I'm not sure what editor is getting at but thinking about what the distribution of the test statistic looks like under the null and thinking about what it looks like under the alternative sheds light on this. If you're making a test you want it to reject the null when the observed data matches more with the idea that the alternative is true and you want it to fail to reject the null when the null is true. Compare where the test statistic is likely to fall in both of those situations and it might make sense why we do things this way.
First of all thank you so much for giving me the answer,
As far as I understand from your answer we select 0.05 area for type 1 error,if test statistic get any extreme value i.e falls in rejection region ,we assume that its due to type 1 error and retain Ho