From OntologPSMW

Jump to: navigation, search
[ ]


Ontolog Invited Speaker Presentation - Adam Cheyer & Tom Gruber on "Siri" - Thu 2010.02.25     (1)

  • Title: "Siri: An Ontology-driven Application for the Masses"     (1B)

Note: to allow us to share the latest in commercial deployment of ontology-based technology, this invited talk will be featured under a special waiver to Siri on the Ontolog IPR Policy.     (1D)

Agenda & Proceedings     (1E)

  • Session Format and Agenda:     (1E1)
    • this will be virtual session over a phone conference setting, augmented by shared computer screen support     (1E1A)
    1. The session will start with a brief self-introduction of the attendees (~15 min.) [We will be skipping this if there are more than 20 participants.]     (1E2A)
    2. Introduction of the invited speakers     (1E2B)
    3. Presentation by our invited speakers - Adam Cheyer & Tom Gruber (~60 min.)     (1E2C)
    4. Q&A and Open discussion (~30 min.) [Kindly identify yourself before speaking.]     (1E2D)
  • Topic: "Siri: An Ontology-driven Application for the Masses"     (1E3)
TomGruber_20100225c.png . . . . . . . . AdamCheyer_20100225c.png     (1E4)

[Dr. Tom Gruber] [Mr. Adam Cheyer]     (1E5)

Siri is a Virtual Personal Assistant ... more specifically, an intelligent software one would download into his mobile device (like an iPhone or a Nexus One phone) and have it start "doing" things for you like an assistant.     (1E7)
Siri was born out of SRI's CALO Project (which Adam presented here before), the largest Artificial Intelligence project in U.S. history. (CALO stands for Cognitive Assistant that Learns and Organizes). Made possible by a $150 million DARPA (Defense Advanced Research Projects Agency) investment, the CALO Project included 25 research organizations and institutions and spanned 5 years.     (1E8)
Among other accolades, Siri has been recognized by the MIT Technology Review as one of their "TR 10" in 2009 - the magazine's annual recognition of the list of top 10 emerging technologies that can change the way we live.     (1E9)
Started in 2007 in a stealth mode, Siri has finally released their product earlier this month. We are glad Adam and Tom (who co-founded Siri, along with Dag Kittlaus) can come and give us a talk and a demo on this brainchild of theirs and their team.     (1E10)
Adam Cheyer is a co-founder and Siri's VP of Engineering. He joined the company from SRI, where he was the Program Director in SRI's Artificial Intelligence Center and Chief Architect of the CALO/PAL project. A pioneer in the areas of distributed computing, intelligent agents, and advanced user interfaces, Adam is the author of more than fifty peer-reviewed publications and nine patents. He was previously the VP of Engineering at Dejima and the VP of Engineering at Verticalnet. Adam is also a founding member of and Genetic Finance, LLC.     (1E12)
Tom Gruber is a co-founder and Siri's Chief Technology Officer. He, of course, is best known to this community as the Stanford scientist who, back in 1992, defined the term "ontology" in the sense we apply to its use in computer science, artificial intelligence and the semantic web these days. Tom is a researcher and entrepreneur with a focus on systems for knowledge sharing, collaboration, and collective intelligence. He did foundational work in ontology engineering with approaches and technologies that are precursors to the infrastructure for today's Semantic Web.     (1E13)

Transcript of the online chat during the session     (1E14)

see raw transcript here.     (1E14A)

(for better clarity, the version below is a re-organized and lightly edited chat-transcript.)     (1E14B)

Participants are welcome to make light edits to their own contributions as they see fit.     (1E14C)

Welcome to the Ontolog Invited Speaker Presentation - Thu 2010.02.25     (1E14E)

  • Title: "Siri: An Ontology-driven Application for the Masses"     (1E14G)

Please refer to details on the session page at:     (1E14H)

anonymous morphed into Jack Ring     (1E14K)

anonymous1 morphed into Jim Disbrow     (1E14L)

anonymous morphed into Hasan Sayani     (1E14M)

anonymous morphed into Joel Bender     (1E14N)

Peter P. Yim: we are running a webex shared-screen session today (info just posted) ... please refresh     (1E14O)

your session page, and follow instructions at the top of that page -     (1E14P)

