Stephen Downes
February 11, 2005
What social networking - and online learning - shouldn't be
When you spend $50 million on e-learning, you expect results. That's not what happened in the Los Angeles Unified School District, which purchased Pearson Education's Waterford Early Reading Program four years ago only to find after a study that the software didn't help, and sometimes hindered, student learning. But as a Pearson spokesperson says, "The findings confirmed what we already knew: you have to turn it on to have an impact." According to studies, teachers didn't have enough time for the computer program because they had to cover a reading curriculum introduced by the district a year before.
This presentation is based entirely on posts from the last week of OLDaily
This same pattern may be observed in the previous week, and the previous week...
Bill Williams forwards this item from last October; it is a description of Scotland's Edinburgh's Interactive University (IU) which, over the past 18 months, "has attracted 75,000 students from more than 23 countries. In sharp contrast to the failure of its English counterpart, UkeU, which had signed up only 900 students when it was scrapped in June, the IU has seen a 75% increase in student numbers over the past year." The differences? The article highlights the personal contact between students and teachers, the ability of students to learn at their own pace, and IU's non-profit structure.
I understand why someone would say this: "To increase the sustainability of portal projects there is a need to 'work towards establishing common frameworks that will enable applications and services, from different sources, to work together.'" After all, it is precisely that failure that accounts for the indifferent success of community portals, the 'field of dreams' scenario, where you build it, and they do not come. But such an enterprise is perhaps best compared with constructing an artificial language: sure, it would make communication easier if evereyone used the standard - but who speaks Esperanto? The growth of community - and hence, community frameworks - is much more organic than that, a product of multiple simultaneous negotiations to create a network of compatible systems rather than a centralized planning department to create a structure.
Exactly the same debate is playing out in the fields of digital rights and authentication.
Publishers Irritated by Google's Digital Library In a press release formatted to look like a news article, Nature Publishing Group is expressing displeasure with Google over its plans to index and excerpt academic articles. "Google has not yet struck any legal agreements with publishers, either individually or collectively... Few publishers would want to opt out of the library scheme, Morris says - but they need to be asked to provide the appropriate permission." This is probably not true; copyright law is not intended to prohibit such use. Google excerpts everything else on the web and provides thumbnails of images. There is no reason to expect permission would be required to cite a few lines of an academic article. But of course, if Google signs an agreement then the publishers can exert pressure on others (such as, say, myself) who don't have "800-pound gorilla" lawyers to defend their rights. Via PC pro, which provides a link to some background.
Scott Matthews proposes DRUMS, a database of creative works and relevant metadata including, crucially, rights and permissions. "The intention is to enrich the environment for creating and consuming digital works by enriching the environment for developing new applications and services to interact with these works." Discussion.
The Open Digital Rights Initiative (ODRL) and the Dublin Core Metadata Initiative (DCMI) have formed a working group to develop an ODRL/DCMI profile. As described in the announcement, "The profile will show how to make combined use of the rights-related DCMI metadata terms and the ODRL rights expression language. This will enable richer rights management information to be captured along with DCMI descriptive metadata and support wider interoperability with DRM and open content licensing systems."
There was a social networking meme that went around a few weeks ago called '43 Things' - the idea is that you would form online groups dedicated to doing one of the 43 things would would like to do. I thought it was interesting but not really applicable to online learning, so I didn't cover it here. Now it turns out that the whole thing was a front for Amazon.com. Well. I don't know what to think, except to observe that companies are becoming increasingly sophisticated in their online efforts.
The pattern explained:
Learning Networks
First pass at a theory of learning environments
We are moving toward a system defined by three major components: a means of organizing input and experience, a mechanism for putting that experience into context, and a means of creation, of becoming part of someone else's experience.While I would disagree with details in just about every one of the twelve points I am in general agreement with the overall thrust of the article.
