Archives for the month of: May, 2017

Well, I accept that we may have (almost) too many books to fit into this apartment. You don’t work in the business without having some sort of bookish interests, and when there are two of you at work on the habit, the result is a lot of sagging shelves. But get rid of them?! (We need Keith Houston’s interrobang at this point. See Shady Characters here and here.)

I have occasionally speculated on the idea of a sort of external hard drive or back-up for your brain accessed via some brain waves or other. That back-up of course already exists in prototype; we call it the “Internet”. As time goes by the interface between brain and external hard drive will no doubt become smoother and smoother, obviating the need to wake up your iPhone. But of course, those of us with those sagging shelves have long had such capability. I was told at university that the purpose of a university education was not to teach you lots of things, but to teach you how to find out the things you need to know later on. Being stayed by masses of books gives you the means to fulfill that aim. Those who suspect that tertiary education becomes less and less important as access to facts becomes easier and easier should reflect on the fact that learning to navigate information sources, the true purpose of such an education, must only become more and more vital as the amount of information to discriminate amongst explodes.

The books you own display a record of the things you’ve thought important at different stages of your life. When I came to America among my luggage was a good number of books. One I know I’ve never read is One Dimensional Man by Herbert Marcuse. The fact that over 40 years later I’ve never read a single word Marcuse wrote doesn’t mean that I never will, and certainly isn’t a reason to discard the book, a reminder of the more activist innocent that I was back then. The Abacus paperback edition set me back 45p! Abacus was an imprint of Sphere Books, which was part of Thomson. It was sold to Pearson in 1985 and was shut down by Penguin in 1990. Fascinatingly this copy, from the eighth printing under Sphere’s aegis (it was originally published by Routledge and Kegan Paul) includes an Erratum slip tipped in to the front of the book indicating that two lines on page 20 should in fact have appeared on page 18. (It would probably cost more than 45p to get this done now!) It always mystifies me how long it takes for glaring errors like this to become known.* I suppose most readers just keep quiet. Nowadays we don’t go out of our way to discover such errors in our books. We just cared more in those days. If a mistake is forced on our attention we’d probably wait for a reprint and hope nobody else noticed. See — if I’d thrown the book out long ago where would all this analysis and reminiscence come from?

In the New York Times Book Review‘s 7 May profile of Penelope Lively, she is quoted as saying “Your books tell you where you’ve been — they’re the story of your own mind. Getting rid of them would be like getting rid of that.” Unsurprisingly she can’t bear to throw away any books even though, at 84, she can’t see the small print in some of them. Now there’s a sensible attitude. If you run out of room for your books, move to a bigger apartment.


* I once had to do a rip and tip on a political science text which contained duplicates of two pages (page 101, say, appeared in its correct location but when you got to page 108 you discovered it was followed by 101 again). To  cure this condition you print up copies of the two pages 109 and 110, cut out the bad ones, leaving a stub in the gutter to which you glue the new leaf. Nothing too surprising in all that — except that I had had charge of the first printing of the book when I was still in the UK, and this meant that it had taken six years for anyone to notice, or anyone who noticed to squeal loud enough for the publisher to pay attention. Lazy student that I was, I always imagined the students happily thinking, “Well, one less page to read”.

Too many kids fall behind in reading early in their school career; indeed probably before they even get to school. Exposure to books, and adults who read books, is important in forming the habit of reading. I guess using the barber-shop, a bit of a social center, as a vehicle could work. Certainly the charitable organization Barbershop Books believes it will. Barbershop Books was founded in Harlem, but has already expanded way beyond New York City having active locations in 10 other states.

The theory behind the initiative is that “African-American boys who don’t often see black men reading a book” should be exposed to books in their regular environment. “Barbershops are some of the only places kids go to on a regular basis . . . There’s already that rapport there, already that relationship with the barber. Why not ask the barber to encourage them to read?”

(If you don’t see a video at this point, because you get this via email, please click on the post’s title in order to see it in your browser.)

