Hacker News Reader

Comment by highace | Hacker News Reader

Use the Google Photos API to build smarter photo and video features for your product... until we deprecate the API in a couple years time and leave you up shit creek.

View on HN

This is a tired meme given that the number of Google services closed has been mostly going down since 2013. See also https://en.wikipedia.org/wiki/List_of_Google_products#Discon...

  2006 3
  2007 4
  2008 5
  2009 7
  2010 4
  2011 16
  2012 19
  2013 11
  2014 7
  2015 6
  2016 11 (actually 8 if you don't count things that got merged into other things)
  2017 4
  2018 2

This wikipedia list is missing a lot of APIs, it seems more related to products -- For example, the QPX API (which I used) was shut down this April and is nowhere in that list. Your numbers are lower than they should be.

I'm pretty sure the qpx API was a purchase, not something Google started. So it's arguable that you can categorize it the same way.

I'm pretty sure shutting down critical parts of acquired products is a major part of the problem.

agreed. Just because they were acquired doesn’t mean that the acquisition customers didn’t feel the pain from being forced deprecated

Notable there would be the recent incident where Twitter bought some ML startup for abuse analysis and shut down customers overnight with almost no warning.

According to your own citation, Google deprecated multiple APIs every year. So the odds of Google cutting off access to an API you use may be decreasing... But it's still nowhere near zero, and it's still a valid concern.

Yeah, but that's not unique to Google. It's a valid concern with any API or service that you don't control.

Does Google shut down stuff at a greater rate relative to anyone else? Are the odds better or worse if X had been run by company Y?

The question is not “is Google better or worse than average?” It’s: what is the risk of building an app on a Google API? The risk is high, and it’s not worth it. Their track record is abysmal. Sure, if you pick a random provider out of a hat, maybe Google is better. But if you look for a solid provider or better just build your own service instead of trying to glom onto someone else’s (someone with no real incentive to see your product succeed).

> Yeah, but that's not unique to Google. It's a valid concern with any API or service that you don't control.

As mentioned by others Amazon seems to have a policy along the lines of "APIs are forever".

That's really nice. Google could make the same pledge. They do well enough and will likely be around for a very long time.

I don't care anymore. I am now in the process of removing Google Maps from every system we run because of their screwing over their users and their devs. I have thousands of dollars of time to spend ahead of me because of the changes to Google Maps this past month alone.

I get it, it's expensive to run maps, it's not free. But this is multiple-times burned by these guys. So the threshold of toleration is lower with Google than other businesses/services.

Nothing free from Google is ever going on anymore of my sites or software unless it is setup with an alternative already in place so when they do _anything_ ridiculous, they can be dropped instantly. But as of right now, the plan is a complete purge of Google.

What are you switching to for web and mobile users for maps?

I'm doing what the parent post is doing, switching to OpenMapTiles which I'll be running on my own. Still using Google maps on Android because it's free and doesn't generate enough volume to cost me much.

I too take caution with Google APIs after the maps price hike. I've also abandoned Twitter API after their streaming API shutdown

I actually don't want to advocate a specific map software right now, as we are likely to switch between multiple systems until we find one we like.

OpenStreetMaps ( https://www.openstreetmap.org ) is the data source for most other mapping systems though.

Suffice to say, I am setting it up so that it can be more easily replaced if need be.

It's not really tired if your app/service was dependent on one of the ones shut down. Granted, most services/APIs have a lifecycle and won't last forever. What makes Google different/frustrating is their habit of shutting things down either in their prime (I.e when goog realizes they won't get 100s of millions of users) or well before the tail end of their lifecycle leaving those who placed a bet on said service high and dry. I view it not so much as a meme as a warning/reminder.

How is this a tired meme? Looks like 99 services were shut down since 2006.. You just proved their point. The consensus here is that you shouldn’t build anything critical with their APIs, and I agree

So if I built anything on Google's platform ... in the last 10 years or so, you are saying "only" 99 products or platforms (by your own numbers) have been shut down, possibly taking your business with it.

That number is crazy high. That there has been a slight trend the last 2 years does in no way make up for the significant risk shown by those other numbers.

You have to be crazy to base your business on the Google platform.

Tired meme indeed.. https://news.ycombinator.com/item?id=17971516

Or raise the API usage costs from under $100 a month on average to over ~$15,000 a month like they just did/are doing with the Google Maps/Places/Geocode API's. Not kidding. They're shutting thousands of websites, businesses down because of this. They gave everyone two months notice to effectively shutting them down.

Or increase the price of the API by ten times.

Seriously, it should be a rule of thumb that if you are not big enough to negotiate with Google for a secured pricing, don't rely on their API exclusively in your project

Reply on news.ycombinator.com