Discussion Series: Incorporating ODK components into existing toolset

Support
1

Hi all,I am the product manager at Medic Mobile, a nonprofit technology
company that builds and deploys mHealth tools
http://medicmobile.org/tools for communities around the world. Using ODK
we were able to quickly build an Android app that health workers can use in
some of the most rural places on earth. Health workers use the app to
register people into health programs, screen for high-risk pregnancies,
submit stock reports, report disease outbreaks, and more. I'm excited to
share the story of how we got there.I was fortunate to meet Gaetano at the
ICTD 2013 conference in Cape Town, South Africa. Our team had been
designing tools for basic phones so that we could reach people with the
lowest common denominator platform. Android phones were become more
prevalent for our target users, so I was keen to chat with the ODK folks
from UW showcasing the new ODK 2.0. The next-generation ODK had promising
features for our users - most notably visualizing data on mobile devices,
which could enable real time, data-driven adaptations to service
delivery.Gaetano hinted, not so subtly, that we should be using ODK. We did
not want to reinvent the wheel, but from what we could tell at the time the
first generation ODK seemed already in the process of being replaced, and
ODK 2.0 was still very much in Alpha.Within a year, we were seeing more
Android phones in the hands of health workers. There was a clear
opportunity to add to the Medic Mobile platform. We wanted to build with
existing components to save time, and since ODK 2.0 was not yet ready we
decided that the benefits of ODK v1 outweighed its potential obsolescence.
Over the course of a few weeks we built a Collect version
http://medicmobile.org/blog/medic-mobile-has-an-android-app-1 that
connected with our webapp via SMS to fill our immediate needs. Then a true
test for “Medic Collect” came up after the horrific earthquake in Nepal
that left over 9,000 people dead. We were able to quickly deploy a disease
outbreak surveillance system with the health workers’ Android phones and
SMS.We also embarked on a longer journey to bring our core webapp directly
to mobile users as an offline-enabled Android app
http://medicmobile.org/blog/we-built-the-mobile-app-health-workers-deserve.
This app presents community health workers with a prioritized list of
upcoming tasks as well as their progress towards their goals. Because this
is a browser based app, we incorporated the ODK-based Enketo Smart Paper as
the form renderer. After about six months of intensive development, we are
deploying our first project with the new Medic Mobile Android app next
week. We’re happy to provide more details on our approach, and we’ll be
sure to share lessons learned. Also, I am curious how everyone on the list
has been planning for ODK 2.0. How have you balanced building features for
a mature ODK v1 toolset while anticipating a v2 which, although not ready
for production, may already include improvements you want to use? Has
anyone else used (or considered using) Enketo on mobile devices?In the past
year our toolset has been dramatically enhanced by taking Gaetano’s advice.
We built these apps faster and more sustainably because of tools built by a
dedicated ODK community. Now we’re excited to be active participants!Kind
regards,Marc

Marc Abbyad | Product Manager
Medic Mobile http://www.medicmobile.org/ @MarcAbbyad
https://twitter.com/marcabbyad
@Medic http://www.twitter.com/medic Facebook
https://www.facebook.com/MedicMobile

2

Dear Marc and the rest of the Medic Mobile team,

Thanks for this great post! It’s great to hear how you’ve incorporated ODK. I have a few questions, though I realize some of them may have not have short answers..

Am I right you all did a fair bit of software development on top of ODK v1? Were you able to contribute some of that back into the core or otherwise make it available? I realize that might be challenging if you customized it for your needs, so just curious to hear thoughts about this.

Is Medic still supporting the SIM application in addition to the Android app? Are there cases where you would recommend the SIM application for users that have Android phones?

Do you have a sense of how often forms are or will be submitted via SMS on the Android app vs. data? Or perhaps it’s too early to say.

I’m also curious to hear the answers to the questions you posed!

thanks again!

neal

image

··· From: opendatakit@googlegroups.com [mailto:opendatakit@googlegroups.com] On Behalf Of Marc Abbyad Sent: Tuesday, November 17, 2015 4:29 PM To: ODK Community Subject: [ODK Community] Discussion Series: Incorporating ODK components into existing toolset

Hi all,

I am the product manager at Medic Mobile, a nonprofit technology company that builds and deploys http://medicmobile.org/tools mHealth tools for communities around the world. Using ODK we were able to quickly build an Android app that health workers can use in some of the most rural places on earth. Health workers use the app to register people into health programs, screen for high-risk pregnancies, submit stock reports, report disease outbreaks, and more. I'm excited to share the story of how we got there.

