For numeric input, offer a slider appearance. Min and max would be determined from constraint?

javarosa has better support for repeats. it's in the core, but we need to build a ui
around it in collect for our next release.

A really cool thing in conjunction with the "form shortcut" would be to have a "send
forms" shortcut.  Then, it would just be a matter of clicking the form to collect the
data and then clicking on "send forms" to see the list and send them.  I'd like to
be able to set up the app and forms on a device and then have someone who may not know
ODK very well use it to collect data.

It doesn't seem like a problem to allow the application to be installed to a SD card,
please allow by adding installLocation to the manifest:

http://developer.android.com/guide/topics/manifest/manifest-element.html#install

Hi, I've installed ODK Collect v1.1.7-RC1 and successfully uploaded data to Aggregate
V1.0 beta rev 3.

However I've noticed that after I've collected data the main menu shows
  * Start New Form
  * Continue Saved Form
  * Send Finished Forms
  * Manage Forms

as I would expect - but the earlier verson of Collect (v1.1.5) would also indicate
that there was outstanding form data to upload (against 'Send Finished Forms').

This is a useful reminder that there is data available to upload.

What version of the product are you using? On what operating system?
 - HTC Wildfire
 - Android 2.2.1
 - UK

when an input field is defined as type "int" the keyboard pops up as the number keyboard.
it would 
be better if it popped up the 9 digit number pad.

it might be a good idea to have the equivalent of a multi-tap keyboard. that way you
can use it for 
integers (one tap) and also text entry for those who prefer it to a qwerty or find
it more familiar.

http://developer.android.com/resources/samples/SoftKeyboard/index.html is an example

What tool are you using?
ODK Collect 1045 (Device LG G2)

What is frustrating, troubling, annoying or lacking?

A new widget to record compass bearings was added to ODK collect in Oct 2013.
See:
https://code.google.com/p/opendatakit/source/detail?r=987078d8b2f6c9973b4b208000366b520bd149e1&name=default&repo=collect

From the above commit, the values for pitch and roll are also read from the device's
sensors (however these 2 values are //commented out in BearingActivity.java)

How would you like it to work?

Similar to the new 'bearing' widget, to have 2 new widgets for "pitch" and "roll  "bearing"
of type decimal, appearance="pitch" and "roll" which use
the device's g sensor to get readings of the orientation of the device.

For surveys that consist of interviewing people, it might be interesting to keep a running
audio of 
the whole interview in parallel to filling the form. Currently ODK Collect only supports
standalone 
audio questions.

Launching an activity to do this is possible, but can we make it easier?

One of our forms was for election results, and was effectively a table of numbers. Is
there any better way to do this besides one question per cell in the table (e.g. "How
many votes for UP in the presidential race?")

On a Galaxy Tab, Environment.getExternalStorageDirectory() returns the internal memory
"/sdcard/". The actual SD card appears at "/sdcard/external_sd/". This is in contrast
to the Galaxy S, which mounts its sdcard at /sdcard/

http://stackoverflow.com/questions/4210154/android-sd-card-characteristics-on-samsung-galaxy

http://stackoverflow.com/questions/5524105/how-could-i-get-the-correct-external-storage-on-samsung-and-all-other-devices

"I am using the DateWidget and its working fine if I am using the
langauge English but facing problem when I am using other than English
as device langauge .
For other langauge its throwing error like

java.text.ParseException: Unparseable date: "Fri Dec 30 00:00:00 GMT
+05:30 2011" (at offset 0)
                                     at
java.text.DateFormat.parse(DateFormat.java:626)"
- Mahesh

http://code.google.com/p/opendatakit/issues/entry

For Android OS (ARM based), we have the driver and ScanAPI for the Futronic FS80 to
capture a fingerprint image, but it has the following requirement:
 1. Have an USB Host port
 2. Have usbfs support in the kernel
 3. Have the root access right
 4. Have a folder "/data/" that can be read/write access.

Attached please find the libusb and ScanAPI libraries for Android, and the console
demo program with source code. To test it, you need to copy the libraries to the system
lib folder, then open the terminal, run as root user:
    #./ftrScanAPI_Ex_fs.

"Problem is now I have at least four servers running in different
places, and perhaps I should get rid of most, but I do change servers
often lately. Is it possible that in Collect's settings I can set up
multiple servers or at least have a sort of "history" selection for
different servers. For example. When not near my home network, I may
rather want to upload to GAE instead for easy access from a different
location.

