Part 2 of 2 parts

The last blog described soft data as data that exists independently of a real world meta language such as a measurement system, and hard data cannot be described without that measurement system.

There are many uses for the hard and soft data paradigm, and it can be applied to any field of knowledge where counting or measurement is integral.

Accounting is a profession where hard and soft data are intertwined. There are results of formulas and counting results. Counting supplies knowledge of income and is hard. Formula provides knowledge of worth and may be very soft.

In music, a tone is counted in cycles per second and that’s hard data. Since time is defined by the rotation of the earth, a note could be hardened by determining the number of cycles per degree of the earth’s rotation. Hard data is the same when measured by any measurement system. The cycles defining a musical note are the same whether counted by the rotation of the earth or my watch.

A musical notation is soft data.  A ‘C-Sharp’ is an indicator of a note. The number of cycles per second needed by the musical instrument to sound that tone varies with the instrument. A C-Sharp note indication on a sheet of music does not specify the hard metric of cycles per interval and hence is soft.

This example may be pushing the usefulness too far, but here goes. Situational ethics are soft guidelines. Actual actions are not described, just a general structure of actions. Laws are hard instructions. Do not speed above a limit. The limit is measurable in miles or kilometers and do not change with the change in measurement system. While technically ethics and laws are not systems of data, they still conform to my definition of hard and soft data.

It is a measurement system that turns soft data into hard data

The meta-language system that turns soft data hard does not have to relate to anything in the real world. One definition of the inch is the distance King David I of Scotland in 1150 said is the width of an average man's thumb. An inch is a commonly accepted unit of distance with no relationship to something in the real world.

Time is a soft concept that our current Gregorian calendar attempts to measure and harden. If we can measure C-Sharp by the rotation of the earth and that rotation occurs in time then an accurate, or at least consistent, measurement of time is needed. Does our Gregorian calendar accurately describe the time it takes for  the earth to  circle the sun? Not even close! 365.2422 days pass from one Winter Solstice to the next, not 365 days. However, the measurement of time is precise or else we could not measure the year to four decimal points. It is precision that allows us to measure C-Sharp, Japanese on-time trains, and the oscillation of atoms. My point is any measurement system will harden soft data and the more precise the measurement the harder the data.

What does soft and hard data tell us?

Soft data doesn’t communicate much information. My living room is a rectangle of 1:2 ratio. That information is enough for an architect to plan my house, but will not provide the architect with knowledge of how large to make my living room. If I add measurements and harden the data by saying that my living room is 10 meters in the short direction, then the architect knows that my 10 x 20 meter living room is very large.

Soft data can be very useful. If the rectangle described above is the ratio of a spot color in an InDesign page, and you make the color rectangle for your monitor then the measurement is in the monitor’s pixels per inch. If the your monitor is 72  ppi, and the width is one inch, then the dimensions of the spot-color rectangle is 72 ppi * 1 (the ratio) = 72 pixels by 144 pixels. On the printed page with a halftone resolution of 150 lpi, the rectangle size is 150 lpi by 300 lpi.

Used properly, InDesign is a data hardening device. Input is soft data consisting of vectors and raster. Output is the same data harden to a page size or a web page.

Icefields hardens soft data.  The device independent color space CIE L*a*b* is harden to SWOP CMYK. Icefields hardens ink color from Pantone 185 red ink to particular RGB coordinates based on the selected color space.

I leave you with this question. Does color management attempt to harden data prematurely and keep it hard throughout the workflow? In my next blog, I will look at a proposed soft to hard image workflow.