Tag naming / checking

[abudden]

Having just started adding my package to the package list, I was looking at the existing tags to see which were most appropriate. I found a lot of duplication. A few examples from the start of the alphabet:

• "2d", "2D"
• "authentication", "authenticator"
• "algorithm", "algorithms"
• "app", "application"
• "benchmark", "benchmarking"
• "bind", "binding", "bindings"
• "Bloom", "Bloom filter"

The list goes on (and even includes "Nim", "nim" and "nimrod"!)...

If these tags are ever selected as the means of trying to search/filter packages, it's going to be very hard to actually find anything useful I suspect.

[Araq]

Stemming and case insensitive searches are not rocket science... I don't see any problems here.

[dom96]

PRs to clean up tags are always welcome!

[achesak]

At the very least, can "nim" and "nimrod" be removed as tags? They're rather pointless as every package is going to involve Nim anyway.

[dom96]

Sure

On Monday, 29 August 2016, Adam Chesak notifications@github.com wrote:

At the very least, can "nim" and "nimrod" be removed as tags? They're
rather pointless as every package is going to involve Nim anyway.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#399 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AAPDezZY0qQkqrLcIaV1gK425xzrc1gfks5qkkkagaJpZM4JmfYM
.

[genotrance]

Can we close this one - tags are really the author's call. If they put in junk tags, it does their own package a disservice since it will show up for random searches.

I presume tags are also reviewed when packages are added.