There are similar initiatives in Ypsilanti, where Fuller Cut offers a discount to any boy who will read aloud while having his hair done, and in Mobile, Columbus, Jackson, Dubuque, Baton Rouge, Muskogee, — all over.

The trouble with music setting is that there is so much going on simultaneously. You have to have your staff lines, then on top of that you have bar lines, notes, time signatures, clefs, slurs accents and so on, with text below if it’s a song. If it has a base line and a treble line, or it’s an orchestral piece, with multiple parts, everything has to be in vertical alignment. In hot metal type, putting a note on top of five horizontal rules was an impossibility: the two elements had to be at the same level so that they’d both print. You can imagine having little bits of type showing a crotchet on or between the five lines of the staff I guess, but this would lead to some pretty intricate work. That it was done can be seen from this photograph of a relatively straight-forward job from

From the Museum of Turnhout

Prohibitively intricate; which is why the manual process of engraving, shown in the video below, lasted until computer setting was sophisticated enough to take over. Apparently early printers would operate with paper carrying preprinted rulings, but this obviously demanded a precise control of registration, always difficult but especially so when the paper has to be dampened before printing. They might alternatively rule in the staves after printing, which again would demand some pretty tight control. Engraving into metal plates, initially copper, was first used for music in 1581.

The Munich music publisher G. Henle Verlag’s website shows a couple of videos of the music engraving process. I think this one is the clearer of them, but if you visit their site you can enjoy a demonstration by their charming operative, Hans Kühner. The manual process, beating notation into soft lead plates using punches and a hammer, continued in operation till the 1990s, by which time an adequate computerized replacement had been developed.

(If you get this post via email, and don’t see a video at this point and at the bottom of the post, please click on the heading of the post to view it in your browser.)

You can see that printing would be via the intaglio process: ink collects in the grooves punched into the plate, and is then transferred to the paper. This of course dictates that the engraver work in reverse images.

Printing music was always a demanding branch of the business. The size of sheet music (conventionally 9″ x 12″) is slightly larger than most presses are built to accommodate economically and the scores need to be bound so that they remain open without attention. Thus the work tended to be done by specialist printers, of whom there remain fewer and fewer. What about an e-reader now that we have crossed the computer barrier in creating scores? Well, of course they are all too small too. Here’s a solution: The Digital Reader sends a note about the Gvido Dual-Screen Music Reader. The Gvido website provides the following lyrical video, showing the device in operation.

On the south side of West 26th Street immediately after the High Line can be found the vast buildings which once housed the successful H. Wolff Book Manufacturing Company. To anyone of a certain age the name H. Wolff Book Manufacturing Company means “big book manufacturer”, with echoes of “the end of book manufacturing in New York City”.

As may be seen from this photo, someone has taken a chisel to the name sign and chopped off an “F”. Seems a pretty silly gesture.

Looking down the street from the High Line gives an impression of the scale of the building(s), 508 and 518 W. 26th, which now seem to be filled with art galleries. This is Chelsea, and the space is great. The  building farther away, #518, was the first one constructed; it was built specifically for the H. Wolff Company, and was completed in 1910. Their expansion led them to construct, in 1926-27, the adjacent building, 508 West 26th, shown in the first photograph and in this one to the east (left) of the original structure; (508 is the grey building, and 518 the pinker one). The expanding company made further acquisitions in the neighborhood, and also subcontracted space to Grosset and Dunlap, George H. Doran, van Reese Press, and Greenwich Lithographers.

Photo: 14 to 26th Street (where you can find a detail which enables one to see the sign a bit more clearly)

And here, viewed from the west is the no-longer visible evidence (you can see a new building going up in the previous picture) of their location in barely legible white paint on the brickwork, immediately below the Grosset and Dunlap sign which just shows below that reddish line.

