Farewell, Google Maps
inderapotheke.de
inderapotheke.de
I think this is one of the most important points in the article - the way they handle these pricing changes destroys trust in Google's other business offerings. How can people use Google products and services as a core piece of their infrastructure when they're willing to bump their prices >10x with only a few months of notice? That could literally be a business-ending event, depending on how core that service is to the business.
In the case of maps, there weren't many great alternatives for a long time, due to Google sucking all the oxygen/profit potential out of the field with their excellent free offerings. Fortunately, their last (sudden) price bump seems to have allowed the creation of some good alternatives.
* shutting down the service entirely because the user base never grew into customers who actually valued the service
* changing your terms of service to forbid an activity that was previously allowed, because someone discovered a use that messed up the price points and the service owners would rather forbid that than offer a reasonable price point allowing it
* moving to opaquely metered service potentially with apparently arbitrary levels of financial exposure to the client
A big price bump with a few months notice is painful (and I'm glad it's bringing competition), but it tells me they're thinking seriously about how to sustain/develop the product and lets both them and me explore the real value of the service.
When it comes to Google, I'm more worried that they might just arbitrarily mothball something on a management roadmap whim.
I mean, there's a real cry-me-a-river aspect here. No one has a Google-granted fundamental right to mapping. It's a free product Google released to drive traffic to their own offerings, and which they happen to make available for free for a lot of purposes.
As it turns out, yeah, there's a lot of value we as a society can derive from low cost pervasive mapping. So let's find a way to share the cost and not just whine about Google, no?
I ended up rolling my own server (in scheme using sxml) and writing some minimal clients for it (tui, Emacs and web). It is running on a raspberry pi 2 and is currently using about 5mb of memory (the guile runtime is a large chunk of that) and has been down once since I started it 5 years ago (power outage).
Reader was to me the absolute best example of the good old web. Since then it seems like the web became insane with the idea of reinventing UI, which was then forced down our throats for both web and desktop, ruining two paradigms that worked just fine.
I am old,I know.
I don't understand why you rolled your own server. Feedly works great. If you did it because it's a fun project, that's awesome! But it certainly isn't necessary if your goal is just to subscribe to RSS feeds.
So I rolled my own. I will probably port it to use SQL so I can do easy searching when my kids move out :)
I might not have explained my point well but I do think you get what I’m saying.
There are reasonable ways to disagree with Microsoft's policy here, but "no alternatives" and "monopolistic" are just laughably silly.
Your comment completely disregarded the possibility that there's a dominant market position. Which there is. It is hard to compete with free. Ask e.g. TomTom or Microsoft.
Never heard of any drug dealer that did that in my adult life, but I do see a lot of large corporations using that business model.
"If Google needs to engage in anticompetitive behavior to keep Android profitable/viable, then they weren't ever profitable."
If startups need a giant subsidized operating system for free or else they weren't profitable, they weren't ever profitable.
Do you see how ridiculous that argument is?
>As it turns out, yeah, there's a lot of value we as a society can derive from low cost pervasive mapping. So let's find a way to share the cost and not just whine about Google, no?
Yeah, Google has vacuumed up incredible amounts of mapping data from users and has now jacked up the prices with the lead this gave them. OSM (or something similar) is the way to share the cost. Google just abused their goodwill and used their other power positions to discourage open source. Whining about Google while moving elsewhere is entirely appropriate.
What? Your counter-counterpoint doesn't make sense at. Lots of startups can afford the per seat cost of OSX or Windows. If your business can't recoup the cost of an os per employee then yeah...maybe it's not a worthwhile venture.
On the desktop side of things that only happens because of Linux, otherwise (if guys like Ballmer would have had their way at the end of the '90s - early 2000s) those startups would now be paying hundreds of dollars per Windows license per each machine (and maybe more depending on the number of cores? the sky would have been the limit, the same as Oracle did in their field). And not to mention the server side of things, I'm pretty sure the costs for server OS licenses would have been prohibitive for the vast majority of startups without the alternative free OS solutions.
You see where this argument leads to?
Google does a whole lot more, and can charge whatever they want for it. OSM provides usable solutions the vast majority of users; i’ve Moved to OSM a couple of years ago in everything for privacy reasons and except for the occasional Waze route or verifications, get nothing from google that I can’t get from OSM.
I mean, kind of, but on the other hand, think about all the companies whose costs are subsidized by having public roads.
They complain that Maps at new prices would be more than the cost of their infrastructure, when actually their entire startup revolves around this data, and 5k to be able to use an amazing piece of high tech software that is ahead of the competition is..peanuts.
No matter what other business interests and strategies Google follows, there is no right to using such valuable tech for less than the monthly salary of an engineer. I find this incredible entitled.
I am not saying that mapping should be free, but such price bumps should be communicated better to allow paying customers prepare for changes and alternative scenarios.
And again, just another reminder, that whenever choosing a provider for anything related to your core business - think of an exit plan!
This is a bit of “thank you sir may I have another”. 10x price increases? It’s not sustainable? This is just a money grab - they’re not going to give a roadmap for how in the future google maps will be worth 10x the value - it seems to me more of a statement from Google that they have no real competition in the space.
"However, from a regulatory view, monopoly power exists when a single firm controls 25% or more of a particular market. For example, De Beers is known to have a monopoly in the diamond industry."
https://www.intelligenteconomist.com/monopoly-market-structu...
It tells me they're trying to minimize the number of users (which reduces their costs) by raising the price while still making it profitable. They raise prices rapidly to see how many people jump ship, and when your revenue begins to decline you've gone a bit too far.
Of course, they've done it very quickly, so they've likely not allowed enough time for the competition to step in or all their customers to abandon your now-expensive product.
Is the marginal cost of serving extra users really a big part of Google's costs? Bandwidth is cheap; I would have expected that building and updating the map data was the expensive part.
Unless the vector information is getting much more expensive (shouldn't be, they own Waze and GMaps mobile products to make their own maps from user data), the cost of running GMaps should be going down on a per user bases, not up.
See this is the important thing. Sustainable. People and companies use these free services that take quite a bit of work to maintain and then when the company decides it needs to stop hemorrhaging money they start charging, often a fairly modest fee, and people lose their damn minds "ermagerd, you're evil, I need this, you can't expect me to pay!". The simple fact that adding a price, or increasing the price, creates such an outrage is indicative, more oft than not, that it is a fair move because the service or good has obvious value to the user.
People lost their damn minds when Netflix increased prices a year or two ago and I'm like "erm, I get way more value out of this than cable and it's a tenth the price, shut up".
>When it comes to Google, I'm more worried that they might just arbitrarily mothball something on a management roadmap whim.
I've been a Project Fi user for 2 and a half years now. This has been my worry every, single, day. Especially since they still owe me like 800$ in statement credits ha.
For your analogy to work you would need to say you are content with Netflix raising their prices >10x (no longer a 10th of cable) because that is what Google just did with their pricing.
Most people I know who stream (ok all people I know) get way more value out of streaming than cable. What is your line on fair pricing for streaming compared to cable, as the service has obvious value to the user?
I don't think anyone, here of all places, would suggest a product requiring engineers and infrastructure be free for commercial use and that the company shouldn't recoup costs.
The good news is, like streaming services, there is competition and pricing will work itself out in the market (assuming no collusion :P)
Netflix raised their prices what they felt they needed to. Google raised their prices what they felt they needed to.
