Below is live coverage of the Developments in Information Retrieval on the Web from the SES Chicago 2009 (official SES Chicago Site) conference.
This coverage is provided by Brian Ussery - Beu Blog & Marty Weintraub from aimClear.
We are using a live blogging tool to provide the real time coverage, please excuse any typos. You can also interact with us and while we are live blogging, so feel free to ask us questions as we blog. We will publish the archive below after the session is completed.
/* Adjust the height to best suite your page */
/* Remove height property to have all text show without scrollbars. */
#cil_page .mainchat { position: relative; overflow: auto; }
scrollpos[1]) { divtop = document.getElementById('cil_mainholder').scrollTop + 40; } else { divtop = scrollpos[1] - document.getElementById('cil_mainholder').offsetTop + document.getElementById('cil_mainholder').scrollTop + 40; } document.getElementById('cil_modalitem').style.left = divleft + 'px'; document.getElementById('cil_modalitem').style.top = divtop + 'px'; document.getElementById('cil_modalitem').style.display = 'inline'; }
function cil_closeLayer(){ if (document.getElementById('cil_modalitem').style.display != 'none') { document.getElementById('cil_divsrc').innerHTML = ''; document.getElementById('cil_modalitem').style.display = 'none'; } }
function cil_replaceString(oldS, newS, fullS) { for (var i = 0; i 0) { position = [document.documentElement.scrollLeft,document.documentElement.scrollTop]; } else if (typeof document.body.scrollTop != 'undefined') { position = [document.body.scrollLeft,document.body.scrollTop]; } return position; }
function cil_getCurrentStyle(elem, prop) { if( elem.currentStyle ) { var ar = prop.match(/w[^-]*/g); var s = ar[0]; for(var i = 1; i
| | Developments in Information Retrieval on the Web | (12/08/2009) |
| 10:36 | | Brian Ussery (@beussery):
Good morning from SES Chicago!
Developments in Information Retrieval on the Web
Hyperlinked web documents are being described by Internet researchers as the "tip of the iceberg." In this session we will cover the important concepts that are pushing semantic technology such as linked data and RDFa (rich snippets) and will give practical examples as they relate to recent applications. This session will demonstrate how Internet researchers are changing the WWW in a fundamental and powerful way. With Google announcing its support of RDFa and with other large companies pushing support for linked data, it is important for search marketers to be aware of these new developments. This session will showcase practical examples that search marketers and web developers can use to begin participating in the linked data movement.
* Moderator:
Sean Golliher, Founder and Publisher, SEMJ.org
Speakers:
- Jamie Taylor, PhD., Author, "Programming the Semantic Web" and Minister of Information, Freebase.com
- Martin Hepp, Professor, General Management & E-Business, Universität der Bundeswehr München
- Jay Myers, Lead Web Development Engineer, Best Buy Co.
- Nick Cox, Senior Product Manager, Yahoo!
|
Tuesday December 8, 2009 10:36 Brian Ussery (@beussery) |
| 10:43 | | Brian Ussery (@beussery):
Jamie leads off
Jamie starts off with an illustration of how PageRank works through IR eyes. Next illustration looks similar to Google Wonder Wheel. Both illustrations show relationships between topics and pages. Jamie works for Freebase where over 10k apps are available to visualize relationships. Next slide depicts "linked data cloud" but Jamie mentions that some knowledge is involved that these are more than nodes on a graph.
Jamie mentions Jeff Jarvis's "What would Google Do?" and it's "link economy".
Jamie hopes to help labeled links become more a vocabulary of sorts….
For example twitter's @ tags or # tags. These are labels that relate to other content in terms of linked vocabulary. That said, in HTML links aren't annotated this way.
|
Tuesday December 8, 2009 10:43 Brian Ussery (@beussery) |
| 10:46 | | Brian Ussery (@beussery):
Jamie talks about marking up content so it can be easier used for ecommerce.
