3 ms·
I'll contribute by telling an anecdotal story. Our company was hired to take on a project that was failing. It was an order export and update system between a
by redleggedfrog 4y ago
I'll contribute by telling an anecdotal story.
Our company was hired to take on a project that was failing. It was an order export and update system between a website and an ERP system, neither to be named for the obvious reasons, and of fairly significant mismatch between the data in the two systems.
It started with some visual tools for the data flow, got a shit-ton of node and random libraries piled on, running on a hosted service with a deployment system for faster updates cause it broke constantly. It lost data 5% of the time or just failed to export the orders. It was shuffled around to different groups which changed up how it worked.
If I had to work on it I'd have quit. Instead I rewrote in 3 days in .NET Core C# as a console app running on a schedule using Json.net, XElement and hand written transforms and it was fun. Copied it up, got it running, buffed off a handful of rough edges for about 2 more days of work and it's been running fine since. No updates needed no data or orders lost.
My point? Reject all that shit that is just bad languages and bad systems and poor choices. Do it with what makes you and the client happy. Clients don't care what it's written in the just want it to work. So choose what you like and gets the job done and lets you move onto the next fun thing. If you don't enjoy your work you're trading your life for money.