The High Line we think of as a sort of aerial park, but it isn’t all that long ago that it was functioning as a means of hauling industrial goods to the railway station, or to the docks. H. Wolff leased and in 1957 purchased 259 Tenth Avenue (just along 26th Street) which had a private rail siding linking in to the High Line. Traffic along the High Line was halted in 1980, when its northern extension was demolished to make way for the Javits Center. A 5-block section at the southern end was demolished as recently as 1991.

Certainly by 1962 when members of the Guild of Book Workers toured the plant this New York location was only involved in binding books. Printing was done in their plant in New Jersey, or came from different printers. They’d turn out 100,000 bound volumes a day. The Guild members were shown around by Jerry Bloom who I knew in a different capacity later on!

In 1968 the company was sold to American Book-Stratford Press who had just constructed a large plant out of the city in Saddle Brook, New Jersey. Harry Wolff, who succeeded his father in the business, died in 2013 aged 86. It’s surprising how little information about this huge company is readily available online.

We have been going on about this for years. Why can’t we do anything about what looks like a self-harming habit? A large proportion of the books which are bought in quantity by bookstores are returned to the publishers for credit a few months later, and a distressingly high proportion of these will be in unsalable condition. Worse are “special sales” to non-conventional sellers. Brooke Warner, at Huffington Post, describes a deal made with Target for 30,000 copies at a 60% (larger than normal) discount, 25,000 of which were returned and pulped. Of course many special sales do work out, and not all returns are as crazy as this, but it does seem that there are times when bookstores almost appear to “pay” for new inventory by returning older books to the same value. Occasionally smaller publishers have to refuse to supply the full quantities ordered by big retailers because supplying the full order would put the book out of stock, forcing a reprint which the publisher knows will prove unnecessary when the over-ordered books duly come back. There are enough ways to lose money on books — why do we voluntarily add to them? Why do we keep on offering return privileges? Is it just that antitrust law can be read as preventing our colluding over anything? Or are we just stupid?

Here’s a post from The Polished Publishing Group, via Linked-in’s Book Publishing Professionals Group trying to get us to do something about it. But rather than stop the nonsense we seem happy to allow the situation to get crazier. Not only will we accept the return of a print-on-demand book, a book printed specifically for the customer, but you can also get credit for an ebook you decide you don’t want!

Of course there are reasons for allowing returns. We do it to encourage bookstores to have massive quantities of our books spilling out of their front doors, so that passers-by may think “Wow. That must be a great book if they need to have so many copies on hand. I’d better buy one.” This may be silly, but it is certainly true that if the book isn’t available in bookstores many impulse sales will fail to take place. How many? Nobody has the guts to try to find out, though from time to time one or two publishers have experimented with non-return policies, or offered discount incentives for no return purchasing. Amazon, the last place to need to show passers-by the size of their inventory, take advantage of such non-return discount incentives quite a lot. It’s not that the industry doesn’t see that there’s a problem: it’s just that we haven’t yet managed to wean ourselves off the idea that large piles of books generate sales.

It was not always thus. Bookstores used to sell off any over-ordered books at a discount. I remember one great sale in White Plains when I first came to USA, where the price of all the books was halved every week: you had to gamble on how low they’d go before the last copy of that book you sort of wanted disappeared. It was as part of the attempts by big publishing conglomerates to make trade books into a mass-sale item that we discovered returns as a sales incentive This lead to the rapid adoption of returns on all books which has taken place over the past fifty years. Now that everyone is doing it, it is surprisingly hard for us to stop: nobody wants to be at a commercial disadvantage.

Now the commentariat is getting exercised about Amazon’s buy buttons becoming potentially available to third-party sales channels — Amazon says they’ll assign, via some algorithm, the buy button to the source which can deliver the product fastest and cheapest. Brooke Warner, again at Huffington Post, was the first to raise the alarm. The New Republic has an article about it. (Link via The Passive Voice which takes its usual exultantly anti-publishing stance. TeleRead and The Digital Reader have also posted about this.) The worry appears to be that those resellers who offer new books for 1¢ will garner all the sales — which of course will mean the author gets no royalty.

