


Few writers have been around and working for as long as I have. Same is true of computer specialists. Endurance doesn’t mean I do great work, but it does give me some perspective on the quality of work that is being done.
For instance, Apple claims there are over 200,000 apps (applications) for the new iPad, rather shocking to an old guy who once knew about pretty much every application in the world back in 1956. What’s most shocking about this enormous figure is that there must be at least 50,000 programmers behind it–and probably more than 100,000. My impression is that they can’t all be doing a quality job–or can they?
Something similar is happening with writing, or perhaps I should say “publishing.” Bowker reports that about 300,000 books are published annually in the USA. Others report more than 600,000, but no need to argue about the numbers. In either case, there are a lot of “writers” out there. My impression is that they can’t all be doing a quality job–or can they?
How can we answer the quality question? Well, some people measure quality (or, rather, lack of quality) by the number of errors they find in a book or website (or bugs in a program). That’s not the only measure, but it’s perhaps the easiest one to get our hands around. Easiest, but not necessarily easy.
Rather than ask the questions in general, let me take the risk of asking them about myself. I have records and memories of how well I did 50+ years ago, and how well I’m doing now. For example, the first person who read the published version of my latest novel, Earth’s Endless Effort, liked the book but sent me a list of 19 typos. Here’s the first item on the list:
Page 6. She thought her only damage was soiled jean, until she grabbed a small tree and pulled herself upright.
Did you see the typo? A missing plural s. Clearly an error that slipped by five first readers.
Okay, here’s the next one:
page 8. She remained standing and cautiously and limped ahead.
Did you see this one? An extra “and.” Again, missed by five readers (and me).
Before continuing in this vein, let’s define “quality”: Quality is value to some person.
So, you be the person. You’ve seen these two clear errors. Have they reduced the value of the book to you?
I know that for me, I’m bothered by typos. If I read a published book and find two typos on the first 8 pages, I’d be rather wary about the quality of the book. Typos on 25% of the pages? Not so good.
But my suspicions weren’t confirmed. Earth’s Endless Effort is 324 pages long, so 19 typos on 324 pages means typos on 19/324. That’s typos on less than 6% of the pages. So, if I kept reading, I would have stopped worrying about the typos. (Of course, as author, I worried about them immensely, and immediately set about repairing them.)
My first thought upon receiving this list was, “I didn’t make this many errors in the old days.” But I was wrong. I did make this many errors, and always have. But in the old days, I had the services of a professional copy editor on each of my books–in addition to five or more first readers. So, I serious asked myself, “Do I need to hire a professional copy editor for each new book?”
To answer that question, I had to ask myself what was the value of this extra stop in the publishing process? What was the cost? I investigated the cost and found a fine copy editor conveniently here in town (Albuquerque) who would edit the book for $10 per page for copy editing fiction. A little web research showed me that charging per page is not typical. The majority of copy editors seem to charge by the hour, ranging from about $25 to $120 per hour.
So how fast do they edit? A typical figure was 6 to 8 pages per hour. Using the lower figure, I get rates between $4+/page and $20/page. So, the quote I received was somewhere in the middle, and for 324x$10 = $3,240, I could have had a professional copy-edit of Earth’s Endless Effort. So, assuming my editor caught all 19 of these typos (and maybe a few more), would it be worth it?
Since I earn about $4 for each book sold, either in paperback or eBook, I’d have to sell over 800 books just to pay the copy editor. Well, would my readers would pay, say, one dollar more for a book without these 19 typos? Would you? Some would, but most wouldn’t. As it is, I’m not likely to sell 800 copies of Earth’s Endless Effort unless it somehow takes off through viral marketing. Not in a year or two, at least.
Every one of my 40 non-fiction books has sold more than 800 copies (more than 8,000, actually), and some have sold more than 250,000 copies. At that sales figure, my copy-editing cost would amount to about a penny a page–well worth the cost to me, just to maintain my reputation as a careful author.
But even those carefully copy-edited non-fiction books still contained errors in their published versions. Because of its top, in The Psychology of Computer Programming I kept track of the typos as they were reported to me. Even after professional copy-editing, more than 19 errors turned up. And even after 35 years, a reader informed me of a typo that nobody else among more than 250,000 readers had ever reported.
The fact is that in 50 years, no reader has ever asked me for their money back because of a typo–or a gaggle of typos (is gaggle the plural word for typos, or is this yet another typo?) The fact seems to be that as long as the number is kept low enough (which differs a bit for each reader), most books of fiction don’t lose value from typographical errors–especially when it’s obvious what the sentence really says. (It’s different for poetry, of course. As Oscar Wilde said, “A poet can survive anything but a misprint.”)
