Kudos to Niantic and the recent changes

I just had some great experiences with ML in the last 24 hours!

1st, I submitted 2 photos for 2 Wayspots at a local church, 1 for the church and 1 for the basketball court there. Emily accepted them within a minute of submission!

2nd, I submitted a new Wayspot submission yesterday for utility box art. I was going to wait 24 hours before upgrading it, but around the same time my photos were accepted, I was notified that Emily accepted my Wayspot!

:clap: :clap: :clap:

4 Likes

I love the Machine Learning model! Really ups the motivation to take a great photo on the chance that it might get accepted in 20-24 hours instead of 20-24 months! And the extra appeals when occasionally it gets the decision wrong.

Wish I could make an “I :heart: ML” badge for my profile.

2 Likes

I agree we should give praise when due. The recent changes have been a massive boon to remote areas. Overall the recent changes have been a massive benefit.

BUT, and I’ve been very vocal about this elsewhere, it has created some issue in the system. Namely, abuse of the system to ‘trick’ in borderline to poor submissions that trigger the AI buzzwords. Overtime I’m sure this can be improved though.

1 Like

i do not know of any buzzwords to trick the ML model

i hope you are posting any ML accepts that should not have been accepted

I’m sorry but I’m not sure what you mean by ‘ML’?

I just know about the new AI system assisting reviewing, designed to speed up the process. I’ve seen a lot of ‘plaques’ and ‘Fish & Chip Shops’ in the UK. Both have wayfarer tags that now help reduce queue time.

The issue with the plaque tag specifically is players are spamming generic benches with personal information on a plaque. The new system has helped them get through. Once players see 1 pop up, it creates a knock on effect.

The issue with fish and chip shops in the UK is their commonness. Some are unique and traditional, but most are being presented as such. I’d imagine in a place like America more would be acceptable though.

ML is the Machine Learning model (AI) that Niantic uses. We called it AI in the beginning and were corrected that actually it is ML so we call it that. Except for those people who give it a human name, which I detest. ML versus AI is already confusing enough.

Where on earth did you get that about tags? Niantic has been very secretive about how this works, as well they should be.

If unacceptable things are getting through approved by “our team” emails, you should be posting about those.

2 Likes

I don’t see that many fish and chip shops on the map, I do see a fair amount in reviews though.

Generic benches are two a penny both in review and on the map.

I haven’t seen an increase in them being accepted because of Emily though.

Good robot :robot:

1 Like

Here ya go!

3 Likes

ty but i still can’t set it - will book mark this post for when/if we can

I’ve also seen a noticeable delay in local nominations review time if they go to the community. My area used to be super fast, sometimes resolving in 2 days, and now things seem to take a month or more.
I am a fan of the changes overall, but it seems I’m not the only one seeing a shift.

I think it is good to keep in mind that there is a balancing. We should never have had a system that allowed certain areas to get 2 day turn around while other areas took 2 years. If the balancing means that some places now take 3 weeks instead of 2 days while other places take 3 months instead of 2 years, i think its better overall.

A friend and i did reviews yesterday and we felt so refreshed that more than 50% were coming from our local area rather than all over the country. I can’t tell you how disheartening it is to know there is a backlog over 1 year in your area and to rarely ever see things to review to clear it out.

It also has a major impact on my desire to submit legitimate things because i don’t want to go add to the morass. Now that there’s a faster turn around, I’m happy to go submit things in areas that need help because they will process without bogging down the whole system.

7 Likes