With Cosmos DB, Microsoft wants to build one database to rule them all
-
Techcrunch[^]:
This project (which at the time was called “Project Florence”) first turned into DocumentDB, Azure’s NoSQL database service, which launched in 2015 and is now morphing into Cosmos DB.
Because data wants to be stored as JSON graphs?
-
Techcrunch[^]:
This project (which at the time was called “Project Florence”) first turned into DocumentDB, Azure’s NoSQL database service, which launched in 2015 and is now morphing into Cosmos DB.
Because data wants to be stored as JSON graphs?
> No data is born relational Then the data is meaningless. Data only has meaning in relationship to other data. A better statement would be: Data is born with the potential of unknown relationships. That is what the power of a NoSQL and Graph DB is all about, putting data into relationship with things that you didn't anticipate. Marc
Latest Article - Merkle Trees Learning to code with python is like learning to swim with those little arm floaties. It gives you undeserved confidence and will eventually drown you. - DangerBunny Artificial intelligence is the only remedy for natural stupidity. - CDP1802
-
> No data is born relational Then the data is meaningless. Data only has meaning in relationship to other data. A better statement would be: Data is born with the potential of unknown relationships. That is what the power of a NoSQL and Graph DB is all about, putting data into relationship with things that you didn't anticipate. Marc
Latest Article - Merkle Trees Learning to code with python is like learning to swim with those little arm floaties. It gives you undeserved confidence and will eventually drown you. - DangerBunny Artificial intelligence is the only remedy for natural stupidity. - CDP1802
Data exist, understandings emerge (from relationships or aggregation functions or MapReduce) I guess NoSQL and Graph DB and so on allow us to delay the modelling rather than constraining the data by the up front designed model.
-
Data exist, understandings emerge (from relationships or aggregation functions or MapReduce) I guess NoSQL and Graph DB and so on allow us to delay the modelling rather than constraining the data by the up front designed model.
Duncan Edwards Jones wrote:
I guess NoSQL and Graph DB and so on allow us to delay the modelling rather than constraining the data by the up front designed model.
Having played with MongoDB and dynamic relationship building, it is easier, but you still need some kind of metadata to inform the application of new relationships, but that often involves more metadata that describes the semantics so that relationships can be created automatically based on the meaning of the data. So it really just pushes the problem around. The truly lazy programmer simply implement the relationships in code. :sigh: Which defeats the whole purpose, IMO. Oh, but wait, it's coded in scripted languages, so the argument can be made that the business logic is easily modified even at runtime. Contrast that with the truly skilled programmer that will either modify the SQL schema on the fly or code the relationships in metadata such that the relationships can be discovered in the schema (whether directly in the SQL schema or in some schema-metadata, either still using SQL or NoSQL) and the program can adapt dynamically without changing the code. :) Marc
Latest Article - Merkle Trees Learning to code with python is like learning to swim with those little arm floaties. It gives you undeserved confidence and will eventually drown you. - DangerBunny Artificial intelligence is the only remedy for natural stupidity. - CDP1802
-
Techcrunch[^]:
This project (which at the time was called “Project Florence”) first turned into DocumentDB, Azure’s NoSQL database service, which launched in 2015 and is now morphing into Cosmos DB.
Because data wants to be stored as JSON graphs?
So if they include it in their effluent interface, will it automatically resolve your tables to turd normal?
I wanna be a eunuchs developer! Pass me a bread knife!