Niantic Says It Had A Very Good Reason For Shutting Down Pokemon GO Trackers

Niantic Says It Had A Very Good Reason For Shutting Down Pokemon GO Trackers

Over the last week, Niantic has been chewed out by the public for not being open enough about what was going on with Pokemon GO. But now, it seems like they’re finally turning all of that around. Not only have they been posting more on social media, today Niantic also blogged about one of the more controversial topics surrounding Pokemon GO: Tracking websites.

The rumour for a long while had been that tracking websites such as Pokevision put too much of a strain on Niantic’s servers, and today, Niantic confirmed that in straightforward terms:

As some of you may have noticed we recently rolled out Pokemon GO to Latin America including Brazil. We were very excited to finally be able to take this step. We were delayed in doing that due to aggressive efforts by third parties to access our servers outside of the Pokemon GO game client and our terms of service. We blocked some more of those attempts yesterday. Since there has been some public discussion about this, we wanted to shed some more light on why we did this and why these seemingly innocuous sites and apps actually hurt our ability to deliver the game to new and existing players. The chart below shows the drop in server resources consumed when we blocked scrapers. Freeing those resources allowed us to proceed with the Latin America launch.

Niantic Says It Had A Very Good Reason For Shutting Down Pokemon GO Trackers

A vague chart, which is not ideal, but still. It’s something.

Niantic continued:

In addition to hampering our ability to bring Pokemon GO to new markets, dealing with this issue also has opportunity cost. Developers have to spend time controlling this problem vs. building new features. It’s worth noting that some of the tools used to access servers to scrape data have also served as platforms for bots and cheating which negatively impact all Trainers. There is a range of motives here from blatant commercial ventures to enthusiastic fans but the negative impact on game resources is the same.
Of course, there are also outright hackers out there attempting to break into systems, hijack social media accounts, and even bring down the service. Some of them have posted publicly about their attempts.
We don’t expect these attempts to stop. But we do want you to understand why we have taken the steps we have and why we will continue to take steps to maintain the stability and integrity of the game.

You can read the full post here.

Bravo, Niantic. This is exactly the sort of communication that players were hoping for. Yes, the game might still be in an iffy state thanks to a variety of bugs, but just having a sense of what is actually going on goes a long way toward quelling anger and frustration.

Niantic even assured people that they have “heard feedback about the Nearby feature in the game and are actively working on it.” Excellent! Here’s hoping whatever they come up with next is more usable than the footsteps system.


      • Yeah, without a baseline that could just be a really zoomed in chart that shows a less than 1% drop. Who knows

          • They didn’t say that freeing the resources was the only thing that allowed them to proceed, only that it allowed them to proceed. Twisty turning PR-speak!

        • Just taking in the report straight from Pokevision themselves that they had over half of all pokemon go users accessing it… at the same time while playing it adds up, that was just one third party.

      • I know, I scoffed at the remarkably barren graph.

        Could be anything, player drop off, consumer excitement, potential earnings loss, a rough scetch of the cliffs of Dover, a bad attempt at following the line before somebody bumped the writers elbow.

    • If only it came out Day Zero of the action. “In the patch that is going live…” “In the next update…” “Due to issues with something we will be doing this…”

      Internet community and game developers need proactive PR not reactionary. Changes have to be communicated and sold to fans.

  • I think there’s a common thread when we abuse corporate entities for apparently not being ‘fast’ enough to communicate, but we’re not being realistic here. That’s a lot of moving parts and to say The Right Thing is one thing, but how it’s interpreted is another. It’s not political reporting where every media release needs to be picked apart by pundits. Or at least I hope it’s not.

    Yes Niantic should be communicating with the player base, but it’s a bit rich to tap our collective feet impatiently and then act like they’re hiding something when they flat out parrot what sites like this have reported has been going on – bit of a no-win situation.

    Scraping sites, GPS-hacks, actual DDOS attempts, yes okay but if 4chan or Twitter users were in charge I’m sure it’d be smooth sailing.

    • Yeah, there’s a huge difference between the data used internally to justify decisions compared to what’s sanitized for the public.

      • Notice how ‘the game’ today is ‘let’s criticise a graph’.

        I mean, really. It’s Friday.

    • Actually its not rich. The fact is they launched a game without a community manager KNOWING that this was a problem that was plaguing them SINCE INGRESS launched which meant they (Niantic) not knowing this was going to be a huge problem despite players from Ingress trolling them that this was a problem for over a year is the real rich bullshit here.

      Saying that Pokemon Go wouldn’t need the same communication management that Ingress needed is also pretty high ranking on the BS odometer simply because Pokemon Go was always going to at least be a bigger game than Ingress as far as global audience and global launch.

  • Niantic Says It Had A Very Good Reason For Shutting Down Pokemon GO Trackers

    “We’ll be going pay-to-win! For a monthly subscription, you’ll be able to use the Pokemon GO Tracker”

    • Hey man, just a suggestion, but if you read the article before commenting it can stop you from looking this dumb.

        • but a sarcastic comment that was about something entirely unrelated to the article.
          Of course, the only way a Pokevision type tracker could be allowed in the game was to charge people to use it, to cover the insane server costs.

  • No scale on that chart?

    Sooo each line represents 10 connections? So it dropped by 20 connections…

    • Part of me thinks that there was someone at Niantic who wanted to release the chart with axis labels etc but then PR or someone stopped them because knowing details like that would reveal too much information about total server capacity. Releasing that chart picture just makes Niantic look stupid though.

