Add MIT license and package metadata; tidy README wording

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Oliver Bristow
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -2,16 +2,16 @@
 
 # alternative
 
-A tiny library for managing multiple implementations of the same function — especially when you're iterating toward faster or cleaner versions and want those choices to stay explicit, testable, and safe.
+A tiny, dependency-free library for managing multiple implementations of the same function — especially when you're iterating toward faster or cleaner versions and want those choices to stay explicit, testable, and safe.
 
 ## Why use this?
 
 When optimizing a hot path, it’s common to accumulate:
 
-- a trusted baseline implementation,
-- one or more candidate rewrites,
-- tests to keep them equivalent,
-- and benchmarks to validate wins.
+- a trusted reference implementation
+- one or more candidate rewrites
+- tests to keep them equivalent
+- benchmarks to validate wins
 
 `alternative` keeps that workflow tidy by making implementation registration and selection first-class.
 
@@ -59,7 +59,7 @@ Use `pytest_parametrize_pairs(...)` to compare the reference against each candid
 Use `pytest_parametrize(...)` to run one test body across all implementations.
 
 - Great for benchmark workflows with [`pytest-benchmark`](https://pypi.org/project/pytest-benchmark/): [`examples/test_benchmark.py`](examples/test_benchmark.py)
-- Also useful for validating that every implementation passes one shared test suite.
+- Useful for validating that every implementation passes one shared test suite
 
 ## Safety guarantees
 

pyproject.toml

@@ -7,7 +7,11 @@ authors = [
     { name = "Oliver Bristow", email = "evilumbrella+github@gmail.com" }
 ]
 requires-python = ">=3.12"
+license = "MIT"
 dependencies = []
+classifiers = [
+    "License :: OSI Approved :: MIT License",
+]
 
 [project.optional-dependencies]
 dev = [