SPRING_SLEUTH_SAMPLER_PERCENTAGE is not a percentage

[cer]

Isn't a percentage between 0 and 100%?

[marcingrzejszczak]

Since percents as their name suggests are values divided by 100 you can also represent them as a double ( 80% == 80/100 == 0.8). We've decided to go with the double representation :)

But English is not my native language so maybe we've used a wrong name for the property. @adriancole WDYT?

[codefromthecrypt]

Maybe threshold is a better term for a 0-1 function?

On 11 Sep 2016 14:17, "Marcin Grzejszczak" notifications@github.com wrote:

Since percents as their name suggests are values divided by 100 you can
also represent them as a double ( 80% == 80/100 == 0.8). We've decided to
go with the double representation :)

But English is not my native language so maybe we've used a wrong name for
the property. @adriancole https://github.com/adriancole WDYT?

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#397 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AAD619GAqJ2rqDDaOTS9_bpi1KgwZ-vHks5qo5z5gaJpZM4J52tD
.

[cer]

I think threshold means something else.
Mathematically speaking this value is a probability, which is a number between 0 and 1

[codefromthecrypt]

SGTM! thanks, chris

On Mon, Sep 12, 2016 at 11:22 AM, Chris Richardson <notifications@github.com

wrote:

I think threshold means something else.
Mathematically speaking this value is a probability
https://en.wikipedia.org/wiki/Probability, which is a number between 0
and 1

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#397 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AAD615fqtAsvCpGb_GfqjZ5bVBiAvtYlks5qpMVigaJpZM4J52tD
.

[marcingrzejszczak]

This is a breaking change actually (which I don't like to do). We could allow people to provide entries greater > 0. You could then provide

SPRING_SLEUTH_SAMPLER_PERCENTAGE=0.5

or

SPRING_SLEUTH_SAMPLER_PERCENTAGE=50

and that would be the same. I could also provide a new property for version 1.1 and deprecate the old one (but not remove it). I could remove it for 1.2 or sth like that. WDYT?

[marcingrzejszczak]

ping @cer ?

[cer]

SGTM

[dsyer]

What if you actually wanted 0.5% (not completely unlikely)? I think it might be better that we just live with it and document the mistake. Or else introduce a new notation with "%" in it or something (seems like overkill for such a simple thing).