1
00:00:00,120 --> 00:00:01,480
Thanks for watching Techquickie.

2
00:00:01,480 --> 00:00:02,760
Click the subscribe button,

3
00:00:02,760 --> 00:00:05,040
then enable notifications with the bell icon

4
00:00:05,040 --> 00:00:07,400
so you won't miss any future videos.

5
00:00:07,400 --> 00:00:09,800
It used to be that looking something up on a map

6
00:00:09,800 --> 00:00:11,880
meant unfolding a huge piece of paper

7
00:00:11,880 --> 00:00:13,160
you kept in your glove box

8
00:00:13,160 --> 00:00:15,240
and driving past the turnoff to Des Moines

9
00:00:15,240 --> 00:00:18,880
in the time it took to find section A53.

10
00:00:18,880 --> 00:00:21,240
Fast forward a few years, and now thanks to Google Maps,

11
00:00:21,240 --> 00:00:24,720
you can zoom in so far on nearly any street in the world

12
00:00:24,720 --> 00:00:26,760
that whether you want a bodega in the Big Apple

13
00:00:26,760 --> 00:00:29,720
or just a regular apple in Manhattan, Kansas,

14
00:00:29,720 --> 00:00:31,520
you're not bound to have a problem.

15
00:00:31,520 --> 00:00:33,500
But this unprecedented level of detail

16
00:00:33,500 --> 00:00:35,200
didn't happen overnight.

17
00:00:35,200 --> 00:00:36,820
Google managed to get a foothold

18
00:00:36,820 --> 00:00:39,600
in the online mapping scene back in the mid-2000s

19
00:00:39,600 --> 00:00:42,140
when it acquired WhereTo Technologies,

20
00:00:42,140 --> 00:00:44,760
who had the idea to let people scroll around on a map

21
00:00:44,760 --> 00:00:46,500
without having to refresh,

22
00:00:46,500 --> 00:00:50,180
a novel concept that helped set Google Maps apart early on.

23
00:00:50,180 --> 00:00:52,960
But they needed more than just a pretty interface.

24
00:00:52,960 --> 00:00:56,080
Fortunately, Google is first and foremost a data company,

25
00:00:56,080 --> 00:00:58,020
so they often start by working directly

26
00:00:58,020 --> 00:00:59,700
with national and local government offices

27
00:00:59,700 --> 00:01:03,240
to determine not only exactly where every road is,

28
00:01:03,240 --> 00:01:06,460
but how lanes are spaced and allocated on freeways

29
00:01:06,460 --> 00:01:09,000
to give an accurate base experience.

30
00:01:09,000 --> 00:01:11,320
Unfortunately, oftentimes the government won't

31
00:01:11,320 --> 00:01:14,120
or isn't able to provide a completely accurate picture

32
00:01:14,120 --> 00:01:16,320
of how an area is laid out.

33
00:01:16,320 --> 00:01:18,100
Oh, you.

34
00:01:18,100 --> 00:01:20,940
So this is where Google Street View comes in.

35
00:01:20,940 --> 00:01:22,680
You can learn more about that up here,

36
00:01:22,680 --> 00:01:26,160
but hold on, isn't Street View just a way

37
00:01:26,160 --> 00:01:28,460
for users to virtually sightsee?

38
00:01:28,460 --> 00:01:29,660
The answer, as it turns out, is no.

39
00:01:29,660 --> 00:01:33,640
Google also uses Street View internally

40
00:01:33,640 --> 00:01:35,380
in order to double check and tweak

41
00:01:35,380 --> 00:01:37,060
its existing mapping data.

42
00:01:37,060 --> 00:01:39,820
As the Street View car takes photos of its surroundings,

43
00:01:39,820 --> 00:01:43,600
special AI can recognize any signs caught by the camera

44
00:01:43,600 --> 00:01:46,260
and send them to be analyzed by Google's servers

45
00:01:46,260 --> 00:01:48,180
to determine not only lane placement

46
00:01:48,180 --> 00:01:49,660
and traffic restrictions,

47
00:01:49,660 --> 00:01:51,180
but among other things,

48
00:01:51,180 --> 00:01:54,420
even the proper addresses of local businesses.

49
00:01:54,420 --> 00:01:56,500
All of this information is then also used

50
00:01:56,500 --> 00:01:59,380
to prevent Maps from giving you bad directions by,

51
00:01:59,660 --> 00:02:01,740
for example, sending you the wrong way

52
00:02:01,740 --> 00:02:03,160
down a one-way street.

