I'm confused by the replies saying that more dense labels would somehow harm usability for driving or navigation. When you're driving, you know where you're going, the map knows where you're going, and it's easy to see the route you need to take. When you're not on a route, denser labels help you to orient yourself with the map and to know when to zoom into a particular location.
That said, I also think the focus on paper maps is misplaced. Old-style road maps had to answer the question, "how do I get there from here?" New-style digital maps don't need to answer that question any more! Questions new-style maps need to answer include:
* I know the name of a place or street; where should I zoom in to see more things around that place or street?
* I need to go to a (gas station / rest stop / hospital); where's the closest one?
* How would I get home from where I am?
* I'm in an unfamiliar place and would like to go "downtown" (where there are restaurants and things to do); where is "downtown"?
* Where is my car right now?
Roads help you to orient yourself with the map, but they aren't as fundamentally important to digital maps as they were to old-style road maps. The visual space of the map might be better spent helping answer questions like these.
This kind of crowded display is fairly common now. I use Here Maps now which is much better + I like supporting a second option to avoid one service getting 2 monopolistic.
My theory of why google maps is getting worse is there must be designers dedicated to this that need to constantly look for 'new' things to add even if its to the detriment as they can hardly say its all good now and twiddle their thumbs. Maybe that's unfair but I've seen this effect on other digital products first hand.
Indeed. I'm that idiot driving in circles around the roundabout because Google Maps gives me the turn right as I'm passing it.
Also, it's forking is completely fucked. If there are 2 right turns coming up it does not indicate which lane I should be in. In once case I had to drive 60kms, 30 kms there and back, because the interchange had 3 right exits and I chose wrong.
Now I use Google Maps to find the location and copy the coordinates into Apple Maps. Or use Waze.
The big problem is that routes used to be useful without turn-by-turn navigation - back when there were bubbles on each turn you could click on to see. Now you're stuck trying to read a long list of directions, or you use turn-by-turn and have no idea where you're going until the last second before each turn.
I agree. Apple Maps and Here maps are both a much better navigation experience now in my opinion. Google maps still has a bit of an advantage on POI, but for 98% of my uses Apple and Nokia do a better job.
That seems oddly bad for being by Google and their skilled engineers. It's really just about factoring in your speed and distance... We've done this sort of thing ourselves in our apps. :/ It wasn't obvious in theory as for what parameters felt "just right", but these insights came soon enough after just a few trial runs in different speeds.
For me this is not a problem. I glance at the top left of the screen to see the direction of the next turn, left or right, then I get in the left or right lane. Also, when I miss a turn the rerouting is fast enough it's hardly a problem.
My big complaint is that Google maps announces I'm on the fastest route even when I know I'm not!
Maybe they are taking notes from Waze. Waze by default is so crowded your entire map is just icons of random stuff you don't care about. Then every once in a while half your screen is filled by an add that you need to dismiss, while driving, to see where you are going.
It's ads like, there's a KFC en route to my destination.
Touch here to dismiss, touch there to set as destination. It takes like 30 seconds for it to automatically disappear. If I'm on a route I'm familiar with I just let it disappear...but times where I'm going somewhere new I frantically lean forward to my dashboard to dismiss their ad.
Which platform for you use for Waze? I use Android and have never seen an ad while moving. Sometimes, the ads show if I am stopped at a stoplight, but they disappear when the car begins moving.
There are also one too many buttons when you choose a route. If I want to drive, then I type in the address, select it, press a button to activate it, then press a similar button to start the route.
Personally I like this UX. I often pullup the directions to a destination without actually starting turn-by-turn navigation: I'd like to see whether Google recommends taking highway X or Y, and then I can drive the rest of the route myself.
One thing I still use Google maps and the GPS one on Android is to learn an area. For that, it is incredibly useful to map the real-world and digital context of where you are in relation to your surroundings, not just have it laser-honed on your route.
I've noticed when purely following GPS directions that I can't remember how to get there, even after a few times following it. Knowing the area helps you take your own informed alternate routes, know what things are on the way that you might want to detour (oh, I needed to go to the post office anyway), recognize if the GPS is going to the wrong destination because of user error, etc.
To expand on your second point, I believe Google Maps is optimized for the user to be following directions. Labels are irrelevant when you've got a blue line to follow.
In their perfect world you don't orient yourself with the map. You tell the map what you want and it tells you how to get there.
Actually I had to switch to another provider due to sparse labelling, need for zooming/unzooming especially during navigation where you have inconsistent connectivity.
Google maps labeling is wrong and dangerous for drivers due to the need to keep your finger there.
Google Maps, like several other google products (like Groups, even Search), has seen a constant degrade in user experience since 2005 or so.
Maybe people forgot, but google maps was /blazing fast/ in the beginning.
Nowdays, it brings my browser down to a crawl even before the images are shown. Maybe people are just stuck with this "google is the best" mentality, but this has stopped being universally true since many years.
Use OpenStreetMap. It's data is way superior. It's _your_ data. Cannot strett this enough.
Want a fancy browser? Nokia maps have always been incredibly sleek to use:
https://maps.here.com/
Heck, even bing maps are /so much faster/. The imagery is also higher quality in several regions.
Google has still the lead with street view, but for the actual maps I really encourage you to look for alternatives. They've destroyed their interface as far I'm concerned.
OpenSteetMap is very difficult to compare to Google Maps as it uses pre-rendered raster map images at fixed resolutions, rather than a dynamic vector-based approach. For me at least, this makes it far slower to load in non-cached areas, and the lack of precise zooming is super annoying.
For me Google Maps is slower than OpenStreetMap anywhere in the world, cache or no cache. Might have something to do with not using Chrome, where Google makes their websites work the best of course.
