From OntologPSMW

Revision as of 03:29, 9 January 2016 by KennethBaclawski (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
[ ]

OntologySummit2015 Symposium Chat Transcript for Monday 2015-04-13     (1)

[8:37] Michael Grüninger: Welcome to Ontology Summit 2015 Symposium!     (1A)

[9:21] Mark Underwood: All - If you are inclined, the Twitter hashtag for the meeting is #ontologysummit     (1B)

[9:31] JackRing1: deal with dynamic time varying content of the IoT     (1C)

[9:33] JackRing1: ontology of IoT pathologies     (1D)

[9:34] Peter P. Yim: @LeoObrst, Ram D. Sriram & chairs ... I've just posted links to your Track A Synthesis slides onto the wiki page, please refresh page     (1E)

[9:34] Peter P. Yim: @Chairs: are slides to the keynote speakers available?     (1F)

[9:35] Mark Underwood: @PeterYim - thx for tweaking the web performance     (1G)

[9:36] Michael Grüninger: Keynote speaker slides available at http://bit.ly/MIT-IOT     (1H)

[9:36] Michael Grüninger: Starting Keynote Speaker: Dr. Shoumen Palit Austin Datta     (1I)

[9:43] Mike Bennett: Which of the many decks at that link is the one we are loking at?     (1J)

[9:49] Michael Grüninger: I thought it was MIT IoT - Talk ...     (1K)

[9:50] Bobbin Teegarden: Could you please post the URL for the talk here? Thanks!     (1L)

[9:52] Michael Grüninger: Sorry -- the speaker is using a different set of slides than the one I was given -- we will upload the correct slides at the end of the talk.     (1M)

[9:52] Mike Bennett: Thanks Michael - that might be the one but I don't see references to the things that have been spoken about. What slide are we on? Then we'll know if it's the one.     (1N)

[10:10] Ravi Sharma: I am also at loss to not be able to see the slides?     (1P)

[10:13] Mark Underwood: Ravi- understood     (1Q)

[10:13] Mark Underwood: Glad to see/hear cites for domain-specific languages in their proposed / planned IoT efforts     (1R)

[10:16] Ravi Sharma: But all it takes is a minute to put them on wiki     (1S)

[10:21] Michael Grüninger: @Ravi -- sorry but the speaker is doing a dynamic extraction of a subset of slides, which are themselves a mash-up of a set of other talks.     (1T)

[10:23] Ravi Sharma: Dr. Palit Datta you mentioned only current standards but what we need are next gen and embedded nanosensors type standards hence beyond ICD 10 etc? Also how will decisions be made for triage and diagnostics, same as now or with connected IoT or not?     (1U)

[10:24] Peter P. Yim: at this particular time point, the speaker is at slide#173 of the slide deck at: http://dspace.mit.edu/bitstream/handle/1721.1/86935/MIT%20IoT%20-%20Talk%2010082014.pdf?sequence=115     (1V)

[10:32] Ravi Sharma: @Dr.Datta - what are the new standards likely beyond ipv6 but relating to embedded healthcare?     (1W)

[10:32] Ravi Sharma: @peter -thanks     (1X)

[10:36] Brand Niemann: Proposal to digitize the semantic web using a numerical nomenclature as a sub-layer to URI (utilizing a IPv6 type format).     (1Z)

[10:37] Ravi Sharma: @Dr.Datta - Epic and other sysytems have to be re-engineered to become semantically aware, which way will you think we will go the retrofit or new apps replacing Epic?     (1AA)

[10:39] Gary Berg-Cross: @Joe I disagree that the problem in HC is elsewhere than semantics. The defined semantics in existing ontologies are designed for certain meanings and not all meanings needed. Diff ontologies commit to different things.     (1AB)

[10:39] Ravi Sharma: @Dr.Datta - Similarly more important are probably injection of clinical and parma data from Medline, Snowmed     (1AC)