Oh, by the way, not all the 19 typos were in fact errors–only 16, by my count. For instance, at one point, a character calls skunks “polecats,” which in certain regions of the USA is slang for “skunks.” My reader is German, so probably didn’t know that, but in a way, though it’s not a typo, it may have been more bothersome than a “real” typo because it stopped my reader in his tracks. I know that often happens to me as reader. In fact, the fewer the typos I find in a book, the more actual typos I do find bother me–because I wonder if perhaps I’m wrong about them. So, for me, quality actually worsens if the copy-editing is too good. (It’s never perfect, believe me.)
With the coming of the web with its hundreds of thousands of “programmers,” the question of what’s “good enough” quality has arisen in the computer programming field. But that’s nothing new, because 50 years ago, in the space program, we had to consider the question with extreme care. Those of us who had John Glenn’s life in our hands knew that “a few bugs” wasn’t good enough, so we spent literally millions of dollars to ensure that each potential bug was tracked down and removed. (And yet we still missed a few, though luckily nobody died because of these bugs that got away.)
So what about those 200,000 apps for the iPad? I checked with Apple’s website and found the page featuring some of the best of those apps. They were the following: Popular Sciene Magazine; a “Real Racing” driving game; the periodic table of the elements; an alphabet game to teach children their letters; a singing coach called Glee; a real-estate search app; the Weather Channel; a to-do list; Marvel Comics; The Wall Street Journal; 25,000 recipes; a baseball watcher; real-time stock quotes; another car racing game; Scrabble; a car race watching app; and a labyrinth game.
It’s hard to see how anybody’s life will be endangered by a bug in one of these apps. (Maybe their virtual life while car racing.) Well, maybe somebody would lose money because of a wrong stock quote, but that’s about the worst I can imagine. And, after one or two errors like that, most users would simply stop using that app. I’d guess there are at least a hundred stock quote apps to choose from in the collection.
So, given the different meaning of “quality” to most iPad users, those “programmers” are still doing a quality job. And so are those authors who self-publish books containing 16 typos.
There’s a movement in the computer community today called “good enough software.” (Some people have blamed/credited me with starting this movement. I don’t think it’s important who gets the credit–or blame.) <http://www.laputan.org/pub/sag/Good-Enough-Software.pdf> We are learning to accept what us old-timers would call “lower quality.”
We old-timer writers whose works are being posting on the web will also have to learn to accept what we would call “lower quality.” Book View Café is attempting to be a publisher that maintains those old values, but that objective is fundamentally in opposition to some of the other things we value—freedom from artistic interference; speed from writer’s brain to reader’s eyes; access to our literature at the lowest cost; and probably others.
So, I believe that we need to learn how to be “Good Enough.” According to the computer pundits, “to claim that any given thing is Good Enough is to agree with all of the following propositions:
- It has sufficient benefits.
- It has no critical problems.
- The benefits sufficiently outweigh the problems.
- In the present situation, and all things considered, further improvement would be more harmful than helpful.
Are we “good enough” to do all this?
Gerald M. Weinberg is a member of Book View Café and hopes to blog here regularly. His science fiction “First Stringers: Eyes That Do Not See” by Gerald M. Weinberg is serialized on the front page rotation.
For more about him and his fiction please visit his bookshelf here on BVC:
http://www.bookviewcafe.com/index.php/Gerald-M.-Weinberg/
Or his personal web page, http://geraldmweinberg.com






A woman published a book about looking for Mr. Good Enough instead of Mr. Perfect. I liked the concept… others took offense.
Me? I wrote this…
In various degrees, in various settings, I’ve learned that “Good enough often is… good enough.”
I even wrote a post “Mr. Good Enough — that’s me!”
musingsandmiscthoughts.blogspot.com/2010/03/mr-good-enough-thats-me.html
Not everything is a space flight. And most of the time… when we screw up… nobody dies. (I teach my students the same thing in math. Being wrong is often a needful step in learning to be right.)
Thought provoking post.
Typos, thinkos, and awkward or confusing constructions leap out to my eye. I’m an engineer, so I try to fix stuff. If you have a document on a topic that interests me, or if you are already an author I love or respect, I offer a nitpicking review before publication for free.
If you like the content of my review, I suppose you could send me a copy of the published document or even some money, but that is not the point of the exercise. The example typos did leap out to my eye and when I reviewed Jeff Conklin’s “Dialogue Mapping” he said I caught five times as many small flaws as the publisher’s copy editor.
The authors I review tend to have first names that start with the J/G sound. Computer languages I enjoy and use tend to start with P. Dunno why in either case. Regardless, when a flaw annoys me, I use the energy of that annoyance to attack the flaw. Sometimes that bothers the recipient of my review. Oh well, nobody is universally appreciated.
On software quality, Richard Gabriel explained the success of West Coast over East Cost Lisp (and Linux over pretty much anything) as “Worse is Better”. Provocative. Hope you enjoy it.
There’s a link to it at
http://www.dreamsongs.com/WorseIsBetter.html
Glad you’re feeling better!
