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 On Saturday, August 26, 2017 at 7:58:02 PM UTC-4, l.garulli wrote: > > 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 <http://orientdb.com/> > > On 26 August 2017 at 22:30, <[email protected] <javascript:>> wrote: > >> 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 >> >> -- >> >> --- >> 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 [email protected] <javascript:>. >> For more options, visit https://groups.google.com/d/optout. >> > > -- --- 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 [email protected]. For more options, visit https://groups.google.com/d/optout.
