Jun 2020 | TeamPage uses Solr for faceted navigation and search
Traction® Software is an early believer in the value of scalable, permission aware, faceted navigation and search spanning all Traction® TeamPage content. Traction Software partnered with FAST Search starting in 2006, transitioning to Attivio in 2009, adding new search and navigation capabilities along the way. Traction Software now includes Apache Solr faceted navigation and search as a free capability for Cloud hosted and on premises TeamPage subscriptions. Learn how Solr supports TeamPage's unique work graph model for indexing, visualizing, and using work that crosses boundaries, linking customers, suppliers, partners and internal teams with different permissions to more public or more private business activities on the same TeamPage server.
Solr automatically clips TeamPage navigation links and search results to match permissions of each object versus the access control list permissions of the user making the request. This makes it simple for TeamPage to support many separately permissioned spaces on the same TeamPage instance. For example a law firm could define several public spaces; confidential spaces for internal groups, each with its own access list; and a unique permissioned space for each client. Clients could see and search their own space - plus any spaces the firm shares with all clients for news and support. Members of the firm could use and search across all public, internal, and client spaces they have permission to see. TeamPage permissioned search makes it simple, secure, and efficient to work across spaces that make business sense.
TeamPage Solr search includes the core permissioned search, faceted navigation, and on the fly indexing pioneered with FAST, along with the Social Enterprise Web and other capabilities introduced with Attivio. TeamPage now leverages all of this experience with a transition to the popular, blazing-fast, open source enterprise search Solr engine built on Apache Lucene.
TeamPage Faceted navigation and search (FAST 2006)
By embedding FAST InStream, TeamPage now delivers state of the art search and content drill-down that is as good or better than that of the most capable content and document management systems on the market today and
- Extends the TeamPage standard search capability by adding permission- filtered search across over 370 formats of documents attached to Traction TeamPage blog posts or stored in TeamPage web folders.
- Provides interactive drill-down navigation, including AND and OR selection, enabling users to quickly understand their search set and narrow tens or hundreds of thousands of results down to just a few that are highly relevant using keywords, entities, and metadata attributes.
- Adds a "Find Similar" choice to posts, comments and search results to make it simple to find other relevant content.
- Adds automatically calculated Rank order to Newest First and Oldest First sort options. Rank order is determined by automatic linguistic and link analysis.
- Adds search by content types including posts, pages, comments, emails and files.
"In the era of Enterprise 2.0, smart companies are leveraging tools like Traction TeamPage to empower individuals to create their own content," said John M. Lervik, CEO, FAST. "Now, by leveraging Traction TeamPage together with FAST InStream businesses can offer their employees a new and more powerful package of tools to help them produce, share and then intelligently search this collective intelligence, creating a smarter, faster-moving and more profitable workforce."
TeamPage Social Enterprise Web Features (Attivio 2011)
TeamPage Social Enterprise Web features enable TeamPage to share, tag, comment, task, index and search linked public Web or company intranet pages. TeamPage Web browser extensions, inline badges and comment widgets link external Web sources to TeamPage discussion and provide visual feedback (similar to the way Disqus adds shared comments to pages on the public Web). The content of the linked external page is automatically indexed and by included in Attivio search results, making it simple to find content referenced by a TeamPage external link.
- Use TeamPage to discuss, tag, task or share internal or external sources using search, TeamPage Badges or bookmarking.
- Indexing TeamPage tags, discussion and tasks associated with external content improves relevance ranking for search by adding social business context.
- New TeamPage Web browser bookmarklet can be used to discuss, tag, task, share and index selected pages on the public Web or a private intranet.
- New TeamPage Badges can be added to any Web page to provide a one-click action to discuss, tag, task or share that page.
- TeamPage Badges shows a live count of the number of social actions associated with that page, with one click to open and see the actions in context.
- TeamPage Badge counts and navigation are permission aware – including only what you have permission to see based on TeamPage access control rules.
- TeamPage search results and activity streams include avatars to show who has discussed, tagged, tasked or shared that internal or external content.
- One click expands any search result or activity item inline to quickly review details without leaving the page.
- One click opens external documents or records in a Web browser or native client (depending on the source).
- Metadata shown for each source type, styling and behavior of source-specific detail shown with search results and in activity streams can be easily customized using the Traction's Skin Definition Language plug-in for that source.
