Over the last few weeks I’ve been noticing a distinct trend with Microsoft documentation and examples and often just plain basic functionality.
It’s been particularly acute over the last few weeks as I’ve crammed in a hell of a lot of new stuff. Unfortunately now, I’m almost invariably attempting to avoid looking at any Microsoft docs. The sheer number of rabbit warrens of poorly worded explanations and demos has wasted umpteen days. Invariably, I read the docs. Implement the patterns, doesn’t work, Google, find someone also complaining about the docs with their explanation and how to fix the code.
This seems to be happening too often.
The latest is attempting to work with ML.NET. Firstly, it’s stuck in March 2021 requiring CUDA drivers from 2019 that doesn’t work on the latest two generations of GPUs. Also, their image classification algos can’t deal with in memory images. People have been moaning about that since 2019. I’ve literally walked away from it to go learn Python.
Odd …
Over the last few weeks I’ve been noticing a distinct trend with Microsoft documentation and examples and often just plain basic functionality.
It’s been particularly acute over the last few weeks as I’ve crammed in a hell of a lot of new stuff. Unfortunately now, I’m almost invariably attempting to avoid looking at any Microsoft docs. The sheer number of rabbit warrens of poorly worded explanations and demos has wasted umpteen days. Invariably, I read the docs. Implement the patterns, doesn’t work, Google, find someone also complaining about the docs with their explanation and how to fix the code.
This seems to be happening too often.
The latest is attempting to work with ML.NET. Firstly, it’s stuck in March 2021 requiring CUDA drivers from 2019 that doesn’t work on the latest two generations of GPUs. Also, their image classification algos can’t deal with in memory images. People have been moaning about that since 2019. I’ve literally walked away from it to go learn Python.
Odd …