Scott,
I wrote:
So, I can imagine a few factors that may be relevant:
Connection performance (I have a 300 MBPS cable connection)
I was thinking, and it isn't the raw bits-per-second rate that likely matters. Rather the round-trip time for a message to sent from your browser to the site, the site to process it, and the response to come back to your browser.
This unfortunately is seldom measured or revealed. And can be hard to measure because it is usually even more variable than the effective bit rate. DSLReports includes a "bufferbloat" measurement, which is a measure of how badly your average trip-time is impacted by streaming data. Often the answer is: severely.
The also have a ping test which is a direct measurement of latency in your data connections:
The quality of your internet service matters, of course, but another factor, often overlooked, is the age and quality of your home router. Not too long ago I replaced an aging DSL-class router with one geared for higher transaction performance - it made a world of difference on a bunch of sites, particularly modern sites that tend to cram a lot of stuff onto each page.
I don't think Groups.io's pages are overstuffed (far from it) but the HTML editor could be particularly sensitive to transaction latency (round-trip time) if it is communicating with Groups.io's server on every keystroke.
Shal
Shal Farley