3 ms·

I was there when both react and spring were both and they were both revolutionary tools, especially spring, the only people unhappy with it were the J2EE vendor

I was there when both react and spring were both and they were both revolutionary tools, especially spring, the only people unhappy with it were the J2EE vendors, us plebs were calling out the name of the lord. I think the problem here is the usual bloat and enshitification that these tools eventually devolve to due to their success. People get so used to them they want to use them for everything and create this universe where you now forgot why they were created and the places they were useful.

steveBK123 2y ago
With backwards looking rose tinted glasses, we can remember all the good tech people were slow to adopt. But what about all the mis-adopted or abandoned tech that blew up projects. The amount of stuff ported to NoSQL that had no business doing so. UML/Nocode tools in the 90s. A lot of the push away from Java only for a lot of performance considerations pushing devs back to it. Sorry about the projects that got force migrated to Scala since it was, briefly, The New Thing. The one man webdev app deployed firmwide on some brand new framework which was immediately deprecated. Again, nothing wrong with any of these tech choices if they were driven by actual user requirements which exceeded the current implementation tech. But most of us know that is rarely the actual motivation of devs...
- mlinhares 2y ago
  While UML/Nocode tools were (and remain) a mistake, I don't think NoSQL was one. I might dislike some solutions in general but stuff like Redis, DynamoDB, Cassandra, Parquet and many others have produced a renaissance in databases that we would not have seen if it was all in the hards of the relational database vendors. MongoDB easy (in a way) replication put every single database vendor on watch and led even PostgreSQL to improve, i think we live in a different environment right now due to all the experiments the NoSQL folks made. It might not have all panned out but I think the whole thing was a huge positive for the market, people stated to experiment and try new stuff again instead of "oh it doesn't look like a relational model, it sucks".

steveBK123 2y ago

With backwards looking rose tinted glasses, we can remember all the good tech people were slow to adopt. But what about all the mis-adopted or abandoned tech that blew up projects. The amount of stuff ported to NoSQL that had no business doing so. UML/Nocode tools in the 90s. A lot of the push away from Java only for a lot of performance considerations pushing devs back to it. Sorry about the projects that got force migrated to Scala since it was, briefly, The New Thing. The one man webdev app deployed firmwide on some brand new framework which was immediately deprecated. Again, nothing wrong with any of these tech choices if they were driven by actual user requirements which exceeded the current implementation tech. But most of us know that is rarely the actual motivation of devs...

mlinhares 2y ago
While UML/Nocode tools were (and remain) a mistake, I don't think NoSQL was one. I might dislike some solutions in general but stuff like Redis, DynamoDB, Cassandra, Parquet and many others have produced a renaissance in databases that we would not have seen if it was all in the hards of the relational database vendors. MongoDB easy (in a way) replication put every single database vendor on watch and led even PostgreSQL to improve, i think we live in a different environment right now due to all the experiments the NoSQL folks made. It might not have all panned out but I think the whole thing was a huge positive for the market, people stated to experiment and try new stuff again instead of "oh it doesn't look like a relational model, it sucks".