Chat log for LAK11 Weekly Check In

January 28, 2011

Moderator (Tanya Elias): dragon and kimberly seem to be going in very different directions

Mary McEwen: @ tanya -- exactly, but I think that kimberly was more along the lines of the presentation from week 1 -- right?

Moderator (Tanya Elias): yeah I have trouble tying the two together

Wolfgang Greller (Netherlands): @Jon did you get my connection to Frank Rennie via Facebook? He wanted to get in touch with you.

Moderator (Jon Dron): hi Wolfgang - I seldom check Facebook - will look

Barbara Dieu: whose intelligence, as well - as after all curricula are made by people

Wolfgang Greller (Netherlands): Cheers. I can also send you an email.

Moderator (Jon Dron): that would be good, thanks!

Moderator (dave cormier): learning styles? really?

Moderator (Jon Dron): Sabine Graf has done interesting work on adaptivity using lerning styles

Moderator (Jon Dron): but I remain highly sceptical

ines cambiasso 1: echo

mpaskevi: hi all, i would really love some feedback on my analysis of #lak11 tweets using nodeXL available here: http://www.bluelightdistrict.org/wp/?p=4606

Moderator (dave cormier): i did

Moderator (George Siemens): Amazon does it already, Facebook, etc

Barbara Dieu: filter the data

Moderator (George Siemens): yes, agree, Dave

Moderator (Jon Dron): absolutely

Moderator (George Siemens): exactly

Wolfgang Greller (Netherlands): is the intelligence not in the prediction of what's coming next?

Barbara Dieu: it would have to be extremely fine grained to adapt to each person

Moderator (George Siemens): web of knowledge, web of science, "knowledge is in the connections" type of stuff

Barbara Dieu: expertise is the knowledge in practice

Moderator (Jon Dron): there are other things you might want to predict

Moderator (dave cormier): my paths are often not sensible

Moderator (George Siemens): Peter Norvig looked at prediction and big data

Moderator (George Siemens): last week's reading/resources

Moderator (George Siemens): learning as computation

Moderator (dave cormier): wow

Moderator (Jon Dron): again- helping with what to do next is easy. But if we want to plan a larger learning project it becomes unworkable

Wolfgang Greller (Netherlands): what i dislike is that this prevents discovery (i.e. coming across interesting things by chance)

Moderator (Jon Dron): @Wolfgang - serendipity can be built into the system

Wolfgang Greller (Netherlands): true.

Barbara Dieu: exactly @Wolfgang

Moderator (dave cormier): always makes me nervous when he addresses my questions

Moderator (George Siemens): concept map physics

Barbara Dieu: once you set a path, people normally follow it and do not question

Moderator (Jon Dron): I found that people tend to like to click on things at the top of a list AND things at the bottom - the latter being presumably the least useful - so people adjust to stupid systems that prevent serendipity

Barbara Dieu: agree @jon

Wolfgang Greller (Netherlands): I still am not entirely happy with things like Google I fell lucky style of serendipity.

Moderator (George Siemens): can you build serendipty into the system?

Wolfgang Greller (Netherlands): Sorry it's called Google I feel lucky.

Wolfgang Greller (Netherlands): Yep, agree

Moderator (George Siemens): others?

Wolfgang Greller (Netherlands): User wants to feel in control not being controlled

Barbara Dieu: Low-quality content has some Internet users worried about the relevance of Google search results.

Moderator (Tanya Elias): predictive models are only as good as the theories they are based on

Barbara Dieu: http://www.informationweek.com/news/security/vulnerabilities/showArticle.jhtml?articleID=229100049

Moderator (Jon Dron): control <> choice - people want to delegate control when they don't know enough

Wolfgang Greller (Netherlands): disagree, not being controlled, being guided and organised.

Moderator (Tanya Elias): show me how

Moderator (dave cormier): @wolfgang think Apple.

Moderator (Jon Dron): being in control means giving control to others when we do not have enough knowledge to make a sensible choice

Moderator (dave cormier): @wolfgang their competitive advantage is often about 'lack of choice'

Barbara Dieu: true @Tanya

Moderator (Jon Dron): but also being able to prevent others being in control when we *do* hvae enough information

Moderator (dave cormier): its the epistemic question.

Barbara Dieu: this is frightening @george

Wolfgang Greller (Netherlands): @dave indeed. but Apple don't only makes friends with this type of patronisation it uses.

Moderator (Jon Dron): Judy Kay has done interesting stuff on srutable adaptation that is relevant here - about not only giving choices but making it clear what the effect of those choices might be

Barbara Dieu: ideologies do not belong to the technology but to the social/cultural/political surrounding it

Wolfgang Greller (Netherlands): On the other hand, at some stages users/learners need to feel secure

Barbara Dieu: and the groups in power

ines cambiasso 1: the ideology behind is not clear to me

Wolfgang Greller (Netherlands): secure not controlled.

Moderator (George Siemens): don't invite me to your cocktail party - i'll kill the conversation

Mary McEwen: totally agree w/George, an absolute HUGE need for awareness/understanding of the basics by even the general population.