If people don't like it they're more than welcome to go to the competition... which in both cases generally has inferior product.
It looks like this specific change almost entirely inconveniences businesses and not private individuals. They're allowing 28k FREE requests a month which is still beyond generous. That's enough for a small entity to develop a service or product around the service and not only get it working but develop a decent alpha, or extremely modest beta, base of users. Then, just like other businesses, you get to put on your big boy pants and accept the cost of doing business.
Does the power company hand out free power?
Does the water company hand out free water?
Does the phone company let you have free calls?
No. So Google is still allowing a more than fair FREE level of commercial usage and now want to actually monetize their product. If it's adding value to your business, paying the new rates is a cost of doing business. Build it into your pricing, adjust your budget. This is the real world, not Narnia.
I imagine in the most common cases the application is effectively advertising "here are our locations, come to one" and by the point of showing the map you should have already removed a significant number of non-conversions making it VERY fair pricing.
In instances where you are using it to actually build a product around, it's still likely orders of magnitude cheaper than buying, and maintaining, map data yourself. In fact it wouldn't surprise me if one year's cost is only a few percent of what it would cost you to initially buy all of the data and develop software for navigating and hosting it.
People allow free, and cheap, things to make them feel entitled. It's easy to do, it really is, but it's something people need to be more mindful of before freaking out.
All I read in that blog post is "HOW DARE GOOGLE! They want us to pay for a service we use! Pity us! Shame Google for wanting to not operate at a loss! How dare they! How. Dare. They!"
We’re talking of Google Maps the platform and its B2B offering whereas Netflix, cable and phone companies are services for consumers with fixed and reasonable consumer-level monthly subscriptions.
If you haven’t built an app on top of Google Maps, then you pretty much have no idea how much it costs and have zero valuable input you can give.
Also, this isn’t the first time Google is doing this bait and switch. They did it before with App Engine as well, first fooling early adopters in order to gather popularity and then raising prices enough that it made plenty of startups to move off the platform.
Along with other blunders in their products, it makes one wonder how anyone can trust any of their offerings long term.
Too bad the US has lost its anti-trust teeth btw, because what Google is doing is to subsidize its offerings until they get popular, effectively using their monopoly to gain popularity in other markets, thus hurting their competition unfairly.
It isn't a bait and switch. It's introductory pricing. They attracted developers and businesses to a product people essentially weren't using, they let people work with it at or near a loss and now that they have a customer base they are trying to make it profitable.
Yes, they'll lose some of their users that want a free ride. They'll also retain many users that will happily pay under the new pricing scheme because they recognize the value add and find it to be a worthy business expense.
It's no different than companies like MailChimp offering free-for-so-many-subscriber pricing and then requiring you to pay once you reach a threshold they've determined makes you a billable customer. Or a company like Evernote allowing you to save so much data (what a map tile is) and use so many devices a month for free before requiring you to subscribe at one of the paid tiers. Or a company like Pushbullet allowing you to mirror so many SMS messages a month before charging.
If you can't easily handle the pricing change, then you are probably wasting your time using it in the first place and need to discontinue using maps anyway or reassess your own business model.
And there is absolutely no similarity between your MailChimp example and what Google has done in this case. The customer didn't complain about costs rising as the number of users starts to exceed an existing threshold.
That said, I wouldn't be so quick to call for the regulator. All businesses have a responsibility to select suppliers carefully and manage their dependency on any one of them. It appears to me that the value of contracts has been forgotten.
Yet they spend the entire post whining about it.
They also say "Of course, we always knew that as we grew larger, there would be cost to using Google Maps."
Spot on. I would that if a site / application is hitting Maps' paid tier they can (somehow) monitize that traffic and cover the cost.
But monitization can't help you if / when they pull the plug.
p.s. I sometimes wonder how often Google looks at Slack and regrets mothballing Wave.
Maybe having a searchable historical database of everything imaginable and their prices would keep companies honest and prevent price gouging.
ianal and all that, but isn't somewhere antitrust- or monopoly- adjacent?
I know if walmart moved into a new city and started giving away hardware for free, then waited until all the local hardware stores had been shut down for 10 months, and then jacked up their prices to 100 times what they used to be before going free, that'd be pretty illegal.
unsure if that’s also predatory pricing? seems different to me
Whether their competition is Lyft, Taxis, or public transportation, their subsidized pricing is a big factor to their customers.
I don't know if this qualifies, and I don't know how much appetite there is for prosecuting Google for offering useful things for free. I don't think they've had much of a history of jacking up prices after killing off a market, until relatively recently. Interesting question, though.
Not to mention Internet.org from Facebook, TOM's (free) shoes in Africa.
This is what drug dealers do.
It was also the plot of the James Bond film Live and Let Die.
I don't really think that happens at all. If so where are they?
That's not too say that there's guys on street corners offering free drugs to random passersby. Rather there's lots of dealers who just deal to aquaintances, friends, friends of friends, neighbors, etc. In that case they can spread a few hundred dollars worth of product around to non-users within their social network in hopes of picking up another client or two, who will end up generating reliable profit for a long time.
There was a related discussion[0] on HN a few weeks ago that I found enlightening. Commentator ucaetano's phrasing of the strategy was "create a desert of profitability around you". The discussion of moats vs deserts found here [1] is also useful.
[0] https://news.ycombinator.com/item?id=17047348 "Laws of Tech Economics: Commoditize Your Complement"
I remember like 3 or 4 years ago having this conversation at a hackathon with a bunch of "Machine Learning" / "Sentiment Analysis" APIs present... right as google started offering both for free on their cloud platform.
Wrong, maybe that "space" (feature or product) goes well with Google's business model, but wouldn't support a viable business mode by itself.
Paul Graham, September 2001:
> "And if you manage to write something that takes off, you may find that you were merely doing market research for Microsoft.”
Google is today's Microsoft :)
Ah, the classic Oracle trick. Good to see the tradition is alive.
Very annoying. Especially because by default every GCP service logs a bunch of crap to Stackdriver.
Our log cost was going to be as high as our compute cost.
My issue is with devs that write software that spits out a couple of gigs an hour and refuse to trim it down or exclude things.
It seems like they haven't learned anything at all. Sure it was google that shafted them in this case, but the problem isn't google, it's that their business is entirely dependent on a third party who has the power to do this. They rejected the option that would have prevented this from happening in the future with another company.
Oracle probably has a sales team on the way to their offices now.
I still have to understand openstreetmap mapping solutions. It's named open, not free, so I guess providing your own server with osm data would work. Only with FreeStreetMap the data would be infected also.
There's a great deal of competition and that generally results in a race to the bottom. The market leader just removed themselves from the market and the best way to capitalize is stay low and build your userbase.
We won't see price changes by any of these companies. They all want to be #1.
I'm not sure what you mean with "free"/"open". The OSM licence has a share-alike part if you merge your own map data with OSM data. If you just have a map and put points on it, it'd be fine.
[1] https://operations.osmfoundation.org/policies/tiles/ [2] https://github.com/openstreetmap/chef/pull/78 [3] https://switch2osm.org/serving-tiles/
OSM.org doesn't prohibit commericial usage, per se. But if you're running a business, you shouldn't rely on volunteers from another project which you're not paying for.
What do you mean it's evil? People call Google evil for so many things that are hardly evil at all, that the word has lost its meaning in the context of Google. For example I honestly cannot tell if you're calling them evil for increasing Maps pricing (thereby hurting maps-consuming startups) or for previously asking a low price (thereby hurting startups that offer maps).
