Printing in Context

Printing is an ancient art. Mass circulation of information based on printing seems to be the foundation of the rennaisance and the industrial revolution.

Printing is also a big business.

To give some idea of the scale of the industry worldwide sales of the top 100 forest and paper product companies were $357 billion in 2008  (PwC Global Forest Survey 2009) .  The global computer printer ink and toner market alone was estimated as $32.5bn in 2003 (IDC).

The print industry is undergoing profound technical change. In the 19th century steam and then electric power made paper and printing cheap.

Newspapers and magazines made up the bulk of consumer use of paper but packaging, textbooks, fiction and photographs were all in demand. Businesses used catalogues and billboards as well as financing the newspaper and magazine industry with advertisements.

The scale of printing is easily forgotten. In 1947 six newspapers in the UK had circulations of over a million and the Daily Express had nearly 4 million readers with the Mirror just behind. As of 2010 the Sun still does have 3 million readers.  By world standards this is trivial, circulation of the Times of India is around 79 million.  As well as the national papers each city used to have a newspaper office and printworks at its heart and once or twice every day vanloads of papers would set off for the suburban newsagent shops.

The newspaper is one side of the equation, the other being the journalists. National newspapers would have an office in every region. they also took feeds from "stringers". Local newspapers employed a few people full time and more part time.

The importance of advertising shouldn't be forgotten; it is critical. In a mass production and mass consumption society advertising drives the wheels of industry. 

Ubiquitous Screens

Several kinds of technological change threaten ordinary print.

Screens

Ubiquitous screens mean people can get the information they need quickly and comprehensibly without needing to print it. Almost every desk now has a screen on it, every briefcase contains a notebook computer and everyone has a mobile phone - increasingly web connected. So why does anyone need a printed piece of paper?

Screens aren't perfect. For instance a typical desktop computer displays under 2 million pixels worth of information wheras an A4 page displays at least 8 million. There are suggestions that comprehension from screens is 30% lower than from the equivalent printed page.

Screens are also expensive to make, bulky and use power. Some things will always make more sense printed than on a screen. The balance will change as screens get cheaper. Various kinds of electronic paper used in things like e-readers clearly change the balance away from ordinary print - they are lightweight, run for a month from a small battery and the data can loaded from a card or from the Internet. Furthermore they have digital rights management which encourages publishers to make a wider variety of material available.

Internet

Screens are limited use if they can only get at local information. The Internet has pretty largely removed any friction of distance for desktop machines. Wireless connectivity is still a problem although the mobile coms companies are promising that 4G mobile will be much better. The remaining issues are payment and copyright - for which we need digital rights management, resonable behaviour by publishers and by users.

Hard Disks

Hard disks store information on tracks that measure less than a tenth the wavelength of light. The costs of storing information on paper are uncertain - it's difficult to put a storeage, access and disposal cost on a paperback book in a child's bedroom. Consultants have suggested the lifetimes costs of paper forms held for organisational purposes could be $50 per sheet. The sheet gets filled in, filled, retrieved several times with the possibility of being missfiled and lost and then eventually a decision to dispose is made.

The costs of  disk space are more easily determined. Hard disks currently cost about $50 per terrabyte. A terrabyte stores the equivalent of about 250,000,000 pages of text  - only the world's biggest libraries hold more.

Hard disks aren't perfect either. Ideally about 6 of them (3 each on 2 sites) are needed to give information a reasonable chance of surviving. However duplication and replication need not be an expensive exercise. The huge ammount of information on Google, YouTube and Facebook is being held for free - or just the opportunity to sell advertising.

Advertising is critical. Advertising need not be slogans, boastful self promotion and exaggeration although those techiques are used - and might be needed in a mass medium like newspapers and TV.  An Internet user gives a search engine a short description of what they are interested in and the search engine knows something of their history, so it can send back information and adverts that fit them. What is an advert ? Information that interests you but isn't entirely adequate in itself?

Office work

The bulk of the paper stream has always been business record keeping, tonnes of paper stored in cabinets, ledgers and archive boxes. Until a decade ago the typical office had a wall of filing cabinets holding the purchase and sales ledgers, client records, product designs and suchlike as paper documents.

Double entry book-keeping, spreadsheets and ledgers all were all invented a long time ago.

Paperless Office

To see where we might be going it is often very helpful to look to the past before projecting forward. In the 1970s offices functioned using paper. Government offices would have one phone to a room. Correspondence and working papers were brought by messengers. A rather advanced office might have a computer terminal connected to a mainframe so that people could look things up. A data entry section employing the most skilled typists prepared the data to go into the system. Memory space on computers was expensive, a typical minicomputer had 8 kilobytes of memory and couple of 1 megabyte disk drives. Minicomputers were not common; a government or university department might have one.

The first use of the term "paperless office" is usually credited to George Pake.  Pake was a physicist who's work on magnetism made the MRI scanner possible. In the 1970s he was recruited from academia to be head of the Xerox Palo Alto Research Facility. Xerox PARC became one of the most famous centres of invention of the 20th Century.

