license.
sBooks reinvent the software of the book to make online digital reading better than offline analog reading. sBooks are:
Better than 'just as good.' Most e-book (electronic book) technology aims to create experiences which are 'just as good' as reading an ordinary paper book. sBooks aim to create reading experiences which are more engaging, informative, expansive, and satisfying than even well-designed material books.
Social reading allows readers to make and share notes, tags, or references (collectively called 'glosses') as they read. Glosses can be associated with organized “reading circles”, shared with friends generally (the "personal circle"), or published to “overdocs” made available (for free or at a price) to other readers.
Semantic search uses knowlets — lightweight knowledge bases — to find tagged content within the book. When searching with a knowlet, readers don't need to know the exact language of the author, but can use their own words and learn the author's language as they search. An sBook's knowlet can be used or extended in the glosses created by the reader's friends, communities, or overdocs, allowing collaborative indexing and knowledge creation.
Ambient navigation provides access to the structure of the book while reading online. Physical books naturally disclose aspects of their content and arrangement by their physicality and simple interactions (e.g. judging progress or thumbing pages) readily yield more detailed information. sBooks provide these kinds of cues and interactions in device-appropriate electronic form.
Publishing with the sBooks platform for the Web is as easy as making a few changes to your HTML documents and providing them on your site. sBooks uses the standard HTML organizational markup to make your documents easier to navigate, browse, search, and share. You can read more about converting HTML documents into sBooks in our publishing guide or just try out our automatic conversion tool.
Because of it's simple foundations, sBooks are easily and automatically bundled as e-books for custom e-reader devices and applications. If the e-reader supports sBooks advanced functionality, they just work. Otherwise, automatic tools can flatten a rich and annotated sBook to provide extra context and functionality in legacy readers.
The sBooks platform is an open standards-based platform developed by beingmeta and leveraging beingmeta's knowlets technology, it's BRICO knowledge base.
The web client
for the sBooks platform is licensed free of charge
for non-commercial use under
the GNU
GPL (version 2 or later) and
the Creative
Commons Attribution Share-Alike
license.
Reading an sBook starts out just like reading a regular document in your e-reader and browser. sBooks is based on principles of ambient design, making added functionality as unobtrusive as possible while still being available wherever you are. For example, in web-based sBooks, translucent overlays (HUDs or 'heads-up displays') appear on occasion or demand to provide information about where you are in the book, annotations or references provided by your community or subscriptions, and descriptive tags provided by friends or publishers to describe what you're reading. When you want to search or navigate, these same overlays become active tools for naturally finding the places or content you are seeking.
While reading, you can also add your own comments (called glosses) which will be visible (as you choose) to friends, colleagues, "reading circles," or audiences (if you publish your glosses). Your comments can include brief notes, extended discussion, external references, or "semantic tags" to help yourself or others locate content in the book.
sBooks are all about adding value to content through rich navigation, search, and social interaction. An sBook is composed for four basic pieces:
Analogous to a conventional book, maps correspond to the table of contents, tags correspond to the editorial index, and glosses correspond to marginal notes or (in part) foot notes or end notes which, in a normal book, are only provided by the author or publisher.
Maps organize the book into linearly ordered sections and subsections based on the narrative structure, the author's purposes, and dependencies of explanation or description within the content. Most sBooks have a primary map which is either provided by the author/publisher or extracted from structure (headings, etc) of the book itself.
Tags are descriptive words or phrases assigned to passages within a book. They are used to help browse or search the content and are connected by inference and distinguished by meaning. For example, shepherd:breed would be connected to dog:animal and distinguished from shepherd:person and dog:epithet. Tags and their connections can be provided by content publishers, third parties, or individual communities (through shared glosses). Tagging can be simplified by automatic tools which draw on the structure of language and external knowledge bases.
Glosses are fine-grained annotations on a book, typically at the paragraph level, that include brief notes, descriptive tags, extended commentary, excerpts, or references (internal and external) to other content. Glosses are readily shared between readers and across communities and may also be bundled into overdocs created by third parties to add value to the content.
