Slashdot Mirror
The Myth of the Paperless Office
AdamBa writes: "The New Yorker is running an
interesting review
of the book
'The Myth of the Paperless Office', also discussing
'Scrolling Forward'. Read
it and the ever-informative
Malcolm Gladwell will
explain why paper enables collaborative work much better than computers do, why a messy desk is a sign of productivity, and give a little background on the inventor of the Dewey Decimal System to boot."
It's on citeseer as "Paper as an Analytic Resource for the Design of New Technologies".
"Our constitution was written on paper!"
No.
Not if you are a citizen of the United States of America, anyway. Our constitution was written on parchment, not paper. Parchment is made from animal skin. It has historically been used for important documents because it is considered more durable than paper (and floppy disks).
According to a FAQ at http://www.paper.upm-kymmene.com, "4) How much energy is needed in the paper production? Depending on the grade 2-4 MWh of heat and power is needed to produce one tonne of paper. The power use of an one hour home-PC session is roughly the same as the power used for a copy of a Donald Duck cartoon magazine. 5) How much water is needed in paper production? The fresh water use of a paper mill is about 10-15 cubic meters per tonne of paper depending on the grade and the mill."
["Marge, I agree with you - in theory. In theory, communism works. In theory." - Homer]
If I recall, one of the various user interface paradigms Apple was working on in the 80's-90's (circa Taligent and Pink) was an interface specifically called "Piles" based on some of this research.
While that never saw the light of day, the lessons learned from that research made their way into the Standard Macintosh bibles of user interface design. To wit:
This is, IMHO, one reason why the classic Mac OS interface was so amazing. You (the user) had complete freedom in organizing the documents on your computer however you wished. Spacially, color-based, or sorted. You could store your documents in whatever made sense to you, without the operating system declaring the Right Place for documents (ie, home directory, etc.. a la Windows and Unix).
Some people's Macs made sense only to their user, which is just how it should be - considering that it's a PC
Now with Apple moving to unix underpinnings which, thanks to the rigidness and inflexibility built into unix, don't allow for this type of "personal organization", it's difficult to find a system design that understands this.
This is the NUMBER ONE problem "old-school" Mac OS users have with Mac OS X - being told that they have to organize things in a certain way (ie, "in your home directory") and the thing that people coming to Mac OS X from a standard unix background don't (can't) understand.
Hire a Linux system administrator, systems engineer,
Paper User Interfaces for Paper Documents
I've been working on a product for a few years that uses paper as a user interface , kind of a follow-on to the graphical user interface. I used to joke with friends that I was working on an 8.5x11 inch 400 dpi gray-scale display that costs 2.5 cents.
Document Tokens -- making paper a first class citizen on the network
You scan your documents, and they get stored in a document repository on the network (using WebDAV over HTTP or some other protocol), and it prints out a piece of paper that refers to the electronic document on the network, kind of a like a paper document or a paper URL. I named it a "Document Token". You drop it in your copier, for example, press the big green button, and it automatically recognizes it, retrieves the original, and prints it back. Or if you asked it to e-mail the scanned document instead, it will e-mail the document as an attachment or just a hyperlink.
Cover Sheets as Forms
Another thing you can do is print out a cover sheet with checkboxes on it and some document meta-data built in, so you can drop the cover sheet for your "Legal Contracts" on top of the latest contract you got, check the box for the account you're dealing with, and press the start button. It will scan, store based on the directions embedded in the paper, and associate the document meta-data with the paper.
Situated Meta-Data Capture
One of the most expensive things about scanning is associating the meta-data with the document after scanning. When you have the paper in hand, you know what the document is and where it came from. The file folder or desktop location is right there in front of you, and the physical presence of the document triggers certain kinds of memory as well. In ethnographic terms, the document is what Lucy Suchman calls situated When you try to add meta-data to a document after scanning, you (or worse, someone hired to look at it for you) is staring at a set of bits on a computer screen, completely divorced from its context, and it's expensive to discover where it came from and what it means. If you can associate this information with the paper document when it's in the paper domain, by marking it down on a paper user interface, then you save lots of time and money.
W3C Standardization
For the web to become a truly ubiquitous computing interface, it must move beyond the desktop. We're working with the W3C to standardize an XML representation of forms such that the same form purpose can be expressed in different media -- desktop, pda, mobile phone, and even paper. Take a look at the XForms last-call specification.
Product
The product is called FlowPort
I think you'll be needing the Printer Freindly version of the article to start with...
Any sufficiently advanced man is indistinguishable from God
No, I'm serious.
I had my doubts too, but my current manager was adament about it; that's how we do review and markup of documents (I'm a tech writer; there's a LOT of that).
Works perfectly. Click the little paragraph looking button, and everyone's changes are visible, in a different color. Hover the mouse over a change to see who made it. Right-click to accept or reject a change. Or use a Wizard thingie to search sequentially through the changes, accepting or rejecting each one. If it gets too messy to read with all those changes and strikeouts, just click the paragraph looking button again - changes hidden.
It's a heck of a lot more productive than trying to make out chicken scratches on paper. Every proposed change is legible, and you can see who made it and when. We use Word comments to explain changes.
My 2 cents.
You're comparing a ton of paper (something like 50000 pages) to a one-hour PC session. Let's scale these down to be something similar. I can read about 50 pages/hour of moderately technical writing. 50000 pages/50 pph = 1000 hours to read all that paper. So that 2-4 MWh is really 2-4 KWh amortized over the reading time of the paper. Figure a 200 W power supply, and the paper is higher by only a factor of 10. AND that doesn't include the construction costs of the computer (you think paper uses a lot of water, you should try silicon computer chips).
-JS
Vanity of vanities, all is vanity...