I was fortunate to meet Gaetano at the ICTD 2013 conference in Cape Town, South Africa. Our team had been designing tools for basic phones so that we could reach people with the lowest common denominator platform. Android phones were become more prevalent for our target users, so I was keen to chat with the ODK folks from UW showcasing the new ODK 2.0. The next-generation ODK had promising features for our users - most notably visualizing data on mobile devices, which could enable real time, data-driven adaptations to service delivery.

Gaetano hinted, not so subtly, that we should be using ODK. We did not want to reinvent the wheel, but from what we could tell at the time the first generation ODK seemed already in the process of being replaced, and ODK 2.0 was still very much in Alpha.

Within a year, we were seeing more Android phones in the hands of health workers. There was a clear opportunity to add to the Medic Mobile platform. We wanted to build with existing components to save time, and since ODK 2.0 was not yet ready we decided that the benefits of ODK v1 outweighed its potential obsolescence. Over the course of a few weeks we built a http://medicmobile.org/blog/medic-mobile-has-an-android-app-1 Collect version that connected with our webapp via SMS to fill our immediate needs. Then a true test for “Medic Collect” came up after the horrific earthquake in Nepal that left over 9,000 people dead. We were able to quickly deploy a disease outbreak surveillance system with the health workers’ Android phones and SMS.

We also embarked on a longer journey to bring our core webapp directly to mobile users as an http://medicmobile.org/blog/we-built-the-mobile-app-health-workers-deserve offline-enabled Android app. This app presents community health workers with a prioritized list of upcoming tasks as well as their progress towards their goals. Because this is a browser based app, we incorporated the ODK-based Enketo Smart Paper as the form renderer. After about six months of intensive development, we are deploying our first project with the new Medic Mobile Android app next week.

We’re happy to provide more details on our approach, and we’ll be sure to share lessons learned. Also, I am curious how everyone on the list has been planning for ODK 2.0. How have you balanced building features for a mature ODK v1 toolset while anticipating a v2 which, although not ready for production, may already include improvements you want to use? Has anyone else used (or considered using) Enketo on mobile devices?

In the past year our toolset has been dramatically enhanced by taking Gaetano’s advice. We built these apps faster and more sustainably because of tools built by a dedicated ODK community. Now we’re excited to be active participants!

Kind regards,

Marc

Marc Abbyad | Product Manager
Medic Mobile http://www.medicmobile.org/ @ https://twitter.com/marcabbyad MarcAbbyad

@Medic http://www.twitter.com/medic Facebook https://www.facebook.com/MedicMobile

--

Post: opendatakit@googlegroups.com
Unsubscribe: opendatakit+unsubscribe@googlegroups.com
Options: http://groups.google.com/group/opendatakit?hl=en

You received this message because you are subscribed to the Google Groups "ODK Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opendatakit+unsubscribe@googlegroups.com mailto:opendatakit+unsubscribe@googlegroups.com .
For more options, visit https://groups.google.com/d/optout.

3

Thank you Neal for your great questions. I have answered them inline below.

Very best,
Marc

Dear Marc and the rest of the Medic Mobile team,

Thanks for this great post! It’s great to hear how you’ve incorporated
ODK. I have a few questions, though I realize some of them may have not
have short answers..

Am I right you all did a fair bit of software development on top of ODK
v1? Were you able to contribute some of that back into the core or
otherwise make it available? I realize that might be challenging if you
customized it for your needs, so just curious to hear thoughts about this.

Many of the customizations we made to Collect are useful beyond our use
cases. For instance form submission via SMS, scheduled android
notifications, and building APKs with default settings can be brought back
upstream. We have already been in contact with the ODK folks and look
forward to having those features in future ODK Collect versions.

For the v1 work that we are doing with Enketo, we have been staying in sync
with the Enketo core and contributing as we can as we go.

Is Medic still supporting the SIM application in addition to the Android
app? Are there cases where you would recommend the SIM application for
users that have Android phones?

Yes, we are still supporting the SIM application as we've found it to be
the best option for completing long or complex forms using basic phones -
which are still the best device for users with limited access to
electricity. The SIM application can also be deployed confidently with less
support and training, so there are still some scenarios to use them with
Android phones.

Do you have a sense of how often forms are or will be submitted via SMS on
the Android app vs. data? Or perhaps it’s too early to say.

So far, almost all health workers using Medic Collect are submitting their
forms via SMS. This includes a national scale project with a thousand users
in Senegal, as well as projects in Nepal.

I’m also curious to hear the answers to the questions you posed!

thanks again!

