I have it happen to me over and over again when Waze has my GPS coordinates. Its uncertainty is on the level of where I am in my house, and not which town I'm in.
Yeah, often the guess you have is something you have no confidence in.
Also, I am just speculating as someone who has had similar problems... I did not contribute to this app.
Another common case is the history of your behavior is fully remote and your instantaneous connectivity is out-of-sync. So the map falls back to this kind of primitive state more similar to TomTom rather than being intuitive and catered.
Looks like it prioritized historical locations over nearby locations. The little clock icon usually represents a prior location that was selected.
Could be an interesting design approach to know which locations are "fungible". I.e., one Wawa is as good as another, but your parent's house isn't a replacement for mine [1].
[1] I'm sure your parents are very nice, of course :)
And ironically the OP was complaining that it didn't return him to the same WaWa:
> Last time I asked for directions to a Wawa, it was to the one on Oak Lane, 5.9 miles away. Why doesn't it guess that I might be trying to go back to the same place?
I'm working on b2b software and when working with the designers at http://fairpixels.pro I've learned many things from their approach as to why (a lot of) software sucks. Personal example: I've been measuring qualitative feedback users have been giving us and passing it to them to make design updates. They (design team) would come back with me with questions about data & metrics. They were more interested in what the users DID vs what they told me. This sounds small, but has made a huge difference in the way I look at product design & development. The things users told us in a few cases, didn't match the way they were acting in the product. After some research, we found out why and managed to redesign those features and solve the problems.
The thing with most software, just like ours, is that many of us engineers don't have the deeper design knowledge to really understand how a product needs to be build from a user's perspective.
One thing I've learned is that the user's complaint is always valid, the user's proposals seldom are.
That is to say, users know when they don't like something, but they are often wrong about why they dislike it or what is interrupting their workflow. Our job is to find and fix the root cause starting from user feedback, and fix the product in accordance to what the user is actually trying to accomplish.
Exactly this. What I used to do is listen to the user's proposal and go build that. 80% of the time it's not the correct thing to work on. I know spend much more time talking with users about their problems/pains, and then use metrics to figure out what they are trying to do. The combination of these two + a good set of eyes will point you in the right direction. Very very hard to do, but it's at least a leap better from my previous approach.
What's interesting about this, is that user complaints can be suspect too. I read an anecdote recently about the American auto industry. They'd ask people about what they didn't like about the cars they bought. And they'd get all these complaints, then try to build cars based on those complaints. The problem was, many of the complaints were for things that took them out of the market for the car they bought. For instance, people who bought sports cars would complain about lack of doors, or lack of back seat, and so GM or Ford would build a car that wasn't quite a sports car, because it was too big, and it wasn't quite a sedan (for whatever reason, too sporty? whatever that means?), and so no one bought it.
I honestly don't know what the solution to this problem is, since feedback is invaluable. I guess, always be careful?
Classic case- ui-paradigm shift in the middle of the programm- for once the right click does not present options - but does something else entirely that undoes work.
Its strange- but usually this is found on components that had a developer-UI attached to it and where stand alone. Its easy to craft a pre-mature UI-decision into the working of a component- and run out of time for a redesign.
Another classic there- you have a UI and suddenly it gets terminal- as in you have to enter commandline shortkeys to archive efficient workspeed with it.
Just to add onto this, there's a whole field of people who design things with human psychology in mind. For example, when you got onto Twitter, they _could_ show you how many notifications you have right away, but they implemented an artificial delay so you stare at it intently and when it finally shows you feel gratification [1].
So while a design may not make sense right away, there may be other factors at play.
Unrelated but similar and I feel like venting... On Google Maps you'll search for something, and the results are shown as a list that almost covers the entire map. It's Google MAPS and it hides the map for so many things. I'd rather go to a place that's 3 miles straight down a free way than 2 miles of stop lights, but I don't know that if the map is covered!
If you're talking about the mobile app, you can swipe down on the list to see more of the map.
In many cases, I think defaulting to the list makes sense because of the extra data it has. For example, I'd rather go to a restaurant that has a 5-star rating and is open now as opposed to one that is rated 2 stars and is closed, even if the 2-star restaurant is easier to get to.
> It's way better than just seeing a bunch of pins on a map.
