Does It Compute?

Mark Twain on Thanksgiving

Mark Twain is the quintessential American literary figure. Consciously or not, he echos Montaigne (“Every man bears the whole stamp of the human condition”) when he says in his Autobiography: “The last quarter of a century of my life has been pretty constantly and faithfully devoted to the study of the human race--that is to say, the study of myself, for, in my individual person, I am the entire human race compacted together.”  

He particularly had a grudge against hypocrites, Congressmen and religion. “Irreverence is the champion of liberty and its only sure defense.” His statements about Congress are classic (and all the more true today):

    “Suppose you were an idiot. And suppose you were a member of Congress. But I repeat myself.”
    “It could probably be shown by facts and figures that there is no distinctly native American criminal class except Congress.”

Here’s what he has to say about Thanksgiving (when he wrote, Thanksgiving had not yet been given a fixed date and was renewed every year by Proclamation):

“Thanksgiving Day, a function which originated in New England two or three centuries ago when those people recognized that they really had something to be thankful for– annually, not oftener – if they had succeeded in exterminating their neighbors, the Indians, during the previous twelve months instead of getting exterminated by their neighbors, the Indians. Thanksgiving Day became a habit, for the reason that in the course of time, as the years drifted on, it was perceived that the exterminating had ceased to be mutual and was all on the white man’s side, consequently on the Lord’s side, consequently it was proper to thank the Lord for it and extend the usual annual compliments. The original reason for a Thanksgiving Day has long ago ceased to exist–the Indians have long ago been comprehensively and satisfactorily exterminated and the account closed with Heaven, with the thanks due. But, from old habit, Thanksgiving Day has remained with us, and every year the President of the United States and the Governors of all the several States and the territories set themselves the task, every November, to advertise for something to be thankful for, and then they put those thanks into a few crisp and referent phrases, in the form of a Proclamation, and this is read from all the pulpits in the land, the national consciences is wiped clean with one swipe, and sin is resumed at the old stand.”

Nicholas Carr: "Automation Makes Us Dumb"

Nicholas Carr has an essay in this weekend's Wall Street Journal that summarizes the main points in his recent book, "The Glass Cage."  I blogged about this last week. 

His conclusion: 

"If we let our own skills fade by relying too much on automation, we are going to render ourselves less capable, less resilient and more subservient to our machines. We will create a world more fit for robots than for us."

Check it out at The Wall St. Journal.

Nicholas Carr: "The Shallows. What the Internet is Doing to Our Brains."

Yesterday I published a review of Nicholas Carr's latest book, The Glass Cage, on automation. Four years agon I reviewed his previous book, The Shallows. It is worth reprinting (the memory of the Internet is short). Here it is:

In 2008, Nicholas Carr published an article in The Atlantic titled “Is Google Making Us Stupid?”  His recent book, “The Shallows: What the Internet Is Doing To Our Brains,” expands on that article in a fascinating, massively documented and very disturbing book.

The first half of the book is a history of the brain. The key concept here is that modern studies have shown the brain to have “plasticity,” that is, it can rewire itself in response to external stimulus. This contrasts with the older theory of the brain as mechanical (or bio-electrical) and unchanging once it is fully formed in your late teens or so.  It turns out that neural pathways can be altered and reformed, and that once new pathways are established they do not “bounce back” when the external stimulus is removed. Brains are plastic and mouldable, but not elastic.  Carr demonstrations this with the evolution of printing, maps and clocks and how they have changed our perception of the world permanently.

And although he does not use these examples, this may also help explain why the builders of Stonehenge or the Mayan cities could develop such extraordinarily precise astronomical readings for the seasons, phases of the moon, and do on.  Their brains were focused on different topics than someone used to getting information from the printed page.  They had capabilities that we have lost.

The development of a phonetic alphabet (fully developed by the Greeks around 750 BC) and the advent of reading changed peoples’ brains dramatically.  This has been the dominant intellectual trend ever since: what Carr terms the “intellectual ethic” of the technology of producing written works. You sit down, you read a book in relative concentration and peace.

The Internet changes all this.  People no longer read books or even articles, they read snippets. As Carr puts it (massively supported by numerous studies), “We don’t see the forest when we search the Web. We don’t even see the trees. We see twigs and leaves.” As Carr extensively documents, the tendency to read short passages, to jump from one hyperlink to the next actually changes the way the brain perceives the world. Our “plastic” brain is reshaped. “With the exception of alphabets and number systems, the Net may well be the single most powerful and mind-altering technology that has ever come into general use. At the very least, it’s the most powerful that has come along since the book.”