TeamPage Solr - Blazing-fast, open source enterprise search platform
- Advanced configurable text analysis - Support for most of the widely spoken languages in the world (English, Chinese, Japanese, German, French and many more)
- Rich document parsing - Apache Tika built-in, making it easy to index rich content such as Adobe PDF, Microsoft Word and more.
- Near real-time indexing - as text and documents are posted to TeamPage, Solr queues and indexes the content
- Extensible plug-in architecture - works as a pure TeamPage plug-in
- Built in security - supports clipping search results to match TeamPage ACL rules of the person making the search request
- High performance query optimization with configurable caching
- Comprehensive administration interfaces
See Apache Solr
Type a few words, quoted phrase, or search expression. TeamPage shows permission filtered results below the search box and Facets in the right side bar. Facets show values search result attributes such as: Space, Tags, object Type, Author, Editor, Status, along with Task or Project attributes such as assignee and due date. Click a Facet name such as Space to see a pop-up card showing the Spaces containing one or more search results. Click any Space name to drill down into that Space, or click the + button to add that space to the current Filter.
The Tag facet shows a list of Tags attached to one or more search results, along with the number of matches for each Tag. Like other Facet cards, Tags can be ordered by name or tag count. Click any Tag name to drill down, or click the + button to add that tag to the current Filter.
The Type facet shows the type of object containing one or more search hits. Types include TeamPage articles, comments, status updates, file attachments, shared files, logged email messages, or proxy. A proxy represents a link to an external Web page or file that has been added to TeamPage by reference using the Social Enterprise Web browser extension. Solr parses and indexes the content of the external Web page so that you can get search result based on the content of that page without having to copy and paste it into TeamPage.
Click the down arrow to the right of any search result to expand the search result inline - no need to open the result in a separate window. Click the up arrow to close the expanded result and continue scanning down the list of search results. You can set your own Search Preferences to change the default Search Order - choose Newest First, Oldest First, or By Rank (relevance).
The Work Graph Model: TeamPage style Understand how TeamPage connects people and their work. When Mr. Dithers shouts: "Bumstead! Where are we on the Acme Account?", the most timely, frequently discussed and contextually relevant version of Dagwood's Acme tasks, projects and work should pop up near the top of the result list, along with the cloud of tags and people who have touched or talked about tasks, projects and other related to the Acme account and its associated activity streams.
The important requirement is making tasks, projects, pages, discussions and other work products first class sharable, named objects that can be connected to each other and what you're working on, discussed, tagged, tasked, and navigated as well as found using search. Being able to talk about tasks and projects relating to Acme captures one important part of your interest and behavior graph (activity stream), and links these items to the names and behavior of other people working with or discussing the same objects.
Borders, Spaces, and Places In his Fast Forward 08 Keynote What's Most Important for Success with Enterprise 2.0? Prof Andrew McAfee said that borders are needed in order to use Enterprise 2.0 principals for many valuable business purposes, but it's very important that "Borders seem appropriate to users.
I agree and suggest adding a follow-on principal: "Borders should seem transparent to those with permission to cross them."
For Traction TeamPage this means:
- All content and relevant context are indexed for search, but the search engine delivers the subset of results that the person making the request can read.
- Tag clouds and drill-down navigation present the tags and drill-down paths derived from what that person can read.
- Traction's Dashboard views roll-up content from many spaces based on tag, content or other criteria defined by Section widgets, and automatically shows the subset of content which that person can read.
- RSS feeds, email, IM notifications and cross-reference lists automatically reflect the content and cross references which that person can read.
For example, if you're an employee at the center of a "hub and spoke" collaboration pattern, when you navigate, search or link information, the borders separating different customer spaces and internal spaces you can read become transparent for collaboration.
You can still use the names of spaces created for particular customers to focus your attention on a particular issue or collection of content, but you effectively see one big wiki / weblog. Borders help you visualize the business context and intended audience.
If a customer logs in to your TeamPage server, they see only the rolled-up content, search results, tags, feeds, and space names that they have permission to read. The content - and existence - of private collaboration spaces of other customers or reserved for internal use are hidden.
Traction TeamPage even extends commenting and inline discussion to work transparently across borders.