Sensible Design Principles for New Networks and Services - The abstract states, "Based on an analysis about the reasons for past failures, we propose three rules for sensible network design process to avoid useless development efforts. First, the analysis of customer needs has to concentrate on practical uses that are likely to become everyday routines. Secondly, the development of a new technology must be based on well-defined, carefully selected core principles. Thirdly, during the development process the real experiences in real networks must be continuously taken into account." Or, as the author concludes: "As to the list of core principles, simplicity and realism are essential." Examples abound; counterexamples, sadly, also.
They told me my philosophy courses would never be useful, but this could have come straight from my metaphysics class: "People don't exist in environments, they exist in themselves and their semilattice-esque relationships with other actors (communities, individuals, spaces, inanimate objects...)." Why is this important? When we are creating a learning system, we are creating a mechanism that allows people to interact with, to experience, and to learn from the world. But what is the nature of this system - is it an environment, as Lisa Kimball suggests, or is it the lattice of relationships described by James Farmer? I submit that it is a complex construction that enables, first, a wide variety of experiences akin to (and possibly extending upon) our experiences of the natural world: things to see, touch, do, and otherwise sense; and second, a mechanism for interpreting and comprehending this experience, a syntax, putting it into a structure, a semantics, which assigns it meaning, and a pragmatics, that gives it a context and use.
The pattern explained:
Connectivism
The self-organizing network
Simply blasting a million learning objects - or blog posts - at someone creates nothing but noise. What must happen in the ether is what happens in nature: coherent bits of information self-organize into coherent clusters of related phenomena - a mechanism of emergentism.Access to blog post content informs search results.
From Roland Tanglao (just to prove I'm not the only person saying this): trackback is dead. "Trackback is broken, and I concur. It was broken right from the start, but we didn't know it because it seemed to work, or at least, work the way most people thought it should work." I don't know whether PubSub will be the long-term answer, but something like it will be: the principle is that you search (where you want, when you want) for sites linking to you, rather than depending on them to tell you (with dubious honesty) that they are linking to you.
I liked this paper a lot though I admit to rolling my eyes at the Hobbes-Rousseau opening. Still, this is fundamentally right: humans have developed the ability to reduce what might be called the transaction costs of communication through effective internalization of social conventions, such as the use and recognition of language, behaviours and other forms of interaction. Historically, because of the difficulty of communication, this has limited our social sphere to about 150 people; beyond that, and instead of the informal mechanisms we employ to, say, build trust, a more formalized and usually hierarchal communications system is needed. The development of effective peer-to-peer technologies, however, has the effect of lowering the transaction costs of communication, essentially allowing us to increase our social sphere. Trust me.
Cut this out and paste it to your wall: it's easier to filter for what you do want than what you don't. That's why I agree with this: "Eventually, we'll need to move to use social networks to our advantage to include FOAF in an email solution that filters spam. Most current current filtering systems work on identifying spam and then let everything else through. We need the reverse: a method of authenticating/identifying good email and block everything else."
As Robin Good summarizes, "a growing body of literature on social norms, social capital, common property regimes, and the emergence of peer production, outline the contours of social sharing as a third mode of organizing economic production, alongside markets and the state." Here is the original article in the Economist.
Christian Spatzierer sent me an email yesterday inviting me to try his company's product, Staff Online, with an eye toward e-learning applications. The idea is that by clicking on a link on a web page you are put into a videoconference with a representative - which could be customer support, an instructor, a mentor, or a tutor. I tried the system, which is based on a Flash interface - this means I didn't install any software, didn't need to do anything, in fact. Using my headset (there is a text window if you're not connected for sound, and their video camera works even if yours doesn't) I chatted for a while with an Staff Online representative from the company's office in Montreal. The sound and video quality were great, though there was an echo when I spoke. The service also supports web touring, which means the representative can show me web pages while she speaks. Were I still tutoring for Athabasca University I would have traded in my telephone for a system like this in a minute (it would probably have been cheaper for the university too). The system is already being used by various community colleges in Quebec. Now it seems to me too that because web pages are used to host the connection, such pages should be thought of as learning objects, and made available through e-learning content syndication networks.
