Slashdot Mirror
Google Will Block Access To Its Autocomplete API On August 10
An anonymous reader writes with news reported by VentureBeat that Google will be discontinuing developer access to its unofficial Autocomplete API, as of August 10 of this year. A snippet from the article: Google currently supports more than 80 APIs that developers can use to integrate Google services and data into their applications. The company also has unsupported and unpublished APIs which people outside the company have discovered and leveraged. One of those is the Autocomplete API. The company says it is making this move "in the interest of maintaining the integrity of autocomplete as part of Search," that it wants to "ensure that users experience autocomplete as it was designed to be used," and finally that "this provides the best user experience for both services." I'm sure many will disagree.
If you use undocumented calls you are all going to have a bad time mmm kay.
It seems every 'generation' of programmers gets to re-learn this lesson.
While the general sentiment of your statement is correct - given the plurality of services they have discontinued in the past - do note that this autocomplete API wasn't particularly "offered to the public"; it was never official or particularly supported.
Relying on undocumented / unofficial APIs always carries such a risk.
Except Google didn't offer it to the public. It is an unpublished API that is and was unsupported for external use.
I don't see the problem here. Don't rely on undocumented APIs.
Learning HOW to think is more important than learning WHAT to think.
Looks like I hit a nerve....
This isn't a Google "product", dumbass. If Google ever intended anyone to use it, then it wouldn't be an unpublished API. Complaining about Google pulling actual published APIs a applications is one thing, but Google is 100% within its rights to block access to people using their services in ways that they never authorised or intended.
Using Google service carries such a risk, whether publish, or undocumented, actived, core or non core.... Google will pull the plug at any minute. Their technology is ideal if you want built in obsolescence you are delivering to a customer.
Use Google API/service
Deploy at customer
Wait until it gets cancelled.
Redo with something else
Profit$$$
Except Google didn't offer it to the public. It is an unpublished API that is and was unsupported for external use.
I don't see the problem here.
Actually, they did offer it to the public. This was an undocumented API. However, like the undocumented maps API, it was exposed to the public. As such, it was offered, just not documented.
Don't rely on undocumented APIs
Google actually encourages people to experiment with their public but undocumented APIs as part of their strategy. However, however experimenting with and releasing a product based on it are two different things. Google has a tendency to throw things against the wall and see what sticks. Maps, definitely stuck and they could even monetize it. Likely, this API also stuck, or it wouldn't be news. However, it probably was being used in ways that they couldn't monetize. Which, is why double-speeak of trying to protect the integrity of what it was originally designed for (aka Google Search).
Of course, it is their API and nobody was charged anything to use it, so Google is free to do as they wish with it.
The problem is that because Google does it first and/or best and/or "sufficiently free for adoption", there tend to not be any well known competing products. As such, everyone ends up relying on Google offerings "by default" and doesn't scramble to create replacements until their hands are forced.
Of course maybe this means that its a good investment to build alternatives to all of Google's offerings, just waiting to take an onrush of new business the moment Google loses interest in them. Then again, that's probably far easier in theory than practice.
Typical sexist reaction.
Girlcottt Google!
There, all balanced up again.
A real example: PICASA. Before Google bought it, there was a healthy market for local Image|Media Management Software. Picasa was free (and decent) there were better ones though ---- or at least software that had actual options --- All of them died and are gone, except for a couple majors.
Or Email clients. There was Opera's M2 - dead. And I found "PostBox" last year, but well f' them. It's based on Firefox with "free updates between major versions". Bought in September 2014 - and not a single update was released.... until PostBox 4 - June 2015... with new icons and bugfixes. Pay Again. No Thanks.
Comment removed based on user account deletion
The reason people hate the "new" Google is because they actually liked the OLD Google, that is pre IPO.
The old Google was like this mad scientist company full of engineers just throwing out all these cool ideas and seeing what people liked, the "new" Google cares only about the stock price and I have zero doubt is no longer being run by engineers but by MBAs (Masters of Being Assholes) who show each other PPTs and say things like "our data shows that the crucial 19-35 demographic currently enjoys X", see how they tried to ram G+ down our throats because their data showed kids like Facebook. The old Google would have never done that, hell you used to have to fight for an INVITE to get to play with the newest Google stuff and people did, because it was nearly always cool and innovative takes on some idea, now its just another Charmin, a large corp cranking out products based on marketing data and that makes a lot of us sad pandas :-(
ACs don't waste your time replying, your posts are never seen by me.
No, you just failed at formatting.
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...