Tom Rokicki wrote:
Generally the results from Al Zimmermann's contests tend to be empirical and not definitive, so don't mesh well with the OEIS sequences. It would be really nice if there was a general repository of "best results" for problems such as these that people could contribute to, rather than having random pages collected by random people all over the web on various problems.
Maybe an Online Encyclopedia of Interval Sequences, which can include lower and upper bounds if exact values are unknown? Then, you can see whether two sequences ([a_n, b_n]) and ([c_n, d_n]) are plausibly the same by checking that a_n \leq d_n and c_n \leq b_n for all n. There is, of course, no requirement that the terms be integers. So you could look up the zeta function by evaluating upper and lower bounds for: \sum_i(1 / i^n) for a few values of n and entering the information into the search bar. At no point do you ever need to know that this is called the Riemann zeta function, or even know the formula (you may have found these values by Monte Carlo testing whether n random integers are coprime!). Best wishes, Adam P. Goucher