ALA Midwinter - Denver, CO
I arrived yesterday in Denver for Midwinter. First on the agenda was OCLC's Symposium From Linking to Thinking: How we’ll live when information surrounds us.
David Weinberg and Nova Spivak were the guest keynotes to talk about the future of the web. When it came to the Q & A section, it became a lively discussion with some folks asking about their experiences and knowledge of what libraries do. There was some validity in the questions but I think it the best thing I got out of the talks was the information about the growth and movement of the web and where it's heading. We as librarians need to know this stuff. I'm not hurt or take it personally if they say things like "metatdata will become the data" and "focus on the findings rather than the search" and "humans will be removed from the search process". By staying on top of this information we can bend and stretch to the needs of people for access to information as we always have. I think we are doing a good job of breaking the archaic structure of what a traditional librarian does. The point is that things will continue to change and we can be creative in the ways we manage and organize information and guide users. I'm excited to hear what they had to say. I'm excited to hear where the web is going.
So here are my notes from the session (not proofed for spelling, sorry):
Friday, Jan. 23 – OCLC Symposium
From Linking to Thinking
How we’ll live when information surrounds us
Roy Tennant – moderator
David Weinberger, technologist
Libraries in an age of abundance
Ebooks will help move books from a private activity to a social activity – one moment away from this transformation
We’re now entering an age of abundance – we get an abundance of good stuff and crap – mostly dealing with the crap – manage it ok – but now there’s an abundance of good
Libraries help us navigate through the content…. They’re symbolic, they symbolize – rich or poor you all get access to the same stuff
It makes sense to have a building to collect stuff – collections have value, well organized content – there is a place for knowledge and culture
Not having links out is an act of stinginess these days, books are stripped of their value because there are no links out to other content, more info.
Need to be able to manage this abundance – the information produces more info – metadata, info about the info – it no longer needs to fit on a 3x5 index card, we can add info as much as needed
The metadata connects everything, searches can bring back everything including the kitchen sink, metadata is the level to pry open the data
The old ways of doing things don’t scale, allow the public to add info, can’t do it ourselves, we need to do it together – we have to do it together because it doesn’t scale, of course it’s not going to be quite right, made by humans,
Metadata frees us to learn more
In the world of abundance, good enough is good enough. Some exceptions, medical, air traffic control, good enough is not good enough but most all else, with so much info, good enough is good enough
Knowledge isn’t the highest level, understanding is – we know how to do knowledge, we’ll keep doing it, symbolic roles – equal access, knowledge is scarce
Nova Spivack – semantic web pioneer and publisher of the Twine search/sharing tool
What does the semantic web mean – the social graph just connects people
The semantic graph connects everything
Metadata view - you can reduce everything to little bits of relationships
Everything is becoming metadata
We’ll be building a higher resolution web
The data carries it’s own metadata – embedding it
The web becomes a database – in the semantic web the data is self describing, my program can go in and grab your info without having to go through an api/programmer to request the info
The idea of the semantic web is software can understand the web, no human involvement – we have to get humans out of the loop – if we have to do it all it’s not going to scale
We can leverage that by enabling different applications and people to learn about a program and embed the metadata
The web – web 1.0
Social web – web 2.0 – front end web, create more interactive programs
Semantic web – web 3.0 – entering into this period, the front end isn’t going to change all that much – this will be back end change -
Web os – web-based apps, web starts functioning like an operating system
Intelligent web – web 4.0 – intelligent agents, ubiquitous agents that are embedded into your phone that tell you what to do
Five approaches to semantics:
Tagging – anyone can make a tag – very easy to abuse, massive numbers to help weed out the junk
Statistics – trying to estimate the value of a page
Linguistics – natural language processing – artificial intelligents to retrieve pages
Semantic web – all about metadata, how you get the metadata is whatever
Artificial intelligents – read and interpret the knowledge
Approaches : make the data smarter/make the software smarter
Grow each area to eventually get to A.I.