Overlays are collections of glosses generated by individuals or communities and typically assigned to a particular book. For example, the glosses assigned to a book by a given reader constitute an overlay that may be visible to their friends or colleagues.
Sharing with an sBook happens through glosses made by one individual but visible to others. These glosses can provide any or all of commentary, communication, or utility. As commentary, they may enhance the content itself through explanation, example, correction, or critique. As conversation, they may contribute to a formal or informal discussion around the glossed passage, building personal relationship and mutual learning. As utility, they can provide external links, internal cross references, tags (and tag relations), clarifications, missing context, or possible rephrasings.
Normally, glosses are shared with one's personal circle. These are individuals explicitly invited into relationship or connected through associated social network sites (like Facebook). Membership in one's personal circle is always mutual: if someone is in your personal circle, you are in their. When reading an sBook, glosses added by your personal circle are automatically visible, whether they read the book last month or last year.
Any gloss can be declared private, keeping it from being visible to one's personal circle. It will still be visible to the audience specified for the gloss (which might include members of one's personal circle).
Co-reading is a kind of mutual sharing limited to a particular book. Any reader can invite a friend or colleague (via email) to read a particular book with them. If the invitation is accepted, the two readers will be able to see one another's (non-private) glosses on the book. Co-reading is limited to pairs of individuals; if Maria is co-reading with Ramesh and Jean, Ramesh and Jean won't see one another's glosses.
Reading circles involve more than two individuals in mutual sharing. In a reading circle, any member can contribute a gloss to the circle and those glosses are visible to any members of the circle. Normally, reading circles are associated with a particular book or collection of books. Anyone can create a reading circle and starting inviting people to join it.
Membership in a reading circle is determined in a variety of ways. Many reading circles are open: anyone can join the circle. Some reading circles are viral: anyone in the circle can invite new members. Derivative reading circles are based on existing social groupings, such as Facebook groups or networks. Finally, reading circles can be gated: new members need to be invited by particular individuals or authorized by some external process.
Independently of how membership is determined, joining a reading circle may also involve payment of some kind, either a one-time charge or an ongoing subscription.
Reading tools for sBooks usually provide a way to browse, join, or leave reading circles for a particular book. In addition, some sBook catalogs display available reading circles along with every title and description.
Another kind of sharing is meta publishing: the creation of an overlay (usually around a particular book) by one individual or community. An overlay created in this way is called an overdoc and can originate from the original authors or publishers or (more commonly) a third party. Overdocs can provide readers' guides, supporting or contrary critiques, humorous asides, internal cross-references, external references to related material, or additional media (images, movies, sound) connected to the content.
Overdocs can also include tags, making the book more searchable or easier to browse. Because sBook tags are concept-based, overdocs can also provide organized knowledge which can further help in searching the book or understanding its content.
Dynamic overdocs are manually or automatically updated on a regular basis. A dynamic overdoc on a psychology textbook, for example, might dynamically gloss a particular section (say on schizophrenia) with pointers to news reports or journal articles relating to the topic. Or a dynamic overdoc on a patent document might include references to court cases which reference particular claims or language.
Overdocs can have one or many authors and glosses are identified by both the overdoc and the individual adding the gloss. As with reading circles, subscriptions to an overdoc can be limited. Open overdocs are available to anyone. Viral overdocs require invitation from a current subscriber. Derivative overdocs are based on membership in an existing group, such as being a fan of a Facebook "page". Gated overdocs, like gated reading circles, require invitation from a contributor or authorization by some external process.
Overdocs can also be priced, either a one-time fee or an ongoing subscription.
Every sBook has an embedded knowledge base describing the content of the book and the discussions around it. The terms in this knowledge base are used as tags associated with particular sections or passages within the book. In its simplest application, this knowledge base replaces the printed index of a conventional book. In more advanced applications, the knowledge base supports intelligent search and might support advanced functions like question answering or summarization.
