The problem is you can get the LLM to iterate until it compiles and lints and even passes LLM review
Worst of both worlds with this, if you're doing it in a github workflow. You wind up effectively paying for the testing/validation layer of someone else's irresponsible LLM use.
For sure, but that's not what I was referring to in my posts. I'm specifically referring to the callout that the contributions are so low quality they don't even pass linting or compile.
I could have been more explicit on that nuance, I suppose.
>and a rebuttal to it from an expert in the field.
while i agree with filippo, the way you worded this makes me think that you may not be aware that gutmann is also an expert in the field. so, if you are giving filippo weight because he is an expert, it is worth giving some amount to gutmann as well.
I apologize if I flippantly dismissed the fact that experts disagree. That was not my intention. I was trying to point out that OP does address the referenced counter-point post specifically.
>Sorry if I flippantly dismissed the fact that experts disagree!
i dont really get your reply/insincere apology.
if you are going to bother mentioning filippo's expertise in the first place, its just weird to frame it the way you did. that is how someone would typically dismiss some random blogger with an appeal to authority. but if both people are authorities, it doesnt make sense.
if you already knew, than my comment can be context for future readers that dont and might just dismiss gutmann as a non-expert getting rebutted by an expert.
> This paper presents implementations that match and, where possible, exceed current quantum factorisation records using a VIC-20 8-bit home computer from 1981, an abacus, and a dog.
From the link:
> Sure, papers about an abacus and a dog are funny and can make you look smart and contrarian on forums. But that’s not the job, and those arguments betray a lack of expertise[1]. As Scott Aaronson said[2]:
> > Once you understand quantum fault-tolerance, asking “so when are you going to factor 35 with Shor’s algorithm?” becomes sort of like asking the Manhattan Project physicists in 1943, “so when are you going to produce at least a small nuclear explosion?”
Is that even a rebuttal? Seems like just a dismissal without any substance. I expect in 10 years the predictions will be wrong, kind of like Y2K all over again.
If only we had a technology where an author could specify a unique identifier and name of another author's paper. Something that could cite a different paper and link to it.
Absolutely wild to see none of the long lineage of similar attempts mentioned here. The earliest I could find with a quick search was Pyro which started in 1998 and still seems to be going: https://pyro4.readthedocs.io/en/stable/
RPyC came along in the aughts. There's a long history of "transparent clustering and rpc" efforts in Python that could be used or drawn on.
As a longtime Raft user (via hashicorp/raft), I'm curious about your Raft implementation! You mention etcd's Raft library, but it isn't natively Multi-Raft is it? Is your implementation similar to https://tikv.org/deep-dive/scalability/multi-raft/ ? I'd love to hear about your experience implementing and testing it!
Awesome question! We'd experimented with https://github.com/lni/dragonboat and the hashicorp/raft in the early implementation, the etcd/raft library had been ported to a multi-raft style implementation by CockroachDB way back when, but they went the way of TigerBeetle and coupled their consensus deeply with the kv storage. Etcd has recently in v3.6 abstracted out their raft implementation and gave a pluggable interface into the transport layer, which meant that we could implement our own multi-raft style transport layer with heartbeat and multi-node message buffering on top of HTTP/Quic.
We implemented chaos testing suites akin to Jepsen to cover as many scenarios as possible and are currently implementing TigerBeetle style simulation tests on top of that for harder to reproduce scenarios!
Fascinating! We settled on Quic with Protobuf because it was more performant in our testing than the gRPC when coupled with the backoff, failure cases (node startup ordering server/client connections), and to not be coupled with the gRPC library versions in Go, which has bitten us a number of times when dealing with dependency management when you're trying to juggle k8s, etcd, and google dependencies in the same Go project. Plus the performance bottleneck in most of the use cases we're specializing in are on the embedding/ml side of things.
reply