3 ms·
Like a few others have mentioned, I do wonder how big the marginal return in baking in these features is. From a quick look, stuff like FASTA/BAM parsing, tran
by neeeeees 7y ago
Like a few others have mentioned, I do wonder how big the marginal return in baking in these features is.
From a quick look, stuff like FASTA/BAM parsing, translation, etc can be implemented in C-land a la numpy, and called from Python, right?
A language like Swift would also support the addition of powerful user-defined operators, as in the case of Swift for TensorFlow[1].
Language adoption is hard to drive, and I wonder if having domain-specific library calls built-in is worth the added effort for people in the field.
[1] https://www.tensorflow.org/swift https://www.tensorflow.org/swift
- Someone 7y agoNitpick: Swift for TensorFlow currently is a fork of Swift, not a library implemented in Swift. https://github.com/tensorflow/swift https://github.com/tensorflow/swift: ”Compiler and standard library development happens on the tensorflow branch of the apple/swift repository. […] Swift for TensorFlow is not intended to remain a long-term fork of the official Swift language. Language additions are designed to fit with the direction of Swift and will go through the Swift Evolution process.”
- neeeeees 7y agoThat is correct, and I should have made that clearer. However, a lot of it’s features are implemented in Swift (via custom operators, overloading etc) in addition to compiler intrinsics. The point I had in mind was that a language like Swift may allow for both power and DSL-like expressiveness without requiring learning an entirely new language. In fact, standard arrays are implemented in Swift code (https://github.com/apple/swift/blob/master/stdlib/public/core/Array.swift https://github.com/apple/swift/blob/master/stdlib/public/cor...) yet feel closer to say a Python list than a Java ArrayList from the user’s perspective.