Our promotion video is finished now in a beta version! We got a lot of feedback and are currently in the progress of a few further improvements. Also instruction videos are currently being made. So, on launch, we will have some videos to show!
The first week of this two-week sprint my kids have been sick again and I was not able to do much. In the second week I mostly fixed bugs/issues that arose during our beta testing. The app is now even more reliable, we also fixed a resource leak, some ux improvements also made it in. The app now informs the user when the connection to the server is lost, instead of simply not doing as requested.
We are working on a new illustration for our landing page, so it gets clear from visiting that there is something new, also the page is being revised, the FAQ reworked, mistakes corrected - this process will continue in the current sprint.
Now still ahead of us: Translation and fixing of some more bugs.
gonimo is now more reliable, in that it detects, in all cases, a loss in connectivity to the server. This was not the case previously because of the design of TCP. TCP, for saving resources does not transmit any data on it's own, this means for a baby station no data at all gets transmitted - because the baby station is only waiting for requests.
The problem is, transmitting no data means, you won't notice when the connection is long gone! This can only be detected by trying to send data and getting feedback from the network. So the fix was easy, instead of server-side WebSocket protocol pings that get responded to by the browser itself, we now use application level pings triggered manually by the app, this way, we can detect broken connections and trigger a reconnection procedure.
This one was mean, I noticed that there is no audio on mobile after connecting/disconnecting a few times. On my laptop I saw that pulse audio resources kept being created, but got never released. Turns out, this is a bug in chrome: Resources of video tags won't get freed automatically when the tag gets garbage collected! You have to set the srcObject to null yourself. But this was not the mean thing about it, I got this quite fast actually and had a fix quickly too, but the chrome dev tools made me believe that we still had a leak, so I kept searching for it for 2 days, to finally find out that there is none, the tools just did not report correctly.
Also the penetration test by FH Technikum Wien is in progress since the beginning of April, we hope it will be finished soon!
We are making progress! Production lauch is happening soon - we target the end of this week or worst case end of this sprint, so in two weeks!