An sBook knowledge base is a knowlet where individual terms are distinguished by meaning while connected by inference.
Terms in the knowledge base represent distinct meanings of particular natural language words or phrases (like 'user' or 'law') describing the book's content. Every meaning has a uniquely assigned term which is associated with multiple and overlapping synonyms. Search uses these synonyms (and more) to find the underlying meanings and uses these to find passages in the book. In most cases the unique terms are just natural language words or phrases, because those words or phrases are unambiguous in the context of the book.
A term regulation might be part of a book's knowledge base and associated with the synonyms rule or law or (more informally) reg. Searches using the synonyms will find items tagged with regulation and if the search term is ambiguous (like rule) present opportunities for disambiguation.
These distinct meanings are connected by inference, so that a book's knowledge base might capture the fact that both federal regulations and building codes are kinds of regulations or that a ballad is a kind of poem (for some books). Searches on a more general term will find more particular occurrences, meaning that authors or commentators can use very precise tags without worrying about complicating search. Other relationships (besides generalization) may also be used by search.
Tags and the meta-knowledge can either come from the author or publisher (with the help of automatic tools) or from the overlays applied by a particular reader. sBooks merges together these various sources of information while retaining their original source; this allows the search engine to handle the case where (for example) two commentators use the same tag in different ways.
Tags and knowlets help with search but they also serve two other important roles. First, they constitute a kind of cognitive map of the books' content, describing what it is about (and not about). Second, they reveal aspects of the author's language and thought which are often unspoken or emergent. For example, Darwin's Origin Of Species often makes arguments about evolution from the similarity of distinct species, but Darwin never uses the word similarity. Instead, he uses the word affinity, which was more common at the time. As shown here in the web-based sBook reader, searching the sBook of for similarity finds the corresponding concept and teaches us something about how Darwin used language at the time.
In order to present these ideas in action, we'll take a look at the web-based sBooks reader. Many of the functions and elements are similar in other reader applications.
In the web-based sBooks
interface, users access advanced functions through translucent
icons and overlays (HUDs or "Heads Up Displays") which are
overlaid on the text. The interface is designed to disclose
opportunities for interaction without interrupting the natural
flow of reading and understanding.
When simply reading the text, the HUD
simply appears as translucent icons or regions overlaid on the
margins. Icons on the top activate navigation
(
on the left) or search (on the right
). Small photos or icons in the right
margin indicate glosses added to particular passages.
Selecting (tapping/clicking) the image displays the associated
glosses. The column of icons in the left margin
indicate the sources of overlays applied this book; selecting an icon
displays the overlaid comments from that source.
In general, clicking on an icon brings up a full-sized HUD; clicking the icon again hides it. The standard interface has four major HUDs:
To add your own gloss, just tap or click a passage twice. The
first click targets the passage, coloring it and
displaying its context; the second click brings up
the gloss HUD. (You can also just double click
to go straight to the HUD.) Any text you have selected becomes
the excerpt associated with your gloss. The Gloss HUD
has your picture on the left with some icons underneath it. It
should look something like this:
The first text box (What do you think?) lets you add a brief note or comment. Typing Enter in this box automatically saves the gloss. A second text box, when it is visible, lets you add tags or type the names of groups or friends you think might be interested in the gloss.
Underneath the first text box are controls you can use to specify the exposure of your comment. A drop-down menu specifies the audience for your gloss, which can be just your friends (personal circle), any reading circles you've joined, or any overdocs you are publishing. Right after the drop-down menu, a checkbox declares whether the gloss should be considered 'private' or shared with your friends. If you check this box and specify an overlay at the same time, anyone applying the overlay will see the gloss, but your friends (unless they are applying the overlay) will not.
The two buttons on the far right, "Post" and "Save", both save the current gloss. "Post" also posts your gloss (or a summary) to any external feeds (e.g. Twitter, your Facebook wall, etc) that you've configured.
