this kind of article quickly summons the; science is not repeatable claim, based on the misunderstanding of experiments as absolute serial duplications.

repeating the use of materials and methods is a procedure that generates a field of data suitable for statistical analysis, this is part of a good experimental design.

what really matters is the serial testing of a logical premise[hypothesis],and the independent verification of results. this results in a theoretical body of relationships[a theory], the goal of a research program.

when all the nuances and exceptions are knocked out of validity, you are left with the product of a famous razor, a general relationship that applies in all cases [sour fruit, young fruit, day of the week, was music playing etc.] and a group of parameters that Will effect different results at different values, a reduced model or paradigm with no exceptions [a law, or principle]

hunting down the exact tools, and materials to perform experiments that have already been done is of low value, variations of experiment that test the same logical premise are of high value.