aeadio

Which is why you split the difference. Own your own domain, and host at a reputable, reliable mail service provider like Fastmail. If you get taken down by a TOS violation, you didn’t lose your email address, just the contents of your inbox — which you hopefully have backed up, at least in your local mail client. Update your MX records to a new mail provider, and you haven’t lost what’s essentially your digital identity at many SaaS services.

Everyone in my family has their own domain. The instructions for renewing a domain, a list of reputable mail service providers, and instructions for pointing MX records at a new service provider, are not very complicated for somebody that can work a computer at a basic level of productivity. Many email service providers will even run nameservers on your behalf so you don’t even have to mess with individual DNS records. You can fit that in about one formatted page of instructions to include in your online contingency plan you share with love one(s).

A simple reject policy is showing up as an error, despite validating fine with other DMARC checker apps like MX Toolbox,

    The From domain (...) has an invalid DMARC record.
    ...
    DMARC record found: v=DMARC1; p=reject;
    The DMARC check result is permerror.

No explanation as to what permerror means.

It might also be useful if the tool attempted to circumvent DNS caching, so users can try tests in succession after updating.

I think the parent comment was referring to messaging, ie iMessage vs the others. Ie, in Taiwan and Japan the dominant solution is Line, not iMessage or SMS.

Processes get interrupted and rescheduled between tens and hundreds of times per second (depending on OS, the scheduler may take over somewhere between 100 and 2000 Hz, although it doesn't necessarily deschedule every time). This is how SMP works at its most fundamental level. Every time that happens, it involves a full flush from the hardware perspective. It's always context switching, savings the registers, clearing the buffers, etc.

When a process gets descheduled and then later rescheduled, there is no particular reason to reschedule it on the same core it was on previously. Desktop class hardware doesn't have NUMA. Last level caches might have locality-dependent performance characteristics, but this is usually pretty nuanced, wildly variable between hardware microarchitectures, and not accounted for / not worth accounting for in the OS's scheduler, since there's a good chance that data won't be around next timeslice anyway.

Binding a processor to a single core affinity-wise is not changing any of this.

If you really have a throughput-bound process that needs to squeeze every single microsecond of compute and/or you're super latency bound for some reason, you schedule a process with what's called realtime priority, which changes the scheduling decision of WHEN to deschedule it. Not every operating system has equivalent mechanisms for this, but that's more in line with the phenomena that you're describing. But a process may well cause itself to be descheduled anyway by doing something that puts it to sleep. You have no control over this.

In reality, looking at which core a process is on is mostly the wrong way to be thinking. You're looking at this on a 1-2 second window (the amount of time it takes for Activity Monitor / Task Manager to update its UI). A full second of time is an absolute eternity to a process. Large amounts of work happen on the order of microseconds and quicker. And in that 1-2 second snapshot that Activity Monitor or Task Manager gave you, the process probably spent time on EVERY core -- and you're just seeing wherever it was in the brief moment when Activity Monitor / Task Manager updated its UI.

> gamers have been binding to single processors/cores for years for that reason, it's a huge cause of microstutter type behaviors

It wouldn't be the first time gamers prescribed placebo tweaks because of magical type thinking.

> Well, that's a dumb and misinformed argument. Only the one out of the box is included with macOS - the rest are download on demand.

... and cleaned up automatically when more free space is needed.

> But as it gets wider use fact is the vast majority of systems have a single device and don't want to cut available storage in half for copies=1.

Setting copies=2 for the root filesystem, but segmenting stuff like caches, user downloads, etc into separate datasets with copies=1, is totally reasonable for this purpose. The base system in most distributions only takes a few GiB. 1 TiB SSDs are pretty common even on entry level laptops these days, and it's becoming increasingly difficult to find new machines without at least 512 GiB.

The slowness seems to come from rendering the desktop-style tab bar. With that disabled, the UI seems more or less as snappy as Safari on the iPhone.

This article is completely inept. I'm nowhere near an expert in WiFi tech, but having done a lot of in-office WiFi setups over the past several years, it's easy to see the differences in quality generation over generation.

> Trouble is, the woes listed in the messaging are those that previous versions of services and wireless standards claimed to fix. If they didn't fix them, why should we believe this lot will? If they did, what is Qualcomm actually saying? It's a paradox.