I was under the impression the trust for Google evaporated a long time ago. Sure, alternatives can be few at times. But anyone committed to and in bed with Google has to be sleeping with one eye open.
I don't know how many bait and switch episodes from google I need to see but by 2008 I was done with them -- for the first time. Ad-nauseam.
If you boycotted and didn't use Google they would wither and die and stop being something to worry about: for immediate google specific free offerings that are sucker bait. Also for self aggrandizing standards/inclusions and buying the overt talent. But whatever. The HN community is full of dupes and buy ins, culture whores and fifth columnists. True belief in OSS died when everyone else took it and got rich.
We already have our own aerial tiles, built off the free NAIP imagery, and I just went through rebuilding the 2017 NAIP tiles last week to make them higher quality and fix some seam issues.
This combination is pretty good and lets us take control of our map tile destiny.
References I used in setting things up:
https://github.com/zavpyj/osm-tiles-docker https://switch2osm.org/ https://ircama.github.io/osm-carto-tutorials/tile-server-ubu...
I have set it up on a Centos 7 server using more or less the instructions from here https://switch2osm.org/manually-building-a-tile-server-16-04... (yes they are for Ubuntu but you'll get the idea) and everything works great. Even if you don't really need it, I recommend trying it to understand how it works; it has some very intuitive ideas.
Beyond the tile server I am also proposing GeoServer (http://geoserver.org) for hosting the geo points on the maps (it can integrate with PostGIS and various other datasources and output the geo points in various different formats). You can then use Leaflet (https://leafletjs.com) to actually display the map and points!
Much better than the NAIP aerial imagery, which takes around 3TB for the source data, 280GB for the resulting tiles, and around a full day to generate.
The download is an SQLite database with a standard-ish structure (MBTiles), you can either write a small server or use theirs.
I needed vector tiles in other projections — WGS84 equirectangualar, an Arctic one, and an Antarctic one. Generating those took many hours on a cluster.
With a commercial license, the tiles are from 2 July.
I assume the commercial tiles are a few days behind as it's just not worth most people's effort to stay up-to-the-hour — different map services can offer that, this company is making easy self-hosted map tile packages.
The free license is presumably using older data to encourage an upgrade.
However, as an example, the one-time $40 for Colorado is likely to be much cheaper than generating them yourself, and is probably adequate if you don't need to keep up to date.
Has anybody else been able to try this?
One interesting idea would be to have both raster and vector tiles; then in your client side code you could detect if the client is a mobile device or a desktop pc (I am not sure if you can detect more things about how speedy the device would be) and serve the corresponding tileset (ie the vector tileset if the device is desktop or the raster one if it is mobile). This way you will definitely decrease the load on your server and most clients won't have performance problems!
In any case, the clients will get faster as the time passes so after some years vector tiles would probably be the norm because they have the huge advantage that it is very easy to style them on-the-fly in any way you like (while, to style raster images you need to re-create them all from scratch).
would it be possible to learn more about your setup?
Thanks!
This results in having a tile server that generates & serves raster tiles using:
* carto to generate a proper stylesheet
* mapnik to create the tiles
* apache with the mod_tile module to serve the tiles
* renderd to be used as a bridge between apache/mod_tile and mapnik
* Postgres/PostGIS as a database to host the data
The actual data is an OpenStreetMap pbf file downloaded from http://download.geofabrik.de/ which is inserted to the database through osm2pgsql.
The above are used for the raster tile server only.
In addition to this, I am using geoserver (http://geoserver.org/) which is a Java web app (runs in tomcat) that is used to view/edit and (mainly) serve geospatial data. What I usually do is that when I need to display a new shape (shp) file I use shp2pgsql to insert it to a table in my PostGIS database (using something like this shp2pgsql -s SRID SHAPEFILE.shp SCHEMA.TABLE | psql -h HOST -d DATABASE -U USER) also another good tool is ogr2ogr which can convert between various formats (it also supports kml, for example ogr2ogr -f "PGDump" koko.sql test.kml).
After the shp file is inserted to the database, I can create a datasource for this dataset from GeoServer and publish this dataset to the web (i.e the dataset will have a URL of the form http://1.2.3.4:8080/geoserver/Dataset/wms); it can then be consumed by client applications (in javascipt) using leaflet or openlayers. Notice that GeoServer can easily publish various file formats for example an SHP directly (you don't need to insert it to the PostGIS; this is something I do for better control of my datasources).
PostGIS is the component that supports the vector tiles (http://docs.geoserver.org/latest/en/user/extensions/vectorti...) so I want to integrate it the OpenStreetMap data that is hosted in my PostGIS (which is used for the raster tiles); I haven't found the time yet though.
I think it's a great setup and it works great for the needs of the organization I work for; actually, I like it so much that I want (for a long time) to write a rather comprehensive tutorial on how to set ip up along with some notes on how to do various PostGIS things but I don't know when (and if) I'll find the time for something so large :/
I've been running a large OSM raster rendering stack for a while, for the whole planet. Moving on to a vector based stack right now, which is what you should do if you are starting today. The reason being the server, with vector tiles pre-rendered, can very quickly render raster tiles on the fly. It's fast enough that for most cases, there is no need to cache them. Once I get the new stack fully burned in, we'll swap our raster rendering stack out for the new vector stuff and do raster on the fly. I might cache the rasters depending on throughput, but from what I can tell it should be basically a non-issue.
Source: https://github.com/peermaps/peermaps
Also writeup about the architecture: https://github.com/digidem/osm-p2p-db/blob/master/doc/archit...
And then you need to think of how to do data updates. OSM updates minutely.
One major advantage of not updating your data very often is that the tile-images won't need to be re-generated thus you save a lot of (CPU intensive) resources! I think that another commenter mentioned that he generated the tiles in a CPU intensive system and then just off-loaded the images to not-so-fast system for serving.
How difficult would it be to build a competitor to Mapbox? Well, if all you care about are map tiles, it's no big deal. If you want to build a business around it, that's always work. You're going to need to build the tiles, have infrastructure to serve it, monitoring, staff to manage it, billing and charging infrastructure, etc...
This is the platform Maps.me is based on, correct? I ask because I downloaded Maps.me the other day, after reading HN, and have found it to be missing most businesses that I've attempted to search for. In addition, there seems to be no obvious way to suggest missing destinations.
I too want Google/Apple maps alternatives to succeed, but my limited trial so far has not been promising.
You might try looking at OsmAnd. It too uses OpenStreetMap data, but it does allow you to add some amount of missing data as well (see https://osmand.net/features?id=osm-editing-plugin).
This, of course, is part of the advantage of OpenStreetMap, you notice something missing, you add it, and now it is no longer missing.
Maps is based on Maps.me.
That's strange. I just checked Maps.me on Android, and if you open the hamburger menu there's an "add a place to the map" option.
I rendered all of Estonia from zoom levels 0 to 18 (~16.5M tiles) which is around 20GB and host the tiles on a cheap DigitalOcean instance.
* What is the recommended procedure for backup + restore?
I find a lot of projects in this space make you go through an manual copy-paste setup process like if it was a one-time thing - it is not - updating OSM data is a crucial part in this process and should be part of it.
I haven't investigated it because we are probably moving to Vector tiles over the next month or two, and we don't need up to the minute map updates. On one of our sites, we have been rendering navteq tiles and we only get updates for those quarterly. It was only our public site that we were using Google Maps for.
