Just to clarify, I'm not trying to defend anything. Just stating a fact.
I've seen the same thing with experienced SQL Server developers who come into an Oracle-based project: they gripe about how Oracle is "weird" or "doesn't do what I expect" and then they eventually accept reality and get on with the project.
I honestly think it's a very informative article, clearly SQL Server doesn't adhere to the standard (whichever one you choose) but I'm not aware of any RDBMS that does so I was curious as to whether there was one.
There's lots of amazing databases around, each has their own quirks and as the gp says it's generally difficult to drop in replacements due to those quirks. There's a few listed in the article for SQL Server I'm sure anyone could easily find where others deviate from the spec too. But anyway getting back to what you said:
> He is giving examples of violating the standard
That's absolutely true and it's something every other RDBMS does.
> That's absolutely true and it's something every other RDBMS does.
To various degrees, and there's violating the standard (by adding non-standard construct) and violating the standard (by failing to implement a feature entirely, or implementing it with different syntax or semantics).
http://troels.arvin.dk/db/rdbms/ is generally my go-to page for that stuff as it lists standard solutions (with reference and mention of the features used being optional if applicable) and alternative db-specific solutions (exclusive or inclusive of the database accepting the standard solution). Sadly it is not really up to date, and obviously does not go through the whole standard.
He is giving examples of violating the standard.