View Full Version : How would you recreate desktop environment?

June 10th, 2010, 06:49 PM
Since 1990's I have wondered and experimented with ways to improve the usability of my desktop environment. Since I am not a programmer I have very limited possibilities to make radical changes but some concepts have emerged along the way. Some and probably many of the things below are already realized. I just do not know about them.

What would the user interface and data management of a computer system look like if it was designed now from scratch instead of building a new one. I would start by focus on how to manage documents and focus on the work and work flow of users. I would start the mental exercise by imagining that the computer desktop is equal to traditional table and then add a few differences.

1. User can have an infinite number of tables, thus never needing to clean one.
2. Any document user has can reside simultaneously on any number of tables through out the whole local area network, still there is only one copy of each document in existence.
3. If document is on multiple tables then it is always possible to activate (move to) the related table using the relation.
4. If document refers to some other document then the corresponding document can always be accessed and opened through the reference link. This must be true even though the system containing the original document is inaccesible.
5. User may request the relations between the documents to be represented graphically as lines.
6. User may add relations between documents
7. System tracks and creates the relations between documents automatically
8. Stack concept replaces Folders
9. "Make available" concept replaces copy paste.

Clarification above of points as a use cases
1: User starts a new activity/project say building a garden shed. User creates a new table for this purpose. By default the system offers to add default interfaces to the table, e.g. e-mail, instant messenger and Internet browser. When the table is created the user can add documents, Stacks and application launchers to the table.
2./6. Our imaginary user has two tables, building a shed and tracking financials. Imagine that financial tracking includes a spread sheet containing the information about the savings of the user. This information is relevant while designing and building a garden shed. Therefore it is natural that this document is defined to be available also in the garden shed table. After the document is linked also to the garden shed table then these two tables (eg. activies) are linked. Our user also creates a document named "Garden Shed Materials" and tells the system to relate the Financials spread sheet to the document.
3. Now while designing the garden shed user receives an message that his wife is planning to take the family to an around the world trip. As an attachment of the message the user has a document describing the travel itinerary. Since this information will affect the financials of the family he drops the document to financials table. Unfortunatelly the trave itinerary does not contain information about the cost of the trip. Since the wife is in the same local area network our user can access the corresponding table from where the document originated. By right clicking the document user can select to move to "Around the world" table to locate the relevant financial information that is contained in the document "Travel cost".
4./9. User opens the "Travel cost" document. Paints the relevant numbers and tells the system to make the information available for the items in the financials table. After this the user moves back to the financials table and opens the financials spreadsheet. Since the number data was made available for the items of the financials table the user can now select the relevant portion of the data and attach it to the financials spread sheet. Now at this point the system creates a relation between the "Financials" spread sheet and the "Travel Costs" documents.
5. Our user closes the financials spread sheet and asks for the graphical representation of the relations. Table background (eg. desktopimage) fades away and is replaced by documents connected by lines. "Garden shed materials" connects to "Financials" which connects to "Travel Costs" which connects "Travel Itinerary" which connects directly to 25 other documents . Back ground of the graph is devided in three to represent the locations of the files either in the local computer, local network or Internet. In our case the travel Itinerary would refer multiple web sites.
8. As the garden shed project is defined it will generate more and more documents. At some point it is not feasible to keep all of the documents spreaded on the table and they must be stacked. Stack is just like folder except that when the the stack is opened all of the documents remember the position and size they were when the stack was closed. Also when the stack is opened the graphical relations may be drawn to ease the locating the relevant document (like a mind map). Also same document can be placed to multiple stacks just like it can be placed to multiple tables.
9. When document or part of a document is "made available" for some table the marked data portion and location of the document is stored to local database. If our user browses for roof tiles for garden shed he can concentrate on the task with the web-browser. When se sees something he likes he just marks the relevant portion and presses CTLR+C. Since there is an underlying active table the system knows that this data is meant to be available for this table. After user has browsed through 20 local harware stores he can activate the "Garden Shed" materials document and select the most promising ones from the database and attach the relevant data portions to the document. In this case the relevant data would be image and price. When user later needs to find the correct internet page of the correct hardware store he just clicks the relevant image and asks the system from where the data originated from and the system can open the relevant URL.

Increase efficiency by makin data available collaboratively, preventing documents from duplicating unncessarily and making sure that information once found does not need to be refound.

Below I have drafted patent claims to invalidate at least some patents of hopefully future trolls.

Claims (UI presentation)
1. A graphical computer user interface characterised by effectively infinite number of separate work spaces wherein said work spaces can be interlinked by semantical connections between documents residing on separate workspaces which links can be presented as graphical lines between said documents on 2d plane of the workspace or as 3d lines between documents of the separate work spaces.
2. Computer user interface of claim 1 characterized by that the semantic links between documents can be created manually.
3. Computer user interface of claims 1 and 2 characterized by that the semantic links between documents can be created automatically by the system.

Claims (Inverted copy paste operation)
1. Software for automatically creating semantic links beween documents wherein said links are created automatically by copy paste operation.
2. Software for automatically creating the semantic links of claim 1 beween documents characterized by that the system stores location of the orginal data source to a database along with the copied data as the user executes the copy command.
3. Software defined by claims 1 and 2 characterised by that the location the data was copied to is stored to the database as the user executes the paste command.

Claims (Data management)
1. Software for automatically managing data created by a multitude of users wherein said software maintains only one instance of each document by storing the relevant document data in to a database of each computer and performing synchronizing operation to said documents according to the semantic links greated by user actions.
2. Software of claim 1 charachterized by ability to store data not modifiable through local area network computers to the said databases of local computers according to semantic links created by user actions and thus ensuring it's continued availability.
3. Software of claims 1 and 2 characterized by ability to maintain at least two copies of each data thus ensuring its continued availability. (distributed peer to peer backup)

Creative commons copyrighted version of the text above can be found from
If and when I make edits or updates I will update them to the linked document.