Backup and restore we are just doing via normal backup programs. The tilecache is stored in meta tiles, which is a d duplicated set of 64 tiles in a single file. So we are only dealing with about a million files in a directory structure totaling 25 gigabytes.
If we needed to restore or update our tiles, we would probably just spin up a new box and rerun ansible to deploy it. I have it all encoded Within a Playbook.
Living in Germany, I find the map quality of OSM much better than Google Maps. Sure, it lacks the integration with the other Google services like a good search and live traffic, but speaking about the pure map quality, Google is lacking a lot of detail here.
"Which maps should we investigate deeper?
Some options we could reject quickly for various reasons. OpenStreetMap is not supposed to be directly used by commercial sites. Apple Maps, even though we wouldn't mind having Apple logo on our site somewhere, was just released as beta and requires Apple developer account to test properly."
* what is the workflow to update the OSM data?
* what is your recommendation for a good geocoder?https://wiki.openstreetmap.org/wiki/Nominatim is an OK geocoder, definitely worth trying out at least. Though the pay services will definitely be better.
I'm seeing my own costs, for a site which I run on my own, rise from $1,500 / month to around $30,000 / month.
Their rationale is essentially that they did not want commercial users abusing their free tier. Before this change, they had been reaching out to such entities individually, offering their 'premium plan' service.
The pricing for the (old) premium plan was remarkably similar to the new pricing which now applies to everyone.
This pricing change has ended the era of Google Maps mashups / hobby projects.
Though, Mapillary is quite interesting but their pricing model is not really comparable to Google's one: https://www.mapillary.com/pricing
Also, it's a lame excuse because this affects paid users such as yourself who have nothing to do with the abuse.
In Google's defense, a hobby project wouldn't need 750k requests a month. At that point it's more than a hobby. But, the 28k limit is low.
Microsoft invested in it, released a product and the product wasn't as good as the market leader. That's not predatory.
There are several competitors, the issue at hand is they have an inferior product. If your argument held any water, literally everyone would use TMobile because they are so much cheaper. Reality is, a products quality matters and Google Maps were both more cost effective AND a better product.
We're going to need a bit more detail from an account that clearly has an agenda.
Edit: If you meant that you wanted more detail about my antitrust complaint, I contacted them via the antitrust email address. They took a while to call back, but not as long as Google took to get me that Maps quote...
I run a geocoding service for Australia specifically (https://geocodeplus.com.au) and a lot of people I talk to are completely unaware that they're going to see a x10 change in what their forward/reverse geocoding is going to cost them.
With auto complete especially, we would have had a 10-20x increase in pricing (and this is with an 'enterprise' plan pricing). As the article says, we were basically given 2 months notice for an enormous price hike that would likely render the business untenable. Luckily we were able to put our backup plan into action, but changes like this really erode the confidence to use Google (at least for mapping) in the future unfortunately.
Seems like now would be a great time to write a blog post on how you did it. :)
As voltagex_ mentioned, most location providers have a different API flow than Google Maps - rather than doing a total abstraction of providers (which would take too much effort), we went for a Google flow and non-Google flow basically.
For example, location autocomplete with Google requires a two step process (call for suggestions, then call for details of the suggestion) where a couple alternative apis provided the necessary data in the suggestion process. This requires a bit of a different flow on the server/client, so building support for both in was one of the hurdles we had to get over.
As for other takeaways, I guess mainly always have a backup plan for mission-critical service providers :)
https://gitlab.com/IvanSanchez/Leaflet.GridLayer.GoogleMutan...
I'm finding it really odd Google would just fuck their customers over this way. They must have known how it affects startups in particular. What were the conversations with Google like?
Just one-sided "This is the new price, take it or leave it!"? Or something else?
Why is it odd?
If you're on a free tier, you're not really a 'customer', just a user.
If you're a 'startup' you probably have funding. If not, you're not the sort of 'startup' that will fit in their calculations.
Anyone can use Leaflet together with OpenStreetMap for free* for heavy commercial purposes[0] if you make your own mirror/render of OSM tiles. Also simple reverse proxy cache[1] can keep public servers usage acceptable. Styles can be also modified at render stage, e.g. [2].
*The cost is only your one time setup work and then any monthly hosting bill.
I encourage anyone to prefer these open alternatives and of course donate money, time or hosting to help develop them.
[0] https://help.openstreetmap.org/questions/4669/can-i-use-open...
My recent experience in various parts of the UK and one trip to northern Finland is that the OSM is superior to Google maps anyway. And with something like OSMAnd it's trivially easy to use offline.
* is the proxy cache an accepted use case also for commercial projects? Where can I find details about that?
* what is your recommendation for a geocoder?
Also I would like to ask you about your procedure for updating the OSM data and tiles - I found this to be a tedious process. Not at all is this a one-time set-and-forget thing, it is quite a bit of things to understand and maintain.
The higher price allows them to actually get paid for the value they provide to various startups, which - as they pointed out in the blog post - would not exist without these maps.
Of course, the open source approach of OSM with mostly unpaid volunteers is again able to show its potential.
Everyone agrees gmaps has the best data imagery, plus street view - so why not make people pay for it accordingly. I really don't see a problem with this. On top of that people finally receive proper support from google once significant sums are involved.
Were there any significant competing maps APIs for the web in 2005? There may have been (and you seem so certain I'm sure you could name them all :P), but it's crazy to claim that there aren't far more competitors in the last 5 years than there were back then. Just look at the article, which is literally about all the competition available.
Don't forget about services beyond providing Maps APIs. Google Maps basically killed the need for a standalone GPS purchase (Garmin/TomTom) as well as purchasing upgraded maps for existing GPS systems, including those built into a car.
Am I complaining? No, but Google gave away for free something that once would've been hundreds of dollars in software. I'm surprised TomTom and Garmin weren't put out of business yet from it, but they survived on their niche products.
Vimeo has pretty much always had a shockingly* high price for their slightly-better plans, since they insist on covering their costs from subscriptions from the get-go. I'd be pretty confident they wouldn't bait-and-switch to grab some more cash (not the least since they're not the dominant video hosting provider). They compete on service and stability.
[*: in comparison to the other popular video hosting sites]
Will they? If my company is now paying $2k/month (where before it was under a free tier) am I really any more likely to get 'proper support' from google? $2k/month is not even a rounding error on a rounding error for google, and will be even less so if all the other users are also paying $2k/month as well.
It would be nice to think that, and I'll change my tune when I see it, but even after working with people who were paying for google cloud platform, I didn't think they we got terribly good support.
Not if I understand their terms correctly [1] [2]. Maybe there's a distinction between the terms of the map data itself and the software published to interact with it?
And there's several other places that offer this as well. So the advantage of building on OSM data and APIs potentially is that you could switch providers without losing API compatibility or having the data available change out from under you.
(transactions aren't defined anywhere I can easily find tho, so it's possible this isn't correct. but I don't really mind prematurely concluding that they're significantly more expensive if they don't clearly describe their pricing system.)
(Having said this, when I once had to do a one-off mapping project for a colleague, I found the OSM tools pretty daunting as a beginner, even though I appreciated how they'd reward effort with customizability and extensibility. I could certainly imagine that a lot of sites wouldn't be prepared to set up this toolset, and I realize that OSM doesn't provide a simple drop-in replacement for Google Maps that can be used by commercial sites.)