[10:42] Terry Longstreth: Google 'Ontoiop' for info on an effort, currently under OMG auspices, to develop an approach for interoperability of Ontology languages.     (1AD)

[10:42] Ravi Sharma: @DrDatta - great talk thanks will be in email touch.     (1AE)

[10:44] Ravi Sharma: unified theory of relativistic identifications?     (1AF)

[10:44] Brand Niemann: Unified Theory of Relativistic Identification of Information in a Systems Age: Proposed Convergence of Unique Identification with Syntax and Semantics through Internet Protocol version 6: http://dspace.mit.edu/handle/1721.1/41902     (1AG)

[10:45] Antoine Gerardin: @PeterYim - link to UTRI http://dspace.mit.edu/handle/1721.1/41902#files-area     (1AH)

[10:47] Brand Niemann: Tim Berners-Lee on the Future of His Invention: http://www.emc.com/leadership/features/berners-lee.htm     (1AI)

[11:00] Michael Grüninger: Second morning session starting now     (1AJ)

[11:00] Michael Grüninger: Track A Summary (slides available on meeting page)     (1AK)

[11:04] Peter P. Yim: @ShoumenDatta, Brand Niemann and AntoineGerardin - re. UTRI, thank you for the pointer to your (Datta) paper(s)     (1AM)

[11:05] Peter P. Yim: == Track A Synthesis     (1AN)

[11:13] Ravi Sharma: @Brand - thanks     (1AO)

[11:27] Brand Niemann: Sounds like the MITRE-VA work might benefit from Michael Stonebraker's Tamr that we have had presented at our Federal Big Data Working Group Meetup on March 16th: http://www.meetup.com/Federal-Big-Data-Working-Group/events/220121871/     (1AP)

[11:32] Ravi Sharma: @Leo - map syntax first and then semantics for interoperability?     (1AQ)

[11:34] Ravi Sharma: @Leo - does the even framework capture Complex event processing within or outside the framework?     (1AR)

[11:34] Brand Niemann: To turn data (IoT) into value you need: Integration, Analytics, and Event Processing     (1AS)

[11:37] Gary Berg-Cross: "Integration" of models of events may be able to evolve from local models that integrate local sensors that are common with other types of events sensed by different types of sensors.     (1AT)

[11:39] Todd Schneider: Ravi, Regarding your questions about context, it's done today when a device connects but the context is very general and non-specific (i.e., domain independent) and the context is not explicit.     (1AU)

[11:44] Ravi Sharma: @Todd - how do we make it domain specific and foolproof / secure as we deal often with actions after matching context such as administer a drug etc.     (1AV)

[11:44] Bobbin Teegarden: Yes! Agents haven't gone away, they've just been being implemented.     (1AW)

[11:46] Ravi Sharma: @leo and sriram thanks     (1AX)

[11:46] Terry Longstreth: Agents as micro theories?     (1AY)

[11:46] Michael Grüninger: == Track B synthesis (slides available on meeting page)     (1AZ)

[11:54] Todd Schneider: Ravi, First we'd need an understanding of what context is, or at least some way to represent them with sufficiently fidelity to constrain interpretations (for/to the domain).     (1AAA)

[12:03] Ravi Sharma: @Gary - Can we take it like setting the interoperating stage by these standards initiatives and also context setting where automated acceptance or rejection of candidates requesting integration can happen?     (1AAB)

[12:07] Ravi Sharma: @Gary - How would Geospatial and temporal awareness integrate with IP addresses space or provenance and metadata will validarte who and who can not provide or consume data?     (1AAC)

[12:15] Ravi Sharma: @torsten - yes you tried to answer some of my earlier questions, SPARQL is it the only or other similar but richer semantic languages?     (1AAD)

[12:17] Ravi Sharma: @Todd - some of what is being talkedabout such as standards, SPARQL etc should converge to define context does it mean more than awareness?     (1AAE)

[12:22] Todd Schneider: Ravi, It's not clear to me that standards by themselves can define context.     (1AAF)

