...and it's not one they would have built.

Preprint je tu https://www.medrxiv.org/content/10.1101/2020.08.03.20167791v1 a koment od buddinggenetics je fascinujuca ukazka vsetkeho, co je zle s reviews.

Uz zaciatok je mocny: strawman, indikovanie, ze clanok je 'misleading [to] the public and/or scientific community', a staznost, ze ' [... is] low, which is a relative term,'
- Nie, "The principal author has stated in media that the cost per test is $10", asi nie je pravda - co povedal je "we expect that labs will only charge about $10 per sample", kym v clanku "Total minimum reagent cost per sample $1.29-$4.37" Mne osobne pride celkom jasne co presne to hovori a nijako zvlast misleaded sa necitim. (Btw. co alebo kto tu je "the principal author"? wtfmj)
- Neviem kolko clankov buddinggenetics napisal bez pouzitia akychkolvek 'relative terms', ale myslim si, ze vela nie - nie sme totiz retardovani skolkari, s istou mierou neurcitosti sa vieme vysporiadat tam, kde extremna presnost nie je nutna, a celkovo si rozumieme. (Odporucam spocitat 'relative terms' v citatelovom oblubenom clanku.)

Avsak skutocny kamen urazu je tato cast 'recenzie': "[...] there is insufficient evidence to show eliminating the N2 primer set is justified. There needs to be an analysis of how many inconclusive test results (N1 positive and N2 negative/ N1 negative and N2 positive) would now become positive or negative tests as a result of eliminating the N2 primer set." (Bold zvyraznenie je moje.)
Toto je extremne skodlivy pohlad ktory je absolutne nelegitimny a v review nema co robit. Ak je to naschval, tak sa da iba hadat ci to je zakernost, namyslenost, alebo nejake pokrivene vnimanie sveta. - Nie, there needs not to be such an analysis, buddinggenetics by mozno chcel, aby tam taka bola, alebo sa mu nepaci, ze tam nie je, ale to je tak vsetko. Autori hovoria ze nejaky pristup funguje, jedine, co 'needs to be' je argument, ze to funguje tak dobre, ako tvrdia, ze funguje, a v ziadom pripade nie su povinni vysvetlovat, preco netestovali nieco ine alebo ci by nieco ine fungovalo.
Kazdy recenzent, ktory stoji za deravy gros by si mal uvedomovat, ze kazda poziadavka na dalsi material v clanku musi mat sakra jasne uvedene, ci to je 'nice to have', 'adding would improve the paper' , 'might be interesting to consider', alebo 'needs'. A legitimne dovody na 'need' su asi iba dva - ked argument potvrdzujuci klucovy zaver clanku nie je dostatocne silny, alebo ked nie je celkovy prinos clanku dostatocny na dane publikacne medium, ale v druhom pripade to je tiez iba 'needs [to be published here]'.

Cele to je o to komickejsie, ze na tom clanku je co kritizovat - '94% agreement in testing outcomes', (parcialne) p < 0.001, a sample size 67? lol. nehovoriac o tom patetickom promo cez nba.