Just like a good architect knows how the to use the affordances and relationships of physical spaces to help cue behavior, architects of social software should aim to use software affordances to make socializing in the neighborhood, workplace, and commons as natural as possible. I think this will require cues to signal and differentiate as well as connect places. The goal should be to help people read context and act comfortably in different places whose norms they can quickly learn, understand and trust. As Harrison and Dourish write:
"A conference hall and a theatre share many similar spatial features (such as lighting and orientation); and yet we rarely sing or dance when presenting conference papers, and to do so would be regarded as at least slightly odd (or would need to be explained). We wouldn't describe this behaviour as 'out of space'; but it would most certainly be 'out of place' and this feeling is so strong that we might try quite hard to interpret a song or a dance as part of a presentation, if faced with it suddenly. It is a sense of place, not space, which makes it appropriate to dance at a Grateful Dead concert, but not at a Cambridge college high table; to be naked in the bedroom, but not in the street; and to sit at our windows, peering out, rather than at other people's windows, peering in. Place, not space, frames appropriate behaviour." - Re-Place-ing Space: The Roles of Place and Space in Collaborative Systems
Information Foraging at FASTForward '07 Blog and wiki spaces become what Information Foraging researchers like Peter Pirolli and Stuart Card call enriched "information patches". These patches (and the resources they link to) are likely to become rich sources of highly contextualized information because they represent the work product of people engaged in a business process that provides a natural context for guided search.
I showed a few slides introducing Information Foraging theory, then a few screenshot examples using FAST's guided search to navigate Traction content by space (project), label, and automatically recognized keyword or entity (person name, company name, location).
You have a much better chance finding a contextually relevant "Smithers contract" in a client project blog/wiki space where you can easily explore content hits with an "Urgent" label and a pileup of recent comments on your own. Or at least find a space where you can post your own question and expect a highly relevant response.
To download my slides click Search Meets Blogs and Wikis (4.7MB .ppt)
Greg Lloyd, FASTForward '07, Feb 8, 2007 San Diego.
See the book: Information Foraging Theory, Peter Pirolli, Oxford Books Jan 2007.
Greg Lloyd on "Enterprise 2.0 for Intelligence Analysts", FASTForward 08 It’s easy to store and find documents in folders if you know the structure by heart. FAST even makes it easy to search the content of your files, Web pages, and messages and navigate by content. What’s difficult to capture, search, and navigate is the context that makes content relevant for collaboration in a link-poor environment that exists in the typical enterprise. Enterprise 2.0 technology – specifically blogs, wikis, and social tagging – provide immediate value for collaborative work by creating an evolving record that connects external intelligence, internal dialog, and work product. This record can make enterprise search work and scale like the Web. Part 1 (Greg Lloyd, Traction Software) of this session examines how intelligence analysts in business and government are using Enterprise 2.0 tools to help manage this deep, broad, and challenging form of collaboration. Click here for a copy of the slides 7MB .ppt
Traction Software Brings FAST-Enabled Search to Blogs and Wikis Traction Software Brings FAST-Enabled Search to Blogs and Wikis Traction Goes Beyond Blogs and Wikis, Offering Integrated FAST InStream Search Technology to Enable Secure Search, Entity Extraction and Drill Down Navigation PROVIDENCE, RI -- (MARKET WIRE) -- December 18, 2006 -- Traction® Software, the leading developer of products for secure, scalable, web-based collaboration, today announced the general availability of its Traction TeamPage™ FAST Search Module. The Module is an integrated and easily installed option, based on Fast Search & Transfer's FAST InStream™ OEM search technology solution. The Module is priced at USD $15K per Traction TeamPage Server.
Traction Software Selects Attivio to Power Information Access for Enterprise 2.0 Social Software NEWTON, MASS. – October 14, 2009 -- Attivio Active Intelligence Engine’s Permissioning Model and Real-Time Updates Prove Key Differentiators in Replacing Leading Search Solution at Traction Attivio, Inc. today announced that Traction Software, the leading developer of Enterprise 2.0 social software, has chosen the Attivio Active Intelligence Engine ™ (AIE) to replace its legacy enterprise search solution for easier indexing, retrieval of content and an enhanced user experience. As part of the selection process, Traction evaluated several market-leading solutions before choosing Attivio’s unified information access engine for its single, flexible API, full Java support as well as its sophisticated and secure permissioning model.