What do you mean by "the lack of precise zooming is super annoying"? In my experience, zooming in Bing Maps and OpenStreetMap are very fast, and only in Google Maps you're waiting ten seconds for the region to load and browser to respond before you can zoom further.
On Google Maps, your zoom factor is any real number within a range. On Open Street View, you are stuck using one of several predefined zoom levels (integers 1-20).
Google maps also changes the zoom level to the nearest integer. You can't do 12.5 for example. When I change the 10 on the google maps link in your comment to 10.5, it rounds it back down to 10 and rewrites the URL. Have ran into the issue before doing custom maps for sites where 14.5 would be the perfect zoom level, but can only pick 14 or 15 :(
There's vector stacks using OpenStreetMap data, I guess it will be a while before one is in use on openstreetmap.org.
Mapbox has one, Mapzen has one, various apps have their own vector map implementations, there are convertors that take OSM data and output maps in the custom vector formats used by GPS devices and so on.
Google News on mobile has become an awful experience on mobile. They hijacked the scrolling so it doesn't keep momentum like every other scrolling page in the system, and if you accidentally scroll downward with too much of an angle it swipes to a new section. I'm amazed every day it still exists and has not been fixed!!
So I am not crazy after all! The other day I was driving in Palm Springs, and I was using Google Maps. I literally had to zoom in until the road I was on almost completely field the screen before it would show me the name of the road or the roads around it. They did something to their display algorithm where you now have to zoom almost entirely into an area to see anything about the area, very inconvenient.
It drives me crazy that I can't find half of the labels I'm looking for, and when I do, I can't increase the size of the labels enough to make them readable. Frequently I'll have to fight with the map, zoom in and out and scroll around to get it to show me the name of a street at all, but when I finally do coax it into revealing a tiny name far downroad, the font is often too small to read. If I reflexively expand the map to get a closer look, everything expands except the label I'm trying to read.
And there's no way to expand it. I do understand that they don't want it to keep expanding beyond a certain point, especially as they add more details as they zoom. But they could let it expand within some constraints, such as the name of a street within the stripe of the street, and cap the size at 3x the normal size or some such thing.
If the algorithm for letting the labels grow a bit is just too hard for them, then just give me a little slider on the side to add a fixed, default label magnification factor. I could expand it a bit to make the labels easier to read in general, expand it more (but just temporarily) to get a better look at something, or shrink it a bit in cases where the labels were harder to read due to overcrowding.
As it is, they've removed too many useful labels without expanding the ones remaining to make them easier to read and prevented me from expanding them manually. Bother.
He, came here to post the same thing. Since the last 2 or 3 years I had the impression GM went worse (for what I do with it, not even mobile) after every update made to it, and sometimes also slower and more CPU hungry. To the point I stopped using it and went looking for alternatives.
Am I the only one who likes to know the road before my turn? That way I know when I need to change lanes/pay closer attention/whatever. New mapping apps really don't seem to want to share that information with me.
- OpenStreetMap / OsmAnd for general use (about 75% of the time);
- Google Maps for traffic info, street view and businesses (about 20% of the time); and
- Bing Maps for areal imagery (about 5-10% of the time).
Google Maps is always extremely slow, but their business listings are very comprehensive due to their extensive marketing, and traffic info and street view are also expensive things open source cannot really rival.
Bing Maps runs at the speed of light when zooming and panning around the map. So much better than Google Maps, but the map's quality is not great so I just use it for areal imagery (another thing OSM does not have).
OpenStreetMap is as fast as Bing Maps and its map quality is better than both google and microsoft if you disregard business listings (at least in western Europe and the middle of nowhere[1]).
Yep, I find this extra annoying when I'm on the roads and have a bad internet connection, or no connection at all (common when I'm traveling in the less-populated regions of the US). I have to zoom in several levels to see the road, but often I can't due to the connection. This is really frustrating.
I'm surprised the author didn't mention the main difference between paper maps and google maps. That is, google maps is interactive. With a paper map what you see is what you get. You had to cram as much information as it would allow. But this is not the case with google maps. You can zoom in, out, and anywhere in between. You can't compare the two based on the level of information displayed at one fixed zoom level because google maps is 3D whereas traditional paper maps are 2D.
A funny thing about interactive maps while driving. If I'm on my phone and I bring up a road map of the area I'm traveling and I am constantly zooming in and panning the app so that I can see locations around me then the map has failed me.
Landmarks are key to any map, but landmarks that fade in and out of focus are not all that useful. I used to be excited by the prospect of being able to bring up a map of my route and being able to see where I was going on my phone mounted to my dash. But as of late, I've been struggling to find the right zoom level that shows enough detail of the area I'm traveling while showing enough of my route.
On a number of occasions over the last year I've has to pull over and reorient myself on my map due to a failed pan/zoom attempt.
It's funny that this article came out at this time as I've been evaluating ways to mount a larger device (tablet) on my dash as maps on my phone has gotten to be rather cumbersome.
I feel that for the most part the details in this article are accurate, that the attempt by google to make the maps load quicker on mobile have compromised critical details available on the maps.
One of the key areas where this could be addressed is by loading details based on need. For example if I select a travel route between two locations, load more of the details related to that route and reduce the extras that fall outside my concern. Show me roadways that leave my target route, as well as the cities and towns along my route. Making an attempt to provide me the details I need without my need to interact with them as much as possible would be great.
yes I have had a similar experience except needing to zoom in so the Map displays a road name as not all road names are displayed when zoomed out past a certain point.
OpenStreetMap itself has no notion of layers. It has a rather simple data model for geometries and completely freeform tagging (so for example there is no technical enforcement of how a road is labeled a road in the main database). Naturally there is quite a bit of effort to use tags with clear shared meanings, but none of that happens in the database, it's the people building the editors and doing the editing that choose what the tags mean.
