I asked a question - when JSON in SQL Server could be constrained by collection of JSON schemas in the same fashion how Microsoft implemented a typed XML data in...

Why ADO.NET on client side - SQL Server has OOB rich metadata functionality in information views

My sincere apologies if I offended an author of the article. 
IMHO the fact that you can do something it does not mean that you have to do it. 
And...

Complete LUNACY. 
I highly recommend you to read this first https://www.microsoft.com/en-us/research/wp-content/uploads/2006/04/tr-2006-45.pdf
Are you aware of  FILESTREAM capabilities (since SQL Server 2008).This IMHO would be much better approach than store BLOBs...

I agree that adding a GUID to a set of data that already contains a primary key does not make ANY sense (unless it used for replication) 😎

bigint datatype occupies 8 bytes of storage

uniqueidentifier - twice more 16 bytes,

Therefore for each 1 million of records you got an overhead of extra 0.763 Mb for additional...

Guys what is wrong with Auto-incremental identity that exists for ages! Or with Sequence originally existed in Oracle and now ported to SQL Server. Why the itchy need to reinvent...

I agree - an ART for sake of ART because Web designers would like to generate identity in JavaScript (!)

Plus Microsoft table designer in SSMS does not...

My motto - DO NOT use GUIDs as a primary key and do not use them at all (unless they required for replication/synchronisation scenarios between 2 OCCASIONALLY CONNECTED databases). Did...

Clients are ALWAYS right - as long as they are paying!

Manie, we all had our "CLIENT IS ALWAYS RIGHT" moments, even when client was 100% absolutely wrong.

How about user request to implement 5 different data interfaces for 5 different...

Alex,

Thank you for a very interesting article. 😎

Please note that the size of value of the extended property (data type sql_variant, with a default of NULL) cannot be more than 7,500 bytes.

Even with such limitation you...