cases even nowadays where "amateurs" find good solutions.

Good point. Even Numberphile and related YouTube channels have highlighted the answer-hidden-in-the-comments on superpermutations (4)(3)(2)(1). And as @standupmaths likes recommending, give it a go, and make a "Parker Square" (5)(6) of it.

Since the number of 11 step solutions becomes infinite by just adding more 1s it's probably not that unlikely to find a solution with several hundreds or thousands digits.*

*: given enough computing power. :-) I mean, adding one 1 will multiply the number of permutations by 16, and from there it will keep on increasing by factors of (15+i)/i . (edit: remove stray factorial, which belied my statement)

You'd need to calculate the density of possible products of single digits numbers in a number range.(easily done with the sieve approach)

...

Otherwise you'd need to prove why it's impossible. ( Which could be done by showing that the density becomes 0)

Hmm. I'll have to think about this some more. I can imagine how I'd start generating the potentials -- 2**$i * 3**$j * 7**$k -- but without generating all the potentials below a threshold, I cannot see any way to generate just potentials of say 200-300 digits. (for any, using the iterators from Higher Order Perl -- section 6.4 especially -- would be how I would start... and brings this reply back into being remotely perl-related :-) ).

Since the number of 11 step solutions becomes infinite by just adding more 1s

I'm also wondering whether there are other base n=11 solutions that aren't permutations of 277777788888899 and however many 1s you insert. A003001 only lists 277777788888899 (since that's the only minimal one), and the other sites I've seen only list that and some of its permutations -- unless I've just not noticed one that looks similar enough to that one that I didn't notice it was separate. OEIS has Numbers with multiplicative persistence value lists for 1-9, but doesn't list the persistence 10 or 11 numbers. But A003001 has a table that (if I understand it correctly) implies there's a second family of persistence=11.

edit 2: found it. A046150 shows 99999999998777772, which would be the family 27777789999999999.

I think the problem with many amateurs is not that they don't have good ideas but that they don't know how to read the surrounding literature.

They either tend to reinvent the wheel or even worse try to invent things which are proven not to exist.

I know about professors which get letters with "revolutionary" solutions for popular problems every week.

The spamming has an extent that some revolutionary things were only published in unknown publications and had to be discovered there.

Btw I still consider me an amateur here, if I was seriously interested to solve it I'd spend a weekly only searching publications.

Cheers Rolf
_{(addicted to the Perl Programming Language :)
Wikisyntax for the Monastery FootballPerl is like chess, only without the dice}

It has been interesting to read the discussion back and forth following my post. I do not have a degree in anything and failed abysmally at high school pure mathematics (applied mathematics I was good at as I could see a real-world use for what I was trying to learn) so my approach to the problem was purely empirical. I can't even qualify as an amateur mathematician as attempts to read the literature tell me that I don't understand the language being employed, I just hear a whoosh as it goes over my head.

However, I did very much enjoy writing the code and playing around with ideas so I'm grateful to tobyink for starting the thread!

Cheers,

JohnGG