Are rethorical programming questions allowed?
-
Are non-English words such as 'rethorical' allowed in questions on the discussion board, or are these allowed when the language being used is stated clearly in the question?
I would expect that the writer is referring to a SQL design paradigm called "thor". After all, we have "COM" and "comical." "Rethorical" suggests an attempt to normalize thor relationships. The practices guide offers the image of smashing tables under a heavy blockish object tied by leather straps to a handle. Among the side effects are static discharges.
-
I don't really expect a serious answer to this. I just found a few database tables which obviously had been designed by that Russian who's liver I already would have liked to be served (fried, with onions). Now I start to believe that he is actually from Eta Carinae and I'm doing my best to understand alien database design. If I have two tables A and B and a relation table inbetween, that must be a many to many relation. Fine. Nothing unusual. But what do you call it when there are two relation tables in a row between A and B? A many to even more relation? m to m^2? And that's only the beginning! How about a relation table that has three relations to yet more relation tables , which have a relation to tables A, B and C. What's that? A multible m to n^3 relation? Or an omnidirectional star relation to the third degree? The Romulan relation nova?
I have lived with several Zen masters - all of them were cats.
-
I don't really expect a serious answer to this. I just found a few database tables which obviously had been designed by that Russian who's liver I already would have liked to be served (fried, with onions). Now I start to believe that he is actually from Eta Carinae and I'm doing my best to understand alien database design. If I have two tables A and B and a relation table inbetween, that must be a many to many relation. Fine. Nothing unusual. But what do you call it when there are two relation tables in a row between A and B? A many to even more relation? m to m^2? And that's only the beginning! How about a relation table that has three relations to yet more relation tables , which have a relation to tables A, B and C. What's that? A multible m to n^3 relation? Or an omnidirectional star relation to the third degree? The Romulan relation nova?
I have lived with several Zen masters - all of them were cats.
"Intersection" data. Intersection data is data that is (only) relevant in the context of a "relation"; e.g. is a "member" of x; or "has possession of" y; etc. This can create multiple relations between the same entities; that "may" result in multiple physical tables. They may also be used to implement "recursion" in 2 tables; again the "relation" table may have intersection data to make sense of the recursion (as per IBM's "virtual logical child").
"(I) am amazed to see myself here rather than there ... now rather than then". ― Blaise Pascal
-
I don't really expect a serious answer to this. I just found a few database tables which obviously had been designed by that Russian who's liver I already would have liked to be served (fried, with onions). Now I start to believe that he is actually from Eta Carinae and I'm doing my best to understand alien database design. If I have two tables A and B and a relation table inbetween, that must be a many to many relation. Fine. Nothing unusual. But what do you call it when there are two relation tables in a row between A and B? A many to even more relation? m to m^2? And that's only the beginning! How about a relation table that has three relations to yet more relation tables , which have a relation to tables A, B and C. What's that? A multible m to n^3 relation? Or an omnidirectional star relation to the third degree? The Romulan relation nova?
I have lived with several Zen masters - all of them were cats.
CodeWraith wrote:
But what do you call it when there are two relation tables in a row between A and B? A many to even more relation? m to m^2?
Nothing stops two entities from having more than one 'type' of relationship between the the two. But to be fair that can be handled by additional attribute(s) on the first table also. That said, for the example provided, the only cases I have seen of that is where a developer added the second table because they didn't know the first existed.
CodeWraith wrote:
And that's only the beginning! How about a relation table that has three relations to yet more relation tables , which have a relation to tables A, B and C. What's that?
Can't speak to the entire world of all possibilities but normally I would deem that 'someone who didn't know how to do data modeling and certainly didn't know how to map it to a database.'
-
CodeWraith wrote:
But what do you call it when there are two relation tables in a row between A and B? A many to even more relation? m to m^2?
Nothing stops two entities from having more than one 'type' of relationship between the the two. But to be fair that can be handled by additional attribute(s) on the first table also. That said, for the example provided, the only cases I have seen of that is where a developer added the second table because they didn't know the first existed.
CodeWraith wrote:
And that's only the beginning! How about a relation table that has three relations to yet more relation tables , which have a relation to tables A, B and C. What's that?
Can't speak to the entire world of all possibilities but normally I would deem that 'someone who didn't know how to do data modeling and certainly didn't know how to map it to a database.'
I have heard from the boss that they kept making changes every day, sometimes even very contradictary ones. It's very possible that the data model started out orderly, but then degenerated because nobody knew what's wrong or right anymore.
I have lived with several Zen masters - all of them were cats.
-
I have heard from the boss that they kept making changes every day, sometimes even very contradictary ones. It's very possible that the data model started out orderly, but then degenerated because nobody knew what's wrong or right anymore.
I have lived with several Zen masters - all of them were cats.
That doesn't sound like a 'reasonable' explanation to me. As a developer I am not just going to create arbitrary and contradictory entities in the database just because some upstream requirements are unclear. And I believe that myself I am fairly good about guessing about future directions from fuzzy requirements. I have known other developers that just throw up their hands and refuse to do nothing. However if there were multiple developers making multiple updates and not coordinating or even reviewing each others changes that would cause problems. And it would also reflect poor management as well (which I am sure management would not like to hear.)