bsstahl / liquidvictor Goto Github PK

Ideally, one could specify a deck and then select which slides get imported. They should probably all be tagged the same as well for easy retrieval or even removal if needed.

Presentations should be exportable to .ipynb format, perhaps using syntax needed for RISE.

Presentations should be stored as ordered collections of slides and other resources

Each user should be able to specify a default starting point (only 1 for now -- maybe more later) for a presentation -- i.e. Specify a title and it automatically builds your first several slides, the ones that are common to all of your presentations.

Currently, the FullPageFragments layout is the only way to have text appear piece-by-piece on the slide (so that it requires a click to add the next item to the screen). We'd also like to be able to do this with an image on the screen.

Currently, ImageLeft and ImageRight are the primary layouts for slides with images. If we implement ImageLeftFragments and ImageRightFragments that layout the screen in the same way as ImageLeft and ImageRight but add the text bit-by-bit like FullPageFragments, we can achieve this functionality.

The Description form-field for Search pages is not being populated. This means that templates that use this form-field have the value left as "{Description}", which can result in strange behavior for clients.

If the presentation output format supports it, enable the ability to zoom in on content items.

For RevealJS based presentations (including Jupyter Notebooks), zoom.js could be used. I'm not sure if there is a mechanism to zoom in Powerpoint.

For slides created in LV it will hold the creation information. For imported slides, it will hold details of the import source.

Each entity type, SlideDeck, Slide, ContentItem should have a unique Id within that space.

The solution should probably be at the data-tier since that is where it should be enforced and because many DB types will handle it by default.
 
I don't believe it is necessary to make sure that IDs are unique outside of the entity type space since each entity type knows its own children so there shouldn't be any situation where a slide is confused for a content item with the same name, but it is even more unlikely since most of these duplicates would likely come from CopyPasta, the odds of that happening are pretty low.

How can we fail-over on this if it is not supported? Perhaps just treat the slides as normal (in-line)?

For revealjs, this would be the implemented as a slide section

• Add your own images
• Search for image by license

This is a completely valid scenario and needs to be seamlessly handled including being able to add/remove 1 instance without impacting others in the presentation.

Take a hardcoded repository of slides and images, and a hardcoded media collection (i.e. the presentation script) and build a console app that will assembly that into a presentation deck.

Create a default presentation in a default presentation repository that can be included in the code repository. This will allow the project to build and execute when first downloaded and can be used to demonstrate some of the features of the system. It will also create a 1st presentation that people can clone to start building their own presentations.

This is a completely valid scenario and needs to be seamlessly handled including being able to add/remove 1 instance in the same slide, or in the same presentation, without impacting others.