Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
fooyc
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
19 ms
·
151.
▲
LiveJournal Outage
(livejournal.com)
1 points
by
fooyc
12y ago
|
0 comments
152.
▲
by
fooyc
12y ago
What I would really love to see in a solution like this is the ability to automatically rebalance shards (like moving data between shards when adding new instances). Sharding can be achieved quite easily on the application level, or by usin
153.
▲
by
fooyc
12y ago
If your login page is HTTP, your HTTPS receiver is useless. A MITM can just change the form's target URL in the HTTP login page. On inject any scripts.
154.
▲
Parquet
(parquet.io)
1 points
by
fooyc
12y ago
|
0 comments
155.
▲
Google FlatBuffers
(github.com)
1 points
by
fooyc
12y ago
|
0 comments
156.
▲
by
fooyc
12y ago
My point is that if you feel that the email practice should be listed for the reason that it sends passwords in clear text over the network, then the equally bad practice of sending password over plain HTTP should be listed too.
157.
▲
by
fooyc
12y ago
I agree that from a security standpoint it's better not to send the password in plain text at all. But then you can start listing non-HTTPS sites, too. You disclose your password in plaintext to many servers every time you login on tho
158.
▲
by
fooyc
12y ago
For good measure you should list websites allowing you to enter a password on a non-https page, too. That's plain text, too.
159.
▲
by
fooyc
12y ago
Odds are that you registered over plain HTTP anyway
160.
▲
by
fooyc
12y ago
That's really trivial, and many web applications work this way: email = input.email plainPassword = input.password hashed = hash(plainPassword) saveToDb(email, hashed) sendGreetingEmail(email, plainPassword)
161.
▲
by
fooyc
12y ago
Turns out the kernel is very well optimised for kernel compiling workloads.
162.
▲
by
fooyc
12y ago
And this doesn't even speak of the syntax used to express ownerships and lifetimes: http://rustbyexample.com/examples/move/README.html http://rustbyexample.com/examples/borrow/README
163.
▲
by
fooyc
12y ago
Surely "a + b" is nicer to the eye than "a.add(b)", but unless a and b are numbers, a method call with a meaningful name is more straightforward than any abuse of allowed operators. > Take a look at Scala for an argu
164.
▲
by
fooyc
12y ago
Thanks for this insightful answer. About operator overloading, some languages have seemless bignums without allowing users to overload operators in their own classes. For containers, this could have been restricted to the index operator.
165.
▲
by
fooyc
12y ago
Sadly, operator overloading is not a really good way of making costs explicit.
166.
▲
by
fooyc
12y ago
Is there many cases where it's safe to allocate an object on the stack, but you would want to allocate it on the heap instead ? (Appart from the the object being too big.)
167.
▲
by
fooyc
12y ago
Depending on what I'm doing, I do want a low-level language. Still, Rust's memory management feels weird and painful. But above that I'm criticising features such as operator overloading, overridable copy semantics, and trait
168.
▲
by
fooyc
12y ago
> > - Explicit allocations (stack vs heap) and ownership management. > This is a great feature. If you're writing another web app, then sure, you don't care about stack vs heap. If you're writing performance oriented
169.
▲
by
fooyc
12y ago
> Copy is a trait that is automatically implemented by the compiler (and cannot be manually overridden) that basically just says "semantically this type can be copied with memcpy", it doesn't actually control the runtime b
170.
▲
by
fooyc
12y ago
> If you actually believe that, you have not really looked. It adds e.g. algebraic data types, pattern matching, actor based concurrency, it's more functional, has pretty good macros, has the compiler do more heavy lifting etc. My p
171.
▲
by
fooyc
12y ago
> Rust explicitly doesn't allow overriding `=`. In part 25 Copy and clone: "copied by default instead of moved, because these types implement the Copy trait" So implementing the Copy trait allow to modify the semantics of
172.
▲
by
fooyc
12y ago
Rust feels a lot like C++ with smart_ptr/unique_ptr by default, just with a different syntax. Bad things they imported from C++ : - Traits (a.k.a. I don't known where this method is implemented). - Over-complicated and compile-tim
173.
▲
Data vizualisation: do you prefer Pie Chart or Histogram?
(twitter.com)
2 points
by
fooyc
13y ago
|
0 comments
174.
▲
by
fooyc
13y ago
I believe Chrome is much more strategical than that. Google could lose enormous market share if any major browser decided to change the default search engine. So they created their own major browser. Chrome is not an information source for
175.
▲
by
fooyc
13y ago
The browser has a local copy of the blocked sites database; so it doesn't have to do any extra request in order to check if a web site is blocked or not.
176.
▲
Light Table
(lighttable.com)
7 points
by
fooyc
13y ago
|
1 comments
177.
▲
by
fooyc
13y ago
Well, as a dev I'm obviously not a fan or Eventual Consistency at all.
178.
▲
by
fooyc
13y ago
That's only 3000 writes per VM per second. Now I would like to see how many reads it can do at the same time. Cassandra does mostly-sequential writes, however most reads do couples of disk seeks.
179.
▲
by
fooyc
13y ago
If I understand correctly, the goal is to avoid too much calls from being blocked for $timeout seconds at the same time, by failing fast when too much calls have _already_ failed. Am I correct that blocked calls can still accumulate before
180.
▲
by
fooyc
13y ago
AFAIK, `use` were required because PHP has variable-variables: <?php $x = "b"; $b = 42; echo $$x; // equivalent to echo $b; // Here the compiler can't determine all the variables
More ›