[12:26] Torsten Hahmann: @Ravi: we are currently figuring out how rich a language we need. Currently, we don't use SPARQL because it was too limited for the kind of aggregations and reasoning over complex events we want to do. We work with Common Logic (though not the full expressivity) to have more flexibility. The semantics is translated to a combination of SQL queries and procedures (not yet automatically, though using sort of templates). One of the big questions is still how to effectively use richer semantic languages in combination with lots of data.     (1AAG)

[12:29] Brand Niemann: Semantic Big Data Science is more about putting semantics in the data (especially big data, IoT, etc.) than putting semantics in the technology (like the past emphasis)!     (1AAH)

[12:32] Ravi Sharma: @all - also there have to be backups for reliability such as failsafe or backups in case internet channel temporarily fails. especially critical for air and realtime car being driven remotely etc.     (1AAI)

[12:33] Peter P. Yim: -- day-1 morning session adjourned --     (1AAJ)

[12:33] Peter P. Yim: remote participants will need to call in again for the afternoon sesssion. (please end your current call now)     (1AAK)

[12:35] Ravi Sharma: @all - I want to capture the statement made that there is a spectrum of users from raw data to most advanced realtime use of information such as formation flights, actions based on other platforms realtime data and archiving by Authoritative data source, these all have to interoperate as needed. [12:35] Ravi Sharma: Thanks see you at 2PM EST.     (1AAL)

[13:51] Michael Grüninger: The conference bridge is now open ...     (1AAM)

[13:56] DavidBlevins: Receiving "The moderator has not yet joined the conference". FYI     (1AAN)

[13:57] Gary Berg-Cross: @Ravi On this Q of Can we set interoperating stage by context setting where automated acceptance or rejection of candidates requesting integration can happen, yes, there will be different degrees of automation as we go forward and test things but I can imagine that as a distributed loosely couped approach we reach one degree new things will be added to IoT that may not be automated within a current level of interoperations. That is some breaks can be expected in operation till we learn the problem space.     (1AAO)

[13:59] DavidBlevins: @Peter are you attending remotely this year?     (1AAP)

[14:01] Gary Berg-Cross: @Ravi you asked " How would Geospatial and temporal awareness integrate with IP addresses space or provenance and metadata will validarte who and who can not provide or consume data?" Seems like an undecided security issue. My smart car, smart phone and smart house may all have different locations and each an IP address and I may be separated from each of them and inputting info from a work computer.     (1AAQ)

[14:04] Michael Grüninger: We are just uploading the slides for the second keynote -- they will be available on the meeting page     (1AAR)

[14:14] Ravi Sharma: @Gary - then what decides access between me and my driver role to access the steering wheel - is it biometrics that decides interoperation? just jokingly we might as well test for cognitive capabilities and alertness as integral to rules for safe driving and also switch to auto safety mode options?     (1AAS)

[14:15] Ravi Sharma: what slides are being shown?     (1AAT)

[14:15] Gary Berg-Cross: Keynote 2 Keith Marzullo (nsf) on smart infrastructure (e.g. bridge knows status) & Cyber Physical systems.     (1AAU)

[14:23] DavidBlevins: @Michael or @Ken - would it be possible to have the slides uploaded to the ontolog.cim3.net/work/ workspace at least?     (1AAV)

[14:28] Mike Bennett: @Michael I believe Ken is teaching all this afternoon.     (1AAW)

[Added to chat] Ken Baclawski: Yes, I was teaching all afternoon.     (1AAX)

[14:32] DavidBlevins: W.r.t the assertion that many of the listed technologies are not "semantic technologies": IoT is an area with a massive amount of information that (seemingly) would be easier to utilize if the systems utilized or integrated information semantics. The absence of semantics limits our ability to fully utilize these devices and sensors     (1AAY)

[14:33] Eric Simmon (NIST): @Micheal - Could you upload the files to soaphub? It's very hard to follow without the slides.     (1AAZ)

[14:33] Michael Grüninger: stil trying ..........     (1AAAA)

