I'm not gona leave a long rant about Raven 3.0, however the new version looks "tasty", it's streamlined, fast, and the html based management UI, make me want to find probems for this new solution :)

Anyhway with 3.0 Raven has past from "I could use it" to "must use it" for any future projects that are a good fit.

The build up has been great and I'm really looking forward to using it. I really hope the upgrade path is painless.

Cant wait to try it on RavenHQ ... been waiting since early summer...

My thoughts? Well, you've been drawing a big circle around the problem in your multiple blog posts complaining about how there's so much to write about: You tried to do too much in one version.

Ade's comment above highlights another problem with taking years to release a product: Many of us are going to be very concerned about upgrade paths with our existing systems and it will slow adoption on production systems.

These problems are best avoided by releasing more frequently and giving each release a "theme" that we can wrap our heads around. How many features are you shipping in RavenDB 3.0 that have been finished for months that we all could have been using already (and providing feedback on)?

I'm very excited for 3.0. I have enjoyed the feature release posts over the last two weeks—it's like a long Christmas for us. 3.0 is supposed to fix almost all of the problems we've had with 2.5.

Indexing, stats, new studio, better resource usage strategies... it is all stuff we want to put into production as soon as it is stable.

The only thing I don't see in 3.0 that we would like one day is the ability for a Raven-managed system for shard rebalancing instead of having to do it manually document by document. Though overall the sharding story is very good even in 2.5. We had a week to fix a performance problem on a system with a lot of incremental modifications of large documents with a handful of large indexes that fan out a lot. (I know, not an ideal model, but it works surprisingly well considering.) 3 of us managed to implement sharding in 4 days on a very big application. It solved our performance problem beautifully. (With the exception of one of the ops guys making a mistake when manually moving documents to the wrong shards because Raven didn't have a system to do it for us based on the Sharding strategy we wrote.)

The book is looking great too (the version released a few weeks ago but still isn't finished of course), and is very helpful. It needs a little proofreading like any draft work does. But the core information and direction of the book is on the right track, IMO.

well, pretty interesting stuff. am wondering if you can write -when you can- posts that covers CQRS/ES with RavenDB. if not, those topics will be covered in the book you're writing ? many thanks Ayende, Great Work really.

Personally I think that if you're going to do big things like introduce a new storage engine, completely rewrite the Studio UI etc. then these things take a while if you're going to do them properly. I would prefer to give the team the time and space to do great things and make sure the release is feature rich and polished. If the release are too frequent, this in itself creates risk and effort for us because we have to go through a cycle of QA etc. if we want to use the new functionality, I'm happy with the way this has been done in one big hit.

There are many things that make this a big release for us too and we're looking forward to getting stuck in as soon as our own busy roadmap allows!

What I like the most is the improvements to the ops side and the studio. Still curious to the pricing changes and specially if I can use custom bundles with the basic license.

I am excited about RavenDB 3.0 and my only reservation is the fact I have been moving away from the .NET world and more in the node.js world and I am not sure the node.js clients that currently exist will be compatible with 3.0 or even up to par with the new features... I wish there would have been some work done on additional clients other than java.

Great work though no less. Can't wait to give it a go!

Would be great to see a series of posts how you adressed the ility concerncs: so looking more into performance, security, upgrade paths (especially for attachements), your stabilization process of the product, whether it has been reviewed from externals...

RavenDB 3.0 is very exiting in all the details. I like Voron, FS, new studio, operations improvements, map-reduce visualization and better monitoring options. But I am still looking forward for new book and ability to use Linux machines for sharding. You're doing great work!