Joel Bender: Peter, step 3 says "Enter the meeting password: ontology123" and it doesn't seem to need     (1E14R)

the password.     (1E14S)

Peter P. Yim: thansk, Joel ...     (1E14T)

anonymous morphed into Tejas Lagvankar     (1E14U)

anonymous morphed into Jim Disbrow     (1E14V)

anonymous morphed into Geroge Terre     (1E14W)

anonymous morphed into Joel Bender     (1E14X)

anonymous morphed into Tom     (1E14Y)

Tom morphed into Tom Gruber and Adam Cheyer     (1E14Z)

anonymous: Richard Christian Brown (UNF)     (1E14AA)

anonymous morphed into Richard Christian Brown     (1E14AB)

Richard Christian Brown morphed into Richard Christian Brown     (1E14AC)

anonymous1 morphed into Ellie Trevarthen     (1E14AD)

anonymous morphed into TJ     (1E14AE)

anonymous morphed into michael     (1E14AF)

anonymous morphed into John McGrath     (1E14AG)

anonymous morphed into Elizabeth Florescu     (1E14AH)

michael: Is there a limit to the web ex..just got an error saying only 15 people allowed in webex     (1E14AI)

michael: I got a network error for the vnc2     (1E14AJ)

Arturo Sanchez: There is a limit of 15 connections on the Web-Ex site ...     (1E14AK)

Arturo Sanchez: Question for presenters: are you developing an extensible architecture to support     (1E14AL)

changes organically?     (1E14AM)

Arturo Sanchez: Question: does Siri know/learn about personal preferences?     (1E14AN)

Arturo Sanchez: Question: what is the range of the context associated with the questions posed to     (1E14AO)

Siri? Can Siri also learn about new contexts?     (1E14AP)

Arturo Sanchez: Question: if you ad a friend named "Siri" would Siri know it has to ask you who you     (1E14AQ)

are referring to?     (1E14AR)

michael: How does the company intend to make money with this tool?     (1E14AS)

Ellie Trevarthen: How is the speech interface affected by background noise -- i.e. if one is using     (1E14AT)

Siri in a crowded place, vehicle, etc.?     (1E14AU)

Arturo Sanchez: Question: how is the ontology represented in Siri? Do you use your own inference     (1E14AV)

John McGrath: Question: I see the system uses NLP to understand term meaning and     (1E14AX)

intent. Is there a query syntax standard for the various services or is their a custom interface     (1E14AY)

require for each service? Or do they need to create an API to a Siri standard?     (1E14AZ)

Arturo Sanchez: Question: "Find the best restaurant around according to Zagat and Martha Stewart" how     (1E14AAA)

would Siri handle this question?     (1E14AAB)

Richard Christian Brown: Can we take a break to answer some questions?     (1E14AAC)

PeterYim1: @Richard Christian Brown - we'll be doing Q&A after the presentation and demo ... try     (1E14AAD)

using this chat room to type out your question(s) first, if you please     (1E14AAE)

michael: CAn you please tell us the name of the tools they are using? are they home grown or     (1E14AAF)

commercial products?     (1E14AAG)

Peter P. Yim: I am somewhat surprised Adam's voice worked on Tom's phone ... do you need to "train" the     (1E14AAH)

voice recognition?     (1E14AAI)

Richard Christian Brown: *noted     (1E14AAJ)

anonymous1 morphed into Cecil Lynch     (1E14AAL)

Peter P. Yim: @Adam & Tom - how do you manage performance and scalability ... tell us something about     (1E14AAO)

that, please     (1E14AAP)

Sean Barker: When would we be able to try this out in the UK?     (1E14AAQ)

Arturo Sanchez: Pretty cool guys! Thank you for sharing your system with us!     (1E14AAR)

Richard Christian Brown: @Adam & Tom - How have you leveraged available computing power and     (1E14AAS)

resources to promote real-time processing and rendering of results?     (1E14AAT)

Richard Christian Brown: Right on!     (1E14AAU)

Richard Christian Brown: What portions of processing are done on the device?     (1E14AAV)

Richard Christian Brown: Or more specifically, do you utilize the device running the application (be     (1E14AAW)

it a MDA, PDA, notebook, etc...) to do naturally language processing and disambiguation?     (1E14AAX)

Peter P. Yim: Tom: "Big Think, Small Screen"     (1E14AAY)

