Running average?

[altf4]

I started writing a feature to toggle the display of a running average of the pings in the current log. However, it wasn't immediately clear to me how error pings should be handled.

Simply ignoring them doesn't feel accurate, as it would make the average higher than it should be, and probably defeat the purpose of displaying it. (Keeping an eye on your network quality)

But how do you numerically account for the missing ping in an average?

Perhaps a running average feature just doesn't make sense in this context.

[zyphlar]

Yeah my use case for this tool is "why is this sucking? Oh, because the
internet just died. Do I need to reconnect or something?" The graph is
there so you don't feel like you've gotta keep an eye on it
second-by-second. But as you mention, an average doesn't really speak to
quality issues. You could treat E as whatever the TTL is (1000 ms?) but
that's not as useful as "half our packets in the last minute have totally
failed."

Right now my most desired features are cross-compatibility with non-Ubuntu,
and an improved graph (using braille characters results in low information
density but I couldn't find a better option easily)
On Jan 5, 2014 12:39 AM, "AltF4" notifications@github.com wrote:

I started writing a feature to toggle the display of a running average of
the pings in the current log. However, it wasn't immediately clear to me
how error pings should be handled.

Simply ignoring them doesn't feel accurate, as it would make the average
higher than it should be, and probably defeat the purpose of displaying it.
(Keeping an eye on your network quality)

But how do you numerically account for the missing ping in an average?

Perhaps a running average feature just doesn't make sense in this context.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/3
.

[zyphlar]

@altf4 I just implemented a simple running average; it's a simple average so failures count as -1 which is wrong, but the number is still somewhat useful for now.