I suppose we have to believe that there really exist new books on sale for 1¢, do we? Nobody appears to know where these “new” books are coming from, though the most convincing explanation seems to be that they come from bookstore returns to the publisher, with a small number being review copies which typically do get into the second-hand market. I don’t spend a lot of time sculling around the Amazon looking for bargains in new books so it’s not a phenomenon I’ve witnessed. I did just take the time to click through (some of ) the New York Times Best Seller list books at Amazon, and failed to find any 1¢ bargains. I did however find one buy button leading to a second seller, but Shattered was available from Soapsix2 for a disappointing $6.50. The nearest approach to the 1¢ barrier was The Handmaid’s Tale, where there were 117 “new” paperback books from $4.05 — a price which probably means that buying one of these would still deprive the author of her royalty — but this is of course a book published long ago which has just gotten a new lease of life, so there can be all sorts of old stock swilling around out there. With a normal new book there just doesn’t seem to be time for “new” copies to accumulate in the unofficial trade. I suspect that the commentariat’s concern is based on nothing other than fantasy. Sure it would be “bad”, especially for the author, if you could get a current bestseller for 1¢ — but it rather looks like you can’t.

Returns, in a modern, well-run publisher’s warehouse, are diverted to a special section, and may even be farmed out to specialist dealers. The returns area/sub-contractors receive the books and issue credit to the bookseller. A little-known side effect of having a well-run modern warehouse is that it costs a significant amount of money to restock a book. Warehouses are set up to receive truckloads of books from binderies and put them cartoned and palleted into storage locations using large forklift trucks. This doesn’t work too well with a single copy coming back from a bookshop. Returns often/usually come back in mixed cartons which have to be opened to see what different ISBNs are involved, scanned to identify then, inspected to see if they are damaged or can potentially be resold as new. If they are damaged the system has to asses whether they are so badly damaged that they can only go for waste, or slightly bashed so they can be resold as “almost new”, or hurt so they can be sold off in bulk at low, low prices to the remainder trade. No one would assert that within this system there is not room for some jiggery-pokery or malfeasance, though of course nobody knows this. Even without jiggery-pokery large skips of slightly hurt books are being sold into the remainder trade for prices which may put the price of any single copy at less than 1¢ — the skip will be sold for some dollars without any regard as to what books it may contain, though hardbacks and paperbacks will tend to be separately binned.

Now this process means that there is a potential for ever so slightly shopworn books to filter into the regular new book trade, certainly into Amazon’s second seller universe. But it takes time for this to happen. Any bookstore is going to keep the books on the shelf for a while in the hope of selling out; when they give up and return to the publisher there’s a bit of time before the return can be processed (it’s not the publishers’ highest priority to process their returns) and then the dealer has to sort and sell the books into the second seller trade. Thus it is unlikely that books from this source can really be too widely available for current bestsellers.

One has of course heard of cartons of new books falling off the back of the truck, but this can’t be too large a problem; maybe a carton here and a carton there. Any systematic violation would become evident: you paid for paper to print 25,000 copies but the warehouse only received 19,076 — even the most inefficient publisher would notice.

Surely we can expect return policy to change soon. The impulse sale factor will never go away, but given that nowadays so much book purchasing can go on online, there’s no real reason for a book-buyer to go without after they have run into a bookstore just as the last copy is snatched up by the person in front of them.

I’ve seen dead insects printed in books and other unidentifiable splodges. Here from The Collation is a picture of a piece of type which has fallen across the type page and been printed — in 1609.

STC 7470 copy 2. The groove is at the bottom of the foot of the type (upper left in the image), the nick on the side. Photograph by Caroline Duroselle-Melish.

Anything which falls onto the type, or in the case of offset lithography onto the plate or the blanket, can remain in place, pick up ink, and be immortalized in print. The press-minder is meant to notice this sort of thing, but even they are just human. Once they do discover such an error they are meant to discard the last few sheets printed on the assumption that several faulty impressions will have happened before they noticed, but sometimes they take out too few sheets.

