magnawave

The best I can tell from talking with insiders at a couple of his properties, is that he did a beyond bad job at “succession planning”. The light exception being MoPOP of course but that’s a different beast. So without setting up foundations/entities to fund/run your projects and getting people aligned before you pass, odds are your estate won’t.

So want to blame the family. But really Paul Allen’s fault here.

It’s a shame, was such a great museum.

AFCI's feel like NEC's attempt to create tons of service calls under the guise of "safety". And eventually seems like they will be required for new builds everywhere as areas adopt the newer codes that require them slathered everywhere.

Of course its "for safety" but of course these breakers are famous for false tripping and causing expensive service calls.

anecdote: In my case, small server rack, on circuit a with an AFCI - since "bedroom". No problem.

Until washing machine on circuit b (no AFCI) runs, then trips circuit a's AFCI. Repeatedly, every time(debug on the breaker a returns ArcFault detection reason too). So... either a) don't wash your clothes b) don't have tech or c) quietly violate the code.

Indeed electrician quietly suggested(after a bunch of triage) I swap out the breaker with a normal one. But of course he wasn't allowed to do that... lol

Been here a decade - nary an issue since - so clearly the usual case of nuisance tripping and nothing more.

I suspect the switching power supplies were close to annoying the ACFI, and a beefy motor on an adjacent circuit was enough to push it over the line. Incidentally, swapped UPSes, power supplies(quality Seasonic), etc and nothing improved.

I guess the wildcard is price.

AWS's pricing model works kinda at their OMG eyewatering scale - aka all the custom hardware they design is highly cost optimized, but just doing custom hardware has a notable cost. This is easily covered by their scale, to make for their famous margins. [during their low scale times, they did use a good bit of HP/Dell, etc]

Oxide seems to be no different (super custom hardware) only major difference being the "in your datacenter" part. Since you own the cost of your datacenter, Oxide has to come in a lot cheaper to even compete with AWS, but how do you do that with low volume [and from the look of it not-cost optimized, but instead fairly tank-like] bespoke hardware? Feels like the pricing / customer fundamentals are going to be pretty rough here outside perhaps a few verticals.

Oh this is awesome. When I was a kid I really really attempted to write a Keen knock off (doctor who themed) on my 386 and it was ok although never really did enough level development to make it super playable. But TO THIS DAY, never knew how the heck they made it scroll so smoothly. I just ended up just jump scrolling. Figured it was some mad assembler skills, never imagined there was a hw register!

Did figure out the double buffering stuff on my own though ;)

Seems so obvious, learn something new everyday, even if it would have been more useful, say maybe ~30yrs ago ;)

…and sadly the doctors have been fleeing too, all my doctors have left. Their replacements aren’t bad, but not to the same top notch level of care. The Polyclinic was honestly off the charts excellent, and now it’s just “fine”. In a few more years who knows. The alternatives are unfortunately not better.

Indeed doing what they have done is impressive, and talented folks at play for sure. Not sure, even vaguely commercially viable(heck 0xide as a whole), but nerdy interesting for sure. But I can’t see practically why this is so much more useful than what society has built together and call the x86 platform over the last few decades.

One of the hard parts of quantifying that, is you have people who wear many hats. So sure you have Cassandra gurus, and probably a decent number of them. But this is in the league where really hardcore automation kicks in to keep staffing sane, and operations possible. But outside that, how do you count datacenter folks, client folks, networking folks, etc who only spend a little fraction of time on the database parts.

But I think I can say with sufficient knowledge, all things considered still, “way fewer than you might think”.

Correct, that’s what object stores are for. However, metadata on said files, is probably very handy to have in a database.

I’m quite sure not all this Cassandra capacity is just file/photo metadata storage either.

Personally I find it funny that on the same day there is an article about Quora, there is also a link for plugin for blocking search results(https://github.com/iorate/uBlacklist). Immediately, I thought, oh that would be great for the spam Quora links. Although lately I’ve seen those far less, they still piss me off every time I hit one. ;)

Thought the exact thing the first time I saw one in a Chicago full service parking garage. And they had very stern signs about being for employees only. So ya that’s likely how they solve the liability problem. But man they looked really cool. So wanted to try it out ;)

The result was "an electrical island in the United States," Bill Magness, CEO of ERCOT, said. "That independence has been jealously guarded, I think both by policy makers and the industry."

https://www.statesman.com/story/news/2021/02/16/texas-power-...

Maybe it’s time to rethink that. HaI has an interesting take on something similar: Japan https://www.youtube.com/watch?v=Mo88zA5nq4Q

Running your own grid seems cool Texas style, until you have a regional problem and have no where to turn.

It’s definitely comically expensive. But I’ve had mine 15 years and it’s still the best least fuss humidifier I’ve ever owned. Oh and I never use their additive. Little dish soap (drop of two) in each refilling keep the water “sticking” to the disks nicely, nothing else to replace. Never had a mold problem with them either.

The declarative functionality comes from K8s not helm. (Of course K8s has imperative tooling too but that’s not the point here).

The main reason folks dislike helm are 2 fold: 1) using string templating on a space formatted language is just messy and ugly for no good reason. If you don’t do anything terribly complicated it’s not too bad but quickly can be pretty fragile and unreadable. 2) golang(which I like generally) but the templating is messy and adds to the hard to read later problem

Personally find helm’s opportunity for code reuse to be restrictive and folks just end up cut and pasting a lot of declarations all over which is ... so terrible. But the first bullet above is pretty a fundamental design flaw.

