I took a course in American political thought and I would say the best source of information on the Constitution would be the writings of James Madison followed by The Federalist and The Anti-Federalist papers. Reading those gives a deep insight into the reasoning behind the U.S. Constitution.

There is DBMail.

Here you go. Now use pdftotext and read it however you like.

I've been on /. for nearly half my life (I finally registered only 8 years later). While I was growing up, I read Slashdot almost daily and every time I would learn something interesting from either the FA or the comments. It is the variety of stories, diversity of opinions, and the signal-to-noise ratio that keeps me coming back here as opposed to than any other discussion site.

I can truly say that this site has changed my life. Thank you, Rob and Jeff. And thanks to all the other editors and users that have made this a positive experience and hopefully will continue to do so.

- Zaki

P.S.: Great timing with the 20th anniversary of Linux. Now I have two things to think about as my own 21st birthday approaches. :-)

Yes, it turns out great when the author has thought all that through. Unfortunately, a lot of papers I grab online are missing that.

I am currently using a Kindle 3 with the Duokan firmware and find the PDF readability much better in terms of the interface. It can also handle ePub, HTML, and DjVU. Possibly more formats, but I haven't tried those.

It has a rudimentary column splitting feature that lets you read the common two-column document format easily without having to continually zoom and pan.

I particularly like that it actually uses the filesystem to browse for documents, so I can organize the files my own way.

It also can play back OGG and FLAC in addition to the MP3s you get with the regular firmware.

I've been thinking about the sorry state of document readers for a while and have surmised that there are several places where the current software is lacking.

Content and document layout analysis
Scientific writing is by nature highly structured. So far, I have not come across anything that takes advantage of that and pulls out the semantics of the text. For example, I would love to be able to click on a citation and have that open up either my browser to look it up or grab the document from storage. Other places where the software could improve is in the automatic generation of table of contents if there isn't one or in the recognition of a floating figure/table along with its caption and then allowing the reader to zoom to just that part of the document.

Annotations
As I am reading, I must have the ability to take annotations easily and quickly. These annotations must be exportable and editable on any of my other devices. The annotations must be able to be searchable and cross-referenced with other documents and annotations.

Interoperability
Each e-book reader use different databases to organize their data into categories or mark a document as read. This should really be open. Furthermore, why is it that I have to plug my device into my computer to transfer large numbers of documents? Why can't I use the LAN to control that particular aspect? I want to be able to search for papers on my computer then just sit back and read through a couple of papers. Having to manage files breaks my flow.

Other areas for improvement include organization of large libraries of documents, bookmarking (down to at least a paragraph level), and the ability to view multiple documents (or different parts of the same document) at once.

Frozen Bubble.

So, now that we can correlate what people are talking about to topics, can we get proper threading? The code doesn't work in some overloaded cases such as categorizing the phrase "Big Apple" under "Apple, Inc.".

I'm actually working on that. It would be nice to have a Slashdot API.

On the other hand, Reddit is actually very easy to work with. They have a JSON URL of the comment thread available for every story.

You should try out Maxima. I think it may be one of the backends for Sage, but if you use it directly you can program the symbolic engine to simplify how ever you want.

Darn! I'm 20. And I've got bookshelves of my own (non-fiction) books that I have either bought or been given, most of them are not for uni. I estimate the total page count to be around 30,000. I love reading in dead-tree form because of the feel you can get for each page. For example, after reading through a paper book, I can usually flip to within a few pages of where a specific piece of information is located. However, at the same time, I prefer electronic books because they are greppable and can be viewed on portable electronics. The only issue I have with eBooks is that many of the technical books I want to read are not available digitally, so I usually end up borrowing these from the library.

Of course, I'm on /., so I'm probably not going to be a representative sample of my peer group.

I would like a pointer to where this can be obtained. I try to avoid clichéd phrases, but identifying them can be a pain.

No, they didn't. Text files were always readily available on different networks. It's just that the general public would rather get a dead tree copy than use up paper on printing them out or read them sitting in front the computer screen. What we see now is a less tech-savvy public that would rather pull all their media from central distributors anyway, because they are ignorant of the alternatives. This is why DRM is being thrust upon us without a mass uproar.