53
00:02:03,160 --> 00:02:04,420
That would be bad.

54
00:02:04,420 --> 00:02:07,160
Adding another layer to Maps is Google's access

55
00:02:07,160 --> 00:02:09,420
to a huge amount of satellite data,

56
00:02:09,420 --> 00:02:10,760
partly from its acquisition

57
00:02:10,760 --> 00:02:13,700
of satellite imaging service Keyhole in 2004,

58
00:02:13,700 --> 00:02:15,800
and partly from government sources.

59
00:02:15,800 --> 00:02:18,100
They use top-down views of street markers

60
00:02:18,100 --> 00:02:19,820
to further improve accuracy,

61
00:02:19,820 --> 00:02:21,240
and they've even got to the point

62
00:02:21,240 --> 00:02:22,880
where they can use satellite imagery

63
00:02:22,880 --> 00:02:26,140
to extrapolate the shapes and footprints of buildings,

64
00:02:26,140 --> 00:02:28,640
not only to give you 3D views in Google Earth,

65
00:02:28,640 --> 00:02:29,660
which are super cool,

66
00:02:29,660 --> 00:02:32,340
but also to enable you to see outlines of these buildings

67
00:02:32,340 --> 00:02:33,940
in the regular Maps view.

68
00:02:33,940 --> 00:02:36,120
So by intelligently combining data

69
00:02:36,120 --> 00:02:37,420
from all of these sources,

70
00:02:37,420 --> 00:02:39,700
Maps can construct quite an accurate picture

71
00:02:39,700 --> 00:02:41,920
of any given patch of land pretty well,

72
00:02:41,920 --> 00:02:46,540
but even Google can't completely automate things, yet.

73
00:02:46,540 --> 00:02:48,300
They also use human employees

74
00:02:48,300 --> 00:02:51,720
in a project called Atlas to make necessary adjustments.

75
00:02:51,720 --> 00:02:53,340
One of this team's common tasks

76
00:02:53,340 --> 00:02:55,800
is to investigate user error reports

77
00:02:55,800 --> 00:02:57,600
and tweak routes on the map layer

78
00:02:57,600 --> 00:02:58,840
that clearly don't match up

79
00:02:58,840 --> 00:03:01,820
with a recent satellite image or street view capture.

80
00:03:01,820 --> 00:03:03,340
And this kind of user reporting

81
00:03:03,340 --> 00:03:06,300
reflects how crowdsourcing is another major piston

82
00:03:06,300 --> 00:03:07,520
in Google Maps' engine.

83
00:03:07,520 --> 00:03:09,240
And I'm not just talking about the photos you see

84
00:03:09,240 --> 00:03:10,680
on business listings, either.

85
00:03:10,680 --> 00:03:12,920
Google relies partially on volunteers

86
00:03:12,920 --> 00:03:15,440
to map areas it may have limited data on,

87
00:03:15,440 --> 00:03:16,880
something that's especially important

88
00:03:16,880 --> 00:03:18,000
in less developed countries,

89
00:03:18,000 --> 00:03:19,940
where it's harder for Google to send vehicles

90
00:03:19,940 --> 00:03:22,220
or get accurate mapping data from a government.

91
00:03:22,220 --> 00:03:24,320
Of course, Google uses other tricks, too,

92
00:03:24,320 --> 00:03:26,080
to flesh out the experience.

93
00:03:26,080 --> 00:03:28,120
Once upon a time, they licensed databases

94
00:03:28,840 --> 00:03:32,400
to dramatically expand how many businesses Maps would show,

95
00:03:32,400 --> 00:03:34,320
but these days, listings are pulled

96
00:03:34,320 --> 00:03:36,660
from other public sources and from business owners

97
00:03:36,660 --> 00:03:38,080
who are eager to take advantage

98
00:03:38,080 --> 00:03:40,520
of the visibility that Google Maps provides.

99
00:03:40,520 --> 00:03:43,520
Owners of large facilities have even submitted floor plans

100
00:03:43,520 --> 00:03:45,860
and granted Google permission to do site surveys

101
00:03:45,860 --> 00:03:47,520
to help you find your way around.

102
00:03:47,520 --> 00:03:49,840
This is particularly useful for sports arenas,

103
00:03:49,840 --> 00:03:51,500
shopping malls, and airports,

104
00:03:51,500 --> 00:03:53,660
and Google even works with a third-party service

105
00:03:53,660 --> 00:03:55,460
that keeps track of gas prices

106
00:03:55,460 --> 00:03:56,920
by looking at credit card data

107
00:03:56,920 --> 00:03:58,620
and even getting direct data feeds

108
00:03:58,840 --> 00:04:00,720
from major gas station chains.