Consumers of OSM data generally do perform a data extraction step where the freeform tags are grouped together and perhaps normalized to some extent. So at that point you sort of start to have layers, but different apps will use different systems and rules for that step, so the layers are coming from the app maker, not from OpenStreetMap itself.
OpenStreetMap is just a database. It has no concept of layers (except the `layers=` tag used for stacking order of objects e.g. bridges). Products which use OpenStreetMap data are free to apply whatever "layer" abstractions in their UIs they like.
Maps, especially when used for real time navigation, are the poster boy example of what Bret Victor means when he says "Interactivity Considered Harmful". [1]
Interactivity is a failure state of software that could have predicted what the user wants from context and past behavior, and tailored the display and presented it to them without being asked.
"Further, the user might prefer to learn information while using her hands for other purposes, such as writing or eating or stroking a cat [or driving the car]. Each time software demands the user’s hands, this activity must be interrupted [at the risk of causing an accident]. Finally, the growing prevalence of computer-related repetitive stress injuries [and using smartphones while driving] suggests that indiscriminate interactivity may be considerably harmful in a literal, physical sense."
That is not true. A good paper (road) map has three zoom levels. 1) A major map covering most of one side of the paper, with enough information to get you easily across the focus area of the map. 2) Often on the side a smaller, sparser macro area map, showing only the freeways and the major cities/towns, often spanning a slight larger area than the main map. 3) A few downtown/focus maps, that are far more detailed than the main map, showing more rodes (sometimes walking paths, and bike lanes, house outlines, etc.). Often on the side along with the macro level map, or on the back next to the index. And 4) a really good paper map has also a macro level public transportation map (often on the back as well) at a zoom level between (1) and (2).
Yes, the interactivity (search, zoom, pan, satellite, topo, traffic, earth, streetview) of google maps acts very much like other dimensions to the map and makes it QUALITATIVELY different from paper maps.
The complaints of the OP are perfectly valid but they're more a matter of taste than an example of bad design decisions.
Maybe the OP would be satisfied with a couple of sliders?
Less Streets-------------|--More Streets
Less Cities-------------|--More Cities
>But this is not the case with google maps. You can zoom in, out, and anywhere in between.
The problem I have Google maps lately, is that i have to zoom so far in to see the names of some roads. To the point where nothing but the road is on screen before the name appears. It's a major UX issue.
Not only zooming, Google Maps also has search. If you're looking for Oakland, CA, you go to the search bar and type "Oakland, CA", and it shows up on the map. No need to scan the map for the thing you're looking for, that's the computer's job!
I guess the main complaint of OP is not about finding a particular location, but to create a mental map with the area of concern. To create mental map efficiently, the map has to show large enough to cover the area of concern, yet has to show enough landmarks to associate with. It's crucial to have both information simultaneously in a single sight---if you have to do something to see one information at a time it becomes very inefficient.
That said, I suspect that there's a fundamental difference of processing maps among people; I like to store a map in my head and trace mentally. When I drive, I prefer to see the map always north up, so it's easy to synchronize with my mental map. However, most car navigation systems orient the map as your direction---I find it very uncomfortable, but apparently that's most people want.
I really wish someone would add an east-is-up mode, since that's the traditional orientation for maps — and the root of the word 'orientation' to begin with.
Because the sun rises in the east. At least in Indo-European cultures, which were once reliant solely on agriculture and transhumance, the sun was extremely important. Without a compass, it's also extremely convenient, being the easiest celestial body to spot. As such, ancient Indo-Europeans stood facing east when performing religious rites. The reconstructed proto-Indo-European words for "north" and "south" are also the words for "left" and "right" respectively, because the north is to your left when you're facing east, etc.
I was going to a location yesterday that you couldn't search for. "The junction which splits the road going to A and B". I was very confused by the new Google Maps which essentially didn't show enough detail for a good overview and orientation.
Search is of course entirely useless, worse than useless, if you don't know in advance what you are looking for. Needless to say, search is not an adequate substitute for intelligent design and proper levels of detail on maps.
Can you explain your use case here? Normal Google search is great for finding the name of films you can't remember by typing vague descriptions of the plot. Maps has similar functionality, like "local coffeeshops" or "<business> in <city>".
My technique: When I'm in an unfamiliar city, I search for restaurants and then looks for clusters. When you see a cluster of restaurants, this tends to be a downtown and is usually somewhere 'interesting'.
> If you don't actually know what you're even looking for, NO MAP whatsoever will be able to help you.
I use OpenStreetMap for this use case so I can only answer for them, and they seem to do a great job at it. If I am looking for the general shopping area, or office areas, or anything like that, I can zoom and pan through a city and find it very quickly. That Google doesn't show labels anymore makes it completely useless (and I've run into that a few times).
You can't zoom out if the information is only showed on a tight zoom. You need to keep zooming in and out.
Here's an example of how tight the zoom has to be to get eg bus stops to show up. For this example the user has to zoom in really close to get the bus stop shown, then either very careful start zooming out while keeping an eye on where their bus stop was OR keep panning.
Good luck trying to zoom in/out interactive map while driving motorcycle at 150 km/h.. I stopped using Google Maps for navigation years ago. I'd like to hear Google's explanation for their maps design decisions, because in my eyes (and all people I've talked to about this topic) they look spectacularly bad. I literary haven't encountered a single person that would say: "Google maps are cool" or "Google maps are getting better and better". Most of the times discussion about maps/navigation ends in sharing experience about alternatives. Lately I use Navigator on Android; it's simple, but gets the job done and as a bonus, supports offline navigation.
That's true, but they seen to cram in roads, at least visually. I would understand it if they optimized the display for navigation, where voice is the primary interface, and the display only is for support. For example, if the voice says "take exit X to Y" Y must be on the map, regardless of its size. And I would think (?of course?), it should not just jump onto the map and disappear a few seconds later.