Three critical inventions are: the laser printer which was based on the photocopier mechanism so directly relevant to Xerox's main business Ethernet networking as a way to share the printer. Ethernet is almost universally used today. Personal Computing using graphical icons on a screen - the precursor of modern operating systems First a bit of context. IBM and Xerox were the two giants of the office world in 1975 (As Microsoft and Intel are today).

IBM was the world's dominant computer maker with 80% of the market in Western countries. IBM can trace its roots back to Herman Hollerith's invention of punch card tabulating machinery in the 1880s. IBM's presence reached deep into most offices, their most popular product was the Selectric golfball typewriter.

Xerox made almost all office copiers. Chester Carlson invented and patented the xerographic process in the 1930s, the ideas were commercialised by Haloid in the 1950s. Haloid renamed themselves "Xerox" and made a vast fortune from copiers. It looked as though Xerox and IBM should dominate the future.

Businessweek produced an Executive Briefing  called "The Office of the Future" (June 30th 1975). In it top people from various firms involved in office equipment looked forward to how technology would develop.

The article is subtitled: "An in-depth analysis of how word processing will reshape the corporate office".  Word processing was then a rather new idea, coined by IBM to describe what Selectric typewriters could do, particularly if combined with magnetic tapes and cards. Xerox had a division called Redactron making word processors and bought the Diablo company who had invented the daisywheel printer.

A working lifetime later the article is both prescient, quaint and worrying; we still haven't solved the problems of automating office work.

The heart of the article is "The Struggle for Cost Control" and the problem it sees is the rising cost of office work, for instance: "Many offices are not even held accountable for productivity," notes David L. Holzman, Xerox' market development manager. "In studies we've made, 50% of all offices are just a part of the overhead." Further, the shift of the U.S. economy to service-based industries (they will employ 47% of all U.S. workers by 1980) and the growth of clerical employees are colliding with soaring clerical labor costs, growing shortages of skilled personnel, and changing social attitudes. "

By 2009 service industries employed 77% of US workers (73% in the EU, 75% in Britain). Perhaps attempts to contain office costs haven't worked. Does it really take 7 people in the office to tell 3 people in the yard what to do?  Or are most of us employed in a giant make-work scheme?

The article is basically a series of quotes from  luminaries in the newly emerging field of office automation.

Pake leads off, visionaries were important in the 1960s: "There is absolutely no question that there will be a revolution in the office over the next 20 years. What we are doing will change the office like the jet plane revolutionized travel and the way that TV has altered family life."

Pake says that in 1995 his office will be completely different; there will be a TV-display terminal with keyboard sitting on his desk. "I'll be able to call up documents from my files on the screen, or by pressing a button," he says. "I can get my mail or any messages. I don't know how much hard copy [printed paper] I'll want in this world."

Vincent E. Giuliano of Arthur D. Little, Inc., figures that the use of paper in business for records and correspondence should be declining by 1980, "and by 1990, most record-handling will be electronic."

The idea of a paperless office was beginning to take shape. It didn't always work out though. In 1977 the UK the government invested in a computer project lead by Information International Inc (Triple-I) to read all the social security documentation using a computer. The project used a high resolution microfilm camera, a special purpose processor developed for the US navy and a DEC PDP-10 and could read handrwiting at 150 characters per second.  The idea of eliminating paper and scanning documents took hold quickly. The system specification was so impressive there was a mass strike by the 10,000 staff who feared for their jobs.  In practice the system didn't work all that well. One of the aims was to read the digits on cheques - which the machine wasn't actually all that good at doing. The III OCR installation had shifts of about 10 specialist staff and in retrospect it wouldn't have taken many more skilled typists to beat the machine.

By the end of the 20th century there was a screen in the form of a computer or terminal on most desks but the display was limited to part of one page. The information available on any screen was also quite limited, most information was still on paper. Screens are steadily improving, there are becomming ubiquitous with one in every briefcase, a small one in every pocket

Computer printing has several roles: home print,  business invoicing, business presentation, design, and fine art "giclée" print. Wide bodied plotters are used to produce banners. Fast laser printers are used as production printers for runs of catalogues and print on demand books.Home printing - student essays, a bit of correspondence, photography and suchlike.  A printer usually has to be cheap to buy and with photography in the mix of purposes the best answer is an inkjet. A scanner making the printer into a multifunction copier (and perhaps a )fax is built in on all but the cheapest models. Low end models use tri-colour cartridges which have a small capacity for each ink and some will be wasted. Models with a separate cartridge for each colour are likely to be a bit more expensive to buy but a bit cheaper to run.

It is generally reckoned that such printers are sold at something like 70% of manufacturing cost with the aim of getting ink sales - a technique known as "Gillette Marketing".

Inkjet printers often can use the same cheap conventional office copier paper used for laser printers. Where a substantial amount of colour is used and particularly for high resolution photographs inkjets commonly need special papers which may be expensive. To get the manufacturers boasted colour range and fade resistance the paper has to match the specific ink.

