I believe much of the future of text should be in the form of documents, since documents, rather than text on servers or in databases, is more robust in terms of longevity and addressability.
Another important issue is that users own their documents. They do not own the data on corporate servers, such as with social media and even servers for private use, such as WordPress, are only available for as long as domains are maintained. This ownership makes the information in them more robust since they can move them and store them at will.
Imagine reading a document and coming across a citation and being able to click on it to load the cited document directly, not load a web page with a download link.
Doug Engelbart used the analogy of a skier: When you are going downhill the interactions need to be immediate or you will simply crash and fall. With knowledge work there is also a significant hindrance for following hunches and viewing information from different points of view when there is a delay.
issues with document connections
Inviting you to imagine the above might not seem like much since this should be already present. However, there are several issues with document based text, particularly in academia and science:
- Currently documents are not directly connected. Links go to servers which the user can then choose to download, or re-download the document from, if the server is currently online
- Addressability does not allow for linking to a specific section of a document
- Making citations is often error prone and students can cheat by copying, thus adding further errors
- Citation trees are not available to users directly, even if they have all the documents, this is something publishers keep private and Reference Managers have access to, to an extent, but it is not possible for the user to devise ways of seeing how the documents connect
issues with document interactions
There are many interesting and worthwhile writing applications/word processors out there, such as Byword, Scrivener, iA Writer, Mellel and Ulysses in addition to the giants of Word, Pages and Docs.
- Each have their own useful interactions inside the application, but once a document is published it is either to HTML or PDF:
– If it’s exported to HTML then there is the issue of keeping up the server or the document dies and interactions are based on what browsers will allow.
– If it’s PDF then the document is flatted to remove metadata for interactions, such as structure (headings) and connections (references).
- Server based interactions, such as social media Facebook and Twitter, can offer interesting and useful interactions but in those cases the information is owned by the companies providing the services.
My feeling is that if we can imbue published documents with rich metadata then we open up the opportunities for richer interactions.