Andreas Link: @mpaskevi: it looks interesing. after a closer look a will write a comment ...

Moderator (George Siemens): exactly

Wolfgang Greller (Netherlands): Yes, big worry

Moderator (George Siemens): Dave is a semantic web denialist

Wolfgang Greller (Netherlands):

Moderator (Jon Dron): semantic web is designed for messiness - not like DC

ines cambiasso 1: echo

Moderator (Jon Dron): but can have multiple and conflicting structures

Adam Weisblatt: Are you saying that the general public needs to understand statistics? But pundits et al base their power on that lack of understanding.

Moderator (George Siemens): @ines - log out of one of your elluminate profiles -you have two windows open

Moderator (dave cormier): @adam not necessarily statistics...

Barbara Dieu: There is an excellent paper by Langdon Winer called Do Artifacts Have Politics?

Barbara Dieu: http://www.jstor.org/stable/20024652 .

ines cambiasso 1: than you George

Moderator (George Siemens): no prob, Ines

Moderator (Jon Dron): Hendler?

Moderator (George Siemens): no worries Jon - dave is King Monopoler

Mary McEwen: as far as serendipity (sorry-much older thread): can't we take a page from biology, and inject (semi-random?) "mutations" that may bring us to something new and better (or new and worse) and leave it to natural selection as to whether it grows/reproduces or not?

Moderator (George Siemens): exactly!

Barbara Dieu: human beings should always be central

Moderator (dave cormier): lol nice description of LOR there :P

Adam Weisblatt: I'm confused. MY personal data in the Census database is used for demogrtaphic analysis without my ethical concerns. Why can't learning data be analysed without harming learners

Moderator (Tanya Elias): may work in social but might be dangerous in learning

Adam Weisblatt: In both scenarios, it is humans that are teaching the machines right?

Moderator (Tanya Elias): yep

Moderator (George Siemens): http://richard.cyganiak.de/2007/10/lod/

Barbara Dieu: "intelligent machines"

Barbara Dieu: this is a game for kids that learns from our answers http://www.20q.net/

Adam Weisblatt: Semantic Web sounds like just tagging on steroids

Mary McEwen: tags don't have relationships (explicitly defined)

Adam Weisblatt: I like the idea of triplets: Object, predicate, subject. I always thought it was a limitation of relational databases that you needed a new table if you wanted metadata about relationships

Adam Weisblatt: That's what I thought. It is hte relational data that takes semantic web beyond tagging

Moderator (George Siemens): ok

mpaskevi: Thanks everyone, enjoy!

Adam Weisblatt: I always felt that learning was entirely about integrating an understanding of the relationships between ideas, so I can see the relevance here.

Mary McEwen: but don't limit it to "understanding" relationships - but also discovering/creating new relationships.

Adam Weisblatt: right

Moderator (dave cormier): I think i now know what my problem is...

Mary McEwen:

Moderator (Jon Dron):

Moderator (Jon Dron): absolutely

Moderator (Jon Dron): the technology is entirely about machines talking to machines

Moderator (Jon Dron): but of course technology is not just tool but what we do with it - that's what matters here

Adam Weisblatt: But how does the metadata of relationships in people's heads get into the machines?

Moderator (dave cormier): http://www.elearnspace.org/blog/ he would show you this

Moderator (Jon Dron): @Adam - yes, that's the biggest problem with the semantic web

Barbara Dieu: A nice person to have to explain what Semantic Web means is Professor Ivan Herman

Barbara Dieu: I saw a presentation of his here in Brazil on the Semantic Web - clear and concise

Adam Weisblatt: So if I could invent the next Delicious but instead of tagging, I get people excited about contributing semantics, that would be cool.

Barbara Dieu: http://homepages.cwi.nl/~ivan/CV/Publ.html

Moderator (George Siemens): @Barbara - thx for the link!

Wolfgang Greller (Netherlands): Annotation tools can partly automate the process.

Moderator (George Siemens): good point, Jon

Moderator (dave cormier): i've always thought that connectivism was a good learning model

Moderator (dave cormier): :P

Barbara Dieu each one of us will connect the pieces in a different way - depending on our background, interest, lived experience and moment in time and place

Adam Weisblatt: In a corporate talent mgmt system for a decentralized organization, could semantic web technology be used to align competencies?

Moderator (Jon Dron): @Barbara - absolutely true. But sometimes it's handy to get assistance from someone who knows how to teach - the problem with most RDF approaches is that they are simply content experts - they don't know how to teach

Barbara Dieu: @Jon - patterns work for tutorials, training

Wolfgang Greller (Netherlands): Trouble is that ontologies are static and so only provide a snapshot.

Moderator (dave cormier): nope. v. useful talk for me

Adam Weisblatt: Thanks all

Moderator (dave cormier): hope it was for others

Barbara Dieu: thank you

Moderator (Jon Dron): nope, good discussion

Adam Weisblatt: very useful