Future of information retrieval:
Productivity/amount of data
Moving towards leveraging the social web – we use people as a search enginge as we move to web 3.0 we want to combine that with the web then move to reasoning (web 4.0)
How to add semantics:
Manual: Hard/Precise – individual hand – tagging by experts – easy/fuzzy: large group collective tagging by non-experts
Automatic: hard/pprecise: database integration-taxonomy management easy/fuzzy:
Libraries that learn
Digitize everytning and putit all online
Automatically self-assemble and self-organize
Personalized interface for each individual and group – giving everyone a personal card catalog – and then learning from that – recommendations and filtering becomes very important
Intelligent assistance (human and automated)
Detect and adapt to changing needs and usage
Q & A:
What are your impressions of libraries, conceptions?
-David: Libraries are, increasingly, maintain their value. Given money to spend is proven support. The tools they use, I’m somewhat familiar.
-Nova: I think catalogs are irrelevant, my observation, is I’m an information junkie, it had tremendous value in college, today, it’s inconvenient to get to when going to google is good enough, bringing the experience of discovery to the web is invaluable, having libraries do that is where it’s at. Exploring on the web – today, exploring the stacks – back then. My experience was focused on knowledge and discovery.
-David: going through stacks, discoveries, it’s a good thing, physical things can be frustrating
Nova: keyword searching is extremely inefficient. If there could be some kind of agent or automated process. Natural language interface would be great, dialoging would be better. Need to put the emphasis on finding not the search.
David: will typing words into a little box ever go away? Maybe, eventually, but it will get better. It will uncover better knowing what we want to find and know what we didn’t want to find.
Nova: hunt and peck v. map and compass, we need a little bit of both – this is what we are moving to more, keyword searching with faceted taxonomy. Give people a sense of location and place.
What role will mobile phones/technology be?
Nova: gps/locationware with data overlays – augmented realities, resources relevant to location – location awareness – overlaying data – geo tagging – talked about embedding mobile chip in contact lens, now, not so great but in 20 yrs from now, we’ll have photo quality visualization in the contact lens
David: information put in by social networks is a good way of filtering relevancy of info
Nova: the mobile device platform is already larger than the pc platform. How should mobile devices change to use them the way we need to? That’s what we should be thinking about. They’re wallets, they’re ebooks, they’re games, this notion of being able to carry around with you everything you need is great – having a mobile assistant. Acess might be very different, we need to think about dialog-based speech interfaces, mobile devices are ideal for that.
I would encourage you to take a broad look at what is meant by search than what is found.
David: we focus on the search box when we’re searching, provides us with endless links to traverse. Sometimes we find, sometimes no. we link and click and on to the next thing, that isn’t finding info it’s exploring. The real searching we’re doing is link to link to lnik, not the search box.
Nova: the inefficiency is deliberate (google) because there is money involved in the number of clicks per answer, might lose revenue, user.
David: aren’t ontologies put together by humans suffer the same effect as sticking with your same social network?
Nova: google is nothing more than a statistically based pile of data, most popular. Social search says don’t just search your own network search the networks that have the expert places or specialized networks. Ontologies are very biased, no way around it,
Sematic web – oil – ontology integrated language, mapping – what does it mean. Say want is inferred. There can many ontologies as desired and can map between them then link and integrate
What are the skills sets needed in the info system?
Nova: one of the good things is you won’t need too much about programming, you will be able to create metadata without, there will be people underlaying that but it will become more available. Discovery, social content is moving more to the individual. I think we will see twitter and other social places be the place of info discovery.
David: skills are required, typing, its familiarity with the social tools, yes but also being involved in smart social networks, being in the right twitter stream, you’re smart because you’re increasingly smart becase you’re in a network.
Nova: soon we will reach a point we’ll have more computing power than the brain has and do some amazing things and augment humans. Today the burden is to be like machines and we need to get rid of that. To go forward the intelligence in not inside, it happens elsewhere, leverage a network.
David: Andy Clark – Being There – really good book.
What can libraries do to become more relevant?
David: I actually do research for things I write, I have access to Harvard library, digital content, things that would make things easier is – scan the books already, databases, ejournals are a nightmare, can’t search across the whole thing, need more open access. If we could get this life for researchers would become easier
Nova: if I had a digital library and wanted more people to use it I would link in to google because ubiquitous is important.