Paper Gives No Feedback

No matter how good a writer I am, no matter how many times I revise, no matter how often I treat a misunderstanding of a document as a bug report, I can't get away from this fundamental fact: PaperGivesNoFeedback.

When you sit with me and explain something, I can say "do you mean ..." and you can say "sort of, but actually ...", and I can say "oh, so then ...", and you can say "yes".

The document can't do that. Books, documents, memos, wiki pages - these are all subject to this problem. They may be necessary, they may be very valuable. But they can't replace two people talking with each other.
This reminded me of a passage in a book on Human Computer Interfaces. I can't remember the reference - it's sitting on my bookshelf at home right now. The chapter was to do with collaborative working, and different methods of communication.

Essentially, the book suggested that methods of communication could be measured on two axes - time and distance. These refer to the distance in space/time between the participants.

For example, face to face communication is characterised by being in the same place at the same time; a phone conversation is same time/different location; etc.

This made me wonder: Where does paper documentation fit into this plane? All of these methods of communication differ in permanence. Paper communications provide a record of the conversation, as does e-mail. Face to face communication doesn't.

A long way away on both axes. If you write a book, your feedback comes in the form of letters from readers, but mostly I expect from sales figures. As axes go, these are pretty blunt ones.

Maybe we need three axes?

Oh yeah, right, we sure need more axes to grind here on Wiki.
Sometimes paper is feedback. As when you don't realise how little you understand something until you try to explain it to someone else. If you're working solo, the discipline of writing something down can help you get your thoughts in order and do you some good even if nobody else ever reads it. -- DaveHarris

Concur. I have proven this to myself any number of times. It's really tested when you go to explain it to somebody else. Often I test my TechnicalMemos on co-workers. Often I waste a good three or four minutes of their precious time just trying to convey what my TechnicalMemo says, then I realize that either the writing sucks or the idea itself is no durn good. Either way I can't explain what it is I'm writing about, so that means I can't build it. BackToTheDrawingBoard -- MartySchrader

See CardboardProgrammer.
Paper can serve as an accurate and detailed record of facts.

If you care to go somewhere, would you like to look at a map? Or, instead, you can talk with someone, have them give verbal directions (and take no notes). Each approach has its limitations: There are many important details that a paper map will >not< show you -- like "move over to the right lanes early, because traffic usually backs up down there." But there's a lot that verbal directions don't cover too -- like, what to do if you get off the planned route, or would like to change the planned route. There's a lot to be said for getting out a map, and talking through the planned route, with your partner. -- JeffGrigg
Agree -- PaperGivesNoFeedback, but WritingDoesGiveFeedback?. In my experience explaining something in prose often makes you wonder why you did something in such a complicated way, leading to unexpected simplifications. -- ArieVanDeursen

PaperGivesFeedback?, just about different things. Paper can tell you whether your ideas are organized in an attractive way. PaperGivesMoreFeedbackThanComputerMonitors?. I often print documents just to see how they look on real paper. The other day I made great progress by printing my documents, cutting them into strips, and playing with the strips for a while.

Just as a side note, it really is easier to spot errors on paper than on a screen whether it's code, prose, or poetry. Last I saw, dominant scientific opinion is that it has to do with the writing on the paper looking subtly different as the angles change. Probably something to do with our lizard brains still attaching more attention to things which move.

Interesting: this exactly what the story cards do: they're paper, and they are the requirements document cut into strips, so that you can play with them and order them -- ArieVanDeursen

But still the paper doesn't tell me I used the wrong word, or failed to convey my meaning clearly.

Must we quibble over every last detail about the use of written instructions and records? It seems that those who so vehemently oppose the use of writing are trying to dig up any lame excuse possible to avoid writing. Come on, folks -- are we not professionals here? Just suck it up and write it down already!
If you want to teach a framework or API, put it on the web/intranet. Watch the server logs for each page. This is quite an interesting experience if you host the Java core API for your students for example.
When I write on paper (or virtual paper) it feeds back to me the same symbols I wrote. Isn't that feedback?

More like regurgitation.

Hardly a valid comparison.
PaperGivesNoFeedback can be due to the fact that it is usually the record of thinking or the record of what has happened or what is envisioned as possible. Whereas Speech is usually the attempt to communicate about actions (past, present and future). Speech is directed to a visually and audibly active audience, while Writing is directed to an imagined and passive audience. Speech allows feedback to occur in that the speaker can interrupt and interact with the audience. A writer can only imagine what will be the impression of the words written on paper for others to read. The writer cannot react with the reader in the present tense, reaction comes in a later edition.

This WikiIsNotPaper?, it is more like a recording of speech in a time frame called the WikiNow. Feedback is present not only at one time, but in a continuim which bridges time.

--DonaldNoyes
CategoryDocumentation

EditText of this page (last edited April 10, 2012) or FindPage with title or text search