Digital Signposts

Looking inside video...

Within current discourse on video for learning, there are many concurrent ideas on how to use video in educational contexts. These cover a wide spectrum, ranging from recorded lectures,  or the Khan Academy video tutorials, (which have inadvertently given rise to the flipped classroom meme), to dedicated educational platforms that aggregate online video, for example  Watchknow.org  Sometimes video is mixed with other content using APIs, as on the HistoryPin website.

Other platforms such as EdMediaShare which uses the Dial-e framework, (developed at the University of Hull), are designed to promote an alternative way of using video. However the most frequent use of video in learning  is still to illustrate or amplify subject matter or to present didactic information in the form of a recorded  lecture or presentation.

This is understandable, one of the drawbacks of using video, as opposed to text, is that it cannot be searched in as easily as texts.  Whilst it is straightforward to search search a library of documents for a word - sentence or paragraph but with video search relies on title or associated metadata. This may be about to change.  Research at the Hasso Platter institute has come close to making 'semantic video' a reality. Using  semantic media analysis;  in the near future, it will be possible to conduct searches  within video, (collections, and individual videos), directly by the content within them

Here are a few thoughts on how this might work, using some clips from EdMediaShare to illustrate:

•  OCR, (search for text in shots of maps or signposts) 
• Audio  (the stirring music, so common in old newsreels,  used to illustrate the Berengeria clip)
• Speech recogniton (look for topics with  specific terminology, eg, Mobile learning.
• Motion Vectors  (Tacoma Narrows)
• Visual analysis (identify structural  changes in video;  tone/colour/pace  as in this JISC resource on copyright)
• Face recognition (identify specific speakers in the  JISC collection) 

It seems likely that such tools will completely redefine the ways in which we can learn through video. Hopefully some of these tools them will also be found within YouTube in the not to distant future.

The Slideshare presentation from Hasso Platter gives a very detailed overview of the underlying process and there is also a demo screencast here.

Domesday Reloaded (Part 2)

Domesday System by ljw Licence CC  BY-NC-SA

Riding on the wave of excitement generated through BBC Domesday Reloaded project makes this a good time to take an objective look at some of the wider issues arising out of Domesday for learning technology, and if possible learn some lessons for the future.

“In 1986 the BBC launched an ambitious project to record a snapshot of everyday life across the UK for future generations. A million volunteers took part…”

This an interesting statement, because it was never realised through the original project format; but now may be with the release of the Reloaded project, and even more so if the data on the National Disc is finally released. Unless folk were associated with a school or had access at a library it is unlikely they would have had  hands-on experience of the Domesday system.

Many blogs and tweets suggest that Domesday Reloaded is a complete reconstruction of the original 1984-86 project, but that is quite far from the truth. It does not contain any of the rich data from the National Disc. The National Disc contains a great deal of, (authoritative), media and statistics. Just how much can be gauged in the first  section of this promotional video,  specifically from 03:15 > 9:32, on the timeline. It is clear that the Community disc data is culturally significant and  the National data disc completes that anecdotal snapshot of the UK picture with an even and focused coverage.

Neither is it true to suggest that the only reason for the loss of the Domesday data is obsolescent technology. Human shortcomings are just as much, if not more to blame. Although obsolescent technology may have contributed to the the loss of the Domesday system, it was not the only cause. This is evidenced by the fact that researchers based at Leeds University and the University of Michigan working on the Camileon Project were successful in accessing and emulating the BBC's Domesday project on a PC.

Although the systems were superseded by digital optical media, many of the surviving systems, (c.1000 sold), were just left in cupboards or as I heard through trusted anecdotal evidence - in one LEA (Local Education Authority), 2 were simply thrown in the skip when moving offices. Again the human factor... one wonders how many times this was replicated?

Evidence, from Charlesworth. University of Hull, (2002); suggests that the greatest obstacle was clearing Copyright. Nearly all the content, including both data and software has IP rights attached. It appears to be an extremely complex web of rights. Charlesworth’s report also indicates, even at the time of writing (2002),  some rights holders were untraceable and it could be 2090 before all data would be in the public domain.

Should ever such a venture be undertaken in the future (using public funds and contributions) it would make sense to ensure that content providers and contributors were asked to make their work available under a Creative Commons licence, or at least the BBC Creative Archive Licence.  Unfortunately this is not a requirement of the Reloaded initiative and there is no real licencing guidance for contributors other than the BBC’s normal T&Cs. (Even so, everyone should contribute because of the cultural significance of the project).  Similarly it would be hoped any future venture would be developed using Open Source software.

In addition to the  Domesday Reloaded will achieve is to generate awareness and of Domesday, (including the original Domesday Book, which is also online), judging by blog post and tweets this is already happening.  In order to fully appreciate the nature of the BBC Domesday Project it is worth reading this detailed description of Domesday and also the work of  Darlington, Finney and Pearce describing the recovery of the data on the discs. Sadly; Adrian Pearce died in 2008 and his original site is no longer available but the http://www.domesday1986.com/ URL now links to the Domesday Special Interest Group.

Update
There are indications from the Centre for Computing History that they are working with the National Disc and hoping to make it live in the near future. Great news if it happens.

 

If there are any factual errors in this post please let me know and I will amend as appropriate.

 

Visual Searching

Via Robin Good on Twitter

This looks stunning, just waiting for my invite now.

Bitmap to Vector Conversion (Yes online!)