So, if you plan a route, do the maps change depending on your destination?
There's a broad trend of tech companies disrupting traditional industries and then making rookie mistakes and generally not living up to the standards of the tradition.
Google, hire cartographers. Amazon, hire librarians and typesetters. Spotify, hire musicologists.
Isn’t it more a problem of having algorithms do what used to be done by humans?
I.e. when you look at a Google map it is using the latest GIS data to construct a “good” representation for your screen resolution using your specified zoom factor and possibly even with dynamic overlays or highlights.
Likewise, when you read an e-book you can read it on many different display sizes, pick your own font, and font size. So where we used to have the book typeset once by a person, it is now done on-the-fly using user and device specific parameters.
The result won’t be as good as when you have a human do it, and even when a human does it, there might be limitations in the e-book markup language, just like with HTML.
That's part of the problem, but you can at least design the algorithm with the benefit of expertise about how humans do it, either by studying the field yourself or hiring people who do. The way Knuth approached algorithmic typesetting is a good example.
Strongly disagree with that. For one thing, Spotify already has an army of tastemakers spending all day assembling curated playlists. For another, they have the Echonest data which relies heavily on manual labeling.
But the major innovation in Discover Weekly was to use machine learning directly on mel spectrograms to figure out meaningful features for human taste. They still want to rely on their experts as much as possible (and hey, I don't blame them for not wanting to fire people), so they try to combine their expert's features with the algorithm's. But this introduces human biases again.
The problem is that when it comes to music, everyone's a missionary. Everyone wants the world to listen to the music
they are excited about. Professional opinion-haver about music is the dream job for many adults, much like chocolate factory taste QA expert is for 6-year olds. And they just can't separate their own opinions from objective truths very well. It's hard to be objective about something you love.
The real great thing about AI in recommendations isn't really the intelligence part. It's the "AO" - artificial objectivity. The algorithm is probably inferior to humans in some aspects (it can't interpret the themes in lyrics very well, for instance), but the advantage is that you have full confidence about
1. What information it actually might use, and
2. What it tries to optimize.
From point one, you can be sure that it's opinion on Smashing Pumpkins isn't affected by that annoying kid in 8th grade that used to listen to them. For point two, you can be sure it's really trying to find the music you will love, not what it thinks you should love.
To get it slightly back on track: I can't wait until an AI can do music history, or etymology, or economics, or history. Or matchmaking in dating! It will be useful long before it can match humans on intelligence. How great wouldn't it be to get results in those fields which you could trust were from a disinterested party.
I don't imagine the musicologists charged with personally recommending music.
A more important task would be to restructure the "information architecture," for example to improve the experience of looking for classical music or jazz. There's a lot to do that isn't just based on opinions.
The dream of AI doing music history seems kind of bizarre to me... as well as the whole idea that human knowledge is bad because it's biased...
Information architecture of music is what Echonest was all about, and I'm pretty sure they keep doing what they were doing when Spotify bought them.
To the degree that judging good jazz or classical is different from judging other types of music, I think that yes, it's based a lot on opinion. In particular the opinion of authorities - critics and other performers.
It's not that this is entirely unreasonable. With music as a social phenomenon, you might prefer to not be "into" the wrong kind of music, even if you would like it for the music itself. Spotify and Echonest have actually talked a bit about how listening patterns can reveal "shameful" tastes, different from the tastes we would like to project.
The job of a recommendation system then, if we should look cynically at it, is to show you the "right" kinds of music that you would like to like, but actually like too - and to not tempt you with "wrong" kinds of music that you would like despite yourself.
And yes, I'd bet you'd need musicologists (or human analysts) for that. It would by definition be very hard to figure out from listening patterns or acoustic features. But isn't this a bit cynical as I said? Shouldn't we try to not be ashamed of what we actually like?
Well, I assume Google has a big staff of skilled people working on GIS, but I'm curious to know what kind of input they get from people trained in the art of cartographical mapmaking...
My hunch is that the tech industry is bad at taking advantage of this kind of traditional trade knowledge, and the essay in question here is an indicator.
As for Amazon, the Kindle e-books very often have horrific typesetting that would be laughed out of any traditional book printing shop.
And as for Spotify, their catalogue mechanisms and information design are clearly awful from a musicology standpoint.
Something that has become interesting to me recently is the amount of knowledge about maratime weather and sailing that isn't readily accessible in a some kind of neat database, app or website. You actually have to talk to people and wander through museums to find out some of it.
It's an example of an industry and body of knowledge that still grows and is passed around outside of the internet.
Weather patterns for a particular area, places to anchor and dock, safe passages through reefs. You can't always google the answer and much of it is gleaned from other sailors who used experience and their tools to figure it out.
Having grown up pretty much constantly online, with the answer to any question I have being reasonably available with a quick search; that some common information about the world can still be discovered and shared is is really fascinating.
It sometimes feels like there isn't much outside of STEM fields left to explore.
> Something that has become interesting to me recently is the amount of knowledge about maratime weather and sailing that isn't readily accessible in a some kind of neat database, app or website. You actually have to talk to people and wander through museums to find out some of it.
I've been online and windsurfing for decades and I've watched online info about weather/wave conditions at various locations grow then peak, and now it's been declining and drying up in recent years.
It used to be curated and published on websites and talked about in public discussion forums. Even usenet and mailing list archives dating back to the early 90s would appear in searches. In the last 5 yrs or so it's all been slowly evaporating into ephemeral Facebook group posts hidden behind a login form.
I suspect this is happening to all sorts of other communities too as they drop below critical mass. I'm not thrilled about it.