[14:33] DavidBlevins: Plug for RPI's Smart Lighting center (in response to a comment made from the room) : https://smartlighting.rpi.edu/     (1AAAB)

[14:37] Eric Simmon (NIST): @Micheal - I tested in a separate chatroom works just fine - if you email it to me I can try uploading (eric.simmon@nist.gov)     (1AAAC)

[14:38] DavidBlevins: @Eric - I believe he's trying to upload to the conference page, rather than the chatroom     (1AAAD)

[14:40] Eric Simmon (NIST): success     (1AAAF)

[14:43] DavidBlevins: @Peter or Michael - Which slide are we on?     (1AAAG)

[14:43] Peter P. Yim: to get to the newly posted slides (for the current KeithMarzullo keynote tale), please refresh the wiki page first     (1AAAH)

[14:44] Peter P. Yim: can anyone (onsite) tell us what slide number we are on now, please?     (1AAAJ)

[14:45] Michael Grüninger: now on slide 24     (1AAAK)

[14:45] DavidBlevins: slide 25     (1AAAL)

[14:45] DavidBlevins: @Michael - the audio has become very quiet     (1AAAN)

[14:45] DavidBlevins: sounds like Keith may have stepped back or away from the microphone     (1AAAO)

[14:45] Eric Simmon (NIST): looks like Keith's microphone failed... CPS attack???     (1AAAP)

[Added to chat] Ken Baclawski: In the audio recording, I amplified the parts of the talk where the microphone was turned off. Those parts do not have the best quality, but they are audible.     (1AAAQ)

[14:46] Eric Simmon (NIST): that's good!     (1AAAR)

[14:46] DavidBlevins: hah     (1AAAS)

[14:46] Mark Underwood: @Eric, thx     (1AAAT)

[14:47] Mark Underwood: @Ann - this is the NSF enterprise we have talked about in the NBDPWG-S&P     (1AAAU)

[14:47] Ravi Sharma: @Dr. Keith Marzullo - since I still can not access - as a former architect of OnStar I was not clear whether monitoring the baby on OnStar was a good or bad thing? kindly enlighten?     (1AAAV)

[14:51] DavidBlevins: @Dr. Marzullo - What constitutes "Platforms" in the context of NSF's CPS initiatives     (1AAAW)

[14:52] Gary Berg-Cross: Sound? Maybe a battery issue??     (1AAAX)

[14:52] DavidBlevins: @Gary - Check headset or phone? I did not notice a drop in volume or quality on my line...     (1AAAY)

[14:53] Mark Underwood: @Eric late LOL re: CPS attack     (1AAAZ)

[14:54] DavidBlevins: @Mark - he seems to have stepped away again     (1AAAAA)

[14:54] DavidBlevins: or turned his head     (1AAAAB)

[14:55] Peter P. Yim: thought the speaker has a lapel mic (or something) ... can only barely hear him now     (1AAAAC)

[14:55] DavidBlevins: Fixed!     (1AAAAD)

[14:56] Terry Longstreth: His pocket control unit was muting itself     (1AAAAE)

[14:59] DavidBlevins: Who asked the question about resiliency?     (1AAAAF)

[14:59] Ravi Sharma: I can hear well     (1AAAAG)

[15:01] Gary Berg-Cross: @Ravi +1 fail safe needed...     (1AAAAH)

[15:02] Ravi Sharma: Resilent systems would not allow that     (1AAAAI)

[15:07] Michael Grüninger: === Track C Summary by Mike Bennett     (1AAAAJ)

[15:08] Ravi Sharma: @ProfKeith - Security encryption would solve that intruders access to car's controllers, when it was designed some encryptions did not have embedded aspects only realtime operating systems?     (1AAAAK)

[15:08] Mark Underwood: @DavidBlevins - Sorry, that person's moved offline & away from the room     (1AAAAL)

[15:17] Mark Underwood: @DavidB - this you? @dblevins     (1AAAAM)