The icons arrayed beneath your picture allow you to add extended
content. Selecting these icons reveals a display which allows you
to add tags
(
),
extended text
(
),
or external references
(
).
To add external references, simply type in URLs and hit Enter.
When adding extended text (details), just type. In either case,
you need to save to actually record your additions.
The tagging interface helps you to choose tags in two ways:
The presented cues are based on tags already associated with the passage, tags you've used in the past, and other potential associations. The tags also include section titles from the book (with a white background) and the names of friends who have added comments of their own. (Tagging with a friend is invisible to everyone else, but brings your comment to their particular attention). Typing in the text field selects only those cues which match whatever you're typing:
Tags are semantic: they can be distinguished by meaning and connected by inference. As you type, the completions will show tags already applied to the document, including synonyms. For example, if you're tagging a passage with the term 'tiger':
you see the completion tiger= tyger indicating that the book uses the tag tyger with the synonym tiger (Blake, the author, spelled 'tiger' with a 'y'.) This knowledge is embedded in the book's knowlet, a small knowledge base describing the content and commentary on the book.
When you add a tag, you can extend the knowlet by specifying synonyms or related tags together with the tag itself. For example, you can add your own tyger tag carrying additional information, such as tyger|tiger|^wild animal|-lion|_bengal tiger indicating that the tiger is a wild animal which is not a lion and that one kind of tiger is the bengal tiger.
Tag inference uses relationships among tags to help connect searches to content. For example, individuals searching for references to wild animals in Blake can find them even though Blake never uses that phrase. The tag inference for an sBook is done in the sBook application itself, so there is no need to contact external services or databases.
Navigation uses a hierarchical map of the book, usually
corresponding to a table of contents, to find or browse content.
This is how the author or publisher organized the book and can
be a way to find things quickly as well as understanding it's
coverage, intent, or biases.
In the web-based interface, the navigation HUD displays a stack of horizontal stripes which reflect levels of organization in the content map.
Each stripe includes the name of a section and a segmented bar indicating its subsections. The background of the section title is a progress bar indicates your position within the section. In the segmented bar describing it's subsections, the white segment indicates the current section and a red tick indicate the current subsection within the subsection.
As you move the pointer over (or touch) elements of the map, corresponding parts of the book will be previewed. Moving away from an element previews different content or returns to the current location. Selecting the element jumps to the previewed content. The preview function allows you to 'flip through' the book by simply moving the pointer over the displayed contents.
Search in sBooks uses semantic tags which are based on interconnected 'concepts' rather than keywords. As with adding tags, this means that you can search on a synonym and find the matching tag, as in:
where our search on 'tiger' reveals the tag 'tyger'. It also means that we can search on a more general term, like 'wild animal' which will find passages about tigers but also passages about other wild animals like wolves.
As you type in the text field of the search HUD, matching tags are automatically displayed: clicking on a tag selects it. Tags in the text field are separated by semicolons — typing a semicolon when there are multiple completions selects the first (and most common) tag in the list. If you are typing a synonym of a tag, you'll see both the synonym and the tag: the tag appears in yellow and the synonym in white. For example, searching on 'baby' brings up the tag for infant:
When a new tag is added to the search, the completions (called the search cloud) are updated to reflect tags associated (in the book) with the results of your current query. In the upper right of the search cloud, the number of matching results and refining tags is displayed. The search on infant (baby) brings up tags associated with babies in the text:
As before, typing displays matching tags and clicking selects an individual tag to add to the search. In the upper right of the search cloud is a report on the number or associated tags (called 'refining tags' because they refine your search) and the number of results. Clicking on that display or hitting Enter brings you to the search results. For instance, after clicking on the weep tag (which includes many synonyms) and hitting Enter we find:
You can return to the search could by clicking the text box and
the 
to the right of the text box clears the current
search.
