Merge pull request #18 from nellshamrell/memory-safe-by-default-best-…

…practices

adds initial content for Rust and Go

docs/best-practice-memory-safe-by-default-languages.md

@@ -0,0 +1,27 @@
+# Best Practices - Memory-Safe By Default Languages
+
+While Memory-Safe by default languages are, by definition, memory safe by default, there are ways that the can be used unsafely (or less safely). Here are best developer practices for using these languages.
+
+## Python
+
+TO DO
+
+## Rust
+
+* Use unsafe blocks sparingly and follow careful practices - see [The Rustonomicon](https://doc.rust-lang.org/nomicon/intro.html)
+* When using unsafe code in Rust, the safety boundary is the module boundary. When examining snippets of unsafe code in Rust, you generally need to assess the entire module in which the unsafe code appears. [Source](https://github.com/ossf/Memory-Safety/issues/15#issuecomment-1847939439)
+* Avoid using the [no_mangle](https://github.com/rust-lang/rust/issues/28179) attribute
+
+## Go
+
+* Using the [unsafe package](https://pkg.go.dev/unsafe#pkg-overview) is sometimes necessary when using C from Go. Be careful (TO DO: Add more about how to be careful)
+* Use the [Go data race detector](https://go.dev/doc/articles/race_detector) to check for data race conditions
+* Follow the [Security Best Practices for Go Developers](https://go.dev/doc/security/best-practices)
+
+## .NET
+
+TO DO
+
+## Java
+
+TO DO