Is it? Showing the pins on the map seems like it would show you roughly the same distances but with the added information of direction and area.
At worst you might eyeball the distance wrong, but does it matter if something is 3.5 miles away or 3.6 miles away?
Most of the time I search something in Google maps I want to see the relative area so I can make a choice based on route and whatever else is near the location.
I guess it comes down to how likely you are to rely on Google map ratings/info. If I'm searching for a restaurant, I generally trust Google's location information, but otherwise I don't trust their ratings and store times to be accurate. Other people could be different though.
Yes, because the list shows you stuff like the name of the place, its address, ratings, hours and all sorts of other stuff. Plus the full screen map view is a short swipe away.
That bugs me the swipe though...sometimes it will show part of the map and I tap the map expecting to see it full screen. No, it just erases, goes to a blank search.
Often, I will search for a string such as "Cost"; I'll probably be referring to Costco, but in the map, it will also show me results for Cost Plus World Market, for example. If I'm only looking at pins, I can't see the names without clicking on each pin, which isn't really feasible in the middle of driving. However, in the list, it's easy to differentiate between the two, in addition to by distance (since that's the information the list shows).
Conversely, if I'm on the way somewhere, and I want to add a stop using the "add stop" feature, such as a gas station, then it sorts by the amount of extra time it will add for each stop, as well as showing the pins on the map. Since I don't really care where I go, as long as I get gas, this seems like a reasonable compromise. (It also puts the gas prices on the map, which is nice.)
However, perhaps I am a happy-path user, or have been trained over time to use the application as the developers intended. Regardless, the current behavior works well for me.
> which isn't really feasible in the middle of driving.
That makes a lot of sense to me. I typically only use my GPS while walking, but now I'm thinking of a use case where I couldn't stop for a few seconds and pan around the map with a finger, and I think I would want a list in that scenario.
The place literally across the road from me is a screen down the list (on mobile) because it's got lower reviews (by about one star or so) than the place 7km away which is at the top of the list.
Like most things Google, I don't think it's that simple. It seems to rank on multiple factors including distance, rating, if you've been there before, if you've saved it, popularity and surely lots of other signals. My search results from your location are probably different than your search results.
Or why's it so hard to get it to show a street name? It will show some, and it won't show others. If the one you want is in the latter category you've got a lot of zooming before you see it. Ugh.
It is so difficult now to get Google Maps to just show the actual map! It's trying so hard to add value by showing you pictures of sandwiches that you can't just see the map when you need to.
> I'd rather go to a place that's 3 miles straight down a free way than 2 miles of stop lights
This is a big pet peev of mine. I optimize for easier driving; less turns and less stops over saving a minute or two. But as smart as they try to be, they never seem to learn anything from _the route I actually take_. You'd think the Google lady would catch on that; when he drives home from work (5+ times a week) we always recommend that he take this Avenue but for some reason he always goes out of the way to take the Highway and I have to recalculate. I'd think they'd analyze every "recalculating" as potential bad advice and learn from it.
I think the clock icon next to the entries they're complaining about means that it's surfacing them because the user had looked at those particular stores' search results before, and would likely want to revisit those search results.
It's a static site (built with Bloxsom, as far as I recall). I can't imagine HN could hug it to death and it's working fine for me right now. I'd want to blame something else for being unable to reach it.
Definitely not. My site hits the front page of Reddit a few times a year and handles it just fine. Today's HN hit was only about twice as much traffic as a typical day.
Case in point: my article about traffic control systems (https://blog.plover.com/tech/highway-stuff.html) is on the front page and has already received twice as much traffic as this one, and the West Coast hasn't even come to work yet.
I feel that the worst new "feature" in Google Maps is when you're on your way to your destination and it suddenly suggests you might first want to go to a random place you've never heard of before. You then have 3 seconds to tap "No" before it silently adds the destination (and 15-30 minutes of extra driving) to your route. I have found no way to disable this feature.
I'm worried that there could be people who were in a hurry to get somewhere and perhaps wound up in a grave instead because they felt the urgent need to look away from the road and fiddle with their phone to reset their map destination while on the highway.
I've seen it reroute me to save time (based on live traffic) but never seen this behavior. Perhaps in your case it's this feature behaving poorly, or some other bug.
