Hacker News Reader

Comment by ardit33 | Hacker News Reader

duh. Hibernate has been doing this for ages, before even rails existed. And it is a mess. You are trying to shoehorn a solution (ORM) so the database will fit the way you would like that data back.

How about actually having the database start evolving just like programing languages have?

View on HN

I'm sure that databases need to evolve to some extent, but is the relational-ness really the problem? When I start with a data model, I find that thinking in relations (and the various normal forms) is a very natural way to approach the problem. I also like the way that sql and relational databases mirror (to some extent) relational algebra - of course, I haven't written relational algebra since school, but it's actually interesting, and shows a solid mathematical foundation to rdbms systems.

What I hate (and I get the feeling you hate it too) is object-relational mapping. This is truly, exceptionally tedious stuff. Hibernate was no fun, but I find Rails to be far, far less of a hassle.

And there's plenty to object to in the various implementations of relational algebra (ie., the different RDBMs's). But maybe the programming language side is exactly where the innovation should be occuring.

"How about actually having the database start evolving just like programing languages have?"

What's stopping you from solving this problem? Sounds like a great entrepreneurial opportunity. Maybe you could submit a YC application.

Haha. Maybe next year. It has taken decades to reach to the RDMS that we have today. They have improved efficienncy by time, but the inovation has stoped.

Any new system, (OOD) will take years before reaches the maturity level to be used in commercial solutions. Kinda hard for a cash strapped startup.

Also, there have been attempts to build OODs but they have failed, mainly b/c they way ahead of their time.

What I would like to see a distributed Object Oriented Database system, that can scale at will.

I think Ellison and the Oracle guys consulted for years in order to eat while working on their relational technology until they thought they were close enough to go for it and drop the consulting and bet everything on the database as a product.

Maybe you could take a similar approach?

Check out Gemstone/S. A distributed, persistent and transactional Smalltalk image. It's flat out cool. As far as I know, they have even ported Seaside to run on top of it, which would make it the ne plus ultra for rich object model web app development. Just like every other odb or network db, the problems only surface when you need to query along paths you didn't design for up front (i.e. when you can no longer chase pointers). That's where rdbms' really shine. And when people not intimately familiar with your apps' object model need some of its data. Expecting some 'analyst' somewhere, to fire up a Smalltalk image, in order to create a report making his boss look good, is probably a bit much.

Reply on news.ycombinator.com