Of course downloading identical forms from different servers could be
a problem, but as for me I would not mind selection all four servers
and uploading in one go when possible. (Or perhaps I need a lecture on
how to handle this info.)"
-- bergie

in collect's time/date widgets, change the code to stop writing our custom strings,
but instead writing a iso8601 string. yyyy-mm-ddTHH:MM:SSZ±hh:mm seems reasonable.
to support partial datetime (in case someone wants just a date as a default value and
for backwards compatibility), might be worth it to look at http://joda-time.sourceforge.net/index.html
to do the parsing.

i have no idea how datetime is done in aggregate, so i'll leave that to mitch. the
problem i see there is that aggregate exports to different systems so lots of bad things
might happen... 

We should definitely add some unit tests for this in.  Yaw - do we have a place / folder
in the repo 
where can start to get unit tests in there?

The two common ways would be either

src/org/odk/collect/android/tests

or we have both

src
tests

and mirror the director structures with the tests for class 
org.odk.collect.android.database.FileDbAdapater in 
tests/odk/collect/android/dabase/FileDbAdapterTests.java

Also, add back the counts, especially for Send Finalized Form. Those counts helped the
team know that they had data that had not been reported.

-- Alex

just like location, i'd like to get the direction i'm pointing...

What steps will reproduce the problem?
1. Upload a form to any auth-enabled server
2.
3.

What is the expected output? What do you see instead?
One would expect one HEAD to get the 401, then another HEAD with the proper credentials,
and then one POST to do the upload. Instead two POSTs are being sent. So evidently
the credentials are not being sent preemptively.

What version of the product are you using? On what operating system?
1.1.7RC, Android 2.3.4

Please provide any additional information below.

Steps to reproduce:
1.Open a form in ODK Collect v1.1.4 while running Android 2.2

The form should go into landscape mode, but instead the form crashes with a
popup window saying "The application ODK Collect (process
org.odk.collect.android) has stopped unexpectedly. Please try again."

I am running Android 2.2 on a Nexus One with the latest ODK Collect from
the svn repository (which says v1.1.4). I have tested this out on the
Android 2.2 emulator with an identical result. Here is the log from the
phone from the point of switching to landscape while a form is open:

05-23 23:59:02.528: INFO/WindowManager(88): Setting rotation to 1, animFlags=0
05-23 23:59:02.548: INFO/ActivityManager(88): Config changed: { scale=1.0
imsi=310/260 loc=en_US touch=3 keys=1/1/2 nav=3/1 orien=2 layout=34
uiMode=17 seq=92}
05-23 23:59:02.598: DEBUG/FormEntryActivity(8162): onPause
05-23 23:59:02.608: INFO/UsageStats(88): Unexpected resume of
org.odk.collect.android while already resumed in org.odk.collect.android
05-23 23:59:02.668: INFO/PropertyManager(8162): calling constructor
05-23 23:59:02.688: DEBUG/AndroidRuntime(8162): Shutting down VM
05-23 23:59:02.688: WARN/dalvikvm(8162): threadid=1: thread exiting with
uncaught exception (group=0x4001d7e8)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162): FATAL EXCEPTION: main
05-23 23:59:02.698: ERROR/AndroidRuntime(8162): java.lang.RuntimeException:
Unable to start activity
ComponentInfo{org.odk.collect.android/org.odk.collect.android.activities.FormEntryActivity}:
java.lang.NullPointerException
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.app.ActivityThread.performLaunchActivity(ActivityThread.java:2663)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.app.ActivityThread.handleLaunchActivity(ActivityThread.java:2679)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.app.ActivityThread.handleRelaunchActivity(ActivityThread.java:3815)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.app.ActivityThread.access$2400(ActivityThread.java:125)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.app.ActivityThread$H.handleMessage(ActivityThread.java:2037)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.os.Handler.dispatchMessage(Handler.java:99)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.os.Looper.loop(Looper.java:123)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.app.ActivityThread.main(ActivityThread.java:4627)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
java.lang.reflect.Method.invokeNative(Native Method)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
java.lang.reflect.Method.invoke(Method.java:521)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:868)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
com.android.internal.os.ZygoteInit.main(ZygoteInit.java:626)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
dalvik.system.NativeStart.main(Native Method)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162): Caused by:
java.lang.NullPointerException
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
org.odk.collect.android.activities.FormEntryActivity.refreshCurrentView(FormEntryActivity.java:284)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
org.odk.collect.android.activities.FormEntryActivity.onCreate(FormEntryActivity.java:174)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.app.Instrumentation.callActivityOnCreate(Instrumentation.java:1047)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     at
android.app.ActivityThread.performLaunchActivity(ActivityThread.java:2627)
05-23 23:59:02.698: ERROR/AndroidRuntime(8162):     ... 12 more
05-23 23:59:02.718: WARN/ActivityManager(88):   Force finishing activity
org.odk.collect.android/.activities.FormEntryActivity
05-23 23:59:02.908: DEBUG/dalvikvm(144): GC_EXTERNAL_ALLOC freed 1459
objects / 67320 bytes in 113ms
05-23 23:59:03.225: WARN/ActivityManager(88): Activity pause timeout for
HistoryRecord{44bff5a0 org.odk.collect.android/.activities.FormEntryActivity}
05-23 23:59:05.008: DEBUG/dalvikvm(88): GC_EXPLICIT freed 3463 objects /
161320 bytes in 101ms
05-23 23:59:12.028: DEBUG/dalvikvm(7811): GC_FOR_MALLOC freed 4289 objects
/ 632496 bytes in 58ms
05-23 23:59:12.726: WARN/ActivityManager(88): Launch timeout has expired,
giving up wake lock!
05-23 23:59:16.678: INFO/Process(8162): Sending signal. PID: 8162 SIG: 9
05-23 23:59:16.708: INFO/ActivityManager(88): Process
org.odk.collect.android (pid 8162) has died.
05-23 23:59:16.708: INFO/WindowManager(88): WIN DEATH: Window{44c56590
org.odk.collect.android/org.odk.collect.android.activities.MainMenuActivity
paused=false}
05-23 23:59:16.718: INFO/ActivityManager(88): Start proc
org.odk.collect.android for activity
org.odk.collect.android/.activities.MainMenuActivity: pid=8175 uid=10083
gids={3003, 1015}

Sorry if I am posting in the wrong location.

I am interested in whether it would be possible to visualize completed forms in a map
on the mobile device (through google earth or any other app).

Cheers,
Jeremy

Even if users enter their credentials in the prefs screen, a prompt always appears on
the first request after ODK is opened. This is both confusing and slows things down
as two request/responses must take place instead of one. Is there any reason for the
prompt or could we just send the saved credentials with the first request and only
show the dialog if there is an auth failure?

What steps will reproduce the problem?
1. install the attached form (.xml file) on a device
2. open Collect and start a new form, choose JavaRosa-Test
3. browse past the first page

What is the expected output? What do you see instead?
expect to see the last page. Instead, Collect crashes.

What version of the product are you using? On what operating system?
Collect 1.1.7-RC2 on a Samsung Galaxy Tab P1000

Please provide any additional information below.

odk.org is taken, so we should probably refactor to opendatakit

Can we get ODK Collect to work in WiFi connections where a proxy server is 
present? I noticed that the presence of a proxy server prevents the app from 
uploading the file/form. 

It would be helpful if ODK Collect could automatically download the available set of
forms on a regular basis. This would keep the local cache of forms up to date, and
would also make sense for enumerators who had less than ideal data connections.

What version of the product are you using? On what operating system?

1.1.4 on Android 1.5 and 1.6

What steps will reproduce the problem?
1. Set up authenticated access to a server
2. Download a form
3. Change user password
4. Access server again.

What is the expected output? What do you see instead?

One would expect that the change of password would invalidate the authentication set
up with the server, prompting for a new authorization.

In fact, it does nothing, as the connection uses the established credentials.  This
causes failures later on when the app is stopped and restarted.

Very non-transparent.

Enhancement

I would love to see a day of the week added to the date input question, or something
that shows up as a calendar where a date can be selected.

While the date works well for birthdates, we often get answers like 'last friday' or
'2 weeks ago'.  We've done single select options for this, but transforming the data
later into actual dates is not easy.  It would be great if an enumerator got an answer
like 'Friday' and be able to have a way for this to be translated into the date that
was last friday (referenced from today).

It would be great if it was possible to introduce incremental numbering in ODK Collect;
for instance when adding a new household item that the correct number 1,2,3, will automatically
be assigned.

