Open ITwrx opened 3 years ago
Interesting idea. Unfortunately the statistics would require manual filtering due to noise from spambots and so on. If anybody is interested in reviewing the searches and publishing statistics please let me know.
Here's a listing of the most common searches in the last 14 days:
24 http
24 gui
23 game
15 web
13 sdl
12 opengl
9 terminal
9 raylib
9 plugin
9 image
8 pdf
8 html
8 gtk
8 cli
7 zip
7 ui
7 template
7 regex
7 json
7 jester
7 graphics
6 Web
6 test
6 png
6 discord
6 3d
5 xml
5 vulkan
5 ssh
5 scrape
5 rest
5 python
5 https
5 hash
5 dns
5 csv
5 audio
5 2d
4 youtube
4 windows
4 uuid
4 unpack
4 tui
4 svg
4 ssl
4 sql
4 shell
4 redis
4 pixie
4 parser
(Multiple searches from the same IP address are deduplicated and counted as one)
I think it would be good for nim community and library devs to know what libraries people are searching for.
So, this issue proposes that every search get saved and the list of searches be viewable, and/or searchable, preferably with a searched count so that library devs can take user interest into account when they are deciding what library to write next.
Maybe just save the search string and nothing about the searcher's machine/connection to protect/respect their privacy.
perhaps to avoid moderation requirements, the strings could just be saved and this package author could just release a sanitized list periodically with search strings and their respective counts/ranked by searches, etc. Anything would be better than just not knowing what people needed for their projects.
thanks