Michael Kirk

This would be a good one to revisit in light of #1181 - it seems like we could soon get an implementation that works across metric spaces (Euclidean, Haversine, Geodesic,...

In case it's helpful, here's some notes when I faced some cross-platform scrolling abnormalities: > "In the beginning" a spinnable mouse wheel was the only input hardware for scrolling. Each...

This is really interesting and I think worth pursuing. My understanding is that you've done at least 2 big things to make things faster: 1. avoided allocating a bunch of...

I was thinking let's merge the string allocation improvement first, since it's strictly an improvement.

Would you mind rebasing and running against the new benchmarks now that #42 is merged? I think/hope they'll be a little more stable.

Is this LUT implementation something you're still interested in merging @mattiZed? Since we've, in the meanwhile merged a subset of the changes, the remaining work was some conflicts that need...

I'd like to see it merged if it's significantly faster. Now that the other perf improvements have been merged separately, we should be able to see if the LUT itself...

Addendum: ``` bench encode: 1000 coordinates time: [55.068 µs 55.138 µs 55.227 µs] change: [-0.7790% -0.4607% -0.1246%] (p = 0.00 < 0.05) Change within noise threshold. Found 9 outliers among...

I gave up on figuring out *why* they are so noisy and just simplified them a bit in: https://github.com/georust/polyline/issues/42

@mattiZed - as a sanity check, are you also seeing medium sized (+/- 10%) swings `in the bench HUGE polyline6 decoding` bench?