Weather patterns for a particular area, places to anchor and dock, safe passages through reefs. You can't always google the answer and much of it is gleaned from other sailors who used experience and their tools to figure it out.
Lots of this information is collated in Almanacks and pilot guides (eg. Reeds Natuical Almanac and the Shell Channel Pilot for UK sailors, plus the books from the Cruising Association and Royal Cruising Club). There's not a lot of incentive for an online version, as the information needs to be accesible offline for reference and in case plans change. However, there is a lot of information in apps: there's an iPad version of Reeds Natuical Almanac [1], Imray has a worldwide Tides planner app [2], and there are apps with charts from both the UKHO and Imray [3].
The best application I've found that can be used offline is ActiveCaptain. The database is free and actively updated. The mobile app is garbage. An old Adobe AIR concoction that sometimes crashes or freezes arbitrarily.
Other apps I've used that have a slick interface are unfortunately online only. Which is useless to me. When I'm coming into a new, unknown, anchorage I need the information now, not after I've already tied up.
> My hunch is that the tech industry is bad at taking advantage of this kind of traditional trade knowledge, and the essay in question here is an indicator.
As is the fact that other than Knuth, no-one seems to have learnt the first thing about how to lay out a page of text in an attractive fashion. And then even given Knuth, essentially no-one bothers to use his magnum opus (I honestly think that TeX is even more important the The Art of Computer Programming, and is what Knuth will be known for in two centuries).
The way I use Google Maps a lot is for discovery. I look at the map of New York to see cities around it. When I look at a more zoomed in level of the city I want to see different boroughs and major roads. More zoomed in you want to discover shops and businesses. The bareness of the current Google Maps makes it very unsuitable for these functions. For example, even at full zoom level it only shows a few (<5%) of the shops and bars at the city center where i live. If i want to get a feel for a city (where are the most restaurants, where are the shopping centres, etc) the maps are really bad for that unless i go searching for the specific terms in the search bar. But that is the down side of the search bar: you never finf something you didn't know you were looking for.
There's much more wrong with Google maps on the desktop than what the article mentions.
My biggest gripe is contrast, rather the lack thereof. Zooming in and out doesn't help. There's a lack of contrast at all levels!
And the algorithm for displaying place names sucks. You'll see certain names at one level, zoom in and they disappear, zoom in some more and they finally reappear.
Paper maps are unquestionably more ergonomic (but much less convenient) than Google maps. But it's not just Google. I find other online maps equally bad. It's quite sad that a paper Rand McNally map is so much better at actually presenting the geography of an area.
Perhaps other posters here are right, it seems like Google maps is designed for point-to-point navigation, nothing more.
The font is so small I had to zoom in 3 times to 150% get a comfortable size. And then zoomed in one more time to 175% to make the font a little thicker - as you say, due to the light color. The irony made me giggle a little.
I wouldn't mind too much if the website styling didn't break zooming in the browser. The paragraph width is fixed as a percentage of the viewport, which means zooming in only makes the font bigger, keeping the same narrow column for its contents.
On the whole the HTML and CSS on that blog are a mess though.
Firefox's Reader View helps here by bypassing the author's styling completely.
Fully agree. That's why I generally browse internet on mobile in Opera which supports text reflow and force-zoom, contrary to Chrome which only supports the latter.
Google know more than the author about how Google Maps is used by end users. The author is grading Google Maps based on the number of cities and roads displayed, not how the users use it.
Google provide an alternative mapping product, Google Earth, for satisfying curiosity about the planet. Google maps is primarily a navigation tool. They have very distinct use cases.
I'm sure Google knows more than the author about their own maps, but please give the author his due: he wrote a successful blog called 41Latitude that did a lot of in-depth analysis of Google Maps's display of information, and later led the Apple Maps team. If he has something to say about information density on maps, I'm inclined to listen.
I don't think the fact that the author prefers the map he designed means anything about which one is actually better or more usable, regardless of his credentials.
Don't you think it's possible to benefit from someone's expertise while recognizing and taking into account their inherent bias? It's generally difficult to find the former without some degree of the latter.
I'm not sure that puts him in a position to be trusted to criticize google maps. Apple maps has been considered inferior to google maps since its launch.
Just because he has ideas about how to pack an overwhelming amount of information onto a map doesn't mean it's a good idea for normal users.
Apple Maps was considered inferior at launch, not since its launch. There's a difference. Apple Maps caught up pretty fast. There's obviously areas where Google Maps is still better, but there are areas where Apple Maps is better, and it's simply not true to say that it's an inferior product today.
Apple Maps has actually significantly improved since its launch, and I wish people would give it its due. I, for one, often prefer Apple Maps to Google Maps nowadays.
My main problem with Apple Maps has always been that it just shows too little information on the screen, exactly what the author now criticizes about Google Maps. That doesn't seem to have changes with the current Apple Maps.
My main (remaining) issue with Apple Maps today is it does not give lane information for upcoming maneuvers. Google Maps does this and I find it invaluable.
Don't mix up data and map design. Google maps is great because the data is unique, but I personally prefer other designs and think that Google Maps has very bad UX.
Apple Maps today is much better than Google Maps. In the context of the article, for example, Apple shows all the cities/towns that Google Maps did in 2010 but no longer shows in 2016.
I think the author's point is that as a navigational tool a more balanced map would be superior. I agree. Even on a mobile device I often find myself frustrated at how far in I have to zoom on a Google map to find details that could be included at higher levels. And as he illustrates, the map could be more balanced without being cluttered.
Google Map's directions doesn't work between postal codes in the UK, unless you specify the space. And it hasn't worked for about a year. I must say, I find Google Maps quite difficult to use.
You could try alternatives like OpenStreetMap. In the Netherlands, Germany, France and nearby countries the map quality is often better (and more up to date) than Google Maps, but I don't know about the UK.
