Namely I’d ask when not if. My opinion is that short of the one time pad, we won’t come up with provably unbreakable schemes.
277 karma · joined May 10, 2017
Also https://twitter.com/galeaspablo
luis ([.@]) ambar [(.)] cloud
Namely I’d ask when not if. My opinion is that short of the one time pad, we won’t come up with provably unbreakable schemes.
In any case, the overall point remains. Short of the one time pad you can’t build a provably flawless scheme.
Have a look at this post, which illustrates this reality being true for hash functions (where similar principles as symmetric and asymmetric encryption apply). https://valerieaurora.org/hash.html
Notice Valerie specifically calls out, “Long semi-mathematical posts comparing the complexity of the attack to the number of protons in the universe”.
Dangerous oversimplification of Order Preserving Encryption (OPE). In structured data like dates of birth, an attacker can infer age ranges, age distributions, and possibly even individual records.
There’s nuance to it of course. But for an article that started out with nuance, the details of the solution and its tradeoffs left me wanting.
The paper Enquo bases itself on could also do a better job of describing what’s missing. Here’s the link to that, https://eprint.iacr.org/2016/612.pdf
One safety mechanism I can think of is that the replicas will detect the leader is down and trigger leader election themselves. Or that upon restart the leader realized it restarted and triggers leader election in a way that B ends up as the leader. (not sure either is being done)
As I think about it more, even if there’s a solution I think I’ll stick to running Redpanda or running Kafka with fsync.
The reason KRaft turns out to be fine is because the KRaft topic does fsync! Source: https://cwiki.apache.org/confluence/display/KAFKA/KIP-595%3A...
KRaft is used for metadata replication in the same way that Zookeeper is used for metadata. I.e., in a very meaningful way.
1) What about Kafka + KRaft, doesn't that suffer the same problem you point out in Redpanda? If so, recommending to your customers to run KRaft without fsync would be like recommending running with a Zookeeper that sometimes doesn't work. Or do I fundamentally misunderstand KRaft?
2) You mention simultaneous power failures deep into the fsync blog post. I think this should be more visible in your benchmark blog post, when you write about turnning off fsyncs.
You seem to know their work, so if you have further publications, I’d love to get them please.
The figures (eg $21M) and names dropped (eg Stanford) are an appeal to authority, which does make me curious.
I’d love to see some papers, I went through all of Balaji Prabhakar‘s publications (titles only), and didn’t see a single paper that sounded like “NTP but machine learning.
If anyone else knows more about this, I’d love to hear from you. Surely $21M doesn’t get dropped with at least someone doing due dill on the tech?
Lamport timestamps and vectors help you get causality ordering. This is not enough for Spanner.
Spanner needs total global ordering. A perfect real time clock would let you achieve total ordering. As you’ve pointed out, Spanner relies on a clock. However, no clock is perfect, so Spanner has to account for that margin of error. And because of its clock (TrueTime) that margin of error is small enough to keep latency low.
E.g. It is possible to make a complaint at a restaurant in a factual way and with a smile.
E.g.2. Giving hard feedback with a smile, and constructively.
It comes down to empathy, and the basic rule of treating others as we wish to be treated.
Will be adding Sipser to my reading list.
I am interested in knowing more about this. Could you expand on why you don’t think any of those books are theory? If it isn’t too much to ask can you mention why each one of the books recommended for the different areas aren’t “theory”?
Pushfor is an instant messaging and content sharing platform designed specifically for secure business use. We are looking for backend, frontend, and/or full stack engineers.
Why work with us? Competence above politics. Humor over stress. Collaboration instead of pointing fingers. Star wars over star trek. Kidding on the last one, or am I? ...From day one, we will value your strengths and give you autonomy on the features you develop. You'll get the chance to work on exciting problems (think heatmaps, distributed systems, real time analytics, event sourcing, functional programming). And last but not least, you'll enjoy all this at our relaxed office (we're all very friendly); we have a terrace overlooking the city which we take advantage of.
Important: good cs fundamentals, always learning, and personable attitude. Specific technologies and development practices can be learned on the job.
Our backend stack is based on PHP (Symfony), Python, Scala, MongoDB, SQL, S3, and Docker. Our frontend stack is based on React, Elm, Typescript, Redux, Docker. We encourage learning, so if you want to move on to something new, we're happy for you to learn on the job.
The interview process consists of an initial phone call, a technical interview, and an interview with management. Contact [luis] [at] [pushfor] [dot] [com]
I didn’t do CS in university. Following the above got me up to speed. I’ve seen it several times on HN. This time it’s me promoting it.
Machine learning however, can use past behavior to make an educated guess, regarding how/when/if an individual will repay and/or how vulnerable said individual is to unforeseen external events.
Machine learning might not be able to go as deep as evaluating someone’s chances of getting hit by lightning, for the purposes of credit ratings. However it does figure out if an individual has the chance to get out of sticky situations.
Summarizing, it’s not about predicting irrational behavior. It’s about using past behavior to make a better educated guess than was possible before.