I've been becoming rather blase lately about online image editing. With a good number of freely available tools such as Picnik, (my personal favourite), and Adobe due to launch the free online version of Photoshop before the end of the year, there is certainly plenty to experiment with.

However, VectorMagic (beta) from Stanford, made my day, and I think this Web 2.0  service opens up a new dimension in Web 2.0 graphic editing. Up till now I have found it necessary to use Adobe Illustrator to convert bit maps into vectors for editing, but I suspect this tool will make my use of Illustrator for basic tasks much less frequent, just in the same way that Picnik changed my dependence on Photoshop.

Not only does the tool deliver an excellent  conversion, it is fast and intuitive, making it a delight to use. With some good  basic editing tools, context sensitive help, and the ability to download you images as an .eps or .SVG this is a really useful tool.  When you add the ability to store your finished graphics online and share them, then this becomes a very powerful tool indeed. The potential for creating re-usable vector 'assets or learning objects' is not to be sniffed at.

In my experience, working with schools,  teachers often shied away from vector graphics. So if it is necessary to explain " the difference between bitmaps and vectors" in  enjoyable way, then this is the bees knees.

Rethinking information handling

Exploring the fundamental changes in the way we work with information, Information R/evolution (above) is the new video from Michael Wesch, whose other works include the viral The Machine Is Using Us, and more recently,  A vision of Students Today. Combined, these videos set out some fundamental and inescapable challenges that education systems must address. It is worth noting some business professionals are already aware of the implications. (via The Obvious)

Incidentally, I installed beta version of Leap on my Mac today and the need for folders is becoming less critical by the minute. T

Seam Carving - It really works!

Rsizr The intelligent image resizing tool,  I mentioned in my August 31st  post is now available as an online application.  It takes bit of getting used to, but it really does work. The secret is to use the selection tools to choose which areas of an image to protect, or even remove. My first attempt is shown above. I look forward to spending some time exploring the creative possibilities, and seeing the work of others. Its also quite fascinating to watch the image processing take place before your very eyes.

Digital imaging moves to the next level

Just when digital imaging applications seem to have reached a high water mark, new developments appear at Siggraph. The video above (via O'Reilly Radar) shows some stunning developments in resizing images without cropping and scaling. The last sequence will add completely new perspective to the debate around 'removing' data from images.

Every cloud could have a silver lining?

Image credit Old Shoe Woman, CC License.

With large numbers of schools, in the North of England  unusable or severely damaged due to the recent floods, perhaps this is a good time  for education authorities and schools to explore how new technologies and social software might help alleviate or avoid some the problems, and provide a meaningful learning experience for affected pupils in the months to come.

It appears that up to 10 million pounds has been made available by the government's  Department for Schools, Children and Families, to support; "temporary accommodation, extra surveyors and summer activities for children flooded out of their homes."

Local authorities could  also consider using some of the 10m for developing a strategic approach, rather than  just fire fighting, bringing together stakeholders and interested parties to explore the possibilities; for example:

• make and laptops, (school stocks and buy extra), available, install wireless networks and connectivity in centres and homes.
  
• develop an 'emergency' online curriculum using appropriate web 2.0 technologies and social collaborative  software. There is a great deal of experience available locally, regionally and   internationally upon which they to draw.

I don't claim to have all, (or even any of the answers - yet), but I am sure with a degree of will and vision, then this, must surely be something worth trying, and could also serve as a real test-bed for 21st century learning. The alternatives could be pretty grim.

Busy day at Google

Yesterday was a busy day at Google with the release of My Maps and a Mac version of Google Desktop.  My Maps makes it easy for anyone to create their personalised Google maps, by simply positioning a marker and adding text or media to create an interactive map.  I made this in ten minutes. The marker listings on the left are neat, but can't be rearranged, which is a shame.

My Maps includes line and shape drawing tools and the maps can be saved as KLM files for use with Google Earth. However, I think there are there are a some features missing which prevent it from being as useful as it might be. The most important is the absence of any embed code, it would also be nice to  measure distances between markers. An RSS feed for updated maps, tags and  tools that enable collaborative work on maps, (think classes), are also be on my wish list.

There are some problems with  video from third party sites. Google Video and YouTube work fine, but Videoegg did not, and the HTML disappeared, others have also commented that Google strips some HTML elements. Having said that clips from Veoh did work, so I suppose it's trial and error. A final worry is that this may have detrimental effects on some mapping services such as QuikMaps which provide  excellent  tools and support. Whatever the benefits of My Maps we need these third party innovators

I believe My Maps will strengthen the case for using Web 2.0 and social software  in schools. Because it is so quick and easy to use, teachers and students should find plenty of uses, in all subjects and areas of study.

Distant Voices

Screencast compilation from voicebook examples

Having just spent a few hours exploring  voicethread beta site, I am convinced that this is a digital storytelling  application with enormous potential for education and learning. What makes the service special is, it combines personal creativity and social interaction. You build a voicebook digital slideshow by combining your images with an accompanying narration. Whilst listening to the narration viewers can zoom and pan around the image. The whole thing is superbly structured with really useful interactive features including:

• audio comments contributed by viewers
• facility to add text, drawn annotations and hyperlinks
• sharing with private and public settings
• full integration with Flickr

The icing on the cake is that potential classroom use has been carefully considered by the developer, who has provided information and advice on setting up a class Voicebook. I recommend having a look at some of the excellent voicebooks already on the site.