For navigation there are separate websites that use OSM data, like yournavigation.org (not sure that's the best one, I just know of it). They often work a bit differently than Google Maps so there might be a slight learning curve.
There is also Bing Maps which has great areal imagery if you're looking for that, only their map quality isn't as good. Certainly not terrible either, though.
I think it's safe to say that the Google Maps team hasn't cared about end-user convenience and utility for quite some time. The Google Maps of 2016 is utterly terrible compared to the same service ten years ago.
The new maps look clearer and less cluttered and more useful to me. I would guess that the maps were simply designed to follow their primary utility function which is navigation.
Old style printed maps had cities on them, because the map didn't know where you were going! You had to find your city or your location on the map. Now the map knows where you're going, so it can show that place extra-clearly while hiding a lot of detail that's not relevant. Roads are relatively more relevant than cities, since you travel along them to get from one place to another: displaying a road shows the user that they have a primary thoroughfare between locations. You might not care about the name of a city if you're just passing through; and the city that is your destination will be specifically shown.
My guess is that they display only as many cities as needed to help people orient themselves while looking at the map, to understand what they're seeing. More than that is irrelevant to the primary use-case of navigation.
> Google Maps of 2016 has a surplus of roads — but not enough cities. It's also out of balance. So what is the ideal? Balance.
The ideal is utility, and the key use-case for Google Maps at that zoom level is driving navigation. The user's going to input their own destination into maps anyway, most of the time, and they'll expect it to appear, so it's no surprise when it does.
Google would have data on this: how many users use Google Maps while driving regularly, multiple times on a trip (at that zoom level), while not having a destination entered (and with no destination, obviously no turn-by-turn directions)? Probably not many. Now imagine overlaying your route with current position and destination on the maps - it's going to be easier to scan the new ones. Edit: Navigation is the primary use-case for a map, and I'd guess usage motivated by that purpose dwarfs the rest by an order of magnitude, and so it's a good default.
What you write is true for navigating when driving a car. For those who try to use Google Maps for other purposes, the changes are somewhat less useful.
I'd actually argue the contrary - viewing the map in transit is for the most part useless (not to mention dangerous if you're also the driver), and I'd assume most people rely on the audio cues rather than actually using the map. I would also guess that users would perhaps only look at it either prior to departure to get a general idea of what their route is like or when they're very close to their destination to look for something like parking or other landmarks to help navigate once near the location. (e.g., trying to navigate a new city and dealing with one-way streets)
You could be planning a trip - deciding where to stay for example.
I personally have found the new google maps very frustrating and slow - I'm constantly zooming in and out. This article finally made me realize what I'm missing, and why I'm having so much trouble with it.
Google Maps is used as the tiling layer for many non-navigation web apps: FlightRadar24 for example. Or those "find us" boxes on organisations' websites. In both those examples the map is used for geospatial orientation not navigation.
A week or two ago I was trying to determine what cities were underneath a flight path. It was particularly difficult, whereas it wouldn't have posed a problem prior to these changes.
It's not surprising that paper maps often do a better job than Google maps, especially when you consider that people have been optimizing the format for several hundred years. For instance, you can't beat a decent topo map for browsing terrain looking for new places to go. Having a large map with high information density is just the ticket.
The place where Google maps really excel is applications that combine a search result of some kind with geographic data. For ordinary navigation, especially if you want geographical context rather than just directions from point A to B, traditional maps still are much better in my opinion.
Agree. I have no doubt the decision was based on lots of data and that Google got it right. And for my part I agree. I would imagine 95% of the people zoomed in around NYC are looking specifically for NYC. Having other, smaller cities labeled on the map only makes it more confusing.
If they are looking "specifically for NYC", what is so confusing exactly? Also your "95%" seems way off. Also the concernsus here seems that Google has all the data, they know best. This is a very naive look at things.
Google is a data company and the data available has changed over the last few years. Maps are both a way to present data and a way to collect data.
The new data that Google has comes from Android handsets and from users using Google maps and Waze on Android and non-Android handsets.
This data is all about users in motion. At the scale shown in this article, it's almost exclusively people driving. As a result, it makes a lot more sense to focus on the connections over the places they connect. This becomes clearer when you view the roads as more active entities by including congestion and other real-time data.
This may not be the best presentation for everyone, but it seems to be the presentation that fits best with Google's current mission and capabilities.
The difference between today's Google maps and the authors 1960 paper map is easy to understand when you stop to think about how those maps are used.
The paper map was used to navigate from place to place. That never happens with a google map. Sure, you navigate with them, but by telling Google where you want to go and letting them draw a line on your map. You don't need all that extra information if your phone is navigating you from place to place. You just need something clean that you can glance at to get a sense of where you are.
So that's what they've designed their maps to give you.
Keep in mind that the tiles shown by default on the OpenStreetMap website are just an example rendering. Anyone can take the OpenStreetMap data and use it to render a map with different cartographic choices.
There's even several other example renderings that can be easily accessed on the website, for example
The Google part happened in Maps. It was a mapping product, and now it's a geographical fronted to search. The v2 was all about services on a cute (and sluggish) rendering substrate. It's now usable these days, still way slower than OSM or bing. I miss the old presentation but alas ...
ps: I recently discovered the 'my timelime' feature. Surprising to say the least.
This criticism reminds me a bit of when people criticize Google search results for a query like "insurance" or "shoes" having too many ads. Searching for vague terms is useless, so they just display ads instead.
At the zoom level the screenshots are taken at, maps are essentially useless. The most important information they can convey is "there's lots of roads here" or "this region is densely populated". The maps aren't optimized for accuracy, they're displaying a summary. The long island example really struck me - the old map displayed the primary route only, the new map conveys the fact that there are multiple options. If you're stuck in traffic and you pull up the map, you can see there's another decent route and ask the app to provide you with directions on an alternate route. If you're using the old maps, you'd just see the single primary route highlightedand assume you should stick with the route you're on.