[15:18] DavidBlevins: @Mark - No. David Blevins seems to be an unusually common name for software developers.     (1AAAAN)

[15:29] Terry Longstreth: @Matthew slide 10 can you elaborate xxx vs categories?     (1AAAAO)

[15:30] Ravi Sharma: @Mike and Ken - What are the desirable attributes of Language beyond SPARQL, things like time awareness and data provenance or other decision logic areas? beyond UML2+?     (1AAAAP)

[15:31] Terry Longstreth: Sorry...I meant Mike Bennett     (1AAAAQ)

[15:31] Mark Underwood: slide 14     (1AAAAR)

[15:32] Ravi Sharma: @Gary - my remark was about OnStar, I agree with your vote that we need fail safe resilient systems.     (1AAAAS)

[15:33] Mark Underwood: Now slide 14 (13 was misnumbered) sorry     (1AAAAT)

[15:38] Ravi Sharma: @ProfKeith - just as we are service (SOA) enabling legacy mainframe apps for new features, the 20+ year old GM OnStar System needed Security layers starting from satellite analog and digital transponders, embedded command sequences for car controllers all this needs resources which they probably did not have when coming out of bankruptcy. Yet even today GM cars have ubiquitous internet access and by the way, the system was primarily Human oriented assisted by computer and GIS GPS separate modules and later integration was not during my tenure as Enterprise Architect at GM.     (1AAAAU)

[15:39] Gary Berg-Cross: @Mike B Great summary of your session, as usual from you..     (1AAAAV)

[15:42] Mark Underwood: +1 Gary re: MikeB     (1AAAAW)

[15:42] Mark Underwood: +1 for Enterprise Architect also     (1AAAAX)

[15:45] Terry Longstreth: Restating ques for Mike topic vs type hierarchies     (1AAAAY)

[15:47] Gary Berg-Cross: @Mike B But people are things in most ontologies, just not artifacts. Some artifacts are agents too.     (1AAAAZ)

[15:47] Michael Grüninger: Taking a brief break. Restarting in 15 minutes (a 1600 EDT) with the Track D Summary. Slides are available on the meeting page     (1AAAAAA)

[15:47] DavidBlevins: Building on Topic vs Type - there's plenty of literature on the Type-Token and Subject-Object problems that are relevant to the issue     (1AAAAAB)

[15:48] Ravi Sharma: @all speakers or others - Some languages abbreviate based on context and are brief - what would that be for Onto-languages?     (1AAAAAD)

[15:49] Ravi Sharma: who is speaking? Stu?     (1AAAAAF)

[15:49] Mike Bennett: Joe Kopena was speaking in answer to that last question     (1AAAAAG)

[16:01] Mike Bennett: @Gary [hh:47] Quite so. But unless the perspn is reporting in themselves (e.g. pain) there is an aboutness relation between the observer and the observed, e.g. spotter and troop concentrations. Similarly one might have a taxonomy of kinds of pressure sensor and a taxonomy of kinds of pressurized vessel.     (1AAAAAH)

[16:01] Michael Grüninger: == Track D Summary by Mark Underwood     (1AAAAAI)

[16:03] DavidBlevins: @MichaelGruninger - I can't seem to locate slides for Track D     (1AAAAAJ)

[16:03] Michael Grüninger: @David Blevins: see note from 15 minutes ago     (1AAAAAL)

[16:04] Peter P. Yim: refresh page first ... direct link also posted above [xx:47] Michael Grüninger:     (1AAAAAM)

[16:04] Joe Kopena: Just wanted to add: That end point in the last presentation about the difficulties applying SPARQL to our applications is a long-ish conversation (already had a bunch at break). Would be happy to discuss more.     (1AAAAAN)

[16:04] DavidBlevins: @Michael & @Peter - Thanks! Seems I needed an additional refresh. Much better layout now.     (1AAAAAO)

[16:25] Ravi Sharma: Provenance appears to be important for seting the context can this be made an ontology pattern?     (1AAAAAP)

