This is the same conversation we have when we add any specialized data store. Shortly after MongoDB, Cassandra, Couchbase, Elasticsearch, and Solr, critics asked couldn’t we just contort RDBMS to handle documents for flexible queries or searches. The history tells a different story. The scaling properties and feature expectations of the specialized stores culminating in tens of billions of market cap, and much more in terms of infrastructure spend.
Could the incumbents simply tack on vector features? Sure, that’s the JSONB story of Postgres. It’s the regex story of all RDBMS offerings after Oracle’s acquisition of Endeca suggested a real commercial opportunity for search-specific databases.
Vector-first storage engines have a place in the market as much as the tack on solutions do. PGVector will be good enough for most users, and Weaviate (or Milvus) will be better suited for the most ambitious, or those seeking the best dev ex.
Benchmarks/data volumes aside, PGVector (or similar extensions to more general purpose data stores) is a much more palatable way for an organization to explore adding vector search functionality/features to their product(s) than adding new technology to their stack.
- Best way to point to a csv, local files, local postgres/mysql and iteratively build a highly accurate system to chat with the data.
- We auto generate yaml files so you can transparently edit them and watch real time how the accuracy improves.
- You can point to local pdf's and auto generate reports from dashboards.
Overall we felt this is a light weight framework to create a highly accurate personal AI data analyst without any vendor lock-in.
At first glance, this security incident looks like a non-issue for the most part. It's common that companies' customer forums or support portals are compromised. It seems like a common attack vector. While, I'm relatively certain some very nasty breaches will come with time and the proliferation of vendors, has there already been a catastrophic security incident that happened as a result of using proprietary, closed-source LLMs?
Could the incumbents simply tack on vector features? Sure, that’s the JSONB story of Postgres. It’s the regex story of all RDBMS offerings after Oracle’s acquisition of Endeca suggested a real commercial opportunity for search-specific databases.
Vector-first storage engines have a place in the market as much as the tack on solutions do. PGVector will be good enough for most users, and Weaviate (or Milvus) will be better suited for the most ambitious, or those seeking the best dev ex.