It's the speed/lag that really annoys me nowadays. If I pan across the map or zoom in, I know I'm waiting a good 5-6 seconds before the new tiles will be loaded. And that's on desktop, not mobile. It's so irritating I've just about abandoned google.
Google really won't care about any comments here, they are all about the data. So simply stop using Google maps. Tell your family to stop using Google maps. Tell all your friends to stop using Google maps. Blog about not using Google Maps.
Find alternatives, there are several mentioned in these comments for starters.
When/if Google start seeing a reduction in their map use, only then will they start paying attention.
Right, giving away too many labels at once would make it too easy for people to copy the map data. By requiring much higher zoom levels, people would have to do many more requests to grab everything.
Not applicable at the zoom scale used in the article, but on higher zoom levels this is certainly a factor.
There might be something in that, yes. Obviously I don't know the details of Google's internal tech, but certainly that's true of the OSM equivalents: Mapnik (produces raster tiles) is much better at label placement than Mapbox GL (renders vector tiles).
Do you mean they're focusing more effort on reducing the complexity of the maps, to save bandwidth, because vectors derive much larger bandwidth savings from reduced complexity than bitmaps would?
I've noticed the disappearing detail from Google Maps as well, and I find it really annoying - especially when perusing a rural area and having to zoom in to ridiculous levels to see town names.
Another thing I'd like to see is making the "avoid tolls" setting easier to get to. Northern Illinois is toll road central, and I-355 in particular is a huge ripoff when you pay cash. Since I don't need any of the tollways for commuting, I can't justify getting an I-Pass.
It doesn't even remember the setting, you have to hit the three dots for menu before you search, choose "Route Options", then tick "Avoid tolls", then hit OK. EVERY TIME.
And if you forget to go through this and start navigation, there's no way to change it, you have to hit back, then do the above. Why not have a way to change route options from the navigation screen??
With open-source tools and services for OSM data from Mapzen and Mapbox, you can make your own map styles: light or heavy in detail, highlighting cities, highways or footpaths.
I suspect part of the idea is that you will use the search to find cities instead of finding them visually based on their names on the map. The issue with large cities that are near other large cities being omitted from maps is 'The Baltimore Phenomenon'
I've been noticing this but could never quite put my finger on what it was. This is exactly it.
On google maps I can never find what I want. I thought it was because I've been using OpenStreetMap, and had gotten used to a different display style. Seeing a place once in OSM anywhere in the world, and zooming out from it to continent view, I can almost always find it again later. On Google Maps I always got lost. Now I finally get what the problem is.
I noticed the same thing a few months ago. I live in a larger city, largest in our county, but we don't exist on the map unless you really zoom. A small town right next-door that's a lot more affluent shows up even when zoomed out to the tristate region. My guess was Google is stepping up their advertisement business for cities.
That or people just find whitespace aesthetically pleasing and Google designers went kind of crazy with it.
I've only ever been using Google Maps for the sattelite imagery, street view, and routin. For those things, Google Maps is still great. OpenStreetMap is what I use for orienting, find cycling paths, trails etc.
So I'm not too bothered by this change. What I don't understand is why hasn't anybody taken the Google Maps routing and use it in OSM apps? Might not be legal, but similar non-commercial projects it should be fine.
There's a lot of effort right now going into making OSM itself better for car navigation. (Take a look at the OSRM changelog: pretty much every feature in the last 6 months seems to be targeted to this use case.) And I'd say OSM is already better than Google for cycling and walking routing.
While the analysis is fine, the conclusion is almost certainly false. At Google's scale, nearly any iteration made on core products is backed by an immense understanding of their end users and a near unlimited supply of user data.
There can be a million things wrong with user testing, and while I am certain they know more about it than me, I think it's a safe bet to say they're not perfect either and may make mistakes in such a complex topic.
(Small example of what I mean: many people like the new style because it's prettier, but when I use OSM I get to places a lot faster than people using Google Maps, so they might be asking the wrong question even if they did user testing with a million people. But that's just one example of everything that could go wrong.)
> At Google's scale, nearly any iteration made on core products is backed by an immense understanding of their end users and a near unlimited supply of user data.
The "Google, where did I leave my keys?" joke is getting closer to feature status, and that's probably what's going on with Google Maps.
Anecdatum: last month I was driving out for a weekend in some rural bungalows a few miles outside a small city (Elvas, Portugal). The address was a bit vague, the place name too common, so much so that I had GPS coords stowed away in a message pic (don't ask ;-).
So, when I pop out Google Maps in the old faithful iPad2 (which happens to be the 3G version, and therefore GPS chipped, good for navigation), and zoom in the area ... amidst thin local roads and lots of blank space, there's the place name and the days we're staying there.
Turns out the Gmail app in that iPad was also used to send or review the emails with the reservations.
Even in my 'desktop' I've been noticing Google Maps marking out city places which seem small compared to other landmarks, but where I often go or mention in emails.
(Thanks, I do know where I left my keys today, I'm good.)
"The primary route across Long Island — Interstate 495"
Off-topic, but as a native NYCer, we would never call it that. It's the LIE. I once had a woman ask me in the parking lot of a Walgreens how to get on the 278. I was puzzled for a second, then I realized she was talking about the BQE. Living out in California now, I miss the days of calling highways by name.
I think what Google is doing makes sense given smartphone adoption and the fact the maps are use for fundamentally different functions now.
You don't need a hulking great map with loads of detail at a high level to get from point A to point B, you now just use your smartphone for that.
I assume Google spotted a trend of people searching place names as opposed to picking points between two separate locations.
So how does that change the function of the map?