      • They could have easily fudged the number by using percentages instead of total connections. With the lack of labels, I’ll declare their rate of Shakespeare dropped by 20 bananas.

        • You forgot to factor in the ratio of Hitlers divided by the aggregate total of Spains. So it actually dropped by 26.666666666666666 bananas with a slight increase in potato clocks.

  • I havnt looked into this very much but do niantic release patch notes? do they contain any insight?

  • The biggest issue is that it doesn’t show if the y axis starts at zero. It is clear there was a reduction in resources being used and it is significant enough to be tracked. But if y starts at 0 then it is a 66% drop in resource usage. If it is only showing the ‘relevant’ section this it could be a 1% reduction in usage, or anywhere in between those 2 values

    • Yeah, the graph is crap. What it comes down to is whether you accept what they’ve said – that it increased the server load and caused issues or you think they’re lying. The graph being there or not makes no difference.

      Personally I’m just happy they’ve started communicating. Makes me not regret spending $30 on the game because I’m confident they’re doing something to resolve the issues now and are listening to the community or at the least they’re aware of what the community is complaining about.

      • There is no doubt at all there was a significant impact on the servers. Didn’t pokevision state they had 40 million people using their app. it is impossible for that not to have a major impact

        • Not to mention, the app would pull down Pokemon for a certain max distance away from you.
          This pokevision, you could scroll around all over the map, using up many more resources than a single user of the app.

      • Anyone who trusted Niantic already knew and accepted the insane server issues Pokevision caused. People who think it all runs on magic and Niantic are meanies who want to make them sad still won’t believe them.

  • Its not hard to imagine that those scrapers work by spamming a ‘player’ at every location simultaneously to get a map of where the pokemon are, and how that could really hammer server resources.

  • Man, what a great idea! Next time someone attacks any of my questionable decisions, I’ll draw a squiggle on a notebook and pass it as a chart that “proves” correlation of causes and results to whatever reasoning I want to use as a rationale. “As you can see here, the moment I started punching you in the face, the rate of prenatal mortality in Kampala, Uganda fell drastically.”

  • Worst. Chart. Ever.

    Edit: Oh they actually updated it hahaha, but no y-axis scale so it’s still useless.

  • I’m surprised at the flak they’re copping for the graph.

    It’s not actually any of our business to know the down-to-the-connection specifics. They don’t have to tell us and in fact probably shouldn’t, in case it aids those with malicious intent.

    All they really needed to say was that those scraper apps increase their server load by a ridiculous factor. The reason that they work is the same reason that they are bad.

    • I agree with you which is why they probably should just have left the graph out.
      Either put a graph in that is useful or just leave it out completely.
      Otherwise you open yourself up for this sort of criticism.

      • It seriously looks as though haphazardly drawn with a pen on a random notebook. A pink pen. Looks like something a 6-year old girl drew to show her parents how her happiness dropped the moment they told her that they wouldn’t be getting her the latest My Little Pony palace.

        • I’d like to think it’s a psychological thing, their version of an inkblot/Rorschach test… One they’re showing to millions of people to discover what those people see when looking upon its glory.

          To me it represents what happens on my graphics card monitor when I stop playing a game.

    • The issue as I said above is they are presenting that as evidence when it doesn’t actually prove anything. Depending on how zoomed in on the Y axis it could either show a 66% reduction or a 0.66%
      I can see why they don’t have specific numbers but even a 0% to max would help. And I think they should have because from the labeling I think that it actually was a 66% drop in resource usage

    • As a developer I’m thoroughly annoyed at the reactions to the graph. It’s obviously some form of server load representation (I’d guess requests per second), and I’d wager the drop in activity is as significant as it suggests, even if the axis aren’t labelled.

      A normal Pokemon GO client would only query the server once every so often for nearby Pokemon. But a scraper needs the location for the entire world, and would be equal to thousands of official clients, not unlike a DDOS.

      Wouldn’t at all be surprised if blocking them gave the servers a massive respite.

  • went out for 20 min after tracking stopped working…. got a kahn, scyther and pikachu

    all is not lost

    • Prior to whatever updates have been made, I only caught doduos. The last few days, I’ve gained four levels from filling the pokedex. I don’t care what everyone else is saying, it’s working brilliantly for me.

  • Remember, a lack of evidence for isn’t evidence to the contrary. The vagueness of the graph doesn’t implicitly mean they are hiding something or being disingenuous. Don’t ascribe to malice that which is adequately explained by ignorance.

  • So let’s say that the first value was 100% of average connections. There’s no distinction between what is player loss and tracker loss, so let’s say it’s tracker loss.
    There’s a pretty solid dip in connections of about 30% following that, meaning what – after losing tracking they lost about 20-30% of their user base?

  • I absolutely LOVE that chart.
    It’s the most fantastic “We give no shits” graph I’ve ever seen in my whole life and it’s remarkable.

  • People need things spelled out in big letters and have a desperate need to believe in conspiracies. No wonder Trump, Boris Johnson and Pauline Hanson have such widespread support.

  • I still call bullshit. This isn’t going to stop the bots, whilst it has temporarily. it will only take a few days for devs to figure out some of the responses are now in tuples. I mean the API is still there and accessible…. This is a bandaid fix at best and they still need to communicate with the community more i mean jump on reddit or something???

Show more comments

Comments are closed.

Log in to comment on this story!