Ideally, this would even work across filled data instances for the same form (as to
keep your household numbers consistent when doing a survey and starting with a new
household altogether where each enumerator gets a batch of numbers beforehand) but
I am not sure if that is technically possible at all.

Many thanks for looking into this!

as android devices continue to ship with better cameras, we are sending pictures that
are unnecessarily huge. 

there is no easy way to do this in the intent (especially with 1.5/1.6 support), so
in the submission activity, we should likely compress before we send. 

it's not clear if this should be a form or device setting.

I'm not sure if this is the right place for feature requests, but it would be cool to
have a widget to collect a path of gps coordinates. Repeating a geopoint question is
quite cumbersome. Some folks at AfSIS have had good luck with the Distance and Area
Measurement Android app.

Andrew

Add an integer widget with plus/minus sign

"I need to count cars, buses, trucks for lets say 15 minutes... all in
one "group" so it will show in one page only."
--  [email protected]

Relatedly, I noticed that there's a text box at the end of the survey to set the "name"
of the form, which is useful in identifying it for submission or later continuation.

I was wondering whether it might be possible to fill this in at the beginning of the
survey instead of at the end? Or to base it on a value in the instance data, selected
by the form designer (such as the household ID in our case), at least as a default
value?

-- chris

For <group> sections, the group label is shown as a header above the question labels.
The font 
used for the group header is pretty small, so users tend to overlook it and get confused.

It would be good to include as part of the ODK collect app location page, a small component
that gives visual feedback on the status of the GPS location gathering operation (ie
like the GPS test app) and show the number of satelites in view, how many are active
etc.

I was testing my auth code on an HTC Droid Eris phone earlier, and noticed that it seems
to be 
uploading photos of zero length to my server.  This doesn't seem related to the code
I added --- I 
dropped back to the release apk and it has the same behavior.  (It all works fine under
the emulator 
as well.)  I put it under the debugger and JPEG file being attached to the form post
is reporting a 
content length of zero bytes in the InstanceUploaderTask.  This is kind of weird, because
the photos 
are being taken and show up in the camera app.

It's a Verizon phone running 1.5.  I don't think it's an issue on the server side,
since I see the 
problem both on my local server and on open-data-kit.appspot.com, and my stepping through
the 
code makes it look like an empty file is being sent up from the client.

Randomising layout or structure of a questionnaire to minimise bias is a standard best
practice. It would be great to therefore have support for random().

A suggestion that ODK collect could incorporate EGNOS Differential corrections (European
and North African). They have an
Android/iPhone Library available.

http://egnos-portal.gsa.europa.eu/developer-platform/egnos-toolkits/egnos-sdk/sdk-core-installation-procedure

What steps will reproduce the problem?
1. Have a form that includes at least an image, a video and an audio field
2. Start new form, select that form, record the video, audio and image,
3. Save the form instance and mark it as finished
4. Browse to the folders where media files are recorded by default

What is the expected output? What do you see instead?
the video audio and image should be deleeted from thei location where they are originally
recorded, and MOVED (as opposed to copied) to the instance folder. This at least should
be the case by default when recording on the fly. It may be goood to have a choice
in the setttings, but this should be the default when you record on the fly (not when
you choose an existing file)

Instead, media are copied and the original kept. Not sure about audio because i dont
know where it s saved but this happens with fotos and video

What version of the product are you using? On what operating system?


Please provide any additional information below.

I'm trying to implement a custom server, so there's almost surely an issue on the server
side; however, even in the case of wrong or even malicious response from the server,
ODK collect should never crash.

It looks like it crashes when retrieving the formList over https if the server doesn't
have a valid certificate (I'm only guessing)

What steps will reproduce the problem?
1. get the form list from a server with https without a certificate

What is the expected output? What do you see instead?

