Semantic Library

We’re well into 2009 now, but there’s always time for trends!

Kathryn Greenhill at Librarians Matter recently compiled a few of the trends that have been making the rounds on library blogs last month.

Cloud, Semantic Web, and Linked Data abound in predictions and in posts emerging through the first two months of the year. It’s looking to be a big year for the Semantic Web in libraries. Add this to the growth in data curation, digital preservation, and eResearch in libraries and there are interesting times ahead. Most of these issues and trends are also mentioned by other contributors to LITA’s 2009 Top Tech Trends.

So what are my predictions for the rest of the year? I’m going to be conservative and just name four. What are yours?

1. The cloud goes desktop

People are getting increasingly worried about losing what they have created. Whether it’s Facebook changing Terms of Service so that even stuff you delete is still owned by them, or Ma.gnolia taking a catastrophic dive and losing most of its data, slowly but surely I think we will start to see people want to be able to synchronise the data they keep online either with other services or with their desktop. This is of course, but one possible use of Data Portability. People are still not aware of the need to back up their desktops, but i think they’re going to want a button to push to back up their cloud.

2. The rise and rise of metadata

Metadata is cool again. Whether you are working in research services, preservation, or reference, metadata helps you find and get stuff and that’s pretty essential to sift through all the crud out there. I’m not talking MARC – I’m talking EAD, PREMIS, SEPIA and much much more. I will be interested to see those that continue to move beyond metadata and start to expose collections beyond their institutions using methods other than OAI-PMH, like linked data.

3. The death of mobile sites

If you are building a specific mobile site or looking at buying a mobile module for your OPAC, stop. The number of sites built specifically for mobile devices has plummeted, not only due to the popularity of the iPhone and other smartphones with browsers, but because it’s a pain to have to build and maintain these sites. The .mobi domain seems to be slipping but m. is gaining hold (eg m.facebook.com). Smart sites are using content negotiation and web standards to display sites on mobile devices, without having to build separate sites to display less information. The mobile web isn’t second-best any more.

4. Librarians get personal

Many of us are now so familiar with blogs and wikis it hardly seems worth mentioning them as shiny new things any more. So what’s next? The library blogosphere has been in a bit of a lull of late, with few major shifts or emerging technologies for everyone to band around. But for me, this is the year more libraries have to get personal. – Not only designing websites that can be personalised, allowing you to save favourite resources, pages and references, but also in building tools for you – LibX for you, not just the institutional level, more subject guides that feature the human librarian that you are talking to behind the email, and even allowing you to personalise the physical space. And this is something I think we can all band around.

Cool URIs don’t change, right? But uncool URLs certainly do. For over 7 years, I blogged at blisspix.net and as was the way at the time, my URL was both a handle and identity. Whenever I signed up for a new service, my username was always blisspix