DVMs are more infrastructure imo, I don’t see them as something users will interact with directly (at least for most DVMs). The most used DVM right now is feed sorting algorithms, and clients make this user friendly and call the DVMs in the background (the user may not even know it’s DVMs that are doing the work).

We haven’t really had multi DVM use cases yet (although folks are working on them).

I think of DVMs as modules that could comprise agents, and the space is still so early. For example, we need libraries where you can call a DVM just like you would call an API (a python library for this is on my to do list).

Many people building agents are using lots of APIs right? DVMs would replace these API calls and offer some neat tradeoffs.

 

 Makes sense! Python DVM library seems like a game changer 

 Yes that would be a plus.  I think we might need a lighter version that is easier to get started with and do useful things.

Also I see it going in the direction of cashu integration, is my guess.  And cashu want to charge on every tx.  That's going to make workflows complex.

I see DVMs getting more complex, not less, and they are already very complex.  Perhaps it's a good time to examine the most useful parts of it, and make a simple version.  After all, nostr took off when the protocol was a simple 2 page spec.  Even nostr core is a beast of complexity now which changes too frequently.

Getting back to simplicity I think is the path to success.  

 I’m curious what would be simpler than DVMs as they are now, do you have any ideas? 

 So the idea is, money in, data out.  I think that's a great starting point.

And you can have money=0

So in the base case it's just an API, right?

Then there is the discovery.  That seems to be a separate part.  How to find a machine.

Then there is the payment.  ie on-chain, mint, lightning, cashu

Then there is the need to make it fast, ie sub 400ms.

I think make it into small logical micro sections that can be simply implmented and less monolitic.