Blog Digging
Durl and Blogpulse-- More Link Digging Tools
Scott Wilson pushes the concept forward a notch, discussing the array of services that would be needed to support an online course - or as he has decided to call it, a "shared learning context". After listing the various elements - resources, forum, portfolio - he then matches the requirements against the e-learning framework. Existing specifications - RSS or Atom, FOAF, iCalendar - are sufficient to provide most functions. ELF, of course, interprets each of these in its own way. Where he runs aground is in assembling these services - "it would make quite a long list of little orange XML buttons!" Well yeah, but instead of IMS Enterprise services, or autodiscovery, why not use OPML, which is designed for exactly this task? See, the thing is, all this stuff being 'discovered' by ELF has already been encountered in the RSS and blog worlds - there's no need to reinvent the wheel here.
The relation between the self-organization in nature and the organization of one's own thoughts is not a causal relation, however - what happens is that a process of recognition takes place - because nature is not inherently organized, but can only be seen as being organized.
The pattern explained:
Resource Profiles
Meaning and context
James Farmer notes that an article such as this that points out the downside of educational blogging "is as valuable as 10 papers on why blogging is great." As with any other medium, student "more or less said that they needed the direction of a teacherly assignment to write, and they weren't going to 'just want to write' in a blog space." For this and other reasons, the blogs simply did not generate the community discussion the instructor had hoped.
Perceptual processing, metacognition and metaphor.
CiteULike, a system that "lets you build a 'personal library' recording bibliographic information and enabling you to tag papers for future retrieval and group sharing." Of course, that's how OLDaily started - as the output from the links I was saving for my own reference. A homegrown system. Now such a system is useful - it looks really useful when you see the ten-item version Seb posts - but my database now contains 8,271 items collected over the years. And I have found that, while I use my system, I tend to use it rather less than I might when, say, writing a paper. And consequently, there are many references in my papers that are not in my database, and vice versa. Collecting useful cites is easy - Edu_RSS now has more than a hundred thousand useful cites. Using them is harder - what I am after is an engine that will read what I'm writing as I write it and suggest appropriate references. Metadata? Well, no, I'm not going to tag a hundred thousand objects. What I am after is a way to capture the context in which these references were used, and then to create a cracking good search engine that recognizes when my writing has entered a similar context.
Like this: "The participants become members of the volcano, hurricane, evacuation or communication team during the two-hour, electronic mission. Operation Montserrat engages each participant to work as a scientist in order to solve problems in real-life situations. The mission challenges participants to apply their mathematics and science knowledge to a real-life event."
Folksonomy?The phenomenon known as tagging - that is, the use of user-created keywords (or 'tags') to classify digital content. The resulting (and some say 'emergent') organization of categories is sometimes known as a 'folksonomy'. Tagging is used by sites such as Flick, del.icio.us and Technorati. Worth noting: tag spam. "These 'self-organising effects' aren't always benign. Some bloggers showed that the 'teens' tag on Technorati brought together innocent photos from Flickr and links to pornographic sites." More on folksonomies here and here via George Siemens. And news that Lulu plans to adopt folksonomies to make its catalog more accessible.
"Creating metadata for text has gone from tedious to insignificant." If this isn't obvious, it should be. From where I sit, the text is the metadata - it describes itself. So where is metadata useful - photographs? Video? Well it would be, if people filled it out well. What would be better, proposes the author, would be if there were a moratorium on metadata. Instead, we should embark on a project to differentiate multimedia content without metadata - in effect, letting the pciture be its own metadata just as the text is. Good article, via Scott Leslie.
In a sentence, here's what's wrong with predefined ontologies and taxonomies: "we use existing patterns to search the data, which can't turn up new patterns." In another, here's what's wrong with manual tagging: "his department generates 5 million new objects per month, too much for manual tagging." Now if you take these two points as given, as I do, then what follows? How would you approach metadata and design?