The forme (form in US) is all the hot metal type, blocks (cuts), and furniture to be printed in one impression locked up after imposition in a chase ready to be moved onto the bed of the press.

This example is not too huge and is going to be printed 4 pages to view. Most book work was done 8 or 16 pages to view which made the formes immensely heavy and unwieldy. They would be wheeled from composing room to press room on metal trolleys constructed at the same height as the bed of the press, so that the forme could be slid into position upon arrival. (It looks like this one is being balanced on such trolley.)

In the early days of printing, when type was often scarce a compositor might set in forme order. In this picture, you are looking at, say, pages 7, 2, 6, and 3. So the comp might deal with those pages first, and then set pages 1, 8, 4, and 5 for the back-up of this sheet. After the first forme was printed the type could be distributed and be available for the comp’s next batch of pages. This procedure would call for accurate casting-off (or compromises on page length consistency).


I wonder if the reason Sponge Bob or Bart Simpson have yellow faces doesn’t have more to do with CMYK than with the RGB solution this video suggests. Of course lots of modern cartoon characters have been developed for TV but the conventions may have been established before that in newspapers and comics, where absorbent papers demanded large screens and spot colors, subtractive rather than additive color.

Still, this is interesting. Maybe yellow does stand out against a blue background being complementary on the RGB color wheel. But just because the sky is blue doesn’t have to mean it’s the constant background color in a cartoon. What about green grass? Wouldn’t complementarity make us want the faces to be magenta?

I just bet the convention predates television, so that the explanation is rather more ink-based.

Link thanks to David Crotty at The Scholarly Kitchen.

ink-ballIt is slightly hard for our modern sensibilities to take, but we cannot avoid the knowledge that early printing houses reeked of urine. The ink was applied to the type by a couple of ink balls. These were leather-covered pads, and in order to keep them supple they were stored overnight in a bath of urine. Urine was also handy for cleaning off excess ink. An ink ball is illustrated at the left, and their use shown below. You can also see ink balls in operation in the video at my recent post Printing on a Gutenberg press.









Ink for writing with a pen would be water-based, while for printing it evolved to be oil-based. In the early day of printing, printers made their own inks with lampblack or soot and animal glue or vegetable oil which each boiled up according to their own closely guarded formula. Part of the success of Gutenberg’s printing innovation is due to the special ink he developed for transfer to and from the cast metal type. Ink making became a commercial process in the 17th century, and the first ink factory in America was established in 1742.

Little color was used in inks until the discovery of coal tar types in the middle of the 19th century though early Chinese printers had added some earth elements to their inks even before Gutenberg’s time. Linseed oil (a vegetable oil) was the main vehicle in printing ink until the mid-1930s when new vehicles (oils and resins containing specific chemicals depending upon what the inks are going to be used for) were introduced for letterpress printing in the United States. UV (ultraviolet) and EB (electron beam) curing vehicles for ink and coatings were introduced in the 1970s. More recent developments in inks have been water-based ink for gravure and flexography, and soybean ink for lithography.

In classical times the ink consisted of soot, gum arabic and water. Shady Characters has an interesting piece on the inks used in Roman manuscript work in which he tells of an early use of metallic inks found at Herculaneum, thus dating to 79CE. For those who crave the condition of a scribe here are instruction on how to make your own ink (remarkably simple, though not as simple perhaps as going out and buying a bottle of ink).

Today printing inks are made of four basic components: 1. pigments to color the ink and make it opaque; 2. resins, which bind the ink together into a film and make it stick to the printed surface; 3. solvents to make the ink flow; and 4. additives which alter the physical properties of the ink to make it suitable for different types of printing. It is a two-stage process: first they make the varnish, which is the base/vehicle used for all inks, though its recipe will vary depending on what the ink is to be used for. It is made by mixing the resins, solvents and additives. The resins react together to create some larger molecules which make the varnish more viscous the longer these reactions go on. In the second phase the pigments are mixed into the varnish, a process which can be seen in the rather lyrical video about modern ink making which can be found here. It’s well worth watching.


