Longer history than 30 days would be excellent for troubleshooting performance issues.
There should also be at least some aggregate data that is permanent. Last 6 months has daily summary which is nice... but then can't you give monthly stats going back at least another 6 so we can see at least a year back (should keep all historical data in my opinion). Just something as simple as Jan 2018: No downtime, Dec 2017: 12min downtime, etc.
Hello, raw access is already possible using the API: https://updown.io/api
Using the /metrics endpoint you can get the apdex per host or per hour for any timeframe you want (minimum resolution is hour though). History is kept forever but aggregated, so you get per-hour data for at 1 day, then per day data for a month and then per month data.
Ruth Ivimey-Cook commented
Pretty please? I would love to get the raw access time data for bin size of e.g. 5 to 10 minutes :-)
If the data storage is an issue perhaps the API could be extended to permit download of the last day's times, so we can pull the data and display in a service such as plot.ly, though of course having it integrated would be nicest.
Michel Helms commented
Maybe history/graph of access times per location would be nice so I can see how it improved over the months