Well, if that isn’t good news for any researcher in our domain. There will be a journal of interesting negative results in NLP and ML. Hooray, that’s what I always waited for. Now I can publish even all the stuff that did not help improving error rates. đ