[16:32] Ravi Sharma: @Mark - Sensors generally differentiate for focusing on a domain such as heartbeat, but human cognition also senses vibration and shock before the building collapses? therefore sensors are only fit for purpose not every paatern is relevant. ome sensor networks might post process trends that would have indicated impending collapse but do we know how to program for every consequence or event a-priori? Hence all events can not be pre-patterned?     (1AAAAAQ)

[16:34] DavidBlevins: @Ravi - A few of us discussed this in-person (and briefly over the chat) last year. It seems that one of the biggest barriers isn't necessarily the inability (or failure) to include context, so much as the lack of good tools for reasoning over contexts in OWL. Many people mentioned that there are some expensive tools to reason well over context, and others work context into their systems through novel features present in certain free/open tools; it is difficult to make use of context in reasoning tasks due to computational overhead (combinatorial explosion) in situations where you don't have a pre-computed context, and a lack of flexibility in situations where you do pre-compute answers using a given context     (1AAAAAR)

[16:35] Eric Simmon (NIST): just a small correct - NIST is not a standards committee - we work with standards committees     (1AAAAAS)

[16:35] DavidBlevins: @Ravi - my last response was in regard to the use of provenance for setting context     (1AAAAAT)

[16:43] Ravi Sharma: @David - Would tracing previous sessions and understanding them by a reasoning engine provide both the provenance and context awareness so the communication and conversation going forward in a reasoning engine can be brief and relevant to the situation?     (1AAAAAU)

[16:44] DavidBlevins: @Ravi - We should make that a working group for next year!     (1AAAAAV)

[16:46] DavidBlevins: @Ravi - I agree with your sentiment on this (both a wider adoption of provenance, and the difficulty of examining previous discussions on the matter)     (1AAAAAW)

[16:47] Ravi Sharma: @Mark -that framework is human intervention but only if sensors have smartness of knowing anomaly?     (1AAAAAX)

[16:49] Ravi Sharma: @david - yes     (1AAAAAY)

[16:50] DavidBlevins: @Todd - Agreed. We've spoken about this, and the conclusion I've come to is that we have too many standards, and not enough solutions. Too many people creating ontologies without intelligent agents/code that are immediately usable in a system     (1AAAAAZ)

[16:51] Ravi Sharma: @David - I have advocated an ontology of standards and ontology that uses standards in cognitive ways.     (1AAAAAAA)

[16:51] Gary Berg-Cross: We are talking about agile approaches for IoT apps that employs ontologies in a modular, best practice fashion.     (1AAAAAAB)

[16:52] DavidBlevins: @Ravi - Is that within the means of our current technologies/frameworks? We have automated reasoning, but not necessarily automated decision making, which seems to be a bit of the issue. Fortunately, this has been discussed a few times today     (1AAAAAAC)

[16:53] DavidBlevins: @Gary - are those being covered tomorrow?     (1AAAAAAD)

[16:55] Ravi Sharma: Just as we take measurements - and understand point, area and volume as well as time series of measurements, we ought to burden sensors and IoT with appropriate scope such as assumptions of homogeneity extrapolating point measurement to area etc. This is training set in Machine learning with training approach.     (1AAAAAAE)

[16:55] DavidBlevins: @Gary - or are you referring to the other items that we discussed in both the current presentation, as well as earlier today? I do find it interesting that this topic has come up again, considering the fact that this was the primary focus of last year's Ontology Summit.     (1AAAAAAF)

[16:56] DavidBlevins: correct to my phrasing: was a sub-track under last year's theme     (1AAAAAAG)

[16:59] Gary Berg-Cross: @David B I don't know if this idea of best practices for ontology development is on the agenda tomorrow. Leo and I each talked about ontology design patterns and modular approaches today and this was in Mark U's title today. There was more of this last year as one track and what applies to Big Data here also applies to IoT.     (1AAAAAAH)

[17:00] Mike Bennett: Got to go - thanks all!     (1AAAAAAI)

