Once we start venturing beyond articles we have the source text for, we have a potential issue of catering for the diverse formats and presentations these come in.
Of course, we could just host files, but these won't necessarily be easy for people to browse and access. Often they can be on platforms which don't allow you access to the source files, and/or require you to go via their website, or possibly some embedded player or viewer.
Here are the main ways of embedding video/audio on the web:
Plain old links - these will look like a text or image link, and clicking them will download the file, or possibly open in the browser, depending.
HTML5 media players - these should appear as a simple player in the browser with play/stop buttons and a progress bar, precise details of which depend on your browser / OS.
Platform players embedded in an HTML <iframe> tag - these are embedded content from another site, and will vary depending on the site. You-Tube videos and SoundCloud audio are an example of this.
For the first two cases, we can either host the content:
in the Git repository
on a some web-hosted storage elsewhere
If we embed content in iframe tags, there are many platforms which host audio and video, each of which may have a different looking player, with (most importantly) different requirements to embed them. Some examples off the top of my head:
YouTube
Vimeo
Internet Archive
Soundcloud
Audiable
Other podcast platforms - Anchor, Stitcher, etc.
Broadcasters' platforms - NPR, PBS, BBC, CNN etc.
And certainly many others...
Likewise, for PDF articles and e-books, the ways of embedding these on the web are similar:
plain old links - will like a link, and download or possibly open in the browser
platform readers embedded in iframes - basically embeds content from another site
Except - there is no HTML5 support for PDF as there is for audio/video. Some browsers (Firefox, for instance) have an integral PDF reader, which makes it appear you're reading the PDF online, but this isn't everyone's experience.
Platforms which might host PDF/e-books in an embeddable form:
Internet Archive
Scribd
Issuu
Amazon?
Probably many others
So: I thought about this and wondered how we'd support all that in the Rushkoff Archive site, and decided it would probably be a bit of a mission - I could get overwhelmed with different cases to support!
Even one platform can have its own diversity of format. You-Tube is a case in point - videos can come in funny aspect ratios, and there are various styles of URL which are used in different cases. See here for a list of You-Tube URLs seen in the wild!
On a previous site I worked on which embedded You-Tube videos, using Forestry.io as a UI, when a You-Tube URL was cut-and pasted in, sometimes they wouldn't work on the site because the URLs were in the wrong format to work in an iframe embed. What needs to be done for them to work is to extract the video ID code from the URL and just use that in a suitable HTML embed tag. Therefore this is what the Rushkoff Archive site does currently - it uses a 3rd party library to parse the URLs recorded in the Forestry form, and turn that into something uniform which works reliably. However because it was quite old and did not understand some You-Tube URLs now in circulation, I discovered I needed to fix it up. (More details in the issue #7)
Therefore, in the face of this potential diversity, what I suggest is that we first map out the problem, by collecting a list of those resources Doug would like to include, and then identifying:
Where they are (which platform/hosting)
File type(s).
File size / number per resource (e.g. a comic may be in a single PDF or multiple images).
What format(s) these exist in.
Publishing details such as licensing and ownership.
Details of presentation, which depends on the media, but includes such things as aspect ratios, length, pages, etc.
How to represent the media in the archive index - previews? If so where do we get those? Can it be automated?
What kind of metadata do we need to capture to facilitate the the above?
What options for consolidation exist, e.g. can these be moved or does the platform restrict the way they are delivered?
This will then feed back to the decision on which platforms and/or media types to support, and how.
Once we start venturing beyond articles we have the source text for, we have a potential issue of catering for the diverse formats and presentations these come in.
Of course, we could just host files, but these won't necessarily be easy for people to browse and access. Often they can be on platforms which don't allow you access to the source files, and/or require you to go via their website, or possibly some embedded player or viewer.
Here are the main ways of embedding video/audio on the web:
<iframe>tag - these are embedded content from another site, and will vary depending on the site. You-Tube videos and SoundCloud audio are an example of this.For the first two cases, we can either host the content:
If we embed content in
iframetags, there are many platforms which host audio and video, each of which may have a different looking player, with (most importantly) different requirements to embed them. Some examples off the top of my head:Likewise, for PDF articles and e-books, the ways of embedding these on the web are similar:
Except - there is no HTML5 support for PDF as there is for audio/video. Some browsers (Firefox, for instance) have an integral PDF reader, which makes it appear you're reading the PDF online, but this isn't everyone's experience.
Platforms which might host PDF/e-books in an embeddable form:
So: I thought about this and wondered how we'd support all that in the Rushkoff Archive site, and decided it would probably be a bit of a mission - I could get overwhelmed with different cases to support!
Even one platform can have its own diversity of format. You-Tube is a case in point - videos can come in funny aspect ratios, and there are various styles of URL which are used in different cases. See here for a list of You-Tube URLs seen in the wild!
On a previous site I worked on which embedded You-Tube videos, using Forestry.io as a UI, when a You-Tube URL was cut-and pasted in, sometimes they wouldn't work on the site because the URLs were in the wrong format to work in an
iframeembed. What needs to be done for them to work is to extract the video ID code from the URL and just use that in a suitable HTML embed tag. Therefore this is what the Rushkoff Archive site does currently - it uses a 3rd party library to parse the URLs recorded in the Forestry form, and turn that into something uniform which works reliably. However because it was quite old and did not understand some You-Tube URLs now in circulation, I discovered I needed to fix it up. (More details in the issue #7)Therefore, in the face of this potential diversity, what I suggest is that we first map out the problem, by collecting a list of those resources Doug would like to include, and then identifying:
This will then feed back to the decision on which platforms and/or media types to support, and how.