Telling the world's smartest company how to run their business.

Because why not, right?

At any rate, Google's navigation tool is terrible and they should invest a ton of money to retool it.

It knows where you are, which is a good start.  And when you tell it where you want to go it knows how to get there, which is nice.

But if you want navigation, it opens up a whole new app.  That should be integrated with the maps to make it much easier to toggle in and out of navigation.

However, that's not the problem with Google Maps/Navigation.   No, the problem is that the other functions of Google Maps don't understand the concept of "on the way." 

Say I'm in New York City (I'm in New York City).  And say I want to go to Amish country.  I can pick my destination or my origination or any point along the way and find gas or a restaurant or antique stores.  But there's no way to tell Google to find gas "on the way."  Not only should I be able to do that, Google should ask me how far off the selected route I'm willing to go for gas and then it should reach into one of the cheap gas apps and tell me the optimal place to stop. 

And that's just the basic stuff.  I should be able to tell Google that I'm going to Amish country because I'm antiquing and I'm planning to visit.  Google should then identify antique shops and visitor-friendly farms which are "on the way," again defined by me. 

If you're traveling with your kids, you can ask Google to identify kid-friendly restaurants near your location.  What you should be able to do is ask Google for kid-friendly restaurants that are along your route and which you can get to within the next 40 minutes or whatever. 

Just in case it's not sufficiently arrogant to tell the world's smartest company what they should be doing, here's one way how they should do it.  If they don't feel like developing the expertise in house at least the beginnings of it already live at Garmin.  They've been doing a version of this for years.  For $12 billion Google can pay a 20% premium to Garmin's current stock price and exceed the 5-year high, two important triggers to get board support for a friendly deal.  Along with the "on the way" ability, they'd pick up a lot of great topographic data and other things useful for non-car GPS devices.