why so much hate for orientDB

[ad...@cs.miami.edu]

There seems to be lots of hates towards OrientDB on the web: leaks, rants, etc, etc.  There seems to be more of this for OrientDB than other DBs (even the often hated MySQL).  Why do you think that is?  Do you think there is merit in the hate or do you think it is mostly  people using the database incorrectly (for example not understanding how the JVM works and then thowing one's hands in the error when they get a heap error).

Best,

-Adam

[Manabu Tokunaga]

I am also interested to know why.

When I present the idea of Graph DB to my team, the general reaction seemed to be stemming from the fear of "unknown" especially of the Graph nature of this database. 

It seems most people are very set on MySQL and Document DB (esp. Mongo or Casandra) for Linux based server dev and not much further.

But maybe there are other reasons.... I do want to know.

Manabu Tokunaga   BIO

CEO

tel/txt: (650) 762-8413 email . hangout: man...@wingumd.com

--

---
You received this message because you are subscribed to the Google Groups "OrientDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email to orient-database+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Luca Garulli]

Hi Adam,

Thanks for opening this topic :-)

Most of the leaks, rants, etc. come from the same "anonymous" source. Of course we know who's behind that: he supposed to collaborate with the OrientDB team long time ago, but then we had a divergence of opinions and that was the best way he thought he could has his revenge...

There is a lot of passion behind Technology, Open Source, Communities. Passion can be good and bad. OrientDB is not the big corporation to hate. We're a small team, 99% of the people are developers, that enjoyed working to the revolutionary idea of the Multi-Model database with a Native Graph Database engine. In 2010 most of the people thought we were crazy: providing SQL to a NoSQL database, mixing all models in one. But now everybody is following our vision and MongoDB and DataStax (Cassandra) already announced to be multi-model. So we were visionary long time ago.

That's said, we knew that OrientDB wasn't the most stable DBMS around. We preferred playing with the Multi-Model concept instead of making our DBMS rock solid and then having hard time to upgrade that concept. This was fundamental to quickly build up this Multi-Model idea and experiment with it. We thought that stability would come later, as soon as the model would be final. And we were right. 

Even if we still have users that are running OrientDB v0.9.8 inproduction, It's only with OrientDB 2.2.x that we reached the maturity and stability demanded by our users and clients.

Now, we still have users that are not happy with OrientDB. The biggest mistake we see our users doing is using OrientDB as a RDBMS. We've seen so many users trying to do JOINs with OrientDB. Of course it works, but it's the worst way to use a Native Graph Database. Unfortunately most of these users ask for help one week before going in production and at that point it's impossible to help them without a complete redesign of their model.

But we see also users using OrientDB at the best, without asking one single question on Stackoverflow.

Is OrientDB perfect? Of course not. There is still a lot of work to do. We're working hard to make our DBMS unbreakable, especially in distributed configuration. We already decreased the number of bugs far below the average of other Open Source DBMSs. We've simplified the API in v3.0 (still in beta), we largely increased the number of test cases, we hired QA engineers to find issues before users do and much more.

Today we have many of the Fortune500 companies in production with OrientDB, with so different use cases. This is the beauty of the Multi-Model, you can do so many things with it.

Last point, not strictly technical, is how we run the business around OrientDB. We decided to bootstrap without getting any money from VCs. Our investors are our clients. We believe this is the right model for a software company with a product that must last for years or decades. I've seen NoSQL companies funded with tons of million of dollars not being able to reach the profitability even after many years (MongoDB, Neo4j, etc). Some of these are already failed (RethinkDB, Basho, etc.).

I know that VCs can put a lot of pressure when you can't make a profit after many years. We aren't in this position and we can take all the choices we believe are good for the product, not just to make the VC happy until the next round of funding...

I think this is the main reason why some of our competitors behave unfairly with us by feeding the "anonymous" source mentioned above.

Of course there are also real users that aren't happy with OrientDB for a lot of reasons. Even if a Multi-Model can be a good fit for most of the use cases, maybe it didn't work out for them for multiple reasons. We love constructive feedback, because everytime we are able to understand what didn't work out for a user, we can work to make OrientDB a better product.

Best Regards,

Luca Garulli

Founder & CEO

OrientDB LTD

--

[ad...@cs.miami.edu]

Luca,

Thanks for taking the time to respond.  I think other groups might have considered my post as troll bait, but it was an honest question and you hit it straight on.

I am looking for a JVM embedded document database that could replace mongodb.  I think my usage of the graph features will come later.  I will certainly be giving it a good spin in the near feature, and I am rooting for you all to reach your goals with both product and company.  Perhaps you should add a piece of code that flushes to  stdout harsh insults if the user puts a JOIN command in their sql (perhaps a bit harder to detect if they do an application join, which I am frequently guilty of).

Best,

-Adam

To unsubscribe from this group and stop receiving emails from it, send an email to orient-databa...@googlegroups.com.