The relationship between a publishing company and its freelance workers tends to be small-scale and personal. You know this person does that sort of work; you trust this person; you send them a job; the job gets done on-time and up to expectation; you OK the invoice; the freelancer gets paid. Everyone’s happy, and so it goes.

Now of course the problem with the gig economy is those boring things like pensions and, in US, medical insurance. It’s all very well being a free spirit and leaving the future to take care of itself: only rarely does this work out without serious sacrifice — wealthy parents in the deep background help here, but it will always be hard to put aside some of your irregular earning for those rainy days. Of course many freelancers are seeking a full-time job. Designers and copyeditors who have recently been laid off make up a sizable proportion of the freelance workforce. Many of them can continue taking on freelance projects even after they get a full-time job. Subcontracting work to a full-time employee in your own company flirts with the unethical. I have done this myself, at both ends, and of course I wouldn’t do anything unethical! I’ve never done on a freelance basis work which I would have had to do during work hours, and this seems to me to be the boundary. Still the ethics of all this are so murky that I do think it wise for a company to have a policy against freelance work being done by current employees. There are enough other publishing companies for whom you can freelance after all.

The idea that freelancers should be organized keeps cropping up, but it seems almost to be a contradiction in terms. If you wanted to be organized you’d be trying to get an in-house job wouldn’t you? The Freelancers Union started in 2001, having grown out of Working Today, and now has 350,000 members about 25,000 of whom buy insurance through the union. If you are that free spirit who can tolerate the uncertainty which comes along with forgoing a salary in return for the ability to work when and however you choose, then joining the Freelancers Union does seem to me to be a good idea.

A couple of years ago Publishing Perspectives brought us the story of Whitefox, a UK publishing services company with a large team of freelancers which can be called into play on appropriate projects. I don’t think this is really the freelancer central that the story implies it is. It really looks more like a publishing services company (surprise surprise). Almost all publishing services companies already rely on freelance workers, but Whitefox does seem quite a large one. The idea of a central register of freelances, all vetted by the organizers of the register, does seem like something that might be valuable, but, circling back to the beginning of this post, so much freelance subcontracting is based on personal relationships that publishers might never need to use such a service.

The New Yorker has a piece on the gig economy. Much of the slow increase in employment after the last recession has been in 1099* work rather than in the W2 jobs which were largely the ones lost. More and more people are being tracked into part-time or contract work. And maybe this is the way things are going to be for the foreseeable future? Certainly there are a lot of young, well-educated people who like things that way. In principle the entire publishing process could be handled perfectly satisfactorily with such an employment model. The main motivation of companies who encourage gig working must surely be cost reduction. If you employ only contract employees you save all those large outlays on pensions, time-off and medical coverage: what’s not to like for a company? The main motivation for the workers who like this sort of work? No doubt a feeling of control: you work when you want and do what you chose. Still, there may come a time when the young English graduate tires of cleaning bathrooms in Airbnb apartments. Maybe then the response is just to move on to something different: as long as you keep on keeping on perhaps you can get by. Maybe fulfillment is to be found in other areas of life — as, be it said, it probably should.

Now if we lived in a world where a universal basic income were available to all this sort of problem would evaporate, especially if we managed also to get our (American) heads round the obvious concept that a single-payer health scheme is the way forward. If W2 employment really is going to dwindle with more and more people “gigging” there should in theory grow a constituency for these radical alternatives to the rat-race.


* For the non-US reader; these numbers refer to tax forms. If you are a full-time employee, every year your employer will send you (and the taxman) a W2 form, detailing what you were paid, and what taxes etc. were deducted. A 1099 MISC is the form a part-time, contract employee will receive. Part-time workers no doubt receive several of these which must make tax preparation that much harder. But hard seems to be the way we like tax prep to be over here. Heath-care in America was originally designed to be provided through the employer: most 1099 employees will not be receiving benefits of this kind.