I've experienced similar behavior. For awhile it seemed it would find a "new route" and give me x seconds to say no or else it would default to the new route. It seems the behavior switched to the default being no to the new route...
I wish there was as good as maps paid alternative. I'd gladly pay for something other than the ever slower with each update Google maps.
Funny you mention Waze because they are planning on rolling out real-time suggested stop adverts in Waze now which sounds very similar to what the parent mentioned.
Software does not suck; requirements, especially from non-technical stake-holders suck. Every single piece of software starts with a single and easy to understand idea, then different stakeholders arrive and start adding inputs and often conflicting requirements, and then software starts to suck. None of these would have happened if we knew how to say "No". :(
He should be happy it actually looked up the place he wants. I go to maps.google.com and not only does it give the wrong town for my location (correct GPS coordinates, it just won't acknowledge the town exists even though it is the reservation seat of government), but the first result for "gas station near" is a Catholic Church and then gas stations in the next, next town over. Google doesn't seem to know about businesses on the reservation. Apple, sadly, is worse.
It's showing him Wawas he's been to before. That's what the clocks mean.
For the search "wawa" it doesn't make a lot of sense to prioritize history over distance, because Wawas are nearly identical. But for "farmer's markets" this would probably be a useful feature.
Figuring out how to differentiate between those two examples algorithmically is super non-trivial though
Oof, that's really bad if that's the case; it means that a mis-click can poison your search results and make them less useful.
I get that Google's all about saving clicks now, but I've noticed between maps and other apps (Google Music comes to mind) the extra inference sometimes just makes me scared to click on things.
Can you at least long press or something to remove a location from your history?
This is one of my biggest pet peeves with modern recommendation systems. It's like somebody decided "no, the user can't click a 'show-me-more-like-this' button, it has to be automatic" and now I have to treat every website like I'm doing reinforcement training with a cat.
But at least with AI there's something marginally close to an excuse, because you need a ton of data to make it work. With maps it's just plain and simple bad UX design, I can't think of any reason why locations shouldn't be sorted by last visited instead of last clicked.
> Google has a user history website
From a usability standpoint, if it's not accessible from the same same interface where your results are displayed, then it doesn't exist.
Even if it gives history priority, there's still something a bit peculiar about the arrangement; it's not sorting by distance within history, but rather probably sorting history hits by recency. That's the sort of thing I think of as "programmer logical" rather than "common sense logical." If I search for "tiki bar" when I'm in Oakland, it's all well and good to remember that I also ran that search when I was in Portland, but it should probably show me a few tiki bars in and around Oakland, and give rather low priority to letting me know it's only 632 miles to Hale Pele.
The reason for pretty much any bug found in software driven by a huge company is that there is a large disconnect between fixing bugs and profit. Fixing a bug almost never excites users, and even if it did in this case, they won't see a measurable increase of users of their software, so you can't prove a profit increase.
Even if a bug is fixed, the effort might be wasted when a large overhaul of the system happens because they want to add a huge feature, like a new searching system or some internal backend refactor. Fixing the bug might be much more difficult than you think due to the scale and complexity of their database or some other reason. Then the change needs to flow through the review and testing pipeline.
My point is to explain that software gets harder with scale, which might cause a bug like this to cost tens of thousands of company dollars to fix, while the benefit to each user (and therefore the company) would be a tiny fraction of a cent, so both managers and financially-aware engineers would easily brush this off as "not worth it." Any argument of the form "why not just fix it?" would be countered with "you could say that for any of the 40k open bugs."
> The reason for pretty much any bug found in software driven by a huge company is that there is a large disconnect between fixing bugs and profit.
Exactly. The answer to any complaint like the one described in this article (and I have similar ones about my phone all the time) is "how much would you be willing to pay to fix this?"
Of course, it doesn't help that, even if I were willing to pay to fix this, I have no way of doing so; even if, say, 100,000 or 100 million users would all be willing to pay to fix it, so that the amount in play might actually be enough to compensate a developer to do the work, we have no way of coordinating such a payment. That's not just because the companies involved are large; it's because the software running on, say, my phone is a cost center to begin with, not a profit center. The profits come from locking users into multi-year contracts, and "too many bugs/misfeatures in the software on the phone" is so far down the list of things the phone provider is worried about people switching for that there is no motivation to spend money on it.
