I am currently working on Amazon Alexa via the Amazon Web Developer Console, and they have added a new "personalization" feature that I dont need. It is not enabled for this skill, and I dont intend to do so. Nevertheless, the viewing window for this feature takes up more than half of my screen in the testing window, and I cannot remove it, nor scroll it out of view, which leads to me being unable to properly view the important parts of the testing page, the Skill IO and the display.
Has anyone had this problem? How can I get rid of this personalization window
Screenshot
Terrible design, you can't for now (unless manually removing it on your browser).
Submit a feedback on the page (bottom left) so they will be aware and fix it soon hopefully:
Related
I'm aware that the Amazon Connect CCP is based on the idea that it can be developed and integrated in any website at will without the restrictions of a closed product. However the fact that it is embedded, means it has some agent usability drawbacks. While browsing to multiple tabs you can eventually loose the tab where CCP resides. This means that if the agents needs to mute, hold, ... it might take a while before they find the right tab.
Is it possible to create a CCP that it can be always visible?
Thanks,
I set up an external screen pop as detailed in the AWS blog, and I have the issue on Windows 10 and Chrome. When call coming the pop-up notification apears, when clicking on the pop-up notification, you are not redirected to the CCP browser window.
However, I also tried the same setup on Firefox and the issue disappears: when clicking on the pop-up notification, the CCP browser tab gains focus correctly. Additionally, I set up another screen pop with a different instance and S3 bucket, using the files provided in the blog, and it yielded the same result: the issue appears with Chrome, but not with Firefox. It all points in the direction of it being an issue related to the integration between Chrome and the Windows pop-up notifications (the same problem for MacOS).
For some reason, whenever I scroll through the documentation (particularly, the developer guides) of AWS, the scroll is very flickery. On the other hand, when I save the raw HTML and browse through the docs "offline" (using this Chrome extension for easier replication), it scrolls smoothly. Checking the performance of the actual (non-offline) page render using Chrome, it looks like the brunt of the work is being done during "Painting."
Actual
Offline
Why is this?
I have a Glass application that is activated by a voice command. The user requests a product and the request is then passed on to a node.js server, which handles the bulk logic of my application. When the node server is done, it builds some HTML templates that it passes on to the Mirror API, which then puts those cards into the timeline of the Glass app.
Now the problem I have is this:
When a user requests something from within the application, he just stays within that application but the cards from the Mirror API are put onto the timeline.
Is it possible to navigate the user from within the application to the timeline item that has just been inserted?
As I build the Mirror API timeline item, I know ahead of time what the bundleId will be on the timeline, so can I use that somehow to navigate to that item on the timeline?
I've had a look here on Stackoverflow, the Glass forums and the official documentation and couldn't see anything that would help.. Anybody here have any ideas?
Currently the only way for forcing a specific card to show in the timeline is by using a LiveCard (https://developers.google.com/glass/develop/gdk/reference/com/google/android/glass/timeline/LiveCard). Live cards will automatically be given focus when published and you can force the timeline to give your card focus after publishing by calling LiveCard.navigate().
So unfortunately unless you want to change some of the fundamentals of your app, this isn't going to be much help.
Though from the sounds of it, you may be able to use a LiveCard to show what is returned from your server. If you want to perform inserts via the mirror API, this could still be done in addition.
I am using meteor accounts-ui/accounts-facebook to login and am getting the following error.
You are using a display type of 'popup' in a large browser window or tab. For a better user experience, show this dialog with our JavaScript SDK without specifying an explicit display type. The SDK will choose the best display type for each environment. Alternatively, set height and width on your window.open() call to properly size this dialog if you have special requirements precluding you from using the SDK. This message is only visible to developers of your
http://goo.gl/IBmdjI (screenshot)
I haven't been able to find any information on the fix anywhere. Is this something in meteor core that needs to be changed?
This isn't a issue really its just a warning. Meteor doesn't use the JS Facebook SDK and gives a specified height/width for the facebook OAuth dialog & your browser size may be too big for the small popup (not too sure- looks ok in your screenshot).
If you disable your app from Sandbox mode/another user uses your app they wouldn't see this warning anymore.
The alternative is to use a redirect to facebook or a mobile display. The best option is the popup (what you have now) because it's the fastest: A redirect means your Meteor app has to reconnect up.
I use www.jango.com as PC music player and want to steer it remotly via my smartphone. I have a server program for controling the PC and an app as client.
I want to use the server program for example to press the play/pause or fast forward button in jango.com running inside the chrome webbrowser.
A simple way would be to simulate a windows mouse click at a specified position within the chrome window. But it would have the disadvantage that it would only work if the jango tab is active and will of course fail if the site changes it's layout and probably due to a lot other reasons.
So I'm thinking about generating kind of the same http request as the browser does if one clicks the button for real. Or somehow accessing the html elemnt
Please consider thatm while having a C/++ background, I lack any web/JS programming skills and would be in favor keeping it this way as much as possible.
Thank you for any suggestions!
The solution is to use an user script for the musik player web site which is able to trigger the buttons in question. Than websockets can be used to communicate between the user script and the c++ server program.