[17:01] Ravi Sharma: @mikeB - please review some of my Qs for you in chat, regards, Ravi     (1AAAAAAJ)

[17:01] Michael Grüninger: @MikeBennett: thanks for a great Track C synthesis!     (1AAAAAAK)

[17:01] Ravi Sharma: @MIke - yes thanks as well.     (1AAAAAAL)

[17:02] Eric Simmon (NIST): Do sensors need to know the context in what they are used? Or do they need to communicate their capabilities so the users can determine how that can be used?     (1AAAAAAM)

[17:05] Torsten Hahmann: Hate to leave this interesting discussion, but have to go now. Thanks everyone for the presentations.     (1AAAAAAN)

[17:05] Ravi Sharma: @David - if we do not do this, namely integrating ontologies and standards as well as major industry forums (orgs such as ICAO-FAA) WMO-Vs-Country Weather Programs? etc.?we will again circle around and proliferate rather than converge to knowledge rather than get would up in data?     (1AAAAAAO)

[17:08] Ravi Sharma: @Ram, @David - just as it is wise to do Upper ontologies and then domain based, it is that type of notion that would limit proliferation of standards.     (1AAAAAAP)

[17:09] Ravi Sharma: @David, @Ram - healthcare is trying hard to keep pace - PQRI, HL&, HiPAA, ICD10, costs are increasing     (1AAAAAAQ)

[17:11] DavidBlevins: @Michael - Aah, Didn't realize that we hadn't *actually* switched to the general discussion     (1AAAAAAR)

[17:11] Ravi Sharma: kindly delay AM session for CA folks     (1AAAAAAS)

[17:12] DavidBlevins: @Ravi - This issue has been discussed intensely, and I agree. It's a difficult topic, and it often seems to devolve in a lengthy discussion of what is broken, and a reduced emphasis on how we're able to actually fix it.     (1AAAAAAT)

[17:13] DavidBlevins: @Ravi - Perhaps we need to focus more on this problem of solving interoperability     (1AAAAAAU)

[17:14] Mark Underwood: Joe's reference link is on slide 42     (1AAAAAAV)

[17:15] DavidBlevins: I don't think "interoperability" is even the right word here. It is not enough to make the systems fit together - they need to be able to produce working results by communicating with eachother     (1AAAAAAW)

[17:16] Ravi Sharma: @David - One of the things could come out from healthcare forums for example, would be to do an upfront impact analysis for adopting all the good things, in the process not hurting small individual practioners who are running away from practice or from taking medicare patients and indirectly pushing healthcare costs and similarly another argument for net- neutrality issues and consumer communications costs? I agree that proper forums and advice to Gov's are required but interoperability can go a long way?     (1AAAAAAX)

[17:16] Eric Simmon (NIST): Things in IoT are not always perceived as being physical     (1AAAAAAY)

[17:17] Ravi Sharma: @david, yes they shoud interoperation beyond just exchanging data like the EHR has to be ful range from research use to operational triage in ER in realtime?     (1AAAAAAZ)

[17:17] Brand Niemann: Is the Data Science for USGS Minerals Big Data an IoT? http://semanticommunity.info/Data_Science/Data_Science_for_USGS_Minerals_Big_Data I have been working on this will listening all day so I would have something to show as an example. Thank you for an excellent day of listening!     (1AAAAAAAA)

[17:18] Peter P. Yim: Great content and discussions today! ... Huge THANKS to all the speakers and the organizers ... looking forward to tomorrow's session ... (for those of you who are in-site) have a wonderful Group Dinner!     (1AAAAAAAB)

[17:18] Evan Wallace: @DavidBlevins - Can you say anything more about the competitor to HL7 that you mentioned? You don't mean CTS2 do you?     (1AAAAAAAC)

[17:18] Peter P. Yim: -- day-1 afternoon session adjourned --     (1AAAAAAAD)

