blogs.computerworld.com — The "NoSQL movement" and Cloud based data stores are striving to completely remove developers from a reliance on the SQL language and relational databases. Some developers think this is something new, but object databases appeared in the 1980s and Ray Ozzie started the commercial document-centric datastore business with LotusNotes.
Feb 3, 2010 View in Crawl 4
kalvinbFeb 3, 2010
Company trying to sell new product claims old product is obsolete.In other news, water is wet.Relational databases will be obsolete when there is no more relational data in the world. So...never.There are many attempts to bridge relational databases into the OO world. Linq is the latest attempt by MS. It's slow, majors in minors and is one more obnoxious layer in the way of just getting things done efficiently and effectively. The amount of time I spend writing objects to talk to the database in an OO friendly way is the least of what I do to implement a project.
wilhoitmFeb 3, 2010
I am not sure if LotusNotes is an good example to mention! /s
Closed AccountFeb 4, 2010
I'm huge on the NoSQL bandwagon. I think a lot of the movement is related to the recession. Programmers have ALWAYS hated SQL. And DBAs love it. With the recession a lot of those jobs were merged into one. So programmers, who prefer to do the work on the application side anyways, just found a way to turn databases in big hash tables.
bradleylandFeb 6, 2010
If there's one thing we all should have learned about programming languages right now, it is that they rarely ever die. This is especially true for languages whose utility and usefulness are embedded in nearly everything we build. Outside of C, I can't think of a language with more reach and influence than SQL.That's not to say SQL is the only, or best, way to accomplish today's tasks. When you think about it, ORM is just a way of abstracting away the SQL, because constructing SQL statements for every conceivable interaction is tedious and unnecessary. SQL is the "byte code" of an ORM stack.When you get down to brass tacks, SQL is no more tied to the underlying tables and data than any particular programming language is tied to actual machine code that runs inside a CPU. Just as programs are traverse down through layers of complexity to the physical structures in a CPU, SQL traverses down toward the pushing/pulling of bits to persistent storage. SQL is (currently) the most common language used to express and access the data we store in a database. If someone felt so inclined, they could write a database that understood a language that was far closer to Java/Hibernate or Rails/ActiveRecord, just to pick two.The question is, what do we gain? Large SQL queries can be difficult to decipher, but just like large math formulas, the concepts they're based on are simple, so we can rely on machines to do the dirty work of constructing the actual statements. In this way, SQL acts are an intermediary, allowing other languages the flexibility to grow and change. If anything SQL has shown its greatness in its ability to adapt to a rapidly changing programming landscape. I'm not sure we gain a whole lot by moving newer ORM concepts closer to the actual data.
bingopowerFeb 15, 2010
...and how does it affect your customers when your internet goes down due to a local telecoms engineer cutting the wrong cable.
philathea80Feb 24, 2010
I just installed Cassandra db and Thrift last night. I can't wait to try lazyboy, a python API written by a guy at DIGG.
crgwbrJun 9, 2010
That's because Digg has enough users/traffic/data that it actually needs the performance benefits. But for 90%+ of the world's databases, relational SQL works just fine.