You’ve done your user-centered design, you’ve assessed your constituents, and you’ve determined that an Android application is the best way to deliver your intervention, using technology. Congratulations!
Yet before you hire the Android development team in celebration, make sure your fancy application is designed to be used fully offline too.
All Weather is Local
With all the talk about cloud-based services for development, and all the promise of Android applications that leverage big data and cloud infrastructure, the reality in almost every country is that all weather is extremely local.
Many areas will have enough Internet bandwidth to handle Android applications that need an always-on connection, yet most rural and underserved areas, will not – by definition. Therefore, you need an offline mode to be effective.
Stop Denying Your Constituents
Recently, I worked with an organization that had an amazing Android application. It is a cutting edge ICTforAg app that has real, demonstrated impact with farmers – when it works.
However, its current incarnation requires Internet connectivity to operate, which is a rarity in rural agricultural areas. Therefore, constituents constantly complained that the tool was inaccessible in farmers’ fields.
The organization was hesitant to make an offline version because they felt that it would allow for outdated versions and inaccurate advice, as not everyone would update their application with each release. Yet, the greater problem was a lack of usage of the current tool, regardless of its accuracy, since it required rare and often totally inaccessible Internet connectivity.
Go Offline Any Way You Can
That organization, and others, is exploring the use of coded SMS as a way to relay data back to a server and get analyzed results when only basic mobile connectivity is available. Yet even requiring coded SMS is often asking too much. SMS can get garbled, delayed, and lost, especially in low-bandwidth environments.
A truly effective ICT4D application does not require Internet connectivity. It should be able to operate offline with a solid feature set, which is enhanced when online, not a useless brick without 4G or WiFi.
Word!
Couldn’t agree more, whilst mobile has revolutionised services, coverage is focused on population centres for ROI reasons. Now if only there was a global coverage solution without spending hundreds of dollars on satellite equipment and plans…
May also be symptomatic of web developers moving over to mobile solutions, as early day apps which often did have an offline capabilities were built by developers coming out of mobile vendors and operators.
Did you see Jame’s comment below about “offline first” mobile design?
Thank you for this wonderful insight. I wish all ICT4D enthusiasts can read this. Any good application many for people in developing countries needs an offline version. It is what we have been telling people that our digital learning platform http://www.yaaka.cc invests more time and resources in the offline app and the offline connected solution. Even in 10years, the internet connection wont support everyone in villages to access useful news, information or knowledge off the cloud, hence the need for more offline solutions.
Couldn’t agree more with Wayan, that due to intermittent or low internet connectivity, android apps should work offline.
We are a ICT4D venture working in this field for last 3 years. In all our deployments, Offline apps (Web or Android) is the crucial part. Since offline apps restrict some functionalities and do not provide a seamless or realtime monitoring framework but at the same time in our experience it is very useful and provides a lot of freedom to field executives.
You may check our app portfolio here: https://goo.gl/jBE9hL
Will be happy to share our experience and learn from other fellows on the same.
Thanks,
Swapnil
Thanks for this article Wayon, and my thoughts exactly. A tool that can be flexible offline, can be updated when you get near decent internet, etc. is key. Currenlty working in northern Liberia, and in the past in southern Malawi and western Zambia in entirely rural areas … trying to implement someone else’s whizbang product and watching it meltdown for lack of connection often caused me to meltdown any number of times, because that lack of offline functionality negated the positive effects of the technology.
This post was inspired by me melting down in a number of times too!
The latest design approach in mobile is “offline first” – http://offlinefirst.org/.
Wow! And to think its starting in Berlin of all places.
This is a __really__ good reminder. It is shocking how many people that should know better, never-the-less ignore areas with poor Internet access.
Stephen
Another situation that some developers miss is what we at mWater call “bad internet.” This is when the app thinks the internet is available, but there effectively is no connectivity due to a variety of things, such as poor backhaul connectivity (getting from the tower to the rest of the internet), governments intercepting traffic and messing with packets, or just too much demand on the network. So the app needs to be tolerant to having a connection but not always getting a response when expected from the server.
Thanks for making this point so clearly John. It could have interesting implications for state management in our IEEE ADB specification work and I’ll raise it.. (Look us up if you are interested to contribute directly: standards.ieee.org/develop/indconn/adb/).
Dear Wayan,
Exactly this is what we are saying to the Tech Dev community. Half of the world is not connected so the solution must work offline. Our solution is designed ground up thinking there is no Internet connectivity & power. – Our solution is A portable, solar-powered, learning ecosystem that creates digital classrooms in the middle of nowhere – without the need for electricity, Internet or expensive AV rooms!!
Yes and not only in the developing world! Crowded central stations in Europe (thinking of Hamburg in my case) or sitting in traffic on my daily commute, that is when I mostly usw apps and experience daily connectivity gaps!
Yes! developers must develop apps that can work on smaller scale. More than 60% of people on the African continent are agriculturalists – with no access the the internet. This is why Village Telco’s or Village Webs, working off local mesh networks is so important. A village serve could host all the necessary biggish data while villagers could also still talk to one another — some villages stretch over long distances.