Previous iterations of WiFi DID fix these problems. Modern "WiFi 6" is significantly better at handling many simultaneous clients than the old 802.11b/g were. We've been through multiple generational improvements to the old scheme, such as,

  - 802.11n using 5 GHz with lower penetration, making it easier to create more and smaller cells 
  - 802.11n introducing MIMO, leveraging airspace better 
  - 802.11ac introducing MU-MIMO, allowing simultaneous transmission for multiple clients 
  - 802.11ax expanding MU-MIMO into the frequency domain, making it more effective 
  - Much better and quicker channel reuse during potential collisions
  - Quicker transmission means the time slices clients DO get are more meaningful

WiFi has gotten so much better over the past several years, that I suspect people just don't remember how unreliable it was before. Part of this is because it's been a gradual improvement, as new standards have come out, on-market hardware has slowly taken up these new features, and client devices need to be aged out and replaced before a newer device can make use of these newer features -- which users might just attribute to their new device being "better", and not that the WiFi got better.

It's not hard to look at the WiFi 7 (802.11be) Wikipedia page and see that it offers several improvements to handle increased client load,

  - Multi-Access Point (AP) Coordination (e.g. coordinated and joint transmission)
  - Enhanced link adaptation and retransmission protocol (e.g. Hybrid Automatic Repeat Request (HARQ))
  - Enhanced resource allocation in OFDMA
  - Optimized channel sounding that requires less airtime
  - Implicit channel sounding
  - Support of direct links, managed by an access point

> There is no consumer problem of today, no domestic use case imaginable, where 50Gbps works where 10 will not.

WiFi quality is not dictated by the advertised transmission rate. That's just the number hardware vendors love to advertise, because it keeps getting bigger and bigger generation-on-generation, whereas all these other improvements are much more nuanced and tough to communicate in marketing copy.

> The 6GHz band, which in the US actually goes some way beyond 7GHz, is more easily blocked by walls and other physical stuff necessary for gracious living than 5GHz, so coverage problems won't be fixed.

Which may actually be desirable. Less penetration through walls means smaller cells that don't interfere (need to share airspace) with as many neighbors. This has been a boon in dense settings (office buildings, apartment buildings, public WiFi) during the 2.4 GHz -> 5 GHz transition.

If you have a large house with many walls separating each room, you may now need more APs to cover the same area. But looking at the proliferation of mesh-based systems, it seems the market is already moving that way anyway.

> It would be amazing if one in a hundred households notice an iota of difference if their current router magically sprouted Wi-Fi 7, and this won't change much for the years it takes to get phones, laptops, and everything else with older versions of the standard upgraded. Good luck getting those managed end-to-end services up and running too.

And this is just uselessly defeatist. My current devices don't support the new standard, so why bother? WiFi has seen tremendous advances in quality and capability over the past 15 years, and users have always realized those advances as their devices age out. Back in the 802.11g days, it would've been unheard of to have multiple TVs and iPads in a single apartment within a densely populated apartment building streaming 4K video over WiFi simultaneously.

Oh boy.

I moved to DC about a year ago. I have nothing against automatic traffic cameras. I've seen them used to great effect in NYC, where I'm natively from, to reduce speeding incidents in sensitive areas like high-traffic mixed use zones, school zones, etc.

The traffic cameras in DC are not that. They're primarily placed at chokepoints that seem intended to collect speeding fines. Some of the most prominent speed cameras are placed on major arterial throughput roads/ that turn into highways, which are either not connected to the sidewalk at all, or have no major foot traffic concerns.

In one particularly blatant location, leading right up to US-50, a major road, which has a nominal speed limit of 45, has a speed camera set to 25 mph dead in the middle, over a bridge, which is largely inaccessible to foot traffic. In the hundreds of feet leading up to the camera, the posted speed limit get progressively lower, from 45, to 30, to 30, until settling at 25 where the camera is, and then progressively raises again.

The residents have learned the game, and now this major arterial road is a nonstop traffic jam in the mile or so leading up, as people slow to a crawl to avoid the fines.

DC absolutely has some aggressive and awful drivers. But the real root cause are things like intensely poor city planning -- the criss crossing grid of exceedingly narrow streets meant to look pretty on a map and picturesque during sunset basically is an absolute worst case for gridlock. They've also done an exceedingly poor job of retrofitting modern amenities like bike lanes onto those streets, where premium space is taken up by triple-wide, treelined sidewalks, yet cars do not have room to navigate major arterial streets, and a single taxi letting out a passenger means congestion and danger to everyone around them.

