[openstreetmap/openstreetmap-website] Replace Shortlink custom algorithm with a standard implementation (Issue #4205)

[Brooks Boyd]

While it is up to the maintainers of the project to make the final call on this, I just wanted to note two bits of feedback on the given rationale why this was closed:

> We are required to maintain that code, even if we generated short links with a different algorithm in future

The concept that once created a bit of code needs to be maintained **_forever_** is not one that most software packages follow. Software has features that get deprecated and removed as needs and behaviors shift. If a specific thing is designed to be a standard, then it is good for it to be maintained for a long time, but even then, the concept of a new standard deprecating an older one is not an unheard-of thing.

The "Shortlink" algorithm is documented on the wiki, but not presented as "a standard" in a way that other applications could generate on their own. Hence, I find it odd why it's not open to even consider deprecating it.

Secondly, I've noted several times that it would be helpful to see usage metrics on that feature to tell if it is being used and in what way. There's been declarative statements made that the focus is on supporting "every shortlink that has ever been generated", but no evidence given that old shortlinks are even used (my thought being that a typical use-case would be sharing a link with a friend, and once that friend has visited, that shortlink would be not used again). Is the maintaining team gathering metrics on site feature usage and using that to evaluate development options?

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/4205#issuecomment-1712118934
You are receiving this because you are subscribed to this thread.

Message ID: <openstreetmap/openstreetmap-website/issues/4205/1712118934 at github.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/rails-dev/attachments/20230908/5d66853d/attachment-0001.htm>