Support forums/forms are buried deep and inaccessible via Google search (or were), and for technical support you're required to use some obscure login to contact someone. People may or may not reply to you in at timely manner, and sales reps don't give a shit about you the second you sign the paper.
Problems with their APIs or errors are impossible to get priority for... so you basically pony up many thousands of dollars and you get a product you have to support yourself. Not to mention back in the day the Google Maps JS had a memory leak (we had to write our own pinning layer).
... I will note, I've noticed Google Cloud seems to be better managed with actual semblance of support but it's the only one I've had a good experience with.
I recommend Google's services to no one because the support is so abysmal and now the price is too (at least for maps).
>According to contemporary biologists the premise is false: a frog that is gradually heated will jump out.[3][4] Indeed, thermoregulation by changing location is a fundamentally necessary survival strategy for frogs and other ectotherms.
Off the top of my head, there are a number of companies/services that can serve as alternatives to Google Maps, and one way or another they all contribute to the collective quality of open source and open data map tools:
* geocode.earth (https://geocode.earth) has geocoding including autocomplete based on Pelias(https://github.com/pelias/pelias)
* Interline can help with routing, OSM extracts, and transit data(https://interline.io)
* Stadia Maps also offers map tiles and routing (https://stadiamaps.com/)
* GraphHopper (https://www.graphhopper.com/) offers routing and geocoding
* OpenCage (https://opencagedata.com/) offers geocoding based mainly on Nominatim but also a bunch of other geocoders
* Jawg.io (https://www.jawg.io/en/) offers maps, routing, and geocoding
Roughly speaking, open data allows for lower prices because investing directly in gathering proprietary data is super expensive. Open data coverage is getting pretty good in lots of areas, and is better than proprietary data in a lot of developing cities (like Jakarta). It can often be updated much more quickly. Google is notorious for taking a long time to fix errors.
Likewise open source map tools are definitely behind Google in many respects, but there's the advantage of transparency and ease of migration if one particular provider happens to go away, or customization is needed.
I'm sure I'm missing some services, and am happy to talk more about any of the challenges/features as much as I know.
Disclosures: I am a co-founder of geocode.earth, and a former employee of Mapzen, along with the founders of Interline
- Firebase suddenly changed its pricing, claiming that they were calculating prices incorrectly. A tiny multiplayer game that I made for kids were not fitting the free tier anymore, and my account was locked until I make payment. A sudden e-mail at 6am in the morning was enough to take an app down. Customer service? It doesn’t exist. I lost the data and moved away. The stress didn’t worth it.
- Kozmos (getkozmos.com) extension was taken down although it was a simple, open source bookmarking extension (github.com/kozmos) with good description, many HQ screenshots and also a video. I noticed it when users began reporting they couldn’t find the extension. They didn’t even dare to reach me out after taking it down. I sent so many e-mails, users mentioned Google Chrome on Twitter, no response was given. Not even a single response. Again, I had no contact to reach out and ask what is wrong and how we can fix. I was so frustrated that I even wanted to build a new browser and ship my extension with it.
- Someone began impersonating me using my full name on blogger. I reported it countless times providing fresh copy of my passport every single time, not even a single response was given. No confirmation, nothing. The blogger is still there impersonating me. My friends also reported it, nothing happened.
Here is three experiences. Based on this, I learnt to never trust and rely on any Google service. The last thing is e-mail and search, and I’ll celebrate the day I’m completely free from any Google services.
Meanwhile, I'm essentially banned from Google Wallet for sending a friend $600 with a card I'd had on the account for years, with no way of recovering it without giving Google my SSN, multiple utility bills and passport. This is far beyond what Know Your Customer regulations require, and I still was well below KYC thresholds that their competitors use (PayPal, Zelle (no apparent limits)).
Considering my bank doesn't have my SSN or passport, I am not handing that out like candy to the world's premier data mining company for a highly questionable account review. This right here is why I'm on Zelle :P
Doesn't Firebase have a daily quota, so your app just stops working for the rest of the day if it goes over quota? I though it would reset the next day.
Or in case of a storage quota you can't store more data until you delete some to go below the limit.
Did you try deleting some data, or they locked your Firebase access completely so you couldn't do anything?
There are a lot of people in the world so the odds of someone having the same full name as anyone else isn't that low. Was this person impersonating you specifically beyond claiming that they have the same name?
I don't want to be snarky, but I think many businesses don't think enough about this issue of value creation.
Working on a low margin business is a choice! You can always ditch it and try to find something with higher margins (= higher value created for users).
Google doesn't need to provide a service in that country, but it's understandable that a 20× increase is going to hit harder than in a wealthy country.
I am only pointing out that entrepreneurs need to rethink their business models if they are hit by this price increase, as this is a very strong indication they are not working on a healthy, high-margin business.
This is what happened to OP, and they were able to switch to haf a dozen competitors at a fraction of the price with almost equivalent service offerings for their use-case. The notion they have a fundamentally unworkable business model doesn't apply here.
I get the frustration and pain with this, but let's be honest, if your business was totally reliant on Google Maps it probably wasn't a great business to begin with. Way too many eggs in a basket you don't own.
There where so many "why should we pay/develop X, Google does it for free" conversations at one of my previous employers who competes in the mapping space with Google.
Without warning, the two maps suddenly stopped working properly and were covered in grey boxes. Looking for a solution in the documentation, I noticed the new pricing model (the email notice only came after their final deadline) and decided to try generating new API keys. Unfortunately, Google didn't update their API console in time and I was thrown into an endless redirect loop.
When I was finally able to obtain my new keys some days later, the two embedded maps refused to load with an unspecific "no longer supported" message. Google's documentation also didn't seem to be updated with any breaking changes.
I then decided to move to OpenStreetMap, which also looks much better in my opinion.
I was investigating the Google Traffic API for an app I was working on just as they were making this change. They went from something like 10,000 free requests a day (which would have been plenty for my purposes) to 0 free requests a day. "Don't be evil" my ass. Anyway, the Azure Maps (which uses TomTom data) is much more generous.
Edit: Added geocode.earth.
"A user has set a new billing account for your project in the Google Cloud Platform console. This initiated billing for your project, resulting in an upgrade to your project's Firebase plan."
Issue with that is that I never made such a change. So how could "A user" have made this change?
The resulting frenzy in inspecting admin and user-logs ( and checking if my account was hacked ) just found a new billing entry called "Google Maps Platform Transition Account" that was suddenly linked to this Firebase project.
And I don't even seem to have permissions to inspect it.
Very confusing indeed.
https://groups.google.com/forum/?#!topic/firebase-talk/jWRCd...
The response from the Firebase team [2] was that no additional charges would occur based on this incident, and they've opened an incident report [3] in their dashboard.
- [1] StackOverflow Thread with users facing similar confusion https://stackoverflow.com/questions/51373671/project-upgrade...
- [2] Official Firebase response https://groups.google.com/forum/?hl=en-US#!topic/firebase-ta...
- [3] Firebase incident report for issue "Firebase Users who use Maps API get auto-upgraded to Blaze" https://status.firebase.google.com/incident/Console/18016
No. Google maps is an absolute, washed out turd style wise. We have much better local competitor.
Compare this:
https://www.google.com/maps/@50.081478,14.4272906,13z https://mapy.cz/zakladni?x=14.4068130&y=50.0892746&z=13
You can try various zoom levels, and mapy.cz is always better. Much more detail. Much more contrast. Much more visual information. Tourist version of the maps.
