doc: update readme to regard precision, open to updates

Author: YeungOnion

README.md

@@ -61,14 +61,16 @@ Thanks for your help to improve the project!
 **No contribution is too small and all contributions are valued.**
 
 Suggestions if you don't know where to start,
+- if you're an existing user, file an issue or feature request.
 - [documentation][docs-url] is a great place to start, as you'll be able to identify the value of existing documentation better than its authors.
 - tests are valuable in demonstrating correct behavior, you can review test coverage on the [CodeCov Report][codecov-url]
 - check out some of the issues marked [help wanted](https://github.com/statrs-dev/statrs/issues?q=is%3Aissue+is%3Aopen+label%3A%22help+wanted%22).
-- look at features in other tools you'd like to see in statrs
-  - Math.NET's
+- look at features you'd like to see in statrs
+  - Math.NET
     - [Distributions](https://github.com/mathnet/mathnet-numerics/tree/master/src/Numerics/Distributions)
     - [Statistics](https://github.com/mathnet/mathnet-numerics/tree/master/src/Numerics/Statistics)
   - scipy.stats
+  - KDE, see (issue #193)[https://github.com/statrs-dev/statrs/issues/193]
 
 ### How to contribute
 
@@ -102,24 +104,41 @@ gh pr create --head <your_branch> # with GitHub's cli
 
 Then submit a PR, preferably referencing the relevant issue, if it exists.
 
+### Precision for contributors
+Take a look at the (Precision section)[#precision] to get a sense of what we have built in. Also take a look at tests and various assertions.
+If you'd like to improve precision, please also make existing tests more precise, perhaps up to the crate's default set in the `crate::prec` module or better.
+
 ### Commit messages
 
 Please be explicit and and purposeful with commit messages.
 [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/#summary) encouraged.
 
-#### Bad
+### Communication Expectations
 
-```
-Modify test code
-```
+Please allow at least one week before pinging issues/pr's.
 
-#### Good
+
+## Precision
+Floating point is hard! We should try to not unintentionally misuse it. The below text is from the `approx` crate's README and contains useful links.
+> ## References
+> Floating point is hard! Thanks goes to these links for helping to make things a little easier to understand:
+>
+>    [Comparing Floating Point Numbers, 2012 Edition](https://randomascii.wordpress.com/2012/02/25/comparing-floating-point-numbers-2012-edition/)
+>    [The Floating Point Guide - Comparison](http://floating-point-gui.de/errors/comparison/)
+>    [What Every Computer Scientist Should Know About Floating-Point Arithmetic](https://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html)
+
+We have a set of approximation checks in the `prec` module that wrap those from the `approx` crate so that it's easier to have convey blanket precision expectations when appraising this crate. For granularity, one can search for where the options are set otherwise. If you use `rg`, you can run this below command from the crate root.
 
 ```
-test: Update statrs::distribution::Normal test_cdf
+$ rg --no-config '(?:abs_diff|relative|ulps)_eq\(.+?(?:epsilon|max_relative|ulps)\s?='
 ```
 
-### Communication Expectations
-
-Please allow at least one week before pinging issues/pr's.
+Below uses $\oplus,\ominus,\otimes$ for floating point binary operation rounded to float.
+The approximation macros are enumerated below, we also have `assert_*` variants, often used in testing.
+- `abs_diff_eq(x, y[, epsilon = DEFAULT_EPS])`
+  is $\textrm{float}|x \ominus y| < \epsilon$, which is `abs(x - y) < epsilon` and
+- `relative_eq(x, y[, epsilon = DEFAULT_EPS][, max_relative = DEFAULT_RELATIVE_ACC])`
+  is `abs_diff < cmp::max(fabs(x), fabs(y)) * max_relative` where `abs_diff >= epsilon`
+- `ulps_eq(x, y[, epsilon = DEFAULT_EPS][, max_ulps = DEFAULT_ULPS])`
+  "number of bits away" or "count of representible floats between".