I was talking with Chouki the other day, and what I said was, "Wouldn't it be neat if we could take every page in Wikipedia, do an associative analysis of the contents of each of the million plus articles, and use clustering algorithms to create a genuine dynamic folksonomy, rather than the highly artificial (and hence, unstable) structures generated by tagging." His response was, well, you could do that, but you wouldn't need to analyze a million articles; a subset would do it. And then he outlined some of the algorithms that would support such a system. Well, from where I sit, it appears that the people of Google are thinking much along the same lines, as they are providing funding and equipment to support the world's largest encyclopedia.
The pattern explained:
Public Policy, Research and Online Learning
Roland Tanglao says, "just like VoIP will be part of every app so will Post To Blog." In other words, content authoring tools - such as MS Word, Excel, Powerpoint, and opthers, which now simply save to file or print, will (and should) be able to post content to any online service, such as Blogger or Flickr. These online services have APIs ready to receive content. But who will write the bit in, say, MS Word, that sends the content? Microsoft? Not so long as they're striving for ubiquity. But who would invest in such a tool, when Microsoft could put them out of business with a shrug? Still - the concept is right, and eventually, desktop tools will become syndication tools.
Blogs @ Middlebury College: Blogging in learning as used by Middlebury College in combination with a content management system (CMS). Students get "the sense that there is a real professional holdingeverything together; this is essential in a writing course where, typically, students feel vulnerable and are apprehensive about sharing work."
The ongoing technological revolution described in this article, a revolution that is sweeping through the internet as we watch - podcasts, vidcasts, Skype recording, and more.
Richard Roth sends this link to a nice set of clear instructions on how to create a group mobile audio weblog is well worth a look - the long distance charges will still deter many, but long distance charges are about to become a thing of the past, opening up worlds of opportunities.
Recording Online Audio Interactions - The Easy Way? Well I had a lot of fun trying to make the setup described in this item work, bothered Alan Levine, but ultimately could not make it work - my laptop, an older Dell Latitude, apparently does not support microphone boosting (because it uses the limited Crystal drivers instead of the better Intel ones), so I can record someone else's conversation, but not my own voice. Still, if you have a new system, this might be worth a try.
Good overview article on podcasting with a number of links to definitions and references. The author describes his own use of podcasting, and in particular lists a number of sources for podcasts of scientific talks and shows.
Some rounding up to do in the world of podcasting. In this item Doug Kaye links to some sites offering tips and advice on sound recording (the whole Blogarithms site is well worth the time - and I really like the site design). Matt Pasiewicz also offers some links to Audacity tutorials (Audacity is free sound recording software). You'll also find on the EDUCAUSE blogs site various links to numerous links to talks and interviews recorded at the recent NLII meeting and posted for all to hear - here, for example, and here and here and here (and more - explore the site). I think they've done a really nice job with this - I've been sort of sitting on these links because I wanted to listen to everything - but there's no way
"ANT helps you download and watch video published on the Internet. ANT allows you to organize and manage video playlists. ANT is a video aggregator that allows you to subscribe to RSS 2.0 feeds with video enclosures. ANT seeks to build opensource software tools to enable an emergent, grassroots, bottom-up, video distribution network based on existing technology such as weblogs and RSS."
The pattern explained:
Educational Blogging
The semantic social network: a redux
Three things: content, coherence, communication.This is the sort of model that should be encouraged and developed for e-learning. "Open Journal Systems (OJS) provides online management for journal submissions, peer reviewing, editing, and online publishing and indexing. Open Conference Systems (OCS) manages conference registration, programming and paper submission and publication. The PKP Harvester (PKPH) is used to automatically create an online index of materials from a variety of online sites including journals and repositories."
It's the start of another new school year in Australia and as teachers sit down to plan the new year they can't go wrong if they being with the EdNA newsletter, a fantastic resource that in the space of a few hundred words puts teachers in context, connects them to resources, and gives them something to think about. It's hard to find a better example of online learning than this.