And then you have what is probably the least engaged police force. Traffic cops can be seen casually parked and playing on their phone for hours on end, often parked illegally themselves, while drivers around them brazenly break the law, knowing there's nothing approaching real patrols or enforcement.

I do something similar, but instead of a script with a bunch of `defaults write`, Ansible has an osx_defaults module. https://docs.ansible.com/ansible/latest/collections/communit...

> your rights are not restricted, when I refuse to do business with you

This is the part that I don't think it's so safe to assume. A case can and should be made that my rights are restricted and/or I have experienced genuine harm/damage as a result. You would have to look at the GPL violation and the contract violation separately, each in a vacuum from one another, to argue otherwise.

Courts may want to look at the whole picture. This looks very much like a "strings attached" situation on the license of the software. Courts generally do respect the authors' intent when issues of copyright/license infringement are not clear-cut, and the GPL is extremely wordy about what the intent was -- freedom to share and deal in the software, out in the open.

Red Hat's own corporate materials on open source acknowledge as much, so that's ammunition for any court battle. They can't claim ignorance of the spirit of open source (or the spirit of the licenses in question). They have been including verbiage about the benefits of openness and public access/development of open source software when selling the very product in question.

This seems to be anything but a slam dunk case if really tested in the courts. I think there's a genuinely good chance that somebody with the pockets to challenge this wins and sets a significant precedence for open source.

He makes an offhand comment implying he doesn't feel they're incompatible, but not in any official stance as an Oracle lawyer. He then qualifies that by saying, more or less, Oracle has not come after anyone over OpenZFS, so everyone should be taking that as an indication that it's OK with Oracle.

He's not the only person that feels CDDL and GPL licenses are compatible. Their alleged incompatibility[1] seems to stem from a lot of hair splitting over technicalities that don't align with the spirit of either license. [2][3][4]

It's particularly funny that the community (except Canonical and a handful of individuals) largely settled on the CDDL-GPL issue--deciding they were incompatible and giving up on the possibility of upstreaming ZFS--but is still so highly split on the question of RHEL's GPL compatibility w.r.t. calling source code sharing a violation of their customer agreement.

The RHEL issue stems from the GPL literally saying,

> You may not impose any further restrictions on the recipients' exercise of the rights granted herein.

Conspicuously, you may not impose restrictions on the EXERCISE of these rights, not just further restrictions to the rights themselves. There's a pretty clear argument to be made that RHEL's actions in ending business relationships with any customers would be a restriction stemming from their having exercised their right. There's basically no hair splitting here. This should be done and dusted in the collective open source community conscious.

But the CDDL-GPL issue? There is no smoking gun in either license. The arguments against are much more strenuous.

But the community settled one debate much more decisively than the other.

If ZFS was as desirable to the Linux community as a fully open RHEL, I think there would've be much more enthusiasm in interpreting legal theory around it.

I think the community ought to be coming down much harder around the theory of RHEL incompatibility. It's really not a stretch from the most straightforward reading of the license.

It's also pretty conspicuous that Oracle has done precisely nothing to stop OpenZFS's inertia. If they felt it was incompatible and were going to act, they had their chance.

[1] https://www.fsf.org/licensing/zfs-and-linux

[2] https://blog.hansenpartnership.com/are-gplv2-and-cddl-incomp...

[3] https://opensource.stackexchange.com/a/2106

[4] https://softwarefreedom.org/resources/2016/linux-kernel-cddl...

The issue isn't about them refusing to give sources to their customers. It's around placing restrictions on their ability to exercise their rights with those sources under the GPL.

> You may not impose any further restrictions on the exercise of the rights granted or affirmed under this License.

You cannot argue that ending their business relationship as a direct consequence of distributing those sources wasn't a restriction on the user's rights. They can play coy all they want, but nobody's going to believe these two events are unrelated. Their EULA is effectively a set of "additional permissions" as defined in GPL section 7,

> All other non-permissive additional terms are considered “further restrictions” within the meaning of section 10 [and therefore invalid]. If the Program as you received it, or any part of it, contains a notice stating that it is governed by this License along with a term that is a further restriction, you may remove that term.