Richard Christian Brown: nice try     (1E14AAZ)

Joel Bender: (no iPod Touch support , I'll try this on my iPhone later )     (1E14AAAA)

anonymous2 morphed into Bernard Ulozas     (1E14AAAB)

Arturo Sanchez: Question: what main security concerns did you identify while developing the     (1E14AAAC)

application and how did you address them?     (1E14AAAD)

Brand Niemann: I have used this for several weeks now and it has worked very well. Thanks, Brand     (1E14AAAE)

Tejas Lagvankar: I'm waiting for the Android App !     (1E14AAAF)

GaryBergCross: Do you think that it will difficult to maintain your preferences for various services     (1E14AAAG)

as they change or time? Say Yelp is preferred for reviews but then adds better info for reservations     (1E14AAAH)

and some other service starts to have better reviews.     (1E14AAAI)

Tom Gruber and Adam Cheyer: For other devices, please stay tuned...     (1E14AAAJ)

Evan Wallace: Thanks. I want this.     (1E14AAAK)

michael: what is the business model?     (1E14AAAL)

Richard Christian Brown: gone are the days of viral marketing ... well so to speak     (1E14AAAM)

John McGrath: Very informative. Most interested to understand how search engine     (1E14AAAN)

platforms can best interface to the closud to support service providers like Siri. Seems a more     (1E14AAAO)

robust dialog(vs query) interpretor stand will be required in the long run.     (1E14AAAP)

Ali Hashemi: thanks all     (1E14AAAQ)

Tom Gruber and Adam Cheyer: Thanks     (1E14AAAS)

Peter P. Yim: -- session ended 12:12pm PST --     (1E14AAAU)

-- end of chat session --     (1E14AAAV)

Almost all the questions raised on the chat were addressed by our speakers verbally during Q&A segment of the session.     (1E14AAAW)

Check out the audio recording for the responses.     (1E14AAAX)

Audio Recording of this Session     (1F)

  • To download the [ audio recording of the session], click [ here]     (1F1)
    • the playback of the audio files require the proper setup, and an MP3 compatible player on your computer.     (1F1A)
  • Conference Date and Time: 25-Feb-2010 10:38am ~ 12:12 pm Pacific Standard Time     (1F2)
  • Duration of Recording: 1 Hour 28 Minutes     (1F3)
  • Recording File Size: 10.0 MB (in mp3 format)     (1F4)
  • suggestion: its best that you listen to the session while having the presentation opened in front of you. You'll be prompted to advance slides by the speaker.     (1F5)
  • Take a look, also, at the rich body of knowledge that this community has built together, over the years, by going through the archives of noteworthy past Ontolog events. (References on how to subscribe to our podcast can also be found there.)     (1F6)

For the record ...     (1F7)

How To Join (while the session is in progress)     (1G)

  • Discussions and Q & A:     (1G5F)
    • (Unless the conference host has already muted everyone) Please mute your phone, by pressing "*2" on your phone keypad, when a presentation is in progress. To un-mute, press "*3"     (1G5F1)
    • You can type in your questions or comments through the browser based chat session by:     (1G5F2)
      • instructions: once you got access to the page, click on the "settings" button, and identify yourself (by modifying the Name field). You can indicate that you want to ask a question verbally by clicking on the "hand" button, and wait for the moderator to call on you; or, type and send your question into the chat window at the bottom of the screen.     (1G5F3A)
    • (when everyone is muted) If you want to speak or have questions or remarks to make, please "raise your hand (virtually)" by click on the "hand button" (lower right) on the chat session page. You may speak when acknowledged by the speaker or the session moderator (again, press "*3" on your phone to unmute). Test your voice and introduce yourself first before proceeding with your remarks, please. (Please remember to click on the "hand button" again (to lower your hand) and press "*2" on your phone to mute yourself after you are done speaking.)     (1G5F4)
    • thanks to the folks, one can now use a jabber/xmpp client (e.g. gtalk) to join this chatroom. Just add the room as a buddy - (in our case here) ... Handy for mobile devices!     (1G5F5)
  • Please note that this session will be recorded, and the audio archive is expected to be made available as open content to our community membership and the public at-large under our prevailing open IPR policy.     (1G5J)

Attendees     (1H)

This page has been migrated from the OntologWiki - Click here for original page     (1H4)