> which might cause a bug like this to cost tens of thousands of company dollars to fix, while the benefit to each user (and therefore the company) would be a tiny fraction of a cent
But if there are 100 million users, that might still work out to a tidy profit--if the users could somehow coordinate the payment, and convince the company that the alternative to accepting the payment and fixing the problem was that all 100 million of them switch providers. In the real world, of course, this will never happen.
>Of course, it doesn't help that, even if I were willing to pay to fix this, I have no way of doing so; even if, say, 100,000 or 100 million users would all be willing to pay to fix it, so that the amount in play might actually be enough to compensate a developer to do the work, we have no way of coordinating such a payment.
No, but we can always go and use another app that doesn't have as many problems.
He's trivializing it, but specifically mentions that including one 5.7 miles away would be fine. And then promptly says because he visited it before.
That's some complicated intelligence and pattern matching that he's asking for. What exactly should it be ranking by? Proximity and popularity and traffic density and Google rating and awhole load of other things.
The rules that Google apply obviously work for some scenarios, but not this one. Could it be better, obviously, but that's not a bug.
I feel that Google Maps search quality has gone down in the past few years. Sometimes searching "Coffee shops" will not show anything within a mile and I have to reposition, press "Search this area" and then it will show the Starbucks a block away.
Probably still better than Apple, which pointed me to a 19-hours-away store when asking for "directions to Albertsons" a couple years ago: https://i.imgur.com/jSCrToj.jpg
Having actually programmed search functionality like this I can assure you, it's really not. That's what you start with. Even if you put aside that distance is something that has to be calculated through triangulation as many systems now include that.
Especially when you add in rankings, you start having to think about sorting by multiple layers, circles of distance with ranking sorted by that.
So the really good match 20 miles away is shown lower than the pretty good match 5 miles away, but they're both shown above the bad match 7 miles away.
And that's before you add categories, past history, any paid ranking, etc. And then the queries get really expensive and you've got to decide what to fudge or not to keep the system responsive and stop your search system eating all your CPU.
And you can bet people will still find scenarios where they'll say "why's it doing this? I'd expect the order to be different".
This is pattern matching and value assignment, something humans are amazing at and computers aren't.
>Having actually programmed search functionality like this I can assure you, it's really not. That's what you start with. Even if you put aside that distance is something that has to be calculated through triangulation as many systems now include that.
It already prints a distance.
It could use that. Or, if it is some kind of point to point distance, it could calculate a driving distance and use that if it wanted to be extra smart.
Add the type of place it shows, and be done with it.
>Especially when you add in rankings, you start having to think about sorting by multiple layers, circles of distance with ranking sorted by that.
We've been doing multiple criteria sorting for millennia. It's not rocket science either.
>So the really good match 20 miles away is shown lower than the pretty good match 5 miles away, but they're both shown above the bad match 7 miles away.
Why is one "really good" and the other's (that are also of the same store) not? They are bloody stores of a franchize.
Don't try to be clever and define what's "good" and just show them to me sorted by distance AND type (in case there's a city called that too or something). End of story.
>And you can bet people will still find scenarios where they'll say "why's it doing this? I'd expect the order to be different".
People might or might not say anything. But way fewer will be confused by a distance sorted list + an indication of what the item they saw is (store, city, village etc), than with this clusterfuck of a result.
There's a massive difference to printing 10 distances from your result set, and calculating 10 million. I'm not sure why you had to exaggerate "millennia".
How does a computer know that searching "Wawa" means you want the store and not the town? You haven't even considered that. It's a search term.
And a really good system would clearly takes a load of things you're not considering into account, relevance, rank, personal relevance, historical relevance age, etc. A really good system might even helpfully try to include traffic density, how busy a store was, etc. What if all the stores near him have 1 star google reviews? What if some of those Wawas are paying for adverts and some aren't? What if some of them have opening times set up, and some don't?
As I say, their ranking system might be working great for other things, but something like this it spits out crap.
>There's a massive difference to printing 10 distances from your result set, and calculating 10 million.
Well, no reason to calculate 10 million. If another way to select a list of top suggestions to show is less costly, you can just calculate the distance for the existing suggestions they're already about to show - and then show them sorted by the distance).
