z77dj3kl

I'm extremely interest the general feeling and views societies had in the past: how they perceived the present and the future, as a whole.

Objectively life has become better and more comfortable for the vast majority of humans since then (Hans Rosling does a beautiful job of exploring this).

But I do think that perceptions and feelings matter, and even though material wellbeing is a prerequisite to that, so is also the general feeling and view that those around you hold, and in many ways I feel we've gone backwards in that.

Seems pretty silly? It has opened a bunch of issues and is harassing devs on github repos complaining about missing SECURITY.md files?

https://github.com/search?q=%22Issue+created+by+Allstar.+htt...

No judgement, but I guess this is just not the kind of hole-plugging security I'm used to.

It's a slightly different flavor of the same thing.

VS Code is an amazing tool that's getting huge adoption because of how awesome it is, and how open source and community centric it is, etc. They've gotten a lot of mindshare and dev love, that's the embrace bit.

The next step is the set of closed source addons. Have you noticed that a lot of the new VS Code features are now in addons that are under a different license. This includes the new remote dev and python tooling. Still free to use and awesome tools, of course, but fully under MS exclusive and controlled. That's the extend.

I don't know what the longer-term plan is. I'm hoping that it won't lead to an extinguish. But if that's what they want, then they could e.g. cripple the open source version, moving all their dev effort into a closed-source, Azure/Github only web environment. Who knows?

I thought OpenAI was originally supposed to be some kind of for-the-good, non-profit institution studying AI and its safe use in particular with an effort to make it more accessible and available to all through more open collaboration. This is cool research, sure; but what happened to making models available for use by others instead of just through some opaque APIs?

Maybe I'm just remembering wrong or conflating OpenAI with some other entity? Or maybe I bought too much of the marketing early on.

Emailing patches is a classic git flow, used e.g. by the linux kernel. Git even includes a command to format emails and send them. Nothing wrong with it!

That's not how timing attack mitigation works: just because you're doing some other stuff too doesn't mean the timings can't be exploited.

You can't even use Linux on Intel Macs properly, no chance it'll work on the M1 anytime soon. The kernel might boot in some form, but last time I checked, even the SSD wasn't working on Intel Macs under Linux.

Ah, that makes sense! I always wondered why any search term would just return 8 year old stuff at the top.

Or the case where you work google which goes down and then you can't possibly duckduckgo it, that'd be too humiliating.

I had a hand saw in mind...

A more apt comparison would be to using a saw vs using a saw with gloves on.

With gloves it's safer and more comfortable, but you're not quite as in touch with the wood as you would be without one.

Without gloves you feel badass because you don't care about safety, and you might be able to carve out some slightly more intricate things.

Either way the outcome is roughly the same and it really doesn't matter what your personal preference is.

They're a public employer though, no?

Advertising? Have some articles be part of the bundle as a "preview", and most articles behind an individual subscription?

This idea of building your own newspaper sounds kind of interesting actually: a mix of authors you've picked and an automatically curated content from other authors to help you find new authors.

I think he's referring to the standard, non-hidden microphones.

The social networks are offline. Universities are great at open collaboration, it's just not a kind of online collaboration where anybody can walk in and partake. People spend a lot of time on a different type of communication: going to conferences, attending seminars, and meeting one-on-one. Why should it be online?

Also keep in mind that a lot of mathematicians are also older, less tech-savvy. This might change in the next decade or two!

I've heard of mathematicians being frustrated by physicists who use fancy stuff without full understanding of the underlying principles, and sometimes they just manipulate the notation without fully justifying what that manipulation means (e.g. you may have an integral operator on some space, and you might just swap an infinite sum and that integral operator, but that requires a justification that you might omit).

What do you mean? Mathematics is all about formal verification, i.e. proofs.

Pitting code review against trust and responsibility is just disingenuous. The point of code review is not that you distrust others. It's that everyone makes mistakes and often someone else having a quick look over your changes can come up with some really cheap wins that are safer, more maintainable, faster, etc.

The OSM style is here: https://github.com/gravitystorm/openstreetmap-carto

There is an issue about converting it to vector: https://github.com/gravitystorm/openstreetmap-carto/issues/3...

Doesn't seem like there's any current vector versions of that style, but you could make one yourself using Maputnik or similar; you'd probably want to start with OpenMapTiles and maybe even one of the pre-existing styles for OMT like OSM bright.

Doesn't need to be OMT though. That's just the de facto free & open source tile generator from OSM data.

Yes, you can render raster layers (tile sources).

https://docs.mapbox.com/mapbox-gl-js/example/map-tiles/

No, but they will pay 2% more to fix something if it means 5% of customers don't write them off as "awful, unethical, big corp" and not buy stuff from them.

There are plenty of examples of companies that did or did not react to negative PR, but the effect is in the statistics, not the individual anecdata.

Auction design is a very active research area and what the "right solution" is, is not so easy to figure out. This has huge applications in... adtech.

It's random math words put together and doesn't make sense even in isolation.

Yeah, I've noticed this too. Most of the time when you look up something gRPC related, the first links are for some of Microsoft's gRPC documentation!

gRPC is one of those projects that I'm not sure if Google really wants uptake or if it's just a dump as a means to open source other things they want uptake in (GCP client libraries, tensorflow, etc). The documentation is awful.

It's a tradeoff between software and infra complexity.

I would argue that if you've built a system up from scratch, this is much easier to debug and maintain than a foreign piece of software. Rabbit is just way overkill if you have a couple of jobs per hour, for example.

Thanks! I'll have to investigate Kopia properly.

This wouldn't be the primary backup, but Deep Glacier is just such a good deal that: I'd be happy to pay the $50 per year for a call option on my data, it'd make me sleep better at night!

Part of my calculus is that I have quite strong confidence in AWS in terms of business continuity and reliability/availability. If I dump my files on AWS, I have high confidence in the files (and AWS) being around in 10 years and retrievable for roughly the same price (or at least no more).

Hetzner would have much lower durability. I'm a bit suss on Backblaze, though I do trust them to be more durable than my self-managed disks (and uncorrelated to my failures). I don't know much about Wasabi; but it's not a good sign for me that their landing page touts their latest funding round at the top: seems young and you never know if the price is subsidized with VC money (and won't be in n years) or similar.

> Also, what happens if you have to retrieve data twice from Glacier?

The killer is the egress. I'd just buy a new set of disks and download it straight there.

Suppose I have ~4 TB of data. If I dump it into Deep Glacier, it'll be ~$50/year (free ingress), if I ever need to retrieve the data, it's like $370.

Normal S3 would be ~$1.1k/yr, or around half of that for the infrequent access tier, both of which are way too expensive.

The only thing I need (and is sorely missing from Restic) is that the metadata be kept separate from the actual data. That way I can store the data in AWS S3 Deep Glacier at a cost of nothing per year, and still do incremental backups. Currently the architecture of Restic for instance requires all data to be quickly and cheaply accessible; which makes it impossible for this.

I have terabytes of data that I'd be happy to dump encrypted and compressed in Deep Glacier and happy to pay $500 to retrieve if I were to mess up my hard drives, but otherwise don't want to pay for the costs of normal S3.

Does Kopia separate metadata from the actual encrypted/compressed blobs?