··· On Friday, November 20, 2015 at 3:42:56 PM UTC+3, Neal Lesh wrote: > > neal > > > > > > > > > > *From:* opend...@googlegroups.com [mailto: > opend...@googlegroups.com ] *On Behalf Of *Marc Abbyad > *Sent:* Tuesday, November 17, 2015 4:29 PM > *To:* ODK Community <opend...@googlegroups.com > > *Subject:* [ODK Community] Discussion Series: Incorporating ODK > components into existing toolset > > > > Hi all, > > > > I am the product manager at Medic Mobile, a nonprofit technology company > that builds and deploys mHealth tools for > communities around the world. Using ODK we were able to quickly build an > Android app that health workers can use in some of the most rural places on > earth. Health workers use the app to register people into health programs, > screen for high-risk pregnancies, submit stock reports, report disease > outbreaks, and more. I'm excited to share the story of how we got there. > > > > I was fortunate to meet Gaetano at the ICTD 2013 conference in Cape Town, > South Africa. Our team had been designing tools for basic phones so that we > could reach people with the lowest common denominator platform. Android > phones were become more prevalent for our target users, so I was keen to > chat with the ODK folks from UW showcasing the new ODK 2.0. The > next-generation ODK had promising features for our users - most notably > visualizing data on mobile devices, which could enable real time, > data-driven adaptations to service delivery. > > > > Gaetano hinted, not so subtly, that we should be using ODK. We did not > want to reinvent the wheel, but from what we could tell at the time the > first generation ODK seemed already in the process of being replaced, and > ODK 2.0 was still very much in Alpha. > > > > Within a year, we were seeing more Android phones in the hands of health > workers. There was a clear opportunity to add to the Medic Mobile platform. > We wanted to build with existing components to save time, and since ODK 2.0 > was not yet ready we decided that the benefits of ODK v1 outweighed its > potential obsolescence. Over the course of a few weeks we built a Collect > version > that connected with our webapp via SMS to fill our immediate needs. Then a > true test for “Medic Collect” came up after the horrific earthquake in > Nepal that left over 9,000 people dead. We were able to quickly deploy a > disease outbreak surveillance system with the health workers’ Android > phones and SMS. > > > > We also embarked on a longer journey to bring our core webapp directly to > mobile users as an offline-enabled Android app > . > This app presents community health workers with a prioritized list of > upcoming tasks as well as their progress towards their goals. Because this > is a browser based app, we incorporated the ODK-based Enketo Smart Paper as > the form renderer. After about six months of intensive development, we are > deploying our first project with the new Medic Mobile Android app next > week. > > > > We’re happy to provide more details on our approach, and we’ll be sure to > share lessons learned. Also, I am curious how everyone on the list has been > planning for ODK 2.0. How have you balanced building features for a mature > ODK v1 toolset while anticipating a v2 which, although not ready for > production, may already include improvements you want to use? Has anyone > else used (or considered using) Enketo on mobile devices? > > > In the past year our toolset has been dramatically enhanced by taking > Gaetano’s advice. We built these apps faster and more sustainably because > of tools built by a dedicated ODK community. Now we’re excited to be active > participants! > > Kind regards, > > Marc > > Marc Abbyad | Product Manager > Medic Mobile @MarcAbbyad > > > @Medic Facebook > > > >
4

Hi Marc,

Thanks for sharing your experience! Medic Mobile's new Android app looks
amazing. All the code contributions your team has made to Enketo are very
much appreciated.

I wanted to ask if there are any lessons learned in the process of creating
a mobile app using Enketo's form engine that you wish you had known in
advance. Several groups are currently doing something similar for iOS and
android, and it is something we'd like to encourage and support as much as
possible.

