Open Source Maps
A big problem I have with the current implementation of Google’s products is that if you are not online, your service will be a shadow of what you are able to do. In an age where we have mini computers in our pockets that are capable of so much more than figuring out how to get to go from point A to point B or finding out when the local cafe is open. Do we really need to rely on an internet connection for these items? Having the ability to be offline and still have a functioning device has been great for not only battery life, but worked great when hiking in rural areas, or travelling on the road, or just wanting to be “unplugged”. While this may not mirror your experience or items you may have thought of. It is a big enough differentiation that we need to be able to do something about it. Why do I need an internet connection? Why do companies need to know if I was at my local cafe? These were my original questions when searching for solution.
After a long 6+ months using OSMAnd, (an Android application for OSM) I figured I’d write a review of OSM, and sell its praises on what it does well, and whether you could replace Google Maps (GMaps) as well.
The amount of features GMaps provides is astounding, and attempting to replicate each feature is not something that any one company can do easily, but the Open Street Map Foundation (OSMF) has the best chance to do it. The OSMF operates the OpenStreetMap (OSM) product. OSM is exactly what it says. An open street map, where contributors can create, edit, upload maps and relevant information from their own corner of the world. OSM is open data, and all items that are provided by them are licensed under the Open Data Commons Open Database License (ODbL). This allows anyone to use their product and create their own “map” on top of the open data. While it is a feat of social ingenuity that they are releasing the data for the world in a completely open format where anyone can search and investigate. Without an easy to use interface it can be quite daunting. If a product is done right you probably didn’t even know that you’ve used data provided by OSM. There are so many use cases for these types of datasets and I will try to explain the main points I can think of, that people use GMaps for, and how using OSM can benefit you.
To make it an even playing field, for all comparisons I will assume that you have downloaded the map data for OSM & GMaps for your area, and are both completely offline.
The biggest use case for this data is navigation software, such as the Garmin/TomTom/Navman. While I list these companies, they actually do not use OSM, but they entirely could. They could give their customers completely free and open maps, but then they wouldn’t be able to charge you ~$80 for the map update. I can understand why they would not want to use the OSM dataset. I am sure there are features and other reasons which ensure that this pipe dream is just not possible. As a 3rd party device, these companies I’m sure are taking a hit in their sales, and operational upkeep. With the rise of smartphones, this has given rise to a new solution. We now already have a 3rd party device that is so well connected that we do not need any other device. We jam pack it with everything possible, so it has become our everything device. Why would you need to spend an extra couple hundred dollars on a separate device when you can just use the one you already have?
Nowadays smartphones can be used for “cheap” navigation device alternatives. With free software powering your device, you don’t even need to pay anything extra to turn your old device into a dedicated in-car navigation system. Before smartphones were so prevalent and in-car navigation devices were all the rage, I don’t think there was ever a time that I did not hear a familiar voice saying “recalculating” through all of my trips. Losing GPS signal, thinking it was in the middle of the ocean, or just missing a turn after rocking out to some radio. These primitive devices (by today’s standards) were using their own processors for calculations, a term known as “on-device”. These processors were very inefficient as (possibly) were the original algorithms. They were just not powerful enough to keep up with a car/bike/train/etc going 60km/h when you miss a turn on a freeway. A solution to the recalculating problem was to increase their processing power. With a faster processor you can calculate changes to the route faster, provide graphical updates to the user, and all around have a better user experience. But a faster processor is not a silver bullet that will make everything right with the world (in relation to GPS nav software). It comes with a price. A faster processor means increased battery usage and increased heat creation. Increasing the battery means you either need to charge it at an increased frequency or need to have a proportionally bigger battery. Increasing the amount of heat generated means your device needs to have better heat dissipation/absorption. These are the problems that smartphone vendors or any hardware vendor deal with when designing a newer product. Combine these problems with mobile internet, and you come to what we have today.