Having the unknown criteria they used to pick the suggestions reflected in the order they show them is not useful to anyone though.
>I'm not sure why you had to exaggerate "millennia".
Not exaggerating -- pointing to the fact that we know techniques to do it, and we have been doing it (obviously manually before computers) for millennia.
>How does a computer know that searching "Wawa" means you want the store and not the town? You haven't even considered that.
I did, and I mention it in 2 places in my response above: they could either sort all by distance and display what each entry is next to it (e.g. town, store, etc), or give sorted lists for different types of results (e.g. locations vs businesses).
>And a really good system would clearly takes a load of things you're not considering into account, relevance, rank, personal relevance, historical relevance age, etc.
Whatever the above system "took into account" it failed. And it's a case that's common to 99% of the population (search for a business). Searching for a location is another 99% common case.
Historical relevance and age? Who cares besides tourists (and even them, when they're not searching for a location or a business but for sightseeing)? A system should have sane defaults and hide those other things under user enabled search/sort criteria.
For location based stuff it seems pretty logical functionality to at offer the ability to sort results by travel time or at least distance. I still have to log into god-awful store websites and use their store locators because they'll give me the information I need.
What is hard about ordering a bloody list of places by distance?
>He's trivializing it, but specifically mentions that including one 5.7 miles away would be fine. And then promptly says because he visited it before.
Well, they could first add the bloody sorting, then we can think about whether they should mark or prioritize previously visited places (which itself does not take any "complicated intelligence and pattern matching" -- it's just persisting visited places and consulting their list).
My point is that all the "complicating factors" are still trivial -- each alone, or put together.
My second point is also that even if they weren't (for some exotic cases) , the most common cases and sane defaults, remain trivial -- and anything else could be relegated to settings and the like.
My third point is also that no matter how easy or hard the complicating factors, a screen like the one shown by the poster, is an utter failure.
To continue on this 2nd and 3rd point, it's like people are trying to defend Clippy, the office assistant, by saying that "AI is hard".
It might be, but getting Clippy out of my bloody monitor and not having it disturb me when I try to compose a Word document, is very easy. MS could always do that (and that's what they ended up doing).
The counter-argument is that we’re (mostly) talking about profitable companies making billions a year, which will be wasting tens or hundreds of millions anyway on unneccessary spending, poor strategic decisions, and other standard corporate cash sinks.
Why not gain reputation and good will by making users happy and deliberately cultivating a reputation for low-frustration, responsive, thoughtful and reliable products?
Of course this is the opposite of “Get shit out fast, fix it later” and would surely be very boring to work on.
The problem is that you can't even identify, much less stop, the cash sinks - nobody seems to be able to do it in any sort of large corporation (or any other sort of organisation). Fixing some random bug that you have no impact numbers for therefore has to compete with the sexy new six-month cash sink that someone thinks will bump stats across the board. "Perfect is the enemy of good" seems to be the mantra, which would be better written as "lazy is the enemy of good." Can we not agree that this is a bullshit saying, and that nobody ever argued for actual perfection? We just have different ideas of what constitutes good (or profitable) software.
TSB's systemic failure was a result of trying to clean up their tech debt (and dependency on Lloyds' systems), though -- the old systems were working fine for them (and continue to work fine for Lloyds), not falling over all the time.
Google Maps apparently thinks I'm Prompto from Final Fantasy XV, and will send me alerts of points of interest (usually restaurants) suggesting I take pictures of them.
Uh, no. If I want to go somewhere, I'll ask you, Google Maps. It's shit like this that makes me glad I switched to OsmAnd (despite the latter lacking in features, such as finding an optimal foot/bus/train route).
You can turn off those adnotifications. They are kind of annoying, but one popup plus a button to block future popups is a small price for a powerful app.
For those unfamiliar. On android phones you can just long-press a notification to disable that notification category from that app. Granted most apps lump all their notifications in a single category, but google maps is not one of them.
Budapest, a city of two million has place names only unique per district, there are 23 of them. So this is ubiquitous there, for example if you enter "deak ter" it will bring up some insignificant spot out in the boondocks instead of the central square of the city with three underground lines, the terminal of the airport bus and so on...
92 comments
[ 3.2 ms ] story [ 166 ms ] threadAlso, I am just speculating as someone who has had similar problems... I did not contribute to this app.