I hated using Google maps when traveling in Spain. Sadly they are the recommended way to get public transport info. But I avoid them at all costs. I sometimes print maps when travelling and I can imagine google maps style would be more than useless when printed B/W on a shitty printer, while I could still use mapy.cz maps while low on ink.
That's the single most important point for the map.
I didn't consciously try to compare the maps before, but looking at the differences now, google maps seem thoughtless and completely random at what is highlighted, how parts of the city are named (some use names, some numbers, for no reason), what parts are shown and what are not.
There's a theme called OSM Bright that looks good, and when I figure out how to set it up, I plan to switch.
mapy.cz indeed has more data but visually, Google IMO looks better on my Windows PC. Maybe it’s high DPI support, maybe it’s font rendering, but small labels are barely readable in mapy.cz, compare the screenshots yourself: http://const.me/tmp/maps/mapy.png http://const.me/tmp/maps/gm.png BTW, Microsoft did even better: http://const.me/tmp/maps/ms.png
Not only does the mapy.cz version look terribly blurry compared to Google Maps, it doesn't do nearly as good of a job putting interesting waypoints in the right location (e.g. the label for "Prague" or the river name going along the river).
Google maps does terrible job by missing labels for half of the city parts.
Buildings are actually 3d at closer zoom levels.
You can click on a building to see the list of all businesses in that building.
Bus routes can be shown when clicking on a route number.
The best part, it works completely offline as a phone/desktop app.
Seems like there are several good local competitors to google maps.
They also have an API that's completely free also for commercial purposes (and they have coverage for Poland)
I have to say the satellite imagery on Mapy.cz lacks sharpness in comparison to Google Maps' satellite view. Google's view actually looks over-processed, which I think would lend it greater acuity when zoomed out. They look like they hold a comparable level of detail, just Mapy.cz could use a photoshop massage.
Also, Google Maps seems to have, at least from the small area I'm zoomed in at from your links, better location labels. ie: https://i.imgur.com/yUOlWO0.jpg
The non-satellite view I do like better on Mapy.cz, though -- it looks significantly more readable as far as structures and streets go, at every zoom level. More useful, maybe, more like a map ought to be.
It looks comparable with 3D view on mapy.cz.
Full disclaimer: I’m one of the Geocodio founders.
I mean, imagine a local grocery store who's rent would go up by 28x, 1 month notice. Or his procurement costs? That means literal bankruptcy or closing down his shop, virtually overnight. That's just not okay, not after years and years of service. You need to honour contracts for at least a year at a time and give notice. You can give two months notice on a 5% price hike, not a 3000% one.
Few other businesses operate in this manner, and those that do we tend to call shady, exploitative, inconsiderate etc.
Do they really have such contracts to begin with?
Why would they not increase the price, then?
Something I've realized as I've gotten older: sometimes you can't afford the best. That is why there is a market for the cheaper options.
The day that they enabled part of the new quota system my app stopped working (partially my fault for not reading their email) but it turned out to be the day before Election Day for an election app. I call in for support and they tell me that they are rolling out the new quota system and that I have to pay $10,000 if I need to increase my quota limit before the official new quota pay as you go system launches. After the discussion I find out I’m not even talking to google and it was a google partner. Somewhere down the line google just had a partner call me instead.
Long story short I changed the API key to get me past Election Day. Decided I’m moving off google maps asap as well.
I’m willing to pay but... can’t handle their constantly changing policies, their support was terrible, their ask for $10,000 for a short term slightly increased quota was absurd, their new quota system is too expensive.
Not to mention when I called to ask about Enterprise pricing years ago, it was $7,500/yr. 9 months later it became $25,000/yr.
I have no trust in neither Google Maps nor Mapbox.
I hail from hardware where, for instance, Apple chooses 3 providers for it's SSDs (SanDisk, Samsung, Intel, ...) to shield itself from price gouging, supply issues with any single party, business concerns, etc..
However, I routinely see entire companies based on single providers despite choice existing in the marketplace. The OP should have designed their service around multiple mapping providers from the get go, if a serious business relied on mapping
Yes you could create a compatibility layer for this, but really that seems like a lot of pre-work for something so unforeseen.
If you're under the assumption that Apple and Co don't spend the effort in abstracting out dependencies on their hardware, your deeply mistaken.
A significant portion of engineering in hardware circles is spent in this effort
What you are saying is already done in places where it can be, for eg, use of ORMs allow the application to be connected with a variety of DBs and your DB queries and application become independent of the underlying DB.
Also if it really is such a big part of your product, I don't really think $80 per day (for 15k unique users) as an operational cost is all that much honestly. And if you have a single page app, Google only charges for the first load anyways. They may have greatly increased the prices, but it does reflect the value they are providing and the tone taken in the article comes off as freeloading and ignorant of the effort level and value of providing such a service.
People also need to note that Google's javascript framework and embeddable maps is a different product than their maps.google.com, which seems to be higher quality both in the data and the rendering/features. The one they offer to embed on your site appears to be a much older version (for instance it doesn't support smooth zooming or 512x512 tiles). There are also TOS restrictions with what you can do with their javascript framework and widgets/API's, for instance you generally can't use other data providers or features from other mapping/GIS services on the same page.
What is very hard is coming up for a replacement of either geocoding/reverse geocoding, or routing; many solutions exist but most don't really work or are quite difficult to setup (IMHO).
Best of luck!
I am having trouble finding a good autocomplete/geo search solution though, curious if anyone has found a decent quality provider or solution you can self host like graphhopper.
Same for Google Maps: https://twitter.com/lucb1e/status/522491538912604160
Most of us here are from reasonably rich countries, and here Google is always fine, while OSM is usually fine (but not always). I am curious if you compare a few hundred random populated places on earth, how they compare. Where it is not economical to create a map, as in the tweet above, Google Maps is terrible. OSM is more often terrible in a country where they have just enough money to be worth it to Google, but where OSM just didn't happen to catch on among local mappers. You only need a few dozen for a country of a few million.
The Netherlands is truly the best country on OSM that I know of: 100% road and address coverage, and there are always people who love to jump on new things like a new bridge or tunnel. Somehow almost all the cyclepaths, trails, etc. are also on there.
Turns out, a grand total of 65 monthly active users maintain it.
I was very surprised. Apparently it doesn't take that much to maintain the map. If we get a few hundred people to work on OSM once a month in a given country of average size, it seems that you can maintain a map of a quality higher than Google Maps, at a tiny fraction of the cost (look at the expenses of the OSM Foundation), while being objective and free for everyone (scientists, businesses, consumers) to use.
The limitation is for the openstreetmap.org website, that is not dimensioned to be heavily used.
Yes, plenty of people do that. Be careful "OpenStreetMap API" is only for editing. There is other software in the OSM ecosystem which has it's own API (e.g. graphhopper has an API for routing). But sometimes people get confused and try to use the "OSM [editing] API" for all sorts of things.
This can be understood in two ways:
1. Cost of serving is the same from US to Africa (datacenters, network bandwidth, engineering etc), so prices are same
2. If #1 results in less adoption from developing countries, it is okay because they are not the focus anyway (i.e. SV companies are focussed on US/EU markets first). So getting rid of non-revenue making chunk of traffic doesn't hurt much.
Just like travel agents or passengers find ways to take cheaper flights from countries or city pairs that airlines price differently to help capture more market, even though it doesn't cost them very different to provide the service.
If they were willing and able to do so, they could already make a fortune by reselling Google Maps data.