Cheers,
Martijn

··· On Thursday, November 26, 2015 at 12:51:47 PM UTC-7, Marc Abbyad wrote: > > Thank you Neal for your great questions. I have answered them inline below. > > Very best, > Marc > > > On Friday, November 20, 2015 at 3:42:56 PM UTC+3, Neal Lesh wrote: >> >> Dear Marc and the rest of the Medic Mobile team, >> >> >> >> Thanks for this great post! It’s great to hear how you’ve incorporated >> ODK. I have a few questions, though I realize some of them may have not >> have short answers.. >> >> >> >> Am I right you all did a fair bit of software development on top of ODK >> v1? Were you able to contribute some of that back into the core or >> otherwise make it available? I realize that might be challenging if you >> customized it for your needs, so just curious to hear thoughts about this. >> >> >> > > Many of the customizations we made to Collect are useful beyond our use > cases. For instance form submission via SMS, scheduled android > notifications, and building APKs with default settings can be brought back > upstream. We have already been in contact with the ODK folks and look > forward to having those features in future ODK Collect versions. > > For the v1 work that we are doing with Enketo, we have been staying in > sync with the Enketo core and contributing as we can as we go. > > > >> Is Medic still supporting the SIM application in addition to the Android >> app? Are there cases where you would recommend the SIM application for >> users that have Android phones? >> >> >> > > Yes, we are still supporting the SIM application as we've found it to be > the best option for completing long or complex forms using basic phones - > which are still the best device for users with limited access to > electricity. The SIM application can also be deployed confidently with less > support and training, so there are still some scenarios to use them with > Android phones. > > > >> Do you have a sense of how often forms are or will be submitted via SMS >> on the Android app vs. data? Or perhaps it’s too early to say. >> >> > > So far, almost all health workers using Medic Collect are submitting their > forms via SMS. This includes a national scale project with a thousand users > in Senegal, as well as projects in Nepal. > > >> >> >> I’m also curious to hear the answers to the questions you posed! >> >> >> > thanks again! >> >> neal >> >> >> >> >> >> >> >> >> >> *From:* opend...@googlegroups.com [mailto:opend...@googlegroups.com] *On >> Behalf Of *Marc Abbyad >> *Sent:* Tuesday, November 17, 2015 4:29 PM >> *To:* ODK Community >> *Subject:* [ODK Community] Discussion Series: Incorporating ODK >> components into existing toolset >> >> >> >> Hi all, >> >> >> >> I am the product manager at Medic Mobile, a nonprofit technology company >> that builds and deploys mHealth tools for >> communities around the world. Using ODK we were able to quickly build an >> Android app that health workers can use in some of the most rural places on >> earth. Health workers use the app to register people into health programs, >> screen for high-risk pregnancies, submit stock reports, report disease >> outbreaks, and more. I'm excited to share the story of how we got there. >> >> >> >> I was fortunate to meet Gaetano at the ICTD 2013 conference in Cape Town, >> South Africa. Our team had been designing tools for basic phones so that we >> could reach people with the lowest common denominator platform. Android >> phones were become more prevalent for our target users, so I was keen to >> chat with the ODK folks from UW showcasing the new ODK 2.0. The >> next-generation ODK had promising features for our users - most notably >> visualizing data on mobile devices, which could enable real time, >> data-driven adaptations to service delivery. >> >> >> >> Gaetano hinted, not so subtly, that we should be using ODK. We did not >> want to reinvent the wheel, but from what we could tell at the time the >> first generation ODK seemed already in the process of being replaced, and >> ODK 2.0 was still very much in Alpha. >> >> >> >> Within a year, we were seeing more Android phones in the hands of health >> workers. There was a clear opportunity to add to the Medic Mobile platform. >> We wanted to build with existing components to save time, and since ODK 2.0 >> was not yet ready we decided that the benefits of ODK v1 outweighed its >> potential obsolescence. Over the course of a few weeks we built a Collect >> version >> that connected with our webapp via SMS to fill our immediate needs. Then a >> true test for “Medic Collect” came up after the horrific earthquake in >> Nepal that left over 9,000 people dead. We were able to quickly deploy a >> disease outbreak surveillance system with the health workers’ Android >> phones and SMS. >> >> >> >> We also embarked on a longer journey to bring our core webapp directly to >> mobile users as an offline-enabled Android app >> . >> This app presents community health workers with a prioritized list of >> upcoming tasks as well as their progress towards their goals. Because this >> is a browser based app, we incorporated the ODK-based Enketo Smart Paper as >> the form renderer. After about six months of intensive development, we are >> deploying our first project with the new Medic Mobile Android app next >> week. >> >> >> >> We’re happy to provide more details on our approach, and we’ll be sure to >> share lessons learned. Also, I am curious how everyone on the list has been >> planning for ODK 2.0. How have you balanced building features for a mature >> ODK v1 toolset while anticipating a v2 which, although not ready for >> production, may already include improvements you want to use? Has anyone >> else used (or considered using) Enketo on mobile devices? >> >> >> In the past year our toolset has been dramatically enhanced by taking >> Gaetano’s advice. We built these apps faster and more sustainably because >> of tools built by a dedicated ODK community. Now we’re excited to be active >> participants! >> >> Kind regards, >> >> Marc >> >> Marc Abbyad | Product Manager >> Medic Mobile @MarcAbbyad >> >> >> @Medic Facebook >> >> >> >> >