"irrelevant talking points"
The entire discussion about formatting is a good example. Start from the assumption, justified or not, that other portable recorders do not have an erase format, only a quick format.
I didn't realize that even the cheaper Tascam units have had Full / Erase format as a normal feature! My other portable recorders have been Sony, Marantz, and Zoom which do not.
So, if the assumption is true, then Tascam's implementation of the quick format was deficient. I have only seen a sporadic report of this type of problem with other recorders. If the root cause was failure to do an erase format, then those other recorders, which only have the quick option, should have the same problems. I think a large number of M10 users never remove the card and have been using the same one, in some cases, for years. Surely if the problem were to show up, you might expect it to show up in that case.
If the assumption isn't true, and other portable recorders do have an erase format, then Tascam's failure to initially include it, especially when their other recorders have it, was deficient.
Also, what tondeaf, voltronic and morst were discussing in the past few post is "statistically representative" (sampling) and not "statistically significant" (hypothesis testing). If the target population is "DR-70 users", then the survey is definitely not representative. Even if the target population is "tapers using the DR-70", it probably isn't representative due to known biases in online, self-reported surveys. Significance would require some sort of hypothesis to be tested and either accepted or rejected in favor of the null hypothesis (such as "tapers using the DR-70 experience these errors more frequently than DR-60 users"). The sample is large enough for hypothesis testing if the difference between the null and alternative is large enough.