Since that does not happen, I have doubts regarding your scenario.
If you say that you're in Khartoum when you're actually in New York, you won't get a very useful map back.
Where your unit cost for a product is essentially negligible (amortized against other GM customers and GCP scale infrastructure), why would you decrease exposure in a potential growth market?
The business answer is "because offering discounts cannibalizes revenue from higher-margin markets." But in this case... you literally know exactly where the user is located.
But is exposure really reduced? Google already has a free quota which is enough to bring plenty of exposure.
It was almost ready when Google changed the rules and started charging a lot for the service.
So I learned my lesson, and never again based a project (or a part of a project) on a tier service, because the rules can change too fast and ruin it all.
That what people are experiencing now with Google map.
Three parts of a map are copyrightable.
The map data is © OpenStreetMap under their open license, but OpenMapTiles claim copyright on the cartography of their tiles — since they aren't an exact copy of OSM, they decide what to filter for each zoom level, what to simplify for performance and so on.
Additionally, if you choose to use it, a map style is copyright: the design of showing motorways in blue at 4px and cycle paths in green, for example.
For the tiles you see on OpenStreetMap.org, all three parts are under open licenses. (And there are decisions at all three levels, for example OSM's database contains the water features you see at OpenSeaMap.org, but doesn't render them.)
What did you use to calculate your $1k price/month estimate for your website?
You increase prices over an order of magnitude. You give people warning. You give people credits to cover a few more months without paying extra.
What is the end game. If you just want extra revenue, why are you giving them the grace period to switch? Why are you giving service credits to give people another couple months to switch?
So if they want people to stop using Google Maps (businesses, not consumers), WHY do they want reduced use? Is there an outside source changing pricing or licensing? I don't think so, I think Google owns the data.
As I'm working through implementing google's suggested best practices and ways to cut down on API calls, it's still quite a shocking realization that each ONE map load (aka any user that loads our site) costs us almost a cent a piece (0.7 per Dynamic Maps load to be exact)
https://developers.google.com/maps/billing/understanding-cos...
The pricing tiers in this article do not include the additional levels of volume discounts which are available if your usage is high. You need to contact a partner or the google sales team to get details of those additional tiers.
The small fry like these kind of people are not google's real customers - the real customers were already paying to use the APIs before these changes happened.
The free limits before were hugely generous. There is a cost involved in offering the free service. If you weren't paying for it, then google was losing money. If you stop using google for free now because of this, then they actually save money and can now dedicate those resources to supporting actual paying customers for which there are many. Sure individual small users are a drop in the ocean, but multiply that up by many thousands and thousands...
Mapping is commoditized. There are many free and paid-for competitors as the article states. If you want the extra value that google offers and chose to stick with using them, then you need to pay for it now. I dont think that is unreasonable. If you dont want to pay for the service and continue getting something valuable for nothing, then go and leech MapBox/Here/someone else's free limits instead. See how long a small company without google's resources can support thousands of startups hammering their services & infrastructure without any revenue to pay for it.
Good luck.
Advanced users can totally change the whole map by switching on/off map elements (schema based on OpenStreetMap tagging)[1] or add own data (MBTiles format - both raster/vector)[2].
There is a support for multiple JavaScript APIs like Leaflet, OpenLayers, Mapbox GL JS (source code snippets in the administration), mobile SDKs (Android as well as iOS) and desktop software development (QT + Unity game engine)[3].
Price calculator is available at [4].
[0] https://www.maptiler.com/how-to/make-own-map-design/ [1] https://www.maptiler.com/how-to/completely-change-the-map-de... [2] https://www.maptiler.com/how-to/hosting-on-maptiler-cloud/ [3] https://www.maptiler.com/blog/2018/05/openmaptiles-gives-you... [4] https://www.maptiler.com/google-maps-platform-alternative/
Well yeah. If Google Maps is the most important component in your infrastructure (seems to be, based on the rest of the article), then it'd be surprising for it to not be the most expensive.
If all they do is repackage Google's data then I don't see Google as being particularly evil in this case, why should they give "the competition" a free ride?
I have to guess they do add something (inventory data maybe?) to make it worthy of going to their website though.
MapBox isn't perfect yet but I'm eager to see them deliver on engineering... it's been about a year since I've looked.
In addition to the $200 monthly free credit, all users get:
Free Maps usage for iOS, Android, and Embed
(for displaying maps only)
This change affects the following Google Maps Platform APIs:
Maps JavaScript API, Maps Static API, Street View API, Maps
SDK for Android, and Maps SDK for iOS.
[1] https://cloud.google.com/maps-platform/user-guide/pricing-ch...
[1] https://techcrunch.com/2015/09/01/google-introduces-uncompli...
Is this Google trying to move into a new space?
It's been a while since I had a GMaps enterprise plan, though, so I'm not sure if they're bumping the pricing on that side. At the time, the pricing was closer to the new prices they're rolling out than they were to the old API console billing prices.
[1] - https://techcrunch.com/2018/06/29/apple-is-rebuilding-maps-f...
[2] - https://electrek.co/2017/07/03/tesla-map-navigation-open-sou...
I looked at available options, and have been hesitant to switch to another. The article mentions Mapbox's pricing opacity. I tried Mapbox a while ago, to be surprised that map interactions (zooming and loading new tiles) counted as usage. I used up quite a bit in an afternoon of playing around with maps.
I'm going to try host tiles myself from next month. I've embarked on a Rustification effort where I'm seeing RAM and CPU savings by porting some services to Rust. Hopefully I'll have spare capacity to host map tiles.
"Get Country Data (GCD) A microservice that returns a country's geographical location information.
I developed Get Country Data because I wanted to reduce the amount of costly calls my apps were making to the Google Geocoding/Places API and I wanted to know additional information like currency names, currency acronyms, currency symbols (HTML entities), international telephone calling codes.
The data came out of multiple other projects I was working on. The data is not read from any external source or a local database but rather it's kept in-memory. The code base contains 42 472 lines of code.
Get Country Data is running in Google App Engine on PHP 5.5."
Would someone like to assist me in growing this?
Trying to look at pricing for internal use, it was always insane (starting at $10k/year, IIRC). The price change may make it more cost effective for small internal use cases, at the expense of open/free/public use.
Considering Google Maps is key in Google's business I wouldn't be surprised if this was actually intended to weed out small companies and only keep the big whales around.
I don't know the numbers but I imagine Google Maps app/web users are by far the largest traffic source. Traffic from projects using the API must be peanuts in comparison.
Business is business and if they can see an opportunity to make more revenue, they'll do it, whether it be Google or anyone else. It's just a matter of time.
I'm sure Google is mining all of that sweet, sweet GPS data from people using their navigation. I can imagine some machine learning of aeriel imagery and some sort of GPS path clustering helps them get a lot of the last mile.
Navigation also helps them get real time traffic data and allows them to learn how costly certain road segments are.
Slightly offtopic: I have been trying to get Google to fix the location of my flat on Maps for years now. I have reported it multiple times to them and it is still unfixed. Does anyone know a more direct way I can get in touch with them?
Apple
250K map views / *day*
Mapbox
50K map views / month
Google ($200 credit buys you)
28k dynamic map views / month
According to another comment in this thread who hosts a tile server him/herself on DigitalOcean, that costs about $40. Doesn't seem terrible even for a startup, either.
$100 seems like a bargain compared to those other options.