To the extent that interpretation is required, courts are going to try and divine the author/copyright-holder's intent. And the GPL goes to great lengths (even including a lengthy preamble) to spell out that is intent is specifically to enhance freedom and ensure unrestricted access and exercise of rights by users.

Analyses that Red Hat's UELA is not a GPL violation are wrong, both in the moral intent of the license and the literal wording. They are mostly coming from a place of "well tekknukkally" that split hairs over literal wording and semantics, and that's not how the law works.

It's up to players like Alma, Rocky and others to battle it out in court.

This method dates back to Socrates https://en.m.wikipedia.org/wiki/Socratic_dialogue

They don’t prevent you from using aftermarket chargers. They just display a warning, which can be disabled in the BIOS.

Were you not aware you could transfer that cash to a normal checking account?

> In plain English that means: we own everything you post.

No it doesn’t. It means you’re giving them license to use it in whatever way they see fit. You need to give them some kind of license over the content since they are hosting it for you.

They need to store it. They need to copy it. They may need to modify it, either for moderation purposes or because it becomes a piece of another post via quoting / replying. They arguably even need to sublicense others to use and copy it for the same.

Every service that stores user content has an identical clause. They are always overly broad and catch-all inclusive of all intellectual rights to avoid missing something.

There’s nothing new here. Just a clickbait site that doesn’t understand copyright and can’t do basic fact checking.

MediaTek is not Chinese. You might be thinking of HiSilicon.

Based on the article it looks like a Cooler Master rebrand. Cooler Master is not Chinese. It's also not low quality, they're one of the most well regarded names in PC cooling. Their Hyper 212 line has been one of the most recommended coolers since I can remember -- well over a decade, at least.

Samsung has consistently manufactured their Galaxy S series in two different versions -- a Qualcomm one, and an Exynos one (Samsung's in-house CPU division), with different geographic markets getting different options. America has always gotten the Qualcomm version (which has consistently been faster), but other markets have seen shifts over time.

The S23 Ultra went Qualcomm-only for the first time. In the year running up, they've been slowly expanding Qualcomm to more markets instead of Exynos.

It seems like they're gradually phasing out Exynos, at least from the higher-end phones -- since they've always struggled to produce an SoC competitive with even Qualcomm, which is already behind the curve compared to Apple.

The Snapdragon in the S23 Ultra is already manufactured on TSMC 4 nm.

The S23 Ultra has a Qualcomm CPU (Snapdragon 8 gen 2), which is manufactured on TSMC 4 nm. The iPhone 12 uses an Apple A14, manufactured on TSMC 5 nm.

This is more about Qualcomm being uncompetitive performance-wise, which has been true pretty much since the start of the smartphone wars.

I've worked with some of these responsible recyclers before. Everyone knows the deal.

If you want IT asset disposition, it'll nominally cost you -- for pickup and processing, and optional value-adds like secure data destruction certificates. If the assets have remarket value, that can begin to subsidize the cost, easily zeroing it out, or even returning profit to the IT department. Recyclers are happy to pay IT departments to pick up their old Apple equipment -- and they know to ask ahead of time, "are these activation locked?"

These recyclers often have vertical integration on the refurbishing side, running an Amazon eCommerce department so they can directly capture that remarket value. And once they've got that going, their bottleneck to growth is how much material they can bring in -- which is never enough.

So the recyclers and refurbishers invariably need to buy material from other recyclers. There's an entire economy around selling pallet-fulls of used, minimally-inspected or completely "as is", Apple and PC equipment. Specialized invite-only communities exist for selling this material in bulk, with lots of overseas actors in on the game. Every R2 recycler is involved here, along with lots of actors from India and China in particular.

Having seen inside that sausage factory, fraud and theft is a legitimate problem. There's more than a fair share of "don't ask questions". More than a fair share of "yup we're definitely R2 certified, wink wink". More than a fair share of wining and dining the R2 auditors to get that certificate under the table in the first place.

I'm gonna go with the GP comment here -- the system is working as intended. Industrialized theft of this sort is much more eminently solvable through technological means than through overstretched law enforcement, especially when a non-negligible amount of material comes from overseas.

It's on the IT departments to figure out if they have enough time to activation-unlock their assets before disposition. Do they want to pay the recycler to pick up all that material, or do they want to sell the material and pad their budget?

