4,245 karma · joined December 27, 2011
This is just propaganda. Another way to look at it: Putin, as a head of FSB, sabotaged Yeltsin to get into power.
I understand, but you are still talking about a situation where surviving a volumetric DDoS attack without a global centralized provider was possible. It wasn't smooth for you, but it could have been if things were done a bit differently.
Anyway, here on the other side of the world it's not like that, DDoS protection is more common. Because in the early days of DDoS attacks with all the dreadful blackholing one of the big European providers OVH invested in DDoS protection and kind of pushed the whole market to provide it too instead of blackholing.
But in practice datacenters, uplinks and internet exchanges often are able to do flowspec, firewall rules, block all UDP for a subnet in all networks they have relationships with, etc. So plenty of those nodes can be behind ISPs that mitigate volumetric attacks automatically, so even simple DNS failover might be good enough to protect from such attacks. It's not that hard. Layer 7 is where the hard part is.
Also, it was only up to like very early 2000s when researchers of decentralized systems mostly ignored the existence of malicious actors, but later everyone became well aware of them and started considering how to deal with them.
People are confused because of the comedic level of bullshit (reminds me of "the box" from a Silicon Valley tv show episode).
I'm afraid what people mean by that statement and what the startup is going to try to do are completely different things. As in people mean AWS-like cloud services running on multiple servers they control with all the fault tolerance and geographically distributed. But for the startup it probably means cloud services running on expensive servers in a fast local network.
Disks:
An Analysis of Latent Sector Errors in Disk Drives (2007) https://research.cs.wisc.edu/adsl/Publications/latent-sigmet...
Failure Trends in a Large Disk Drive Population (2007) https://static.googleusercontent.com/media/research.google.c...
Cycles, Cells and Platters: An Empirical Analysis of Hardware Failures on a Million Consumer PCs (2012) https://www.microsoft.com/en-us/research/wp-content/uploads/...
Flash Reliability in Production: The Expected and the Unexpected (2016) https://www.usenix.org/system/files/conference/fast16/fast16...
Filesystems:
IRON File Systems (how filesystems behave on disk errors) (2005) https://research.cs.wisc.edu/wind/Publications/iron-sosp05.p...
EIO: Error Handling is Occasionally Correct (2008) https://www.usenix.org/legacy/event/fast08/tech/full_papers/...
SQCK: A Declarative File System Checker (2008) https://www.usenix.org/legacy/events/osdi08/tech/full_papers...
All File Systems Are Not Created Equal: On the Complexity of Crafting Crash-Consistent Applications (2014) https://www.usenix.org/system/files/conference/osdi14/osdi14...
Filesystem error handling (2017) https://danluu.com/filesystem-errors/
Files are fraught with peril (2019) https://danluu.com/deconstruct-files/
And a DRAM reliability paper, for a more complete picture:
DRAM Errors in the Wild: A Large-Scale Field Study (2009) https://www.cs.toronto.edu/~bianca/papers/sigmetrics09.pdf
Consider that filesystem data cannot be synced to underlying storage in a such way that it can be reliably read later and backups by definition are asynchronous and provide eventual consistency. This means that the best meaningful guarantee an application can get after machine crash is being able to read data written up to some point in the past, but not after every successful fsync before crash. Usually, though, even that guarantee is hard to achieve, as bad blocks happen and redundancy is not there. Given all that it's ok to relax filesystem APIs behavior to proper physical constraints. Like fsync is only meaningful as an ordering operation, no need to actually flush anything to disk immediately on fsync or any operation.
Next is multi process and multi threaded scenarios. Should O_APPEND only work correctly from a single thread, should each write be atomic and to what size, we certainly can't have gigabytes in an atomic append or should there be some synchronization mechanism that blocks others? Same for temporary files and unit files.
And what to do on bad blocks? Should there be redundancy within a single disk, should block device underneath be log structured storage with block remapping and scrubbing and provide reliable storage layer to the outside by sacrificing space? Maybe for desktop machines it should, but not for servers, at least not all of servers, they need a different API.
I'm not even touching performance considerations here, that depend on performance-friendly APIs a lot.
No, most Crimeans do speak both Russian and Ukrainian, just like everyone else in Ukraine, since Ukrainian is the only official language and it's pretty much impossible to not be exposed to both languages, especially in education and media. After the fall of Soviet Union all the ethnic Russians in Ukraine (incl. Crimea) had no way of avoiding it and do speak Ukrainian.
This is misleading and factually incorrect. Everyone who speaks Russian also speaks Ukrainian in Ukraine, except for people who literally moved there recently from Russia and other countries.
I think you got it backwards. I remember using black backgrounds and things like "mc -b" specifically because CRTs I used sucked at light colors, trinitron displays later were nicer though and using white backgrounds was pleasant. Nowadays led backlit displays suck again on light and bright colors and don't seem to improve, hence "dark mode" became a solution.
I'm sure all of them can implement FizzBuzz, just not on the interview under pressure. It has absolutely nothing to do with competence, but understanding that requires just a little bit of background in human psychology.
There is enough research in job markets showing that companies get more picky and advertise more demanding and complicated job requirements as more candidates are available. So your claim is empty, since this is how companies behave regardless of supply. And if you were to measure it, how big artificial barriers or filters are for someone to get hired could be the measure of oversupply, not shortage. With shortage there would be no barriers, everyone applying would be interviewed and likely hired. But given current hazing hours long interview practices it's pretty clear that companies are drowning in oversupply.
In all of those fields projects eventually get finished and involve a lot of bureaucracy and waiting. There is much more breathing room, than in software and more time to recover.