(disclosure: I'm a developer at Esri)
I’m wary of building anything on a proprietary JavaScript api, but if they provide tiles that can be used in leaflet or mapbox gl it might be an attractive option.
Beyond that use this form to request an increase in your limit: https://developer.apple.com/contact/request/mapkitjs/
As for pricing, nothing has been announced at this time.
You would be well-advised to reciprocate. To the extent that third-party services are essential to your survival, making efforts to bring it in-house, get long-term contracts, have backup providers, etc. is simply prudent.
Dependency is control. If you build your business in someone else's sandbox, it shouldn't be a surprise that they can hurt you at their whim, and absolutely will do so if they can make a buck.
Someone you pay to care about your stuff will never care about it as much as someone who works and maintains it themselves.
If I understand correctly, it's not that the database's licence prohibits commercial use, it's just that you're expected to host it yourself, right?
How is this a problem?
Gets rid of the Mapbox logo and is a really smooth experience.
What does this mean? How did Apple Maps use it?
All the while complaining about how expensive Google has become.
Hosting OSM is definitely cheaper than Google Maps, but it's a little ironic that they didn't consider this alternative solution.
It's not true. Ish. On openstreetmap.org there's servers that render the map data into the images you see. That's run by volunteers, on mostly donated hardware and hosting. There is no SLAs. If you run your big commericial app against that, and cause the load to go through the roof, the volunteer sysadmins will be sad, and will block your app. It's not commerical vs non-commerical. Big non-commerical apps do get blocked for misusing it too.
You can download the raw data from ( https://planet.openstreetmap.org/ ) and run your own tileserver and use that commerically.
You can use OSM for commerical sites. You cannot use the volunteer hardware on openstreetmap.org for big usage.
The cost to Google is the same, no matter where the user is.
What you're asking is basicaly that people from Germany to subsidise the cost for Poland by paying more, right?
I think what we will see in the maps arena is multiple-country specific maps sites popping up, like a map site specific to Poland for example. Physical businesses will then have a real hassle making sure they have a presence on ~10 different mapping sites, instead of just Google Maps.
Ironically it will make getting around Europe more difficult. I have lived in Poland and there are a lot of business absent from Google Maps, and a lot of cities not present on Google Maps transit.
And on the other side, big internet corps are implementing joke consent pages with tons of dark patterns that clearly go against the spirit of the regulation. I think they're pretty certain that they will eventually be fined, but they've done the math and found that milking customer data for the X years until the first fines are imposed is more profitable than complying now and not risking a fine.
What the internet is missing though is an alternate funding mechanism. When I visit a site I indirectly give the owner something like $0.002 (0.2 cents) worth of bucks from advertisers who show things to me. Why can't I just pay the website that as a microtransaction directly? Transaction fees would eat it all - but aggregated enough, and perhaps as part of a $10/month 'internet content subscription', it could work in the way that Spotify does.
There have been multiple attempts at offering that. The most prominent is probably the current iteration of flattr [1], but it seems really hard to get people excited for that model which in turn means it isn't worth the hassle of signing up for content creators.
Patreon is probably the closest to a working model, but that means a few people have to really love your service.
The worst thing about targeted ads is that we're building an insane surveillance infrastructure that the government is already eager to tap into, and will be even more eager if one of our governments devolves into dictatorship.
I'm from East Germany; and while I'm too young to have experienced it myself, the Stasi is still fresh in our minds. What we build today in the name of improved advertisement would have brought tears to Mielke's eyes.
The Stasi are fresh in no one's minds or you would eliminate this surveillance.
Certainly it can also be used to dispense useful information, but if someone is spending money to dispense that particular information, they have an agenda. It might be innocuous, "buy our product!". Or it can be malicious. Either way there is always intent and you cannot separate media, especially advertising, from the intent behind the media and the very goal of advertising is to grab (or accost or steal) your attention.
The internet already has alternate funding mechanisms: the host either eats the cost themselves and pays for it out of pocket, sells something to pay for hosting and other expenses, or users pay the host directly for access. Advertising is just one of at least four options.
It is not free. Your users pay for that with their privacy and Google gets your user's browsing history for free. It is unfair and you should not have used it in the first place.
I have refused to use dozens of service which required me to click pictures to prove that "I am not a robot" and sent the service provider an email informing them of my decision. I would urge everyone to do the same.
I've started noticing that if you're sneaky you can get a surprisingly large tolerance for error. I've no idea how effective that actually is at poisoning results.
There is an (admittedly a little nasty) part of my brain that really likes the idea that if I'm being asked to provide free labor for a proprietary company, then I'm not really under any obligation to be a great worker.
Just refusing to use the service is probably more effective though.
https://addons.mozilla.org/en-US/firefox/addon/decentraleyes
It's amazing how much you need to break free from tracking, but I'm on the war path :) Btw, taking the red pill will cut you off Facebook and Google accounts, unless you provide real phone numbers.
Matrix movie: "You take the blue pill – the story ends, you wake up in your bed and believe whatever you want to believe. You take the red pill – you stay in Wonderland, and I show you how deep the rabbit hole goes"
Google is actually devouring Internet as we speak, like a black hole:
- email. Keeping your own server unblocked by Gmail is a loosing battle. I use iCloud, but it still goes through Google for my Gmail and G-Suite peers.
- Search - monopoly. Analytics - monopoly.
- Android. Try make a phone with open source Android without Google apps and licensing.
- Companies and schools move all its people to G Suite (Google).
Ten years later 'Internet' will be replaced by 'Google' in dictionaries, as it happened to 'googling'. GSP = Google Service Provider (former ISP), GETF = Google Engineering Task Force (former IETF) :)
They did pretty nice analysis and found out their whole infrastructure costs $1300 a month and they would have to pay additional $5000 for Google Maps alone (right now GM costs them nothing).
Looks like GM will be ~10x more expensive to them than Mapbox or MapTiler, here's the original article (in Polish) translated by Google (sic!):
https://translate.google.com/translate?sl=pl&tl=en&js=y&prev...
Looks like many folks who leeched on GM's free plan will move to alternatives so maybe this dick move will turn around and became actually a good thing, would love to see more competition in the area and right now Google Maps is years ahead of competition: https://www.justinobeirne.com/google-maps-moat
How much of that $1300 is based on getting some freebies or low pices by staying below some usage threshold? Will they come upon additional costs on other services as they scale up?
Treating every map load the same even though one user can move around the map a lot, zoom out till he can see the earth, etc. is not smart. A much better metric is number of tile loads, so that cost corresponds with usage.
Few things:
* Who's your mapping provider?
* Why do I need to sign up to even see example code?
* Why would I use this over something like MapBox?
* What's your accuracy like in Australia?
At this time, we are a mix of mapbox + our own data.
I don't know why you need to sign up to see sample code. I'll give that feedback.
For mapping you'd use it over mapbox because it comes for free with your navigation package. For navigation you'd use it over mapbox because it has better turn-by-turn instructions, routes and traffic estimates.
The map is accurate throughout the world. The navigation product is specific to North America at this time, but it will expand.
I get that anything I do on my own is likely below the threshold.... but at the same time I feel like this likely will somewhat reduce maps usage and what if any bragging rights I might have with an employer (if I can find one ;)).
For me I maybe wonder if I'm going to make a hobby app that I should look at alternatives instead, or just quit dorking with Google Maps in my spare time...
Granted this is all very individual light stuff, but I wonder if it is a consideration people will have considering how dramatic and sudden the price change has been.