Another common case is the history of your behavior is fully remote and your instantaneous connectivity is out-of-sync. So the map falls back to this kind of primitive state more similar to TomTom rather than being intuitive and catered.
Could be an interesting design approach to know which locations are "fungible". I.e., one Wawa is as good as another, but your parent's house isn't a replacement for mine [1].
[1] I'm sure your parents are very nice, of course :)
> Last time I asked for directions to a Wawa, it was to the one on Oak Lane, 5.9 miles away. Why doesn't it guess that I might be trying to go back to the same place?
The thing with most software, just like ours, is that many of us engineers don't have the deeper design knowledge to really understand how a product needs to be build from a user's perspective.
That is to say, users know when they don't like something, but they are often wrong about why they dislike it or what is interrupting their workflow. Our job is to find and fix the root cause starting from user feedback, and fix the product in accordance to what the user is actually trying to accomplish.
This is often very hard
I honestly don't know what the solution to this problem is, since feedback is invaluable. I guess, always be careful?
http://simpsons.wikia.com/wiki/The_Homer
I suspect there is a lesson there about using aggregate feedback, rather than listening to single individuals.
Its strange- but usually this is found on components that had a developer-UI attached to it and where stand alone. Its easy to craft a pre-mature UI-decision into the working of a component- and run out of time for a redesign.
Another classic there- you have a UI and suddenly it gets terminal- as in you have to enter commandline shortkeys to archive efficient workspeed with it.
So while a design may not make sense right away, there may be other factors at play.
[1] https://www.vice.com/en_ca/article/vv5jkb/the-secret-ways-so...
In many cases, I think defaulting to the list makes sense because of the extra data it has. For example, I'd rather go to a restaurant that has a 5-star rating and is open now as opposed to one that is rated 2 stars and is closed, even if the 2-star restaurant is easier to get to.
Is it? Showing the pins on the map seems like it would show you roughly the same distances but with the added information of direction and area.
At worst you might eyeball the distance wrong, but does it matter if something is 3.5 miles away or 3.6 miles away?
Most of the time I search something in Google maps I want to see the relative area so I can make a choice based on route and whatever else is near the location.
I guess it comes down to how likely you are to rely on Google map ratings/info. If I'm searching for a restaurant, I generally trust Google's location information, but otherwise I don't trust their ratings and store times to be accurate. Other people could be different though.
Conversely, if I'm on the way somewhere, and I want to add a stop using the "add stop" feature, such as a gas station, then it sorts by the amount of extra time it will add for each stop, as well as showing the pins on the map. Since I don't really care where I go, as long as I get gas, this seems like a reasonable compromise. (It also puts the gas prices on the map, which is nice.)
However, perhaps I am a happy-path user, or have been trained over time to use the application as the developers intended. Regardless, the current behavior works well for me.
That makes a lot of sense to me. I typically only use my GPS while walking, but now I'm thinking of a use case where I couldn't stop for a few seconds and pan around the map with a finger, and I think I would want a list in that scenario.
The place literally across the road from me is a screen down the list (on mobile) because it's got lower reviews (by about one star or so) than the place 7km away which is at the top of the list.
For fascinating essays on digital maps, this site: https://www.justinobeirne.com
Related to your question:
Google Maps’s Quiet Transformation (2017), esp. the second half of the article https://www.justinobeirne.com/a-year-of-google-maps-and-appl...
Surpising Changes to Google‘s Cartography (2016) https://www.justinobeirne.com/what-happened-to-google-maps
Why are Google Maps’s city labels more readable than those of its competitors? (2010) https://www.justinobeirne.com/google-maps-label-readability
This is a big pet peev of mine. I optimize for easier driving; less turns and less stops over saving a minute or two. But as smart as they try to be, they never seem to learn anything from _the route I actually take_. You'd think the Google lady would catch on that; when he drives home from work (5+ times a week) we always recommend that he take this Avenue but for some reason he always goes out of the way to take the Highway and I have to recalculate. I'd think they'd analyze every "recalculating" as potential bad advice and learn from it.
Cached version from Google: https://webcache.googleusercontent.com/search?q=cache:E3wiNT...