I stumbled a little over the typos in this post, but I still benefited from reading it. In fact, I think the typos only helped to reinforce your message. Thanks for writing this. /Tobias
I liked the comparision you made between writing books and developing software.
it has always been confusing when to conclude that a product has attained quality…
your definition of quality: “Quality is value to some person”
but the value may change over time which varies the quality of product…for non software products there is a guarantee of 1year which claims if you encounter any problem within 1 year, product will be replaced…but if I am using a software product and incurr a loss there is no guarantee.
do you think we can see this in future, if software products have critical problems then user is eligible for getting compensation for that loss…or may be I am not aware its already implemented in some part of the world.
Well, there is always Sturgeon’s Law: ’90% of everything is crap”. This leaves 10% which is not-crap. Estimating that all professional print publishers print only 1% of all manuscripts submitted (this was the percentage that Baen gave once), this still leaves 9% of all manuscripts that are not-crap and not-print-published (and we all know members of that 1% crowd that we all shake our heads over and wonder how on earth _that_ got by?).
But Baen is only one company. What is the percentage of all manuscripts submitted to all companies that ends up published? I suggest that it is considerably less than 10%; after all, Baen tends not to get Romantic Fantasies, while other companies tend not to get Mercenaries of Doom (stereotyping hugely; Miles Vorkosigan was indeed a Mercenary of Doom, but he is not a cliche). Also, some writers end up not submitting their manuscripts to all companies, because they’d have to live 200 years to do so, given current wait times for submissions. Also, some writers do not have agents, and given the number of houses which are now agent-only, this reduces the number of manuscripts submitted (some say this is a _good_ thing).
So, let’s be generous; let’s say all print publishers accept and publish up to 5% of all manuscripts submitted. That still leaves 5% of all manuscripts submitted which do not suck, and which people might enjoy.
Now, trying to _find_ them in a huge haystack of those which are, er, challenged, that’s the hard part.
Shiv asked:
“for non software products there is a guarantee of 1year which claims if you encounter any problem within 1 year, product will be replaced…but if I am using a software product and incur a loss there is no guarantee.
do you think we can see this in future, if software products have critical problems then user is eligible for getting compensation for that loss…or may be I am not aware its already implemented in some part of the world.”
Shiv, this is such a pertinent question, I think I’ll try to write a blog post about it. It may take me a while to work it up (after all, I’m writing novels and such), but I’ll get to it.
Thanks for asking,
Jerry
Jean Lamb wrote:
Well, there is always Sturgeon’s Law: ’90% of everything is crap”. This leaves 10% which is not-crap. Estimating that all professional print publishers print only 1% of all manuscripts submitted (this was the percentage that Baen gave once), this still leaves 9% of all manuscripts that are not-crap and not-print-published (and we all know members of that 1% crowd that we all shake our heads over and wonder how on earth _that_ got by?).
But Baen is only one company. What is the percentage of all manuscripts submitted to all companies that ends up published? I suggest that it is considerably less than 10%; after all, Baen tends not to get Romantic Fantasies, while other companies tend not to get Mercenaries of Doom (stereotyping hugely; Miles Vorkosigan was indeed a Mercenary of Doom, but he is not a cliche). Also, some writers end up not submitting their manuscripts to all companies, because they’d have to live 200 years to do so, given current wait times for submissions. Also, some writers do not have agents, and given the number of houses which are now agent-only, this reduces the number of manuscripts submitted (some say this is a _good_ thing).
Jerry: Jean, I think Baen must have been talking about fiction mss. only. From many interactions with my non-fiction editors, I’d say it’s not that high for non-fiction (if we don’t include poorly-written-but-editable non-fiction that really has something to say).
[What is "poorly written ..."? Try the previous paragraph as a terrific example.]
Jean:
So, let’s be generous; let’s say all print publishers accept and publish up to 5% of all manuscripts submitted. That still leaves 5% of all manuscripts submitted which do not suck, and which people might enjoy.
Now, trying to _find_ them in a huge haystack of those which are, er, challenged, that’s the hard part.
Jerry:
Precisely! (and much better written than my comment. I’ll try to do better below.
There are other ways to find those needles, ways that do not involve time-consuming slogs through slush by highly-paid editors.
One modern way is to do what BVC (and others) are trying: Put the mss. up in a cheap and accessible way, then capture reader feedback to improve those mss.
To improve them, or perhaps to “improve” them by throwing them away if the feedback is sufficiently devastating. The most important feedback is the popularity of the story. If nobody spreads the word, perhaps that story should just fade away.
And I don’t mean that total-popularity-in-numbers = quality. It does mean one kind of “quality”–probably the kind large print publishers are seeking. But we at BVC, for example, think we ought to pay attention to any story that receives a certain pattern of continued reader interest, even if not numerous to interest large publishers.
Oh, and BTW, this method is not really new. It was used by perhaps the greatest English novelist of all time (in my biased opinion). Charles Dickens didn’t have an internet, but he did have newspapers in which he serialized many (most?) of his novels. [Steampunk authors/readers, take note!]