Expected is some meaningful error message to the user saying that the identity of the
server cannot be verified

Observed is that ODK Collect crashes


I think this is the relevant part of the log:

09-24 18:50:13.152 W/System.err(25941): javax.net.ssl.SSLPeerUnverifiedException: No
peer certificate
09-24 18:50:13.152 W/System.err(25941): at org.apache.harmony.xnet.provider.jsse.SSLSessionImpl.getPeerCertificates(SSLSessionImpl.java:258)
09-24 18:50:13.152 W/System.err(25941): at org.apache.http.conn.ssl.AbstractVerifier.verify(AbstractVerifier.java:93)
09-24 18:50:13.152 W/System.err(25941): at org.apache.http.conn.ssl.SSLSocketFactory.createSocket(SSLSocketFactory.java:381)
09-24 18:50:13.152 W/System.err(25941): at org.apache.http.impl.conn.DefaultClientConnectionOperator.openConnection(DefaultClientConnectionOperator.java:164)
09-24 18:50:13.152 W/System.err(25941): at org.apache.http.impl.conn.AbstractPoolEntry.open(AbstractPoolEntry.java:164)
09-24 18:50:13.162 W/System.err(25941): at org.apache.http.impl.conn.AbstractPooledConnAdapter.open(AbstractPooledConnAdapter.java:119)
09-24 18:50:13.162 W/System.err(25941): at org.apache.http.impl.client.DefaultRequestDirector.execute(DefaultRequestDirector.java:359)
09-24 18:50:13.162 W/System.err(25941): at org.apache.http.impl.client.AbstractHttpClient.execute(AbstractHttpClient.java:555)
09-24 18:50:13.162 W/System.err(25941): at org.apache.http.impl.client.AbstractHttpClient.execute(AbstractHttpClient.java:487)
09-24 18:50:13.162 W/System.err(25941): at org.odk.collect.android.utilities.WebUtils.getXmlDocument(WebUtils.java:216)
09-24 18:50:13.162 W/System.err(25941): at org.odk.collect.android.tasks.DownloadFormListTask.doInBackground(DownloadFormListTask.java:82)
09-24 18:50:13.162 W/System.err(25941): at org.odk.collect.android.tasks.DownloadFormListTask.doInBackground(DownloadFormListTask.java:1)
09-24 18:50:13.162 W/System.err(25941): at android.os.AsyncTask$2.call(AsyncTask.java:185)
09-24 18:50:13.162 W/System.err(25941): at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:306)
09-24 18:50:13.162 W/System.err(25941): at java.util.concurrent.FutureTask.run(FutureTask.java:138)
09-24 18:50:13.172 W/System.err(25941): at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1088)
09-24 18:50:13.172 W/System.err(25941): at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:581)
09-24 18:50:13.172 W/System.err(25941): at java.lang.Thread.run(Thread.java:1027)
09-24 18:50:13.192 W/dalvikvm(25941): threadid=12: thread exiting with uncaught exception
(group=0x4001d5a0)
09-24 18:50:13.213 D/dalvikvm(25941): GC_CONCURRENT freed 418K, 50% free 3067K/6023K,
external 0K/0K, paused 1ms+1ms
09-24 18:50:13.213 E/AndroidRuntime(25941): FATAL EXCEPTION: AsyncTask opendatakit/opendatakit#4
09-24 18:50:13.213 E/AndroidRuntime(25941): java.lang.RuntimeException: An error occured
while executing doInBackground()
09-24 18:50:13.213 E/AndroidRuntime(25941): at android.os.AsyncTask$3.done(AsyncTask.java:200)
09-24 18:50:13.213 E/AndroidRuntime(25941): at java.util.concurrent.FutureTask$Sync.innerSetException(FutureTask.java:274)
09-24 18:50:13.213 E/AndroidRuntime(25941): at java.util.concurrent.FutureTask.setException(FutureTask.java:125)
09-24 18:50:13.213 E/AndroidRuntime(25941): at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:308)
09-24 18:50:13.213 E/AndroidRuntime(25941): at java.util.concurrent.FutureTask.run(FutureTask.java:138)
09-24 18:50:13.213 E/AndroidRuntime(25941): at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1088)
09-24 18:50:13.213 E/AndroidRuntime(25941): at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:581)
09-24 18:50:13.213 E/AndroidRuntime(25941): at java.lang.Thread.run(Thread.java:1027)
09-24 18:50:13.213 E/AndroidRuntime(25941): Caused by: java.lang.NullPointerException
09-24 18:50:13.213 E/AndroidRuntime(25941): at org.odk.collect.android.utilities.WebUtils.getXmlDocument(WebUtils.java:327)
09-24 18:50:13.213 E/AndroidRuntime(25941): at org.odk.collect.android.tasks.DownloadFormListTask.doInBackground(DownloadFormListTask.java:82)
09-24 18:50:13.213 E/AndroidRuntime(25941): at org.odk.collect.android.tasks.DownloadFormListTask.doInBackground(DownloadFormListTask.java:1)
09-24 18:50:13.213 E/AndroidRuntime(25941): at android.os.AsyncTask$2.call(AsyncTask.java:185)
09-24 18:50:13.213 E/AndroidRuntime(25941): at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:306)
09-24 18:50:13.213 E/AndroidRuntime(25941): ... 4 more
09-24 18:50:13.213 W/ActivityManager( 1487):   Force finishing activity org.odk.collect.android/.activities.FormDownloadList

