The shortest answer is “nothing”. You might take a screenshot of each page and then feed those images to OCR software… but that’s really not worth the effort. Also, note that you may need to switch on a “text selection” tool in the document viewer to be able to copy-paste. Also, if I remember correctly, acroread has a “save as text” menu item in the file menu.

The somewhat longer answer is that someone somewhere may have made a program to do that, but you’re asking the wrong person. I don’t know much about software for MS Windows, I never used that operating system for anything else than simple office tasks … and even that was many years ago. By the way, please note that depending on where you live, it may be illegal to break the copy protection of an ebook you paid for. Silly but true.

The ultimate answer is that your computer isn’t doing what you want (converting some ebook format to plain text), but what someone else wants it to do (which is refusing to do what you want). It’s an annoying situation which is a natural result of using software of which you don’t have the human-readable code. If you did had that sourcecode, you (or someone else) could modify it to ignore the “disallow copying” information in ebook files. Unfortunately I’m not sure if such software (known as “free software” or “freedomware” or “open source” ) is available for all ebook formats. PDF is no problem, about the others I don’t know. It’s difficult to create software that reads file formats that are not published standards…

As far as possible, I use only software that I’m allowed to modify. This is one of the reasons I don’t use MS Windows. Another reason is that Windows simply can’t be made to work the way I like (this can be argued to be the same reason in different words). Linux (more precisely GNU/Linux) is what I use. If you’re curious, I recommend this website: http://www.getgnulinux.org/

UNIX is a group of operating systems that are all derived from one operating system designed very long ago, that are certified to match some standards. For example Solaris and BSD are UNIX, and - surprise! - so is MacOS X (if I’m not mistaken. at least it is UNIX-like, see next paragraph).

UNIX-like is a bigger group of operating systems, that all resemble UNIX. Most of these allow you to work in a graphical environment similar to what you’re used to from Windows, but all also have a text based working environment (”command line”). The general method of using the command line is to write a description of what you want in some plain text file(s), run some small program(s), and find what you want in a new file made by said programs. This may seem primitive, but it allows for really advanced automation of tasks. For example a few years ago I was in a situation where images and bits of text were shared between a few documents. I made it so that after modifying some image or text, I would have to type only one word on the command line, and my computer would figure out which documents needed to be updated, would update them, and then upload both pdf and html versions to my website. This automation was quite easy to do!

To convert an ebook to plain text, I would type this on the command line:

pdftotext AliceInWonderland.pdf AliceInWonderland.txt

Super easy, but works only for pdf files. For other filetypes there are other commands (for example antiword or catdoc for MS Word files), but as I said before I don’t know if these are available for all ebook filetypes. Usually a good strategy is to first “print” them to a postscript (ps) file in Windows, and then use ps2txt on Linux. Chances are that if copy-pasting isn’t allowed, printing isn’t allowed either, though.

thanx
Nat

Open the document on your computer, select all text, copy the text, open a plain text editor (on windows this would be notepad), paste the text in that editor, save, and you’re done.

Of course many ebooks are protected against this trick :/ .

On Linux and other UNIX-like operating systems there are many conversion utitilities with names like “pdftotext”.

\documentclass[10pt]{article}
\usepackage[T1]{fontenc}
\usepackage{txfonts}
\usepackage[vcentering,dvips]{geometry}
\geometry{papersize={128pt,160pt},total={108pt,140pt}}
\usepackage{color,graphicx}

I especially wonder what will happen to illustration and special formatting.

Thanks

Enlightenment - yes, it’s nice. I use it. But it’s not truly different from xfce and gnome. I know it’s very customizable, and might be a good prototyping environment for an improved GUI, but it still has its limitations. For example I don’t think splitable scrollbars can be made just by building an edje file.

Some e apps are nice because they are well designed, but at the same time those same apps are NOT nice because they are not even similar to each other. What we need is basic widgets that are “nice” so the whole desktop is both “nice” and consistent.

Consistent interfaces - I’m all for consistent interfaces. But why must they be consistent with a crappy design that just happens to be part of the dominant OS instead of just with themselves?

Interesting factoid: OpenLook was meant as a standard UI that could be implemented on any OS. There were multiple implementations of it that didn’t share any code. Unfortunately the only that was (partially!) released as free software was XView/OpenWindows from Sun (which by the way has nothing to do with MS Windows). It was discontinued so Motif could become the universal GUI for UNIX… a big mistake!

The quest for a better, more efficient GUI is an honest one, but doomed to obscurity. There is already plenty of choice given by existing WMs - like Enlightenment.