ThoughtMesh is an unusual model for publishing and discovering scholarly papers online. It gives readers a tag-based navigation system that uses keywords to connect excerpts of essays published on different Web sites.
View a demo of how to use tags to navigate in a single essay and across the mesh (opens in a new window).
View a demo of how to add your own writing to the mesh (opens in a new window).
ThoughtMesh is an unusual model for publishing and discovering scholarly papers online. It gives readers a tag-based navigation system that uses keywords to connect excerpts of essays published on different Web sites.
Add your essay to the mesh, and ThoughtMesh gives you a traditional navigation menu plus a tag cloud that enables nonlinear access to text excerpts. You can navigate across excerpts both within the original essay and from related essays distributed across the mesh. More...
So let's say you are reading an essay on Modern art. You can pick a single word out of that essay's tag cloud- -say Picasso- -and view a list of all the sections from that essay that relate to Picasso. Or you can view a list of sections of other articles tagged with Picasso, and jump right to one of those sections. You can also combine tags to narrow your search, such as Picasso + Cubism + 1900.
As an author, you can choose to post your essay in a central repository hosted by the Vectors program at USC, the sponsor of this project. Or you can self-archive your essay on your own Web site. (That's the "distributed publication" part.)
Firefox 2+ on Linux, OSX, or Windows
Internet Explorer 7+ on Windows should function correctly.
Safari users will be able to use most of the features, but Safari's JavaScript engine does not support search-as-you-type or remote scripting for outside lexias.
Your essay should be divided into sections of (optimally) 2-5 paragraphs apiece, with headings.
Subsections and subheadings are fine--ThoughtMesh's navigation menu can accommodate up to two levels of headings. You can have a top-level heading without text, and you can have one without any subsections.
This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section.
This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section.
This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section.
This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section.
This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section.
This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section.
This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section...
This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section. This is the text for the first section.
This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section.
This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section.
This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section. This is the text for the second section.
This is the text for the third section. This is the text for the third section. This is the text for the third section. This is the text for the third section.
This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section.
This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section. This is the text for a sub-section...
You can also type an essay directly into ThoughtMesh, but this is not recommended unless you are an ace writer or just editing.
ThoughtMesh offers limited support for hyperlinks, italics, and other HTML markup.
For now, ThoughtMesh does not support essay illustrations, but we hope to support them in a future release.
We are still working out the maximum size for a ThoughtMesh essay--for now, try whatever you want.
You will be limited to 2-3 tags per lexia--ThoughtMesh will tell you when you have too many.


A bunch of keywords in a box. Click on one to see text excerpts related to that theme, or click on several to see excerpts tagged with all of those keywords.
A text excerpt from a longer essay or Web site--usually a couple of paragraphs.
Most of these sites are data-base driven collections of text blocks run off a single server. ThoughtMesh's tag registry (or mesh) can connect articles on different servers across the Internet.
Yes and no. Like the long-term vision of the Semantic Web, ThoughtMesh treats every page on the Web as a potential "database record" to be searched. Unlike the conventional XML-powered vision of the Semantic Web, however, ThoughtMesh's data are only minimally structured in the page itself; instead, a registry of tags housed on a remote host serves to connect all the individual pages. But it's still a model of distributed publication, since in principle the same pages can be navigated via independently operated registries.
Sort of. Del.icio.us's global folksonomy of tags is great, but it only indexes entire pages, which is less efficient for finding relevant passages in long academic papers. ThoughtMesh helps trace thematic connections between particular sections of online essays. And ThoughtMesh's tags (and the meshes that connect them) are determined (or at least validated) by the authors of the pages.
ThoughtMesh exploits participatory media, remote scripting, and lateral navigation. So yeah, you can call it that.
Check out the demos at left, or contact Jon Ippolito at ude.eniam.timu@erutluc.loop.
Conceptual architect, client-side designer, and client-side engineer.
Designer and server-side engineer.
Telamon.js author and remote scripting contributor.
ThoughtMesh uses Mehta's Tagline software.
To learn more or get help using ThoughtMesh, contact Jon Ippolito at ude.eniam.timu@erutluc.loop.