The results of the search are based on both tags from the
publisher and tags from any glosses merged from your community or
subscriptions. The glosses appear with a picture of the source
for the gloss (usually a picture or avatar of the author) and the
short text and tags it assigns; icons will display more extended
text
()
or cross references
()
when available. Holding down the mouse button or control key while
over a gloss previews the glossed text in context.
sBooks is all about adding value to content either by creators, commentators, critics (constructive and otherwise), readers, and communities. sBooks provides some standard ways to share added value.
As we've shown, it's simple to add your own gloss to a book you are reading. This gloss can include a brief note (e.g. 'cool idea'), an extended comment (e.g. a supporting or denying argument), or an external reference (e.g. to an illustrative graphic, a different explanation, a funny perspective, or related content). Glosses can also include tags (e.g. 'dachshund' or 'hot dog') and possible synonyms or related tags (e.g. 'hot dog|frankfurter|wiener' or 'hot dog|^sausage' .vs. 'hot dog|^race'). You should personally try to use tags unambiguously in any single book; the display of matching tags (when you add a gloss) should help with that.
Normally, when you add a gloss, it will be visible to any sBook readers in your "personal circle". This consists of friends you've explicitly invited into your personal circle or "imported" from social networking sites like Facebook. Having a friend in your personal circle is always mutual: if they're in your personal circle, you are in theirs. As we describe below, can also invite someone to share glosses (again, mutually) on a particular book without inviting them into your personal circle. These kinds of personal sharing can be disabled by marking a gloss private.
In addition, a gloss can be associated with a tribe, which makes it visible to everyone who is a member of that tribe and reading the book. Tribes can be reading circles (where anyone can contribute) or overdocs (where only particular individuals can add glosses). Membership in tribes can be open (anyone can join), viral (requiring invitation from a current member), paid (requiring a fee), authorized (by some external process), or some combination of the above.
In the standard web interface, social functions are accessed through
the "Connect" tab of the App HUD. In the web interface, the App
HUD is activated by either the "A" key or by clicking the App
Icon in the left margin. The Connect tab allows you to
change the overlays you see (by joining or leaving tribes),
create new tribes, invite friends or colleagues to engage in
various ways, or update your personal information.
You can join or leave a tribe (and its overlay on the current book) by selecting (clicking/tapping/etc) the button next to the description of the tribe. Selecting the other buttons do pretty much what they say, including creating new tribes (as reading circles or overdocs), or updating personal information and site affiliations.
A reading circle is a group of individuals who are discussing and
sharing comments on a book. Anyone in the circle can contribute a
gloss and anyone can see the glosses contributed by everyone else.
To create a new reading circle for a book, you need to specify a name and a description. You will automatically be made a member of the reading circle that you create. You can also declare the reading circle to be secluded, so that added glosses will be private by default, though their author can always choose to expose them to their personal circle.
When you create a reading circle, you can also place restrictions on how members can join. This can include requiring invitation (by a current member), requiring payment (not yet implemented), or requiring that the reader be validated by an external site (using OpenID).
An overdoc is a set of glosses on a book created by a group of
individuals (and possibly a publisher). Readers can subscribe to an
overdoc to have its glosses added to their version of the book.
To create a new overdoc for a book, you need to specify a name and a description. You will automatically be made a reader and contributor for that overdoc. As a contributor, you can invite anyone else to also be a contributor.
When you create an overdoc, you can place restrictions on how readers can subscribe. This can include requiring invitation (by a current member), requiring payment (not yet implemented), or requiring that the reader be validated by an external site (using OpenID).
The Connect tab allows you to use email to invite friends or colleagues to interact with you using sBooks.
Just type email addresses in the invitation field and use the drop-down menu to invite them to:
When you select the "Invite" button, your mail program will open with a message that includes a URL. When your friends or colleagues receive the message, they can follow the URL to identify themselves to sBooks (if they're not already logged in) and accept the invitation. That's all it takes.
Invitations expire after a week for security reasons.