2025-05-02 03:53:13
Indeed, and SQL isn't even the worst thing about an RDBMS. Being global mutable state and having a fixed data model are particularly pernicious.
I think he would find reading about PStates and paths illuminating.
2025-05-02 03:32:57
RT Lyes
btw for the all the sql drama that is happening recently.
@redplanetlabs is building the most ambitious and realitistic replacement for large scale software.
they are addressing the issue at the core. data structures instead of data models and giving exactly what uncle bob want (APIs for interacting with data).
https://redplanetlabs.com/
2025-04-29 01:44:58
I respect Erik greatly, but it's frustrating to see such misinformed comments like this. I've emphasized all over the docs, blog posts, and website how easy it is to integrate Rama with existing infrastructure. In no world does Rama require you to "drop everything". Most users start with Rama by using it to build a new feature, and they integrate with their existing databases/queues as necessary.
Or perhaps he means you have to "drop everything" conceptually, referring to the traditional techniques for building backends. It is true that programming Rama requires learning a very new way of thinking of backends – event sourcing, dataflow, and indexing in terms of any shape instead of with a fixed set of data models. However, rejecting any tool because it requires a new way of thinking is incredibly self-limiting.
He makes a good point that adoption is based on the perception of users of their own problems compared to their perception of the benefits and pain of using a new tool. There is no doubt that the perception of Rama for many does not match its reality. It's hard to comprehend how much Rama really does reduce complexity – kind of like how it's hard for someone to comprehend the analogous benefits of functional programming when they've never done it before.
Our big challenge at this point is making the perception of Rama match its reality for both its benefits and cost of adoption. For benefits, we're doing that by releasing case studies of actual users and releasing lots of tutorials and examples in a wide variety of domains. The G+D Netcetera case study we released a few days ago is a great example of Rama's benefits, as Rama enabled them to improve the performance of a product used by millions of people on multiple critical metrics by over 100x while also reducing their AWS costs by 55%. As they said, "It was a strange experience realising how much simpler software can be."
There is no doubt Rama has a high learning curve, but the cost of adoption is not nearly as high as it seems. I'm hoping all the tutorials and examples we're releasing are helping to get this across. Something else we're doing, and we'll have more news on this later, is building new developer tools on top of Rama that don't require learning any of Rama's new concepts at all (e.g. PStates, dataflow, event sourcing). This will enable people to use and get the benefits of Rama for specific domains with a very low learning curve.
2025-04-24 01:01:36
What I love most about this case study is how Rama enabled G+D Netcetera to greatly expand the market of their product. This is similar to what we saw in the Multiply case study: the mindset changes from "what does our tooling allow us to do?" to "what's our dream product?".
2025-04-20 05:14:13
The #rama channel on Clojurians has been quite active with lots of good design questions ever since we released the free version of Rama
