Oh, hackernews

I can't think of a compelling reason to use Mongo over Cassandra except "we already use it and don't want to change".

It's a fair bit easier to setup than Cassandra and a document store is simply easier for most people to reason about than column-families.

For something like your typical rails app, Mongo has some real nice properties. It makes it sticky. You start out with Mongo because you can just drop data in and off you go. You keep using it because Mongo, despite all of its drawbacks, is really plenty good enough for more than 80% of the stuff on the web.

What I'm mostly interested to see is the rate of new projects using mongo.

Do they have growth at a rate that justifies their valuation?

If that's the case, is HN living in its own bubble? Because if you read HN you'd think that nobody will use mongodb for a new project.

What should one use then?

At my (admittedly tiny) place of work, we use it in production for literally everything alongside redis. It gets the job done. Not my favorite. I'd rather be using PostgreSQL since our data is 100% relational. But though some wicked twist of fate, I'm the only one who can use SQL dbs, and the Sr. devs like mongodb.

yes?

are you saying it has become ubiquitous?