Next he talks about strong identifiers and how identifiers are used.
|
Tuesday December 8, 2009 10:46 Brian Ussery (@beussery) |
| 10:51 | | Brian Ussery (@beussery):
- Labeled Links
- Strong Identifiers
2 formats
microformats
-easy to use
-small fixed vocab
cons:
no standard parsing pattern
no strong identifiers
- limit to utility
RDFa:
unambiguous identifiers
RDF = Subject Predicate Object } Triple
Shakira, Knows, Penelope Cruz = Triple
HTML5 Microdata
-easy to use
-strong identifiers
-extensible
-easy to parse
Jamie wrote "Semantic Web" OREILLY
Thanks Jamie! :)
|
Tuesday December 8, 2009 10:51 Brian Ussery (@beussery) |
| 10:55 | | Brian Ussery (@beussery):
Up next is Martin from Germany creator of Good Relations vocab which has been adopted by Yahoo Best Buy and others......
Currently little unified view back and forth across data silos. |
Tuesday December 8, 2009 10:55 Brian Ussery (@beussery) |
| 10:59 | Tuesday December 8, 2009 10:59 |
| 11:01 | Tuesday December 8, 2009 11:01 |
| 11:02 | Tuesday December 8, 2009 11:02 |
| 11:03 | | Brian Ussery (@beussery): By linking your data with other data, you increase visibility of your own data. |
Tuesday December 8, 2009 11:03 Brian Ussery (@beussery) |
| 11:08 | Tuesday December 8, 2009 11:08 |
| 11:08 | | Brian Ussery (@beussery): Build links to data, not just documents. Chains of affiliate links are optional |
Tuesday December 8, 2009 11:08 Brian Ussery (@beussery) |
| 11:09 | Tuesday December 8, 2009 11:09 |
| 11:10 | Tuesday December 8, 2009 11:10 |
| 11:10 | Tuesday December 8, 2009 11:10 Brian Ussery (@beussery) |
| 11:11 | Tuesday December 8, 2009 11:11 Brian Ussery (@beussery) |
| 11:13 | | Brian Ussery (@beussery):
Up next is Jay from Best Buy... |
Tuesday December 8, 2009 11:13 Brian Ussery (@beussery) |
| 11:19 | | Brian Ussery (@beussery): ecommerce chanllenges > product data > creates uninformed customers > under utilized product "long tail"
|
Tuesday December 8, 2009 11:19 Brian Ussery (@beussery) |
| 11:22 | | Brian Ussery (@beussery): Best Buy Semantic Product Web URLs indexed via:
[www.google.com] |
Tuesday December 8, 2009 11:22 Brian Ussery (@beussery) |
| 11:23 | | Brian Ussery (@beussery): - human machine readable
- easy html
- easy for engines
- granular |
Tuesday December 8, 2009 11:23 Brian Ussery (@beussery) |
| 11:23 | | Brian Ussery (@beussery): What's next?
Best Buy plans to provide more information into events.... |
Tuesday December 8, 2009 11:23 Brian Ussery (@beussery) |
| 11:24 | | Brian Ussery (@beussery):
In terms of products, explore beta (mentioned previously) |
Tuesday December 8, 2009 11:24 Brian Ussery (@beussery) |
| 11:24 | | Brian Ussery (@beussery): @jaymyers |
Tuesday December 8, 2009 11:24 Brian Ussery (@beussery) |
| 11:25 | | Brian Ussery (@beussery):
Nick is up! |
Tuesday December 8, 2009 11:25 Brian Ussery (@beussery) |
| 11:26 | | Brian Ussery (@beussery): Nick points out time for implementation is short |
Tuesday December 8, 2009 11:26 Brian Ussery (@beussery) |
| 11:27 | | Brian Ussery (@beussery):
Nick points out that search thinks in terms of nouns and not information. |
Tuesday December 8, 2009 11:27 Brian Ussery (@beussery) |
| 11:28 | Tuesday December 8, 2009 11:28 |
| 11:30 | | Brian Ussery (@beussery):
A side result of richer presentation is richer data. Virtual verticals from structured data from publishers. Structure allows intent to be more clear in many cases.
Bing it seems is the other major engine using structured data. |
Tuesday December 8, 2009 11:30 Brian Ussery (@beussery) |
| 11:31 | | Brian Ussery (@beussery):
Google is taking a "much more familiar approach" by using templates on the fly.... |
Tuesday December 8, 2009 11:31 Brian Ussery (@beussery) |
| 11:36 | Tuesday December 8, 2009 11:36 Brian Ussery (@beussery) |
| 11:38 | | Brian Ussery (@beussery): Q&A |
Tuesday December 8, 2009 11:38 Brian Ussery (@beussery) |
cil_setSizes();