Well, we no longer need to have the zoomed out overload of detail, if we need more information about a place we are visiting, we type in the city name, or address, then zoom in close to see the detail we need.
The article kind of skimps over the point that we can interact with those maps now.
"Less is just less. And that's certainly the case here."
Not sure I agree, in my opinion most of us use search & destinations nowadays, even in offline mode. The only reason I look at a map is to gauge distance between me and my destination.
Almost no one I know uses directions – all look at the map, see "3 intersections north, then right, then the 2nd on the left" and drive/bike/walk like that.
I certainly do
And for this use case Google maps is getting pretty useless
Oh we all know that person that just follows the printed directions blindly. They get off the route, they're helpless. And they can't get there a 2nd time without the directions again, because they never paid attention to where they were actually going.
I've had the frustrating experience of looking on google maps and seeing a town I was looking for not be labeled at a zoom level where I could see where it was in relationship to other towns. But I wonder if part of this is about google trying to funnel users into using maps in a particular way. Is this their attempt to get users to search more and scrutinize maps less?
Are they trying to make online interactive maps a different experience than paper maps?
It's way too slow and its missing some stuff buts it's moving in the right direction. Task based (someone mentioned this) is correct. Tasks are easy targets for Google's user-centered design and machine learning. It's not as simple as reproducing a road map from the 1950's, adding pan and zoom and calling it a day. You design for the device (hint: mobile) and the tasks that are used MOST. Do you really think Google doesn't keep track? People don't use Google Maps the same way as a 20in folded road map or a school atlas. If you wonder what tasks are they are designing for just look at the UI of the App. It's dominated by a big SEARCH bar and a big "get DIRECTIONS" button. That's what people do and that's what it's designed for. They are trying to make it glanceable in a car and make room to see search results or to add stuff based on interaction (like traffic, alternative routes, etc.) and not waste time and bandwidth on loading all the extra crap.
The difference between the album-cover paper map and Google Maps is that every "pixel" on the paper map was put there by conscious, functional and esthetic choice. In fact the people who designed and drew that map probably bragged about it to their peers.
No designing human ever sees a rendered Google Map, except for testing. The overall esthetic programmed into the algorithms are designed, but no human looks at the final result and says "it would just look better if this was shifted up and to the right a bit." There are an infinite number of possible renderings on a digital map, but a specific published paper map goes out the door with a sigh of satisfaction.
Which is to say, there's a long way to go before a digitial map is beautiful, and a specific rendering is as good as possible for the person viewing it for a specific purpose.
The only way digital maps win (and they do) is that you can ask them things, and they cover the world.
From my experience, using the Google maps within the city, it tends to ne pretty sparse with labels by default.
This is a good thing because in a place like tokyo there is just too much stuff - if I care for aometgubd I either search for it, or drop a pub and check what is around.
On the other hand, Maps always shows the labels of things I have searched for in the past, yielding a customized legend of landmarks+things that matter to me.
The conclusion seems to undermine the whole piece. If the changes were made to help mobile uses then great, I almost always use it on mobile, and apparently so do the majority of users.
He poses the question of which map you'd want when lost. A mobile phone with Google Maps is clearly the right answer.
The comparison to a paper map is stupid. A paper map is severely limited in that you can't zoom in on it so it has to be packed with enough information to hopefully be useful. An interactive map only needs to give you enough context to know what to zoom in on. If I'm using a touch interface that overloads me with information in one screen, it's a bad interface.
It's like claiming that the new york times should display the entire full front page of the newspaper on a mobile device so you can read several articles without scrolling or loading more content because that's what you used to be able to do with the real paper.
Not stupid at all, actually. The paper map is hand-tuned and shows a great deal of thought and care about precisely what to show, how to show it, and where to show it.
There are many lessons that digital maps can learn from hand-crafted maps.
The article has its points and it is always interesting to analyze those specific readibility and rendering analysis. It helps to understand the choice made by interactive map platforms through static image and dynamic controls.
Justin O'Beirne knows what he is talking about. He has worked for Apple Maps and he also created other interesting and really precise analysis in the past.
It's probably already been said, but comparing Google Maps to a paper map is stupid.
The paper map has to have lots of cities on it, because there's no other way to find where exactly the specific suburb is if it's not on the map. In Google Maps, you can zoom in, you can search, you can have a link for direction sent to you, ...
The author may claim "less is just less", but apart from "printing a map before knowing where I'm going", I can't think of a situation where the "improved" map would be at all useful.
Great article, but man it reads oddly with JavaScript turned off. Huge masses of text repeated, scrolling weirdly broken. It's almost a proof of its own point …
Optimal placement of labels is an NP-hard problem[1]. Since google maps transferred to vector and label placement is now done on the (mobile) client, I am not surprised by what happened.
[1] https://en.m.wikipedia.org/wiki/Automatic_label_placement
When using a map for navigation at a glance, with a route overlaid on the map, I do not need to know the names of the streets, but I do need to see that there are three cross streets before my turn.
Comparing google maps to here maps at the same zoom level (on desktop, not on mobile) here maps has more info. It's like google maps has completely given up on the desktop.
That said, I also think the focus on paper maps is misplaced. Old-style road maps had to answer the question, "how do I get there from here?" New-style digital maps don't need to answer that question any more! Questions new-style maps need to answer include:
* I know the name of a place or street; where should I zoom in to see more things around that place or street?
* I need to go to a (gas station / rest stop / hospital); where's the closest one?
* How would I get home from where I am?
* I'm in an unfamiliar place and would like to go "downtown" (where there are restaurants and things to do); where is "downtown"?
* Where is my car right now?
Roads help you to orient yourself with the map, but they aren't as fundamentally important to digital maps as they were to old-style road maps. The visual space of the map might be better spent helping answer questions like these.