stabbles

39 ms·

481.
by stabbles 6y ago
In my experience they don't often have a configure file in the repo, but only in the release tarball (some repos on github have an action to setup a tarball with configure script). That way the user only needs a shell and make, nothing
482.
by stabbles 6y ago
Today I had to use autotools for the first time, where I previously only used CMake. I thought I should be scared, and m4 looks daunting, but what they at least get right is that it's trivial to generate both shared and static librarie
483.
by stabbles 6y ago
not sure why this was downvoted to oblivion, as it seems to be correct
484.
by stabbles 6y ago
Ah, I have to take that back, since benchmarks run in the order of seconds and they use sockets to start and stop the timer, which likely means compilation time is not included.
485.
by stabbles 6y ago
I think it's more interesting to see what people do with the language instead of focusing on microbenchmarks. There's for instance this great package https://github.com/JuliaSIMD/LoopVectorization.jl which ex
486.
by stabbles 6y ago
They are measuring compile time and runtime speed, not just runtime speed like for statically compiled langauges
487.
by stabbles 6y ago
Github seems to be doing worse
488.
by stabbles 6y ago
From the git history it seems like DataFrames.jl maintainers contributed at least some fixes to the scripts, so I guess that means they aren't opposed to it.
489.
by stabbles 6y ago
password-store has 0: https://reports.exodus-privacy.eu.org/en/reports/dev.msfjarv... but it's commonly used with OpenKeyChain, which does have a tracker: https://reports.exodus-privacy.eu.org&#x2F
490.
by stabbles 6y ago
Github disables secrets for actions triggered from pr's. But users can commit an action that runs on their pr.
491.
by stabbles 6y ago
Well, all functions that use finite precision arithmetic deal with approximations. Errors of 1/2 ulp like trig functions are rather the exception. And machine learning is keen on using low precision floating point type, so errors are r
492.
by stabbles 6y ago
An obvious example here is when you expand e^x with its Taylor series using `n` terms. The derivative of that length-n Taylor expansion is identical but only `n-1` terms long, so you lose precision. More generally if you approximate a smoot
493.
by stabbles 6y ago
Yeah, I think only Power has hardware double of 128 bits, but it's not the same as double-double because the exponent range is equal to double's exponent range. I agree this package is not new conceptually, but it has a convenient
494.
by stabbles 6y ago
Relatedly, there's a relatively new julia package called MultiFloats.jl [1] for doing arithmetic with numbers of the form a_1 + a_2 + ... + a_n where a_{i+1} is of the order of the round off of a_i. So it's fixed precision, but th
495.
by stabbles 6y ago
While you're at it block png and gif too
496.
by stabbles 6y ago
A way to make GitHub's new homepage strictly faster is by removing the dynamic bits
497.
by stabbles 6y ago
Mpi over ssh sounds like a bad idea. Typically hpc clusters have a highly optimized vendor libmpi. E.g. cray has an abi compatible libmpi.so to mpich, so you are likely better of replacing the libmpi.so in your container with the vendor ver
498.
by stabbles 6y ago
The problem with slurm is how it's typically used: ssh into a shared login node with a shared file system, authorization is tightly coupled to linux users on that node, submit jobs with sbatch. Kubernetes deployment feels much more mod
499.
by stabbles 6y ago
We study ants too... Go to the ant, thou sluggard; Consider her ways, and be wise
500.
by stabbles 6y ago
Oh come one now, don't be silly. Think about vectorization and loop unrolling. It _always_ does a memory load with an offset as a single CPU instruction. As an example a 4 times unrolled sum of doubles on AVX looks like this: L64
501.
by stabbles 6y ago
Exactly, there's no silver bullet for ranges
502.
by stabbles 6y ago
Lisp programmers don't care how heat affects food since the 1950s.
503.
by stabbles 6y ago
Now do a closed range in type uint8_t representing [0x00, 0xff] Or how about a closed range in uint64_t that ends at 0xffffffffffffffff Or a range from roughly a = -9223372036854775808 to b = 9223372036854775807 in int64_t. b - a will overf
504.
by stabbles 6y ago
And Rust does 1..13 and 1..=12
505.
by stabbles 6y ago
This might be an unpopular opinion, but I'd say it's more natural for kids to relate the word 'three' to 'third' and 'four' to 'fourth', instead of 'two' to 'third' and &
506.
by stabbles 6y ago
That's great advice for a kid learning python. "Kid, let me tell you about pointers in C, then you'll understand we you have to write range(1, 13)"
507.
by stabbles 6y ago
What you see most is domain-specific languages implemented through macro's. For instance this package, https://jump.dev/JuMP.jl/dev/examples/basic/ , https://github.com/SciML/Mo
508.
by stabbles 6y ago
You can always write julia> begin for i = 1:3 println(i) end end 1 2 3 /s
509.
by stabbles 6y ago
julia> (1:3)[begin+1:end] 2:3 it exists
510.
by stabbles 6y ago
julia> supertype(String) == supertype(SubString) == AbstractString true If you insist just use f(s::AbstractString)

More ›