109
00:04:00,720 --> 00:04:03,560
Finally, whether you realize you've signed up for it or not,

110
00:04:03,560 --> 00:04:05,800
Google also uses you.

111
00:04:05,800 --> 00:04:08,020
Location services and the GPS on your phone

112
00:04:08,020 --> 00:04:10,760
helps Google understand, on top of where it is,

113
00:04:10,760 --> 00:04:11,600
how to get there,

114
00:04:11,600 --> 00:04:14,200
and how many square feet of patio space is available,

115
00:04:14,200 --> 00:04:17,300
what times of day your favorite watering hole gets crowded.

116
00:04:17,300 --> 00:04:18,560
And they don't show you,

117
00:04:18,560 --> 00:04:20,460
but they probably know the demographic makeup

118
00:04:20,460 --> 00:04:22,020
of the clientele, too.

119
00:04:22,020 --> 00:04:22,860
Wow.

120
00:04:22,860 --> 00:04:25,460
So the real answer to how does Google Maps work

121
00:04:25,460 --> 00:04:28,660
is that Google simply has so many information gaps,

122
00:04:28,660 --> 00:04:29,940
and gathering tentacles,

123
00:04:29,940 --> 00:04:33,100
that if they were the star of a Japanese anime,

124
00:04:33,100 --> 00:04:34,860
I probably wouldn't watch it at work.

125
00:04:34,860 --> 00:04:36,840
Are you interested in computer science?

126
00:04:36,840 --> 00:04:38,020
Then check out Brilliant,

127
00:04:38,020 --> 00:04:39,340
a problem-solving website

128
00:04:39,340 --> 00:04:42,260
that teaches you to think like a computer scientist.

129
00:04:42,260 --> 00:04:44,160
Instead of passively listening to lectures,

130
00:04:44,160 --> 00:04:45,480
you get to master concepts

131
00:04:45,480 --> 00:04:47,700
by solving fun and challenging problems,

132
00:04:47,700 --> 00:04:50,540
and Brilliant provides the tools and framework you need

133
00:04:50,540 --> 00:04:52,060
to tackle these challenges.

134
00:04:52,060 --> 00:04:53,840
Brilliant's thought-provoking content,

135
00:04:53,840 --> 00:04:56,000
based around breaking up big complexities

136
00:04:56,000 --> 00:04:58,320
into bite-sized, understandable chunks,

137
00:04:58,320 --> 00:05:01,540
will lead you from curiosity to mastery.

138
00:05:01,540 --> 00:05:02,540
And you'll be in the company

139
00:05:02,540 --> 00:05:04,380
of over five and a half million members

140
00:05:04,380 --> 00:05:07,280
who share your curiosity and love for math and science.

141
00:05:07,280 --> 00:05:09,180
So what are you waiting for?

142
00:05:09,180 --> 00:05:11,700
You can support Techquickie and learn more about Brilliant

143
00:05:11,700 --> 00:05:13,500
by clicking the link in the video description

144
00:05:13,500 --> 00:05:15,960
or going to brilliant.org slash Techquickie

145
00:05:15,960 --> 00:05:17,220
and signing up for free.

146
00:05:17,220 --> 00:05:18,760
The first 200 people to go to the link

147
00:05:18,760 --> 00:05:21,540
will get 20% off an annual premium subscription,

148
00:05:21,540 --> 00:05:23,200
so be sure to check it out.

149
00:05:23,200 --> 00:05:24,620
That's it for this video, guys.

150
00:05:24,620 --> 00:05:25,460
Thanks for watching.

151
00:05:25,460 --> 00:05:26,800
Like the video if you liked it.

152
00:05:26,800 --> 00:05:27,600
Dislike it if you dislike it.

153
00:05:27,600 --> 00:05:29,740
That's pretty straightforward.

154
00:05:29,740 --> 00:05:30,980
Check out our other channels

155
00:05:30,980 --> 00:05:34,600
and comment below with video suggestions if you have those.

156
00:05:34,600 --> 00:05:35,940
Subscribe if you want to.

157
00:05:35,940 --> 00:05:38,560
Follow us on social media or don't.

158
00:05:38,560 --> 00:05:39,660
It's your life.