With mobile internet costing as small as $1 per gigabyte (or even lower), this enables lower spec devices to offload their processing power to any company’s data centers. If you had the power of a data center in your pocket, what could you do with it? Obviously the correct answer (for this post) is to get super fast navigation instructions! There is no need for the 5+ second wait time to figure out how to get home from your friends engagement party at 4 am. Or to wait the 3+ seconds for your device to figure out how to get back onto the planned route after you missed the turn off. Companies have improved the processes so that not only is your route calculated in a matter of seconds, alternate routes with estimated timelines and a color coded description of banked up areas shows just how bad the traffic is. To be honest, GMaps it is an amazing technological feat as a complete application. There is a reason why millions of people use it daily.
What happens when you don’t have internet then? GMaps, when used completely offline, even when picking two spots on the same road fails to find directions.
>.< But of course that happens! How on our big ball of dirt do you expect to find directions between two locations when you have no map data on the device. It’s not like its magic! Once you have downloaded the map data for you area, (if you are even allowed too) GMaps can navigate just like the rest of them. There is no real hindrance between being offline and online for GMaps. Extra features such as current traffic conditions, or if there is current delays will not work for obvious reasons, but their current system is quite solid even offline.
So how does OSM’s app compare in the navigation department.
OSM works backwards when finding an address I’ve found. Just like how taxi drivers ask you for your suburb first, and then the road, and then the number, this is because of OSM. Or the integration for narrowing down addresses in the apps that use the OSM data. Addresses are hard, and there are so many variations how someone might write an address I can understand why the OSM apps have integrated it this way. If your super interested, check out: https://en.wikipedia.org/wiki/Address on all the different format variations that will need to be handled. GMaps on the other hand has some type of natural language interpretation which is amazing. As not being able to click an address and it open up in OSM, or copy paste an address from an email is quite annoying and makes me want to change back to another application. If OSM could implement some type of natural language model like GMaps does then this would solve my only serious grievance I have.
GMaps has a faster and easier to understand navigation system with simple selections and it even works offline. But this only works for car trips. All of their other options walking, cycling, etc do not work. These items are apparently no problem when you have internet, but are “too hard” for the little company called Google to do when offline. Using OSM, I can plan out all facets of my trips; figure out how long I have to go, as well as the estimated time left. It even includes the altitude and steepness of your walking tracks. So I can distinctly say: “phew, now it’s all downhill”. But the only thing holding them back from completely overtaking GMaps is their interface. While it may be a bit busy compared to GMaps, it gives increased functionality. For those of us who like to see as much information as possible about where they are. OSM nudges out the competition here. It’s ability to easily plan out multiple stops and easily rearrange them, has been something I have thoroughly enjoyed.
As Google Maps has the backing of Google search, you know that searching on any of their products would be amazing. This follows through for GMaps. People now do more and more searches for not just addresses but businesses in their area. Search terms “watch repair”, “brunch” and “pizza” have turned into hot topic keywords with companies bidding to be in that number 1 spot. It has blossomed into its own miniature SERP ecosystem. “If your business is not setup for local search on GMaps then you might not be missing out on sales” is what a marketing directory tells you :P. But honestly, it is entirely likely. I cannot count the amount of times that I’ve gone to a local establishment based on the reviews and because its matched some queries. This is a huge point for using GMaps, and I’d be lying to everyone if I did not say it wasn’t handy as all hell.
When you are offline, GMaps has minimal data but still allows you to search the downloaded dataset, which is a nice change to their always online model. The limited items that you can see still allow you to use it to search for related terms as they have some hidden search terms associated with their data. As certain cafe’s show up when “brunch” is searched, not just anything with the search term in it, ie; “Cafe le Brunch”. Of course searching for “watches” or “shoes” will definitely deliver you a different search results if you were online. From my cursory usage you can see: Business Names, Total Review Score, Total No. Reviews, Contact Number, Address, and the business’ Opening Hours. Everything you need to make an educated guess.
OSM on the other hand allows anything possible to be added to a business’ metadata under the guise of “tags”. A Pizza parlour might have extra tags such as: delivery, outdoor seating, cuisine, with the values of those tags as: yes, no, pizza. If they accept Diners card, or Bitcoin or they even offer bank transfers. The possibilities are endless. The validity of these values are all of course the responsibility of the community by and large and depending on how active members are in your area, or how the data set has been imported, or supplied to OSM. Its entirely possible for businesses not to exist as far as OSM is concerned, and this is where GMaps has OSM beat.
