@MaxVernon Thanks a lot for your kind words. I was involved in several time-consuming activities during the last year, but I'm planning on getting more involved in the near future as things become less hectic over here
2
Is there anything I can do to help in solving the top tags issue? I'd be glad to take part if possible
@Vérace for sure they must, otherwise durability would be tricky. Perhaps disk-oriented vs memory-oriented would be better classifiers? That said, I've read opinion that with sufficient replicas and geographic dispersal a purely RAM resident solution gives the same surety as disk-based system. Don't think I'll be trying it, though.
@MichaelGreen Maybe have a Kinsey type scale from 0-5 (1 - exclusively disk --> 5 exclusively memory). You could have D-oriented (Dame...?) or M-oriented (Mr...?) and in dba circles one could ask "Are you M- or D-"? I don't think that marketing departments would be very keen on "Acually, I'm Bi-..." - sorry for the purple patch... chomping at the bit to get out for walk...
if a data presented with 4-dimension in which each dimension is dependent to hierarchical 3-level aggregate like (country, city, street), then we can summarize it into 4096 ways !
we know for cube with n dimension in which none of dimension is
hierarchical we have 2^n summarizing ways, but in th...
@Vérace yah depends on whether or not they edit tags on a question where they answered, and when they performed the edit. "Top Tags" are updated once per day, and may be based on recent contributions to the site.
As each combination would need to contain unique elements (disregarding order), for a four-dimension array we would have: (), (A,B,C,D), (A), (B), (C), (D), (A,B), (A,C), (A,D), (B,C), (B,D), (C,D), (A,B,C), (A,B,D), (A,C,D), (B,C,D) which is 16
I haven't figured out the combinatorics yet, but it would seem at face value to be off
It's possible they're doing super wacky rollups say the rollup is A, B, C, the cube is D, E, F, G, then you could do bonkers ((A,B)D,A(G)) but I'm not sure that makes any sense?
If that's the case then I think I see how this might permutate
If you have a single hierarchichal triple-dimension (which itself internally is orthogonal), then there are 8 ways to ask to slice it - that's given in the diagram. And then you have 4 of those, so it's 8^4.
If instead you treat it as 12 dimensions (which it is), then there are sum_(n=0)^12 binomial(12, n) = 4096 way
we know for cube with n dimension in which none of dimension is hierarchical we have 2^n summarizing ways, but in this case that each dimension has hierarchical 3-level aggregate have 4^4=256 ways.
That's right too, because all the dimensions are independent - even the "hierarchical" dimensions.
The binary power is just choosing which to include or not to include, so I guess that makes sense. Maybe Wolfram ALpha can just prove they are always the same.
@bbaird Yeah, when I first looked at it, I was like, why do they offer (month, week) and (year, week) - that only makes sense if week is week of the year and then you got calendar year issues within months (assuming you haven't adopted some calendar like the Internation Fixed Calendar/Eastman/Cotsworth)
I have to switch brains and go to meeting. I will check back in and answer the question on SO if it hasn't already been closed/answered.
I ran into someone at a SQL Saturday that worked for an insurance company that put all the policies into... Access databases. Of course, they would max out, so each year was its own Access database and... uh, like, we do things terrible but NOT THAT terrible.
@JohnK.N. Seems like it might have been a good place to post musings (database related) that aren't directly relevant to the site (meta) or a question - or a longer treatise on a subject already discussed/answered? It appears to be dead/mort/muerto/marbh/holt/mrtav - pity... maybe in need of ressurection?
at the time of shutdown, our blog site hadn't been updated in over 5 years.
also, if anyone is interested in writing a blog post and doesn't have a blog setup, I could probably be convinced to add an author or two to mine at sqlserverscience.com