A quick look at the Jamf documentation suggests it's pretty easy to mass unlock these devices, so I'm not buying that it's that much of a strain.

I’m no expert here at all, but I wonder how much this translates to real world risk. It seems like they’re irradiating unprotected cells with a direct blast of UV. How much of that UV actually penetrates the top layers of skin? They’re also subjecting cells to 20+ minutes of continuous exposure, while during a gel manicure typically involves 60-120 seconds of exposure. Obviously we’re not seeing individuals walk out of manicure sessions with necrosis of the fingertips. Their values of 20-70% cell death don’t seem meaningfully representative.

I’d love somebody with an informed opinion to weigh in here. Is this bad science? Bad reporting? Or is there genuine danger here that’s been flying under the radar?

My concern is, we see a lot of sensational science reporting that plays up the risks and dangers of potentially routine things, lowering public trust in science.

Spammers will often register paid accounts with stolen credit cards, and opt for lower tier accounts because the small charge is much less likely to trigger fraud protection than, say, attempting to subscribe to Workspace Enterprise.

Google almost certainly has, and has consulted, internal analytics on where these spam campaigns are originating and which account tiers they are abusing most.

For some reason, “I think Apple does this too” always comes up in the discussions about Microsoft forcing online accounts within Windows.

MacOS doesn’t do this. Accounts are machine local. iCloud integration is an optional layer on top. Always has been. Nothing has changed.

Maybe give the OS a try before assuming Apple is following all the same antipatterns just because they also happen to fall into the category of “big tech”.

As far as I can tell, this article is basically a fabrication for clickbait.

Sony has been making high-end portable music players for years now. There's still a lucrative market for personal music players among audiophiles, and Sony is one of the big names. They cost what they do, because Sony dedicates serious time and attention to the playback circuitry in these units. They sound phenomenal, even with extremely sensitive (and expensive) IEMs.

This isn't even a return to the "Walkman" branding. Models like the A105 and ZX507, or the uber-expensive WM1A and WM1Z, have been out for years.

https://electronics.sony.com/audio/walkman-digital-recorders...

This doesn't even appear to be any sort of anniversary edition. I can't find any explicit callback to the original Walkman in the marketing copy. There's no press release from Sony reminding us how far we've come from the original Walkman.

All that happened here is, Sony announced yet another product in a long line of similar products with consistent branding, price and featureset.

> Which major distro allows using ZFS from its official installer…?

That says more about the distributions than about ZFS’s fitness or maturity as a file system.

It’s understandable that many distributions don’t bundle ZFS, since the licensing is still an open debate. Although that’s not so cut-and-dried [1], I wouldn’t expect broad support from distributions unless and until these issues have been tested and proved out one way or the other. No organization wants to insert themselves into a legal skirmish.

The best (and probably only good thing) to come out of Ubuntu’s support for ZFS is the vote of confidence in favor of ZFS’s legality in-base from a major corporate entity. We can hope more distributions pick up on this, but I’m not holding my breath.

But legal issues and technicalities say nothing at all about the relative technical merit of ZFS (or any other file system).

> how many people use it as a filesystem for the linux root? Never seen anyone putting the main OS on zfs

Quite a lot of users, actually. There’s a significant community around ZFS, including an ecosystem of some robust tooling. ZFSBootMenu is evidence enough that people want to run root on ZFS, and that root on ZFS is both feasible and highly desirable. Compare its feature list to what you can achieve with root-on-Btrfs, and ZFS pretty handily comes out ahead as the more powerful file system in terms of real world administrative capabilities.

[1] https://blog.hansenpartnership.com/are-gplv2-and-cddl-incomp...

> Zfs on linux is still not ready for primetime on root partitions

Not sure where you got this from. OpenZFS on Linux is considered stable by the project. It has been rock solid, free of major issues, and considered more than ready for prime time since at least 0.8, and probably much earlier. It’s now at 2.1.

I’ve been using it across multiple environments, including my personal laptop and workstation which maintained the same root-on-ZFS install for 6+ years, including 2 full migrations to new hardware using zfs send. I’ve deployed much larger pools in professional settings and they’ve been rock solid.

Btrfs has had numerous data loss issues in that time. I can only recall one data loss issue in a ZFS on Linux release, several years ago, which was caught very quickly.