HackerTrans
TopNewTrendsCommentsPastAskShowJobs

_joe

no profile record

comments

_joe
·il y a 4 ans·discuss
OP lives in Rome, where the average driver mentality is somewhere between "need for speed" and "carmageddon".

Basically no one is ever willing to wait for anyone else, as OP's reasoning shows pretty adamantly.

Of course they're grossly incorrect.
_joe
·il y a 4 ans·discuss
Acting like what makes Rome's traffic dangerous are cyclists or moped drivers (even if those can often be quite reckless) is basically victim-blaming.

We should never take such posts too seriously.
_joe
·il y a 5 ans·discuss
This is indeed correct. Wikimedia overall uses less than 2000 bare-metal servers, so yes the infrastructure is tiny compared to those.

What can be interesting, I think, is that you have a completely open infrastructure that has to solve problems on a global traffic scale.

If people are interested in knowing more, I suggest you also take a peek at the wikimedia techblog, specifically to the SRE category https://techblog.wikimedia.org/category/site-reliability-eng... and the performance one https://techblog.wikimedia.org/category/performance/
_joe
·il y a 5 ans·discuss
It's both a waste of donor money and a starvation of resources for people actually consulting images on wikimedia commons.
_joe
·il y a 5 ans·discuss
Any for-profit entity hotlinking Commons is unfair. Heck, they have the right to redistribute freely the image as they see fit, instead of consuming resources that are a common good.

But this goes beyond that - it's some blind check of internet connectivity for the app, and doesn't get shown to the user. We're pretty sure of that, given that with the amount of noise that task generated, if there was an app featuring that image at least one of the ~ 90M daily "views" would've been someone reading these posts.

Now, given we want to be nice, we didn't just blindly block the traffic, although making requests without user-agent is against our UA policy https://meta.wikimedia.org/wiki/User-Agent_policy
_joe
·il y a 6 ans·discuss
Purges have been migrated to kafka as a mean of transport, at long last. So now if a purging daemon crashes, purge requests are not lost.

You can see per-server stats on purges happening here:

https://grafana.wikimedia.org/d/RvscY1CZk/purged?orgId=1
_joe
·il y a 6 ans·discuss
The real throwback would be going back to use newsgroups, and only have competent people posting.