[17:18] Mark Underwood: Michael requests add'l use cases from the public vernacular about IoT to help energize relevant discourse for ontology communities     (1AAAAAAAE)

[17:19] Ravi Sharma: @Eric - yes but recently diverted also to physical and wearable and even implantable IoTs not being the original intent.     (1AAAAAAAF)

[17:19] Mark Underwood: That :Elementary: episode "Olympus"     (1AAAAAAAG)

[17:20] Ravi Sharma: Peter has turned us off chat -so capture own chat inputs please.     (1AAAAAAAH)

[17:21] DavidBlevins: @Evan - I misspoke. I was speaking specifically about the Clinical Document Architecture, which has been implemented as the CCD by the HL7 group, and as the CCR by a separate consortium     (1AAAAAAAI)

[17:21] Ravi Sharma: @Evan- Hi, I would also be interested in the answer!     (1AAAAAAAJ)

[17:22] Ravi Sharma: What is CTS2?     (1AAAAAAAK)

[17:23] Ravi Sharma: Peter - thanks Ravi     (1AAAAAAAL)

[17:23] Ravi Sharma: signing off now.     (1AAAAAAAM)

[17:24] DavidBlevins: @Ravi - In essence, we now have 3 standards which exist purely as information products. The CCD standard by HL7 is not accompanied by a comprehensive set of UML diagrams, however the CCR standard is. It was argued, in the discussion, that the presence of this ontology made CCR completely reusable between Electronic Health Record systems, however this is not the case in practice. The issue created, specifically, is that CCR's ontology specifies matters far outside of the scope of it's use as a simple information exchange medium.     (1AAAAAAAN)

[17:25] Ravi Sharma: @David - When will HL7 have such a capability or will it adapt CCR?     (1AAAAAAAO)

[17:26] Bobbin Teegarden: Wouldn't that make it even more reusable as you could pick and choose your own useable subset?     (1AAAAAAAP)

[17:26] DavidBlevins: @Ravi - it will not. CCR is a competing standard. The HL7 consortium is working on a new version of CCD that fixes some of these issues, but it is still deficient (in my opinion)     (1AAAAAAAQ)

[17:27] Ravi Sharma: @David -that is the issue, standards are meant for increasing adoption and ought be like open source i.e. build towards common accessible knowledge?     (1AAAAAAAR)

[17:29] DavidBlevins: @Bobbin - theoretically, yes. CCR works great as a message format standard, but it is not widely used for that. Additionally, the standard's specifications push towards using it as the definition of a record system, rather than an information storage system. Record systems are federally mandated to do more than simply store the inforamtion     (1AAAAAAAS)

[17:29] Ravi Sharma: @David - I can understand business survival, model and industry forum push-pulls but at least open standards bodies have no loss in integrating and collaborating for single standards, except egos do come in picture?     (1AAAAAAAT)

[17:30] DavidBlevins: @Ravi - True. It's a difficult issue to tackle. Unfortunately, I am signing off for today in an attempt to get to the dinner. I'm sure you and Todd could have a lengthy discussion about this over chat, haha     (1AAAAAAAU)

[17:31] Ravi Sharma: @david - you mean System of Record?     (1AAAAAAAV)

[17:31] Ravi Sharma: Todd is gone? I will also again sign-off more for tomorrow regards.     (1AAAAAAAW)

[17:32] DavidBlevins: hah, phrasing. I actually did mean Record System. It *should* be used as a System of Record, but isn't     (1AAAAAAAX)

[17:32] Ravi Sharma: more tomorrow, kindly keep in touch drravisharma@gmail.com.     (1AAAAAAAY)

[17:32] DavidBlevins: Yes, Todd as signed off. I'm off as well for the day. Talk to you tomorrow     (1AAAAAAAZ)

[18:11] Mike Bennett: @Ravi [20:30] > What are the desirable attributes of Language beyond SPARQL >> I don't really know the answer to that, we would need to dig into the details in the slides. Joe Kopena (his slide 14) is where this was brought up.     (1AAAAAAAAA)