It was a great session and we received a lot of questions. We attend different NOG conferences regularly. ISPs are incentivized to help us by providing good data. Although we are agnostic about adversarial geofeeds, ISPs themselves need to work with us to ensure good quality of service to their users.
We already do quite a lot of outreach, in fact, most network engineers in the ISP industry across the world are familiar with us. But if any ISP operator has any feedback for us, we are only an email (or even a social media comment) away.
I can't imagine this would work. Can you not remember the footage of the wall of houses/cars/boats/trees/etc moving across the landscape? Your raft is getting flattened.
SSD (nvme) and TCP (10/100gbe) both exhibit low tens of microseconds of latency. This is ignoring redundancy for both of course, but the penalties should also be similar.
If the storage is far away, then you'll go slower of course because of additional network latency. But since the article is comparing to EFS (aka AWS brand of NFS), it's fitting to talk about a nearby scenarios I think.
It's always funny when I watch stuff about some foreigner visiting my home country and they either focus on something not all that important, or get something completely wrong.
The funniest part is trying to present some dish as "traditional" that everyone here eats, while it's some super niche thing only one region does, occasionally, if you have grandma that remembers how to make it
I appreciate that must have been difficult. If you could set the record straight, where would you have taken him? Love to hear your reccos and thoughts!
Heh, similar thoughts! The main difference that I only used Scheme for SICP, and I've used a bit of Haskell.
I like Haskell in theory, but: just to get a hello world takes a lot of CPU and disk space. The standard library is full of exceptions (you can use a different prelude, that opens a whole different can of worms). The ergonomics of converting between the thousand different string types are awful.
So, you being basically me, I have some recommendations:
Idris (2): good stdlib, has dependent types. A beautiful language. The compiler is self-hosted and bootstrapped by lisp - very elegant! The ecosystem is basically nonexistent though.
PureScript: also improves on Haskell in many ways. But, it's more of a frontend language, and though you can do backend, you're stuck with JavaScript runtime. Oh well.
Traditional marketplaces provide a range of services, from unified delivery to complete logistics management. They also provide all the kyc filtering and fraud screening that quite lowers the merchant risk.
On top of that, many of them provide additional assurances, like vendor screening and easy dispute resolution on fraudulent operations.
I have feeling that microservices improve overall design when they can live on their own, as microapps perhaps, also with their own UI. What is the point of service if it is not usable beyond its original design and just bound to other similar services?
I'm pretty shocked we are still waiting for NVMe hard disk drives (classic spinning rust).
The overhead of SATA doesn't matter a ton, but doesn't seem excellent. It just feels like eventually it's going to be silly having SATA ports on systems at all, be such a very narrow use case, and that PCIe & NVMe really ought be an option. (More so if PCIe switches hadn't skyrocketed in price over the years, alas.)
There have been rumblings about NVMe HDD for years now but still no products. How that would help though, make the SATA ssd feel less necessary.
Talking about why one feels a particular way requires trust in the interlocutor, so I'm not surprised outsiders don't hear anyone taking about it. As for insiders, is it a given they aren't?
The video for that song uses excerpts from the movie version of Johnny Got His Gun. I don't know if the songwriting was inspired by it but the video certainly was. Because the video relied so much on the excerpts the band ended up buying the rights to the movie just to not pay royalties.
Hilariously, I won a writing prize about this connection as a teenager in 1989. Fun to see you had a similar experience
There's an assured win combo in this game, at least against the Hard AI. The combo is as follows:
Take the top left corner. It takes the center. Take the bottom left corner. It flips it's center piece every time. Take either right corner. It takes the remaining corner. Take the center right. Flip one of your left corner pieces to form a diagonal ghost line
If the AI doesn't take the center and waste its flip, then it can definitely be countered.
As for feedback, I'd say i had a few moments where i was confused as to why a whole cluster of pieces flipped to ghosts. Honestly if that was described in a tutorial better it'd probably be a non-issue.
That was actually a great article. For us, that is like a crowdsourced bug hunting program. We actually got duped ourselves, and we appreciate the author.
We added additional features for location hint modeling and selection for IPv6 networks. There are a handful of open engineering tickets to understand more about the entire internet infrastructure of the country. Of course, hosting a probe server out there would be helpful.
Speed is the rationale. I have zero hesitation to deploy and am extremely well practiced at decomposing changes into a series of small safe changes at this point. So maybe it's a single spelling correction, or perhaps it's the backend for a new service integration -- it's all the same to me.
Churn is kind of a loaded word, I'd just call it change. Improvements, efficiencies, additions and yes, of course, fixes.
It may be a little unfair to compare monoliths with distributed services when it comes to deployments. I often deploy three services (sometimes more) to implement a new feature, and that wouldn't be the case with a monolith. So 100% there is a lower number of deploys needed in that world (I know, I've been there). Unfortunately, there is also a natural friction that prevents deploying things as they become available. Google called that latency out in DORA for a reason.
I wasn't aware of the thread local trick, I solve this problem by not setting WS_VISIBLE and calling SetWindowPos & ShowWindow after CreateWindow returns (this solves some other problems as well..)
https://retractionwatch.com/2021/09/14/highly-criticized-pap...