As for the group pop-up, I guess that I would like at least the terminology to change,
avoiding the term "group"  that doesn't make sense except when building the form. 
I would actually like it to use the name of the group or something like that.

For example, in one of the Market screenshots, there is a form record bird observations.
 It would be nice if the title and content of the popup was "Add a new 'observation'"
(and "Add another 'observation'" on subsequent ones).  Then the buttons could just
say, "Add" and "Do Not Add" (or "Continue", maybe).

Making just that change would be great with me.

What would take more work, but I think would be a little better is to add buttons to
handle what the pop-ups do.  I have attached a screenshot mock-up that shows what I
mean.   I just rolled what goes at the top (every page of the group) and what goes
at the bottom (last page of group) into the one page.  It adds a "Skip 'observation'"
button at the top to jump out of the group, and a "Next 'observation'" and "Done" buttons
at the bottom.

Enhancement.

Many answers we get are given as a range: 1-2 weeks, 3-4 days, etc.  Is there some
way that a type of answer can be a range (I'm thinking two sliders of numbers, like
the ones used for dates?)

Photos can't be rotated, so when you take a picture it comes in as 
landscape regardless of the orientation you took the picture, so upon 
importing into fusion forms the default info windows sometimes show 
sideways photos. I've trained myself to not take portrait shots 
anymore. Even though portrait is much better for snapping shots of 
trees. 

Galaxy Tab OS 2.2

when you get a dialog and rotate, the dialog disappears and is not brought back. 

it would be nice to get some visual confirmation (shake left to right) when at the beginning
or end 
of form entry

submissions currently take place over https. support password protected and
fully encrypted data communications when available. 

Slide up/down for ranking of items (perhaps enable/disable individual items for "N/A"?)

What tool are you using?
Annotate widget & Draw widget

What is frustrating, troubling, annoying or lacking?
I'd like to see an option to change the line width or at least have the line width
default to a thinner line.

How would you like it to work?
next to set color have an option for change line width.


NOTE: the core team has limited ability to make software enhancements.
We rely heavily on community-contributed enhancements to the tools. If this
is something you very much need, consider hiring a developer to make the
changes with the stipulation that they work with the core team to specify
the details of the change and contribute the change back to the project
under an Apache 2 license.

Enhancement feature.

In the "Change Settings page" 

The ability to sort or group the saved forms by a common thread.
also the ability to default name the finalized form using a populated form field.

Thanks!

 Keep Up the Good Work !! 
 and I apologize if these features have been asked for .. did not see them in the issues
forum. 

I'm doing fieldwork now in Ghana and we're finding that enumerators
are accidentally deleting blank forms when they intend to delete
instances of data collection.

For the short-term, we will probably just try to quickly re-train them
to avoid this.  But I have two design suggestions.

In the short-to-medium term, I would recommend adding a check box,
hidden under "change settings" that says something like "Protect Blank
Forms From Deletion".  It's easier as a trainer to prevent staff from
messing with the "settings" than it is to stop them from trying to
delete data instances that they've collected and want to delete.

But in the longer term, I think this is a bit of an interface design
problem with ODK Collect.  When looking at the two tabs visible after
clicking:  "Delete Saved Forms" it can be confusing which of the two
tabs is actually active (in bright sun, the black and white tab has
higher contrast, and looks active, when in fact it's the dark-gray on
lighter gray that is active).  I've come close to accidentally
deleting blank forms myself...

What steps will reproduce the problem?
1. Use the widget form from the ODK repo
2. Take a photo in the image question
3. Retake the photo after it displays
4. It should crash


What is the expected output? What do you see instead?
I get this error from the log (http://dpaste.com/223770/) 

What version of the product are you using? On what operating system?
v1.1.4 on Cyanogenmod 5.0.8



Please provide any additional information below.