feature, more variation when using random ranges

[paulcollett]

Currently when using a range between big numbers...

<div data-dummy="100,300"></div>

...the random amount of outputted words will usually land somewhere in the middle of that range.

Ideally we want to output high and lows more often to better test our layouts - which is the overall goal of dummyjs

[paulcollett]

One approach could be to reduce the possible outputted variations to 4 or 5.

Instead of:

<div data-dummy="100,300"></div> // 200 possible outcomes

5 possible outcomes:

<div data-dummy="100,300"></div> // 5 possible outcomes

100 // low
150 // low+(high-low)*0.25
200 // low+(high-low)*0.50
250 // low+(high-low)*0.75
300 // high

Or, use some kind of algo with preference for low + high + middle? Called "inverted bimodal distribution"? https://en.wikipedia.org/wiki/Multimodal_distribution