I'm worried that there could be people who were in a hurry to get somewhere and perhaps wound up in a grave instead because they felt the urgent need to look away from the road and fiddle with their phone to reset their map destination while on the highway.
I wish there was as good as maps paid alternative. I'd gladly pay for something other than the ever slower with each update Google maps.
Are you sure you're not confusing that with rerouting you based on traffic? And that the extra driving was done to get around a traffic jam?
EDIT: Are you tapping the map while driving? https://productforums.google.com/forum/#!topic/maps/LqK7zqz6...
Also, why are people using Google Maps instead of Waze for navigation?
I take it you're not a developer?
For the search "wawa" it doesn't make a lot of sense to prioritize history over distance, because Wawas are nearly identical. But for "farmer's markets" this would probably be a useful feature.
Figuring out how to differentiate between those two examples algorithmically is super non-trivial though
Source: Just tested this in-app.
I get that Google's all about saving clicks now, but I've noticed between maps and other apps (Google Music comes to mind) the extra inference sometimes just makes me scared to click on things.
Can you at least long press or something to remove a location from your history?
Google has a user history website where you can see everything they recorded from Search and other apps, and you can delete the entries
But at least with AI there's something marginally close to an excuse, because you need a ton of data to make it work. With maps it's just plain and simple bad UX design, I can't think of any reason why locations shouldn't be sorted by last visited instead of last clicked.
> Google has a user history website
From a usability standpoint, if it's not accessible from the same same interface where your results are displayed, then it doesn't exist.
Maybe he looked at a location first for information, and to know where it is, and needs to look for it again at a later time to actually go there.
Even if a bug is fixed, the effort might be wasted when a large overhaul of the system happens because they want to add a huge feature, like a new searching system or some internal backend refactor. Fixing the bug might be much more difficult than you think due to the scale and complexity of their database or some other reason. Then the change needs to flow through the review and testing pipeline.
My point is to explain that software gets harder with scale, which might cause a bug like this to cost tens of thousands of company dollars to fix, while the benefit to each user (and therefore the company) would be a tiny fraction of a cent, so both managers and financially-aware engineers would easily brush this off as "not worth it." Any argument of the form "why not just fix it?" would be countered with "you could say that for any of the 40k open bugs."
Exactly. The answer to any complaint like the one described in this article (and I have similar ones about my phone all the time) is "how much would you be willing to pay to fix this?"
Of course, it doesn't help that, even if I were willing to pay to fix this, I have no way of doing so; even if, say, 100,000 or 100 million users would all be willing to pay to fix it, so that the amount in play might actually be enough to compensate a developer to do the work, we have no way of coordinating such a payment. That's not just because the companies involved are large; it's because the software running on, say, my phone is a cost center to begin with, not a profit center. The profits come from locking users into multi-year contracts, and "too many bugs/misfeatures in the software on the phone" is so far down the list of things the phone provider is worried about people switching for that there is no motivation to spend money on it.
> which might cause a bug like this to cost tens of thousands of company dollars to fix, while the benefit to each user (and therefore the company) would be a tiny fraction of a cent
But if there are 100 million users, that might still work out to a tidy profit--if the users could somehow coordinate the payment, and convince the company that the alternative to accepting the payment and fixing the problem was that all 100 million of them switch providers. In the real world, of course, this will never happen.
No, but we can always go and use another app that doesn't have as many problems.
What other app? What apps don't have random bugs or obviously insane behavior popping up when you least expect it?
He's trivializing it, but specifically mentions that including one 5.7 miles away would be fine. And then promptly says because he visited it before.
That's some complicated intelligence and pattern matching that he's asking for. What exactly should it be ranking by? Proximity and popularity and traffic density and Google rating and awhole load of other things.
The rules that Google apply obviously work for some scenarios, but not this one. Could it be better, obviously, but that's not a bug.
Probably still better than Apple, which pointed me to a 19-hours-away store when asking for "directions to Albertsons" a couple years ago: https://i.imgur.com/jSCrToj.jpg
Especially when you add in rankings, you start having to think about sorting by multiple layers, circles of distance with ranking sorted by that.
So the really good match 20 miles away is shown lower than the pretty good match 5 miles away, but they're both shown above the bad match 7 miles away.