Due to the abundance of categories and options for their data to be search, makes it somewhat harder to search. If you do not mind what you want to eat, and just want “Food”, that’s simple enough. But if you search for “Pizza”, very few places actually show up for me. This seems to be due to a couple of reasons. First is that the “Pizza” tag under the category of “Food” needs to be assigned to every restaurant/cafe/place of business that serves Pizza, and the secondary reason is that not all Pizza places have the name “Pizza” in its title. When submitting a restaurant mappers are certainly not going to go through the whole menu and jot down all types of cuisine they are happen to be serving at that point in time. This problem is a hard one to solve and there is not a lot OSM can do about it. Besides from asking mappers to be as meticulous as possible, it might be possible to have an automated web scraping system. Something along the lines of: When an item under “Food” or “Service” is added that has a link to a website. Scrape the website for information to see if it is possible to confirm, extra locations, validity of addresses, as well as anything else that can be evaluated automatically. But I am unsure how the OSM feels about scraping websites as I believe you are only meant to submit information that you have validated in person, as anything you submit will end up as freely available information.
GMaps as a singular product without the abundance use of the automated systems Google has in place would remedy it to be in the same place as OSM is in now. Pretty good to use, but lacks in key areas. OSM in this regard is an adequate product but still needs some work done on it.
I think travel is very much a subjective topic and I will just to be as open minded but everyone’s ideal holiday can differ in such detail that my opinions may be so far off you might close this blog and never come back. Whether you like to backpack between hostels, camp in rainforests, or stay in luxury hotels in a singular city, or hop between cities via a helicopter. GMaps is inferior to OSM when it comes to this category. I’ve said it. Within the first couple lines I have already given up my declaration. Travel for me is getting entrenched in the local areas, trying heaps of new things, and walking off the beaten track. While GMaps can help you find the best single origin espresso within 10km of your position, or where a place is, and the reviews. It can’t help you with the features OSM can.
OSM has the ability to download Wikipedia articles for the areas you are visiting. Thus you can overlay every article for a specific town on your map, zoom into your location, and find out all about the attractions you are visiting. The Colosseum, Notre Dam, or the Empire State Building, or anything that has a wiki article! While this might be a bit overwhelming when there are 100 wikipedia articles in the space of 10m from your when visiting places that have such a rich history (such as anywhere in the EU). As long as you have the space on your phone/device nothing should stop you from trying to learn, or at least have the choice if you want to learn about something. Not only that, they have recently created a new section called “Travel Guides” which is currently in its beta phase.
Honestly, after a recent trip to Singapore this was by far an absolutely amazing wealth of knowledge and really helped me not only getting around, but finding hidden places, and even finding a super cheap sim card for internet service while I was there. It is a complete “how to” for nearly any country. Everything has been categorised into items such as: Getting Around, Talk, See, Do, Buy, Eat, Drink, Sleep, Learn, Work, Stay Safe, Stay Healthy, Respect, and one of my personal favourite categories (besides from food obviously), Connect. It looks to be an integration of the site Wikivoyage. Which upon investigation is a sister site to Wikipedia with a speciality for travel. If you have not heard of it and are going on a trip to another country, or even another town, I suggest download OSM, and get reading. Who knows what you might learn.
The fact that Wikipedia, Wikivoyage, and complete offline navigation for walking is integrated into OSM, it beats the offline version of GMaps hands down, possibly even the online version as well. Whilst you are in a different country, possibly halfway around the world, why wouldn’t you want to learn as much as possible. Knowing how to get around, where to eat, and being able to stay safe are not something that you might want to look up in the moment. As it is completely on your device, no one will judge you for wondering about the history of Ayers Rock, and they certainly will not use that data to sell you some advertisements.
I’m amazed what the OSM community has done and I’ve learnt so many new things that I never would have found out if it was not for the Travel Guides section, as well as the wikipedia integrations. While OSM may have some shortcomings in areas such as searching for locations of businesses, it’s reverse address searching, its ability to work without any type of internet connection (besides from the initial downloads) allows you to be guaranteed that you’ll be able to find your way no matter where you go. It is a great replacement for GMaps for those of you wanting something more privacy friendly, and completely open.
I’d love to hear your thoughts in the comments below!