This is the crux of the matter. Carr cites several studies showing that if two groups of people are given the same article to read, in one case containing numerous hyperlinks to related information, and the other without hyperlinks, the group without the hyperlinks has a better grasp of the content of the article when tested later.  “Research continues to show that people who read linear text comprehend more, remember more, and learn more than those who read text peppered with links.” Or, as Carr puts it in one of his many pithy formulations: “What we’re experiencing is, in a metaphorical sense, a reversal of the early trajectory of civilization: we are evolving from being cultivators of personal knowledge to being hunters and gatherers in the electronic data forest. Google is “training our brains to pay attention to the crap.” or “the more we use the Web, the more we train our brain to be distracted.”

If all this is true, then how did Carr focus enough to write a 200 plus page book filled with hundreds of quotations and references?  Initially, he says, he had a great deal of difficulty. Then he retreated to the wilderness of Boulder, Colorado, where there was no cell phone service, only a relatively slow DSL web connection. He cancelled Twitter and Facebook, mothballed his blog, shut down his RSS feeds and changed his email settings from checking once a minute to checking once a day.  After a while, he was able to concentrate enough to write the book.

I was also gratified to see studies that confirm what I have always thought: if you make software too “easy to use,” you in fact prevent people from actually learning to use it and their ability to tackle any problem out of the ordinary decreases.

An issue that Carr does not address is the quality of information available on the web. With information being “flattened” – everything is available, you do not have to work to get to it – how do you know what information is good, true, valuable, etc? Take Wikipedia: a great resource if you are looking for somebody’s dates of birth, but otherwise notoriously unreliable (some colleges have banned using Wikipedia for research projects). And studies of scholarly research papers show that fewer articles are being cited, that is, the breadth of knowledge and reference has actually shrunk. This is due to the reliance by search engines on popularity: the more something is quoted, the higher it appears in the search engines and therefore the more it will be quoted.  The ease of finding information tends to impede searching further.

This is a must read for anyone interested in the development of the Internet and its effect on the human brain – and therefore the future of humanity.

Nicholas Carr on Automation

I always read Nicholas Carr. His last book, The Shallows, on the effect the Internet has had on thinking processes, was fascinating and disturbing. Carr demonstrated at length that whereas knowledge used to be defined as actually knowing something, with the internet it is defined by knowing how to look it up. People no longer read books or even articles, they read snippets. As Carr puts it (massively supported by numerous studies), “We don’t see the forest when we search the Web. We don’t even see the trees. We see twigs and leaves.”

His new book, The Glass Cage, deals with the effects of automation. The basic thesis, which emerges from many examples, is that automation leads to “de-skilling,” the loss or degeneration of previously acquired skills when they are replace by automated processes.  Some of this is well known: the industrial revolution reduced skilled craftsmen to unskilled machine operators.  Other examples are less obvious.

There is evidence, for example, that the automation of the airline cockpit has led to the loss of skills on the part of pilots and is likely responsible for a number of crashes where pilots reacted wrongly in emergency situations and caused a fatal crash, including the Air France crash in 2009 which killed 228 people.

More disturbingly, it appears that the computerization of the medical profession has contributed to sloppy diagnoses, unnecessary tests and increased billing as doctors’ skills decline when faced with a reductionist set of options on their computer. I know I’ve had more than one doctor stare at her computer and say “why isn’t that option here?” Scary.

What underlies many of the problems with automation in various areas is that it seems the brain can be trained and get stronger with exercise, just like muscles. This has been documented in the brains of London cab drivers, who are famous for being able to find any address anywhere in London. But the converse is also true: when the brain is not required to work as hard, it loses the ability to deal with situations it had previously been trained to do. The brain becomes, for lack of a better word, flabby. This is the key analysis that results from and underlies all of Carr’s examples. It happens in a variety of ways and depending on various functions. For example, heavy reliance on a GPS system instead of “building cognitive maps of space” to navigate by may lead to atrophy of the hippocampus which has been linked to the development of dementia. Can using GPS systems contribute to early-onset Alzheimers?

