Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
denormalfloat
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
61.
▲
by
denormalfloat
8y ago
> Is it using protocol buffers? Do you care? I don't, but other people do. For anything low latency, the cost of encoding/decoding starts to become extremely difficult to avoid. But that said, I'm pretty sure GRPC can use
62.
▲
by
denormalfloat
8y ago
Do you happen to know by chance what scaling algorithm is used?
63.
▲
by
denormalfloat
8y ago
> but my point is that it filters out great engineers who worked and delivered solid products because the interviewers are just not interested in details of past work experience Again, yes. It is hard to fire a bad candidate in case the
64.
▲
by
denormalfloat
8y ago
This would likely result in interviewer bias based on how familiar they are with the project the candidate was working on. How do I fairly compare 10 candidates based on their answer to that question? The interview questions need to be ob
65.
▲
by
denormalfloat
8y ago
Data structures and Algorithms are fun, but more and more I see that general software engineering is something we need. Several useful questions venture too deeply into domain specific things, making them difficult to give to candidates.
66.
▲
by
denormalfloat
8y ago
It's because lying on a Resume is extremely easy. I do interviews for a large software company about once a week. The relationship between what a candidate says they did and actually doing simple programming problems is non-existent.
67.
▲
by
denormalfloat
8y ago
Out of all the string types I have used, I think Go has the best of them. Since it's a GC'd language, you don't have to pay the refcount atomic overhead. The representation is two only words long. Strings are immutable, m
68.
▲
by
denormalfloat
8y ago
Requesting the Desktop site actually works, surprisingly.
69.
▲
by
denormalfloat
8y ago
Alright, so existing debugging tools don't work with QUIC. We will need to make some new tools that can expose the information we need. If the hops between the source and destination are willing to expose the info, (and we can assum
70.
▲
by
denormalfloat
8y ago
I'm saddened to not see Closure listed on there for JS minification. It might be worth mentioning that there are more advanced minification tools. Also, there doesn't seem to be anything on JSON minification, which is a sizable p
71.
▲
by
denormalfloat
8y ago
So, what's the memory model going to be? Without guarantees about the safety of operation, how can anyone reason about the safety of the program?
72.
▲
by
denormalfloat
8y ago
For me, squashing images is most useful when creating thumbnails. This makes pages load much faster due to the smaller file size. In this sense, optimage would be more useful if it could also intelligently crop the image down to a square.
73.
▲
by
denormalfloat
8y ago
Java Puzzlers by Joshua Bloch and Jeremy Manson are pretty entertaining. For a pretty bog standard language, Java has some more exotic corners: https://www.youtube.com/watch?v=wbp-3BJWsU8
74.
▲
by
denormalfloat
8y ago
HTTP/2 effectively forces TLS 1.2+, due to the limited cipher suites in the RFC, and browsers' restrictions. For the sake of argument, TLS is a requirement. To this end: * TLS requires ALPN. This would not be so bad, but so man
75.
▲
by
denormalfloat
8y ago
PUSH PROMISE is great and all, but the use is limited. If the remote side already has the asset you push (because it's cached), you waste your bandwidth and increase latency.
76.
▲
by
denormalfloat
8y ago
(I work with HTTP/2 as part of my job) Downsides: * TLS requirements are complicated and hardly anyone knows how it works. * Several agents partially support it, but the overlapping feature set is useless. An easy example is trailers.
77.
▲
by
denormalfloat
8y ago
Go's scheduler is not preemptive, which means that sometimes Goroutines can hog the processor. If loops don't contain function calls, or allocate memory, or other preemption points, they can prevent other goroutines from running
78.
▲
by
denormalfloat
8y ago
Go may have more threads, but it is not better at concurrency. It lacks several key things that make Java much easier to scale up, even if Go is easier to get started with. On my wish list: * Thread Local storage. A million Goroutines m
79.
▲
by
denormalfloat
8y ago
Yes, it spawns a kernel thread if all the threads are currently blocked. I ran into this issue running Go on a FUSE mounted file system. If the network hung, Go would spawn thousands of threads and OOM. Go can't _not_ spawn them, a
80.
▲
by
denormalfloat
8y ago
Why not just include the Origin on all cross-origin requests? Then the server could deny/allow it without the need for preflight.
81.
▲
by
denormalfloat
8y ago
Using arithmetic average of $420 million a year is incorrect, because inflation is exponential. Conservatively estimating 2% inflation and holding all else equal, that comes out to about $312 million for year 1, $318m for year two, etc.
82.
▲
by
denormalfloat
8y ago
Buying Money Market funds also have similar yields and are frequently backed by treasury bonds. Vanguard's MM fund currently has a yield of 1.96.
83.
▲
Gamma Steganography
(carlmastrangelo.com)
60 points
by
denormalfloat
8y ago
|
8 comments
84.
▲
by
denormalfloat
8y ago
Encrypting or signing cookies quickly becomes painful when you have multiple front ends. Either they all need to coordinate around the signing key, or they all need to share the same key, and they need to synchronize on key rotation.
85.
▲
by
denormalfloat
8y ago
What shared contracts are there for JSON? The only one I know of is OpenAPI/Swagger, and that's already complicated. Wireshark is used because sometimes we don't have the source code for software we use. The only way to d
86.
▲
by
denormalfloat
8y ago
The success of JSON is due to one thing: Browser support. As soon as it's use on a server, or a mobile device, or practically anywhere else almost all the benefits of JSON go away. Can you inspect the payload easily? Well, yeah, if