Protecting distributed content in HTML

[aschechtman]

I've been developing, publishing and selling e-books for iSilo for a few years now. These e-books are written in HTML and are compiled with iSiloX for iSilo's PDA readers.

I'm beginning to get inquiries about users who want to put these e-books on their company servers or to access the e-books content via the web rather than on PDA. I imagine the same situation would apply to other publishers who use iSilo.

My question is this. By what mechanisms can one distribute HTML content on a server or on the web and protect it at the same time? For instance, if I put an e-book on the web or on someone's server, what's to prevent a single user from compiling it themselves with iSiloX and then distributing it freely without any expiration dates, copy protection, etc? Are there features that can be built into a web page that makes it impossible to compile with iSiloX to protect it? Any other ideas?

Andrew

[Voltage Spike]

No. You cannot protect web pages. If you want to see someone else's attempt, look into Oreilly's Safari collection. So far, they have simply been able to slow people down, not stop them.

You could either require people to install iSilo for Windows or switch to another format.



Perhaps you could simply offer your content in an open format? Some companies (such as Baen Publishing (http://www.baen.com/)) have successfully gone this route with great success. If you make things easy for the customer (e.g., allowing them to convert the document to the format of their choice) and offer excellent service, then shouldn't your customers be willing to pay for your product?

Of course, not everyone has the same faith in humanity as I do. :)

[aschechtman]

Thanks for the reply. I like the philosophy and approach of Baen Publishing but don't think the concept translates too well to electronic-only products. They postulate that making a book available for free in electronic format will drive sales of the paper version because people who want to sit down and read something still prefer a hardcopy.

But my work is not "read" per se, it is referenced. It's value is in its hyperlinked organizational structure. A paper version (without hyperlinks) would be useless.

A fiction author can take solace in knowing that no one is going to take their free e-book, print it and try to resell it - cannibalizing their print sales. But with an electronic only product like mine, there's no barrier (other than integrity) to prevent someone from making and distributing unlimited copies.

While I have faith in humanity, I think that "doing the right thing" is reinforced when barriers to theft are present. It has been said that most locks won't stop a thief but are there to keep honest people honest.

Is there a way to include certain components in a webpage that will confuse or block compilation with iSiloX? For example, one could include a lot of graphic components that would make a compiled file too large or unwieldy but to get around this an iSiloX user could just disable inclusion of graphics. I'd imagine there are other means one could use to put up content that couldn't be easily compiled with iSiloX.

Andrew

[iSilo]

One possible way to go about it that will prevent the average person from converting your HTML content directly is to have the server require logging in and having the connection be secure (e.g., https). iSiloX does not have the ability to access content via https.

[aschechtman]

Just had a new idea about a way to protect HTML content from iSiloX compilation that publishers could use to make available an online version of their content for iSilo without giving one and all the opportunity to compile their own versions. I got the idea when seeing how some web developers have put images of the text of their e-mail addresses on their sites instead of the text itself with the idea of preventing spiders from putting them on spam lists.

Possibility #1: The publisher could substitute in place of one or more key pages an image with mapped links instead of text with hyperlinks (ie. in place of a table of contents). Since I believe iSiloX cannot process images where specified coordinates on the image link to different URL's this page would probably result in a dead end if one tried to compile it.

Of course, perhaps at some point iSiloX will be able to include image maps and that would invalidate this strategy.

Possibility #2: The publisher could insert links to very large pages containing garbage text that would be compiled by iSiloX making the overall file unwieldy and large and searches of that file unbearably slow.

Any thoughts?

Andrew Schechtman, M.D.
MeisterMed

[Voltage Spike]

Possibility #2: The publisher could insert links to very large pages containing garbage text that would be compiled by iSiloX making the overall file unwieldy and large and searches of that file unbearably slow.

Note that iSiloX allows for regular expressions when determining which links to follow. Failing that, the user could manually include each individual page.

Good ideas, but you are simply creating hurdles rather than walls.

[aschechtman]

What do you mean by "regular expressions?" I'm not familiar with that term.

I think sufficient hurdles often suffice to keep honest people honest. For instance, manually adding each page may be an option but how many people will bother to add each of the 2000+ pages in some complicated HTML docs?

Andrew

Note that iSiloX allows for regular expressions when determining which links to follow. Failing that, the user could manually include each individual page.

Good ideas, but you are simply creating hurdles rather than walls.

[Voltage Spike]

What do you mean by "regular expressions?" I'm not familiar with that term.

Read the manual under URL Filters (http://isilox.com/support/manual/iSiloX/ReferenceManual/DocumentProperties/Links.htm). For example, if your bad link goes to garbage.html, I can specify that iSiloX not follow links that match that name.

I think sufficient hurdles often suffice to keep honest people honest.

The definition of "honest" varies from person to person, however. :)

I can't imagine a scenario where I have the right to view a web page with one web browser (e.g., Mozilla) but not another (e.g., iSilo). Of course, I have no problem downloading unauthorized electronic copies of books for which I own the "dead tree" version.

[aschechtman]

Read the manual under URL Filters (http://isilox.com/support/manual/iSiloX/ReferenceManual/DocumentProperties/Links.htm). For example, if your bad link goes to garbage.html, I can specify that iSiloX not follow links that match that name.
Thanks for the clarification. I suppose it wouldn't be too difficult to identify the names of the URL's that were garbage because iSiloX would hang for a while compiling their content. But if there were many of them scattered throughout a large document, it would be a real pain to pick them out. I don't see whether there's the ability in iSiloX to filter out URL's linking to pages that exceed a certain size. If that were possible, then this would be a way to exclude garbage files en masse. But if they're not in a particular directory or named in a similar way, the URL filters wouldn't be much help unless the garbage files were picked out one by one.


I can't imagine a scenario where I have the right to view a web page with one web browser (e.g., Mozilla) but not another (e.g., iSilo).

I agree that there's no reason to bar a user from viewing content on the web and on a PDA. But here's the situation. Let's say I'm selling an e-book for iSilo to a user but I want to make the same e-book content available to that user online as well. With iSilo I can protect the e-book content using a password or registration code. With HTML on the web, unless I configure a secure server, the content is out there for anyone with iSiloX to compile, use, or even distribute without any barriers if they know the URL of the HTML content. So, the question isn't about stopping you from using an electronic copy of a text you've already purchased in paper, it's more a question of making sure it's not easy for you - well, not you, someone with malicious intent - to take that electronic copy and mass produce it for friends, family, and beyond.

Andrew