I see similar phenomena all the time in terms of the way people use software. Generally speaking when users are trained on new software (if they even are trained), the maximum they learn is the minimum they need to do their job. To the extent that a firm has invested in making the software easier to use by implementing shortcuts, macros, bookmarks in Worldox or other usability features, people’s ability to use the base features of a program without the productivity improvements declines sharply. Very often, if something disrupts the program users are completely helpless, leading to emails from users with subject lines like “Worldox has stopped working” or (my favorite) “it’s doing it again.” Not exactly enough information to be useful. Even basic Windows troubleshooting like exiting the program or logging off and logging on again never occur to them. As Carr puts it, “whether automation enhances or degrades our performance in a given task, over the long run it may diminish our existing skills or prevent us from acquiring new ones.” And Carr quotes George Dyson: “What if the cost of machines that think is people who don’t.”

In response to this situation various software makers, Microsoft in the forefront, have tried to make programs easier to use and more bullet-proof. But in line with Carr’s analysis, as things are made more automated and easier, the degeneration of skills increases, potentially leading to a downward spiral: Microsoft simplifies software to meet lesser skills; skills decline; software is simplified further, etc. This also penalizes “power users” because in an effort to simplify, advanced features are hidden and made more difficult to use. It tends to reduce peoples willingness to ask for extras: “how do I do X?”Or as Carr puts it, “the cost of accommodating the dilettante is a demeaning of expertise....As we grow more reliant on applications and algorithms, we become less capable of acting without their aid.”

In terms of the software that lawyers use every day, one approach to this dilemma would be to “build your own automation.” If you are using Word, have you customized the Quick Access Toolbar and the Ribbon Bar?  This is one of the main ways of making Word more efficient for you. If you are using Worldox, have you created Bookmarks (like the bookmarks in a browser)? Do you create and save searches? Quick Profiles?  More generally, do you ask or research how to do additional things?  This is a good way to begin offsetting the natural tendencies of skills to degenerate (a version of the famous second law of Theromodynamics).

Carr ends with a lyrical assertion of the value of using tools and technology: “Used thoughtfully and with skill, technology becomes much more than a means of production or consumption. It becomes a means of experience. It gives us more ways to lead rich and engaged lives.... To cede choices about the texture of our daily lives to a grand abstraction called progress is folly.”

I reviewed Carr’s earlier book, The Shallows when it came out several years ago. The Internet has a short memory, so I will republish it again tomorrow.

Copy and Paste from Acrobat

I’ve recently had several clients ask me what the best way is to copy and paste text from a PDF file into Word. It turns out that this is a complex and thorny issue with lots of “gotchas”.

The prerequisite, of course, is that the document has to be run through OCR (“Optical Character Recognition”) so that the text can be searched. OCR processes differ considerably by manufacturer. In my view the Abbyy FineReader is the best (this is the program that ships with the Fujitsu scanners and is also used by Trumpet’s Symphony products for Worldox). It is noticeably better than the native Acrobat OCR (more about that later). So if you are scanning with something that can use this by all means turn on the OCR.

In theory, once a document is OCR’d, copy and paste is fairly simple: Select the desired text using the Select tool in Acrobat (an icon on the second line of Acrobat:

Selecting text can be iffy depending on how it was formatted. Tables are particularly difficult. You may want to position the cursor at the beginning (or end) of text to be selected, hold down the Shift key and use the cursors to select text. This tends to be more reliable than trying to select with the mouse.

Once the text is selected, right-click to copy it and then paste it in to Word. Most often, text will be copied in with a Hard Return at the end of each line of the original pasted text. These have to be removed. It is possible to write a little macro that will do this line by line. Create a find and replace that replaces Paragraph Signs (Hard Returns) with a space. Theoretically you can automate this to replace only Paragraph Signs in selected text but due to a bug in all versions of Word, this does not work properly the majority of the time and you frequently end up replacing all Hard Returns everywhere in the document. You will have fewer errors if you do it manually, assuming the text is relatively short. Alternatively, you could paste the text in to a blank document, run the macro, then move it back in to the document your were working on (I said there were gotchas).

If you use Acrobat to export a document to Word, the result is likely to be unsatisfactory.  In particular, many paragraphs will be converted as text boxes and will be extremely time-consuming to clean up. This is another problem that Abbyy does not have – generally speaking it converts text properly as just text. Therefore you will be best off copying and pasting. Converting the whole document can cause problems and is frequently not time efficient.

There are many alternatives to Acrobat that may give better results in terms of copying and pasting text. I have not tested them all, but they include pdfDocs (a version of which can be added to Worldox along with other features at a modest price), Nuance PDF Converter, Foxit PDF editor, Nitro PDF Professional and CutePDF. Depending on exactly what you want to do, one or another of these programs may be a good option.