I think his point was about drives silently returning corrupt data, not that you shouldn’t have redundancy. Absolutely you should have redundancy and backups.

At the (tens of) thousands of drives scale (and where you treat drives as cattle) having some extra checks and balances for what you write is an excellent idea. Even better is doing that in a distributed way so multiple machines make that same decision. Occasionally you will run into a drive where the firmware has jumped the shark or some such. (Or CPUs or memory or bus issue for that matter).

But generally speaking drives DO know when they are returning bad data and will error before they will do that. The odds are about as good as other forms of hardware errors that will eat your data.

Not anymore. From what I’ve been told from an insider, the latest versions are derived from a common code base now, with platform specific bits separate for each platform.

The latest office for Mac looks quite a bit like the windows version now (for once) so it’s not too surprising.

Yes Oracle did that to Solaris. You are right. But it was already pretty dead before Oracle bought Sun. Which was why Sun was for-sale in the first place.

That's not just Oracle's fault. Place blame where blame is due(java) - but blaming everything on Oracle is kinda silly here. Sun mis-stepped pretty hard in the years after the original .com bubble and what we are seeing now the the final result. I'm shocked it took this long.

At the end of the day, Sun/Oracle has to be able to make revenue from Solaris for them to keep paying those peoples' salaries.

Is your startup / company deploying on Solaris? Nope of course you aren't - who is?! Pretty much nobody nowadays - w3techs.com shows Solaris at 0.0005% for webservers for example. Everyone talks about how great it is(to be sure there are some small bits that are pretty dang amazing) - yet for some reason nearly no one actually DOES use it! Perhaps there is a reason for that. Lord knows I have many battle scars from the many real world rough edges that never seemed to get less fundamentally shitty(for starters: path_to_inst and smf I'm looking right at you!)

I definitely respect the amazing work of some of the engineers on the Solaris team back in the day. ZFS and DTrace being two that have made the tech world by in large a better place either directly or thru inspiration. And Linux and the BSDs are definitely better for those selective code gifts! But nonetheless, the world voted with their feet, and Solaris didn't make the cut.

I definitely wouldn't call Bloomberg a "non-engineering" company if you've ever worked with any of their trading / market data products. Pretty much true of any trading side finance company nowadays.

As commented already, this is actually incorrect. RAID cannot and will not detect data errors unless the drive itself report the errors and issues the "I can't retrieve this data" call back to the raid controller. If it lies, RAID is fat dumb and happy. Sadly, drives lie all the time.

Now in reality in big enterprise SAN arrays, the vendors usually add checksum blocks onto the underlying disk (aka larger sectors in the Clariion, or checksums into the filesystem on NetApp) OR use special certified firmware on the drives that make sure the drives never lie themselves! So largely if you are on a big enterprise SAN, you probably don't seriously need to worry about bit rot.

But outside of that space, most PCI card RAID controllers, and I suspect at least a few "enterprise lite" arrays, probably don't do checksum calculations. So BTRFS and ZFS provide notable value. (and definitely on raw drives)

Actually most SSL based browsers will transmit the domain cleartext as part of SSL handshaking. This was added so virtual hosting webservers can all have independent certs on the same IP

See the following link for the specifics ... but needless to say its easy to block SSL access with a transparent proxy or layer 7 firewall. (which based on the error page looks like a Palo Alto device which definitely can do this...)

https://idea.popcount.org/2012-06-16-dissecting-ssl-handshak...

Yes this is exactly the problem - less so with CFL but hugely so with LED. LEDs should appear as a part of the fixture itself - not as a screw in replacement for the legacy Edison socket. I suspect this will happen eventually when people start to see all the cool lighting options you have with LED fixtures - but its going to take a good while to get the mental shift to happen!

"And the whole point of RAID is to combat that..."

Actually the whole point of RAID is to combat a MISSING drive, not a corrupt drive. Parity works when I'm lacking one bit ... not when any of my bits might be lying.

Half of the reason of ZFS is to protect against the scenarios when a drive's internal checksum should catch a data error - but the drives returns bogus data and says 'oh its good'. RAID5 can't catch that(mathematically impossible) ... RAID6 can but most implementations don't check because its expensive(extra IO and calcs)

This is the reason most of your high end SAN vendors don't just deploy RAID5/6/10 but also checksum the data themselves under the covers. They don't trust even high end SAS/FC drives.

Indeed vary brave. Filesystems ARE NOT and never have been an API to exchange data between multiple OSes - they are a means to persist data efficiently for a single OS. Now if you don't mind breaking a bunch of the FS rules, sure compatibility hacks can always be written - but are you really sure each implementation is FULLY honoring the rules. I can guarantee you, for anything but the simplest filesystems, the answer will be "no". Hell, even FAT has implementation stragglers out there in the wild and its brain dead simple.

Use an API for cross platform sharing layered on top of a filesystem: Either a Virtual Machine sharing protocol (aka 9p virtio) or a real network protocol AFP/SMB/NFS. Anything else is taking your data in your own hands and is antithetical to the design of modern OSes.