And that's before you add categories, past history, any paid ranking, etc. And then the queries get really expensive and you've got to decide what to fudge or not to keep the system responsive and stop your search system eating all your CPU.
And you can bet people will still find scenarios where they'll say "why's it doing this? I'd expect the order to be different".
This is pattern matching and value assignment, something humans are amazing at and computers aren't.
It already prints a distance.
It could use that. Or, if it is some kind of point to point distance, it could calculate a driving distance and use that if it wanted to be extra smart.
Add the type of place it shows, and be done with it.
>Especially when you add in rankings, you start having to think about sorting by multiple layers, circles of distance with ranking sorted by that.
We've been doing multiple criteria sorting for millennia. It's not rocket science either.
>So the really good match 20 miles away is shown lower than the pretty good match 5 miles away, but they're both shown above the bad match 7 miles away.
Why is one "really good" and the other's (that are also of the same store) not? They are bloody stores of a franchize.
Don't try to be clever and define what's "good" and just show them to me sorted by distance AND type (in case there's a city called that too or something). End of story.
>And you can bet people will still find scenarios where they'll say "why's it doing this? I'd expect the order to be different".
People might or might not say anything. But way fewer will be confused by a distance sorted list + an indication of what the item they saw is (store, city, village etc), than with this clusterfuck of a result.
How does a computer know that searching "Wawa" means you want the store and not the town? You haven't even considered that. It's a search term.
And a really good system would clearly takes a load of things you're not considering into account, relevance, rank, personal relevance, historical relevance age, etc. A really good system might even helpfully try to include traffic density, how busy a store was, etc. What if all the stores near him have 1 star google reviews? What if some of those Wawas are paying for adverts and some aren't? What if some of them have opening times set up, and some don't?
As I say, their ranking system might be working great for other things, but something like this it spits out crap.
Well, no reason to calculate 10 million. If another way to select a list of top suggestions to show is less costly, you can just calculate the distance for the existing suggestions they're already about to show - and then show them sorted by the distance).
Having the unknown criteria they used to pick the suggestions reflected in the order they show them is not useful to anyone though.
>I'm not sure why you had to exaggerate "millennia".
Not exaggerating -- pointing to the fact that we know techniques to do it, and we have been doing it (obviously manually before computers) for millennia.
>How does a computer know that searching "Wawa" means you want the store and not the town? You haven't even considered that.
I did, and I mention it in 2 places in my response above: they could either sort all by distance and display what each entry is next to it (e.g. town, store, etc), or give sorted lists for different types of results (e.g. locations vs businesses).
>And a really good system would clearly takes a load of things you're not considering into account, relevance, rank, personal relevance, historical relevance age, etc.
Whatever the above system "took into account" it failed. And it's a case that's common to 99% of the population (search for a business). Searching for a location is another 99% common case.
Historical relevance and age? Who cares besides tourists (and even them, when they're not searching for a location or a business but for sightseeing)? A system should have sane defaults and hide those other things under user enabled search/sort criteria.
What is hard about ordering a bloody list of places by distance?
>He's trivializing it, but specifically mentions that including one 5.7 miles away would be fine. And then promptly says because he visited it before.
Well, they could first add the bloody sorting, then we can think about whether they should mark or prioritize previously visited places (which itself does not take any "complicated intelligence and pattern matching" -- it's just persisting visited places and consulting their list).
My second point is also that even if they weren't (for some exotic cases) , the most common cases and sane defaults, remain trivial -- and anything else could be relegated to settings and the like.
My third point is also that no matter how easy or hard the complicating factors, a screen like the one shown by the poster, is an utter failure.
To continue on this 2nd and 3rd point, it's like people are trying to defend Clippy, the office assistant, by saying that "AI is hard".
It might be, but getting Clippy out of my bloody monitor and not having it disturb me when I try to compose a Word document, is very easy. MS could always do that (and that's what they ended up doing).
Why not gain reputation and good will by making users happy and deliberately cultivating a reputation for low-frustration, responsive, thoughtful and reliable products?
Of course this is the opposite of “Get shit out fast, fix it later” and would surely be very boring to work on.
But even so.
Uh, no. If I want to go somewhere, I'll ask you, Google Maps. It's shit like this that makes me glad I switched to OsmAnd (despite the latter lacking in features, such as finding an optimal foot/bus/train route).