正在加载视频...

视频加载失败

Time is the Best Unique Identifier

44,009 次观看 • 1 年前 •via X (Twitter)

10 条评论

Dr. Brownian Bridge PhD 的头像
Dr. Brownian Bridge PhD1 年前

Tsoding is the greatest stand up comedian of all time. Only a few will appreciate this.

DOGE Bolt⚡️.🇺🇸.🔫. ₿ 的头像
DOGE Bolt⚡️.🇺🇸.🔫. ₿1 年前

Basically UUIDv1

antony 的头像
antony1 年前

now i can't say "distributed systems" in any other way

Cyberistic 的头像
Cyberistic1 年前

why are you using time and not space?

zxq9 的头像
zxq91 年前

"Snowflakes" are suspiciously similar to UUID "version 6" (styled in the link here as vVI() to not conflict in the future if there is ever an official UUIv6). (above is the call, implementation is here: )

Nostalgic southerner 的头像
Nostalgic southerner1 年前

That beat switch on was smoother than butter on a hot desert day.

Scott Neibarger 的头像
Scott Neibarger1 年前

I rather think time is the second best unique identifier.

Jason 的头像
Jason1 年前

Beatboxing beats 🎧

Linuxcel 的头像
Linuxcel1 年前

I do this with everything. I use the timestamp format from the denote.el package. Now everything has an id - files, notes I want to keep a daily log of, todos in my code, etc.

CaptainAmericaTex 的头像
CaptainAmericaTex1 年前

A unique identifier with a time component is always the right way to go. The other approach (which git uses to create unique filenames on the server) is to _pretend_ that a SHA-1 hash is a unique identifier, and accept the small risk of 1 in 10^48 chance of a hash collision.

相关视频