summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorFrederick Yin <fkfd@fkfd.me>2022-09-30 22:33:51 +0800
committerFrederick Yin <fkfd@fkfd.me>2022-09-30 22:33:51 +0800
commit58ae395460d695a67ffbeb4bfa565566efd5c328 (patch)
treeb9ea9358d97ce430c19486c71bc87ee5d71f0884
parentd63a1fb23d9bab707fe0d21f88afa4b51c5f8794 (diff)
ta/vg151_funny_shit: VS incident
-rw-r--r--docs/ta/vg151_funny_shit.md30
1 files changed, 30 insertions, 0 deletions
diff --git a/docs/ta/vg151_funny_shit.md b/docs/ta/vg151_funny_shit.md
index b615c75..07665fb 100644
--- a/docs/ta/vg151_funny_shit.md
+++ b/docs/ta/vg151_funny_shit.md
@@ -102,3 +102,33 @@ assumed they could use any.
Well _in theory_ you could write the project in Python with matplotlib. Or
in C/C++ with OpenGL. But… better not talk about it.
+
+(See followup below)
+
+## 2022-09-30, Gitea
+
+On 16:54 it came to my attention while `git pull`ing all the students'
+homework repos that one branch was 108.26MiB in size. It should not be
+larger than 1MiB.
+
+I smell fuckup fumes from a distance away. Upon close examination, they
+pushed their project code, which should have been strictly kept personal.
+Worse, there are two versions. Worse, neither is in Matlab, but in C++.
+Worse, one of them is built on Qt. Worse, they were written with Visual
+Studio. Worse, they pushed all the vsidx, vcxproj, ipch and all kinds of
+crap.
+
+Manuel predicted that students will do all sorts of stupid things, going
+as far as considering git lfs. Real conversation:
+
+> TA: will lfs be used in this course?
+
+> Manuel: it should not but we never know what they will push by mistake…
+
+Murphy's law worked. Apart from this VS atrocity, students were pushing:
+
+- tarballs / zip files they should be submitting to the OJ
+- batch file automating tar
+- pdf
+- symlinks to the pdf
+- a CC BY 4.0 International license