Buyers need to decide how much they are likely to print and how seriously they are taking photography to decide which model to buy.Business oriented inkjets tend to have separate cartridges with larger capacities than those intended for home use. They also tend to have separate heads and cartridges which gets running costs down further but may sometimes make fault-finding more fiddly. Manufacturers claim some recent inkjets have running costs a third lower than those of laser printers. Low cost inkjets would spell the end of laser printing but at the moment laser printers may still prove more reliable.  It might seem as though home users could save money by trading up and buying business printers and that might to be the case if you print regularly and in fairly large volumes. The heads and cartridges in an inkjet need to be used fairly regularly or there is a risk of stale ink blocking nozzles.

Professional inkjets include wide body printers like the Canon Pixma Photo Pro and HP's Designjet family. Heads and cartridges are separate and the ink cartridges are very large so they are not held on the carriage, they would be too heavy. Ink tubes carry ink from the tanks to the heads.  The ink cartridges may be expensive but they give a much lower copy price than the consumer oriented machines.

Inkjet printers tend to have more problems than laser printers and the cure is usually a new printhead. However printheads are not cheap and trying to eradicate a fault can use a great deal of ink.

Speed is also an issue with inkjets. Manufacturers often claim headline print speeds of 30 pages per minute but that is for a low quality draft print, producing a high quality picture can easily slow the printer down to a tenth of that speed. Laser printers usually perform more in line with user expectations. 

Laser printers tend to preferred over inkjets in business. Laser printers generally have lower running costs, higher reliability and longer lasting cartridges which means less fuss. Laser printers will often print after being left on a shelf for several years.

Basic business processes like invoicing and correspondence tend to use mono laser printers  because they are usually cheapest to run.

People hanker after colour of course and for business presentation such as brochures or catalogues it is invaluable. Some colour laser printers claim to run in black and white as cheaply as mono models.  Claims about low running costs for colour might be treated with scepticism. Even if the colour cartridges are fully disengaged the running cost probably doesn't include replacing the transfer belt and fuser.  

Just looking at the colour process would suggest its costs might be about four times higher than mono printing. The actual costs of colour printing tend to be considerably higher than for mono print because incorporating letterheads and pictures increase the page cover.

Production Printing

Marketing and communication flyers, catalogues, books, magazines, newspapers, have traditionally been produced using machines like  offset litho presses. The cost of printing using machines like this is almost unbeatably low, providing  you want tens of thousands of copies - which is true for mass circulation magazines and best selling authors. The problem is  twofold:

• Professional presses are expensive machines with specialist operators and a factory environment.
• Printing uses plates and the platemaking process can be expensive - it just isn't economic to produce fewer than a thousand copies of anything.

Offset litho operators can reduce their costs using Computer To Plate (CTP) processes where a polyester plate is laser printed and then used to print an ink image.  Conventional printers have stitch, fold and trim machinery that can produce professionally presented material.  Low cost laser printers don't have an option to attach this sort of machinery.

The printing industry might face a difficulty with the ubiquity of screens. The current generation of mobile phones have a small screen that won't hold much text and a battery powered backlight so it's only suited to working for a couple of hours. E-books and a new generation of mobile phones might reduce the need for books. Amazon reported that US sales of electronic titles overtook paper titles in 2010 and some of the big book retailers are clearly struggling.

There is still a possibility for books to fight back. Many of the properties of paper and books are difficult to reproduce electronically. Cheap long term information storage accessible in seconds, no battery, portable and still works when bent or broken - books work rather well compared to computers.

Printer manufacturers are skilled at balancing the "value proposition" they offer. If the printer is low cost then the ink is usually expensive. The more expensive printers take higher capacity cartridges that usually give a much lower price per page.

In the introduction it was suggested that there are about 60 printer makers and that 10 of them are competing on a broad product range.  On a rough count HP have more than 70 "engines" on the market at any time. Lexmark have 30. Canon about 40. Epson 30 and so forth - that's without counting minor variants like those with a network card, duplexer or tray.

It's probably fair to suggest that there are several hundred printer models from major manufacturers on the market at any one time. Then there are label printers from Avery, Dymo, Sato and Zebra and specialised printers from people like Durst, Heidelberg, Mutoh, Océ, Pitney Bowes and Roland. These special printers double the size of the list. Some things like the electrostatic inkjets that put serial numbers on bottles might not be regarded as computer printers by everyone - but they often have a data connection.

For people providing spares and repair the sheer number and variety of printers is of some interest so some guess on the number of printers out there would help.  

Printers can last a long time so there is a substantial back-catalogue of older machines out there. It's generally older machines that need spares. At a rough guess there are about 400 printers on the market at any one time, substantially different models come out at intervals between a year and 5 years and printers can easily be in service for ten years and sometimes longer. There may easily be 4,000 models of printer out there.

Choosing a new printer can be frustrating even for people with a strong interest in IT. Good comparative information is quite difficult to find. Even with good information the prices of printers and consumables vary a lot and that completely changes the value proposition. Most people probably just choose a brand they like and then buy what they can afford.