rtyler
/
brokenco.de
1
0
Fork 0
Code Pull Requests Releases Activity

Fix a couple more typos

This commit is contained in:
R Tyler Croy 2024-05-13 16:11:37 +00:00
parent 34e1707a8c
commit fff1163c67
2 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
_posts/2024-05-12-helpful-code-generation.md
View File

@ -13,7 +13,7 @@ herald a new age of software development efficiency is misleading in most cases
bordering on malicious. While there are a number of other recommendations or
predictive machine learning models which _can_ improve software development
efficiency, LLMs propensity to generate bullshit undermines trust in a way that
makes me question their validity at baseline as a sofwtare development tool.
makes me question their validity at baseline as a software development tool.
A couple months ago I recall using Google's Bard in an attempt to generate some Terraform boilerplate code. It so fantastically failed to do so that the top Google search result from Stack Overflow was almost always _functional_ code whereas Bard consistently produced invalid code.
@ -93,7 +93,7 @@ It seems that most of the people pushing LLMs in software development have an
ulterior motive that is _not_ in your best interest as a developer. Executives
who want to squeeze more productivity out of you, platforms who want to
increase lock-in to their proprietary tools, or hardware manufacturers who want
to sell more chip.
to sell more chips.
This has been said a lot lately, but what makes delivering software hard is not
the "writing code", but typically the "understanding code" part. For which LLMs

2
generate-tags
View File

@ -1,4 +1,4 @@
#!/usr/bin/env python
#!/usr/bin/env python3
import glob
import os