Emotion Markup Language Incubator Group: Phone meeting

04 Sep 2008

See also: IRC log


Bjoern, Idoia and Nestor, marc, Paolo_Baggia, Enrico_Zovato, Ian, Jianhua, AndyBreen, felix, Christian,BillJarrold
Regrets Catherine

Chair's summary

More support was expressed for the idea to continue as a subgroup in MMI.

In the discussion of Core 1, there are two syntactic variants that are preferred, but no agreement yet which one is best.

In the discussion of Core 6, it is not yet clear whether and for what purpose to annotate complex emotions explicitly, and if so, whether that should be done by a joint parent element or by an attribute.


Discussion on continuation

<paolo> Christian: I'd like to continue

<Ian> I also think the sub group would be fine

<Bjoern> so do I and would clearly like to continue.

<paolo> marc: Maybe issues on Patents

<paolo> marc: There is policy in W3C, to disclose patent

<paolo> paolo: Yes there are patent issues,

<paolo> marc: risk people in MMI that already have patents

<paolo> Christian: Can try the MMI way and afterwards to follow another option

Core 1 and Core 6 discussion

Core 1: Type of emotion-related phenomenon

Bjoern presented the different proposals for core 1, as contained in:

http://lists.w3.org/Archives/Public/public-xg-emotion/2008Jul/0013.html and


The concern not to mix type and category was shared in the group; variants 4 and 5 were considered more closely.

Variant 4:


<category set="everyday" name="relaxed" confidence="0.8" />



<mood emotion-related-state-types="Scherer">

<category set="everyday" name="relaxed" confidence="0.8" />


Variant 5:

<emotion type="mood" emotion-related-state-types="Scherer">

<category set="everyday" name="relaxed" confidence="0.8" />


Participants voiced different preferences between these two variants; some considered variant 4 to be simpler and more intuitive, some considered variant 5 to be simpler and more consistent. One explicit comment, shared by several group members, was that in variant 5, it is confusing to call the element "emotion" but then in the type state e.g. "mood", which is different from "emotion".

<Christian> Christian prefers variant 5 with thinking about the name of the 1st level tag (emotion/affect/emotion-related-state/ers...)

Marc: decision might be taken later

Paolo: completely agree we can choose one but keep an Issue Note that there is pending a decision.

Marc: Proposal: Suggest to use the first example variant 5, keep element <emotion> at this stage.

Marc: Is it acceptable?


Core 6: Complex affect

<marc> http://lists.w3.org/Archives/Public/public-xg-emotion/2008Sep/0005.html

Bjoern: <present proposal>

... Variant 0 would be to not annotate complex emotions explicitly; code that needs to know that two emotion annotations are co-occurring would have to analyse time stamps.

... Variant 1 <complex-affect> element with affect elements inside.

Bjoern: Variant 1 has restrictions if affect elements have different timing, and timing would be annotated on the complex-affect element only.

... Variant 2 add cross-links instead.

Different opinions were voiced whether or not to annotate complex emotions or not. Nobody seemed to like the idea of cross-linking.

<Christian> christian wonders why time information is not sufficient. If two emotions overlap in time they overlap is complex

<Christian> we could add an attribut isComplex=true to those emotions to make it more explicit

<Christian> just one more thought from Christian: Making it explicit might make it more easy for human readers to find overlaps

<Christian> and possibly also faster for parsers (looking for an attribute might be less expensive than calculating times)

Marc: It is too early to take a final decision. I believe this point will be difficult to be settle. Need of reality checks.

Ian: I have ideas in term on generation, I can send it to the list.

Marc summarizes:

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/09/04 15:41:27 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.133  of Date: 2008/01/18 18:48:51  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found Scribe: Paolo_Baggia
Found Scribe: paolo1
Inferring ScribeNick: paolo1

WARNING: 1 scribe lines found (out of 160 total lines.)
Are you sure you specified a correct ScribeNick?

Scribes: Paolo_Baggia, paolo1
Default Present: +04989aaaa, Bjoern, Myriam_Arrue, +0160372aabb, marc, Paolo_Baggia, Enrico_Zovato, Ian, Jianhua, AndyBreen, felix, +49.381.402.aacc, Christian, +1.512.567.aadd, BillJarrold
Present: +04989aaaa Bjoern Myriam_Arrue +0160372aabb marc Paolo_Baggia Enrico_Zovato Ian Jianhua AndyBreen felix +49.381.402.aacc Christian +1.512.567.aadd BillJarrold

WARNING: No meeting title found!
You should specify the meeting title like this:
<dbooth> Meeting: Weekly Baking Club Meeting

Got date from IRC log name: 04 Sep 2008
Guessing minutes URL: http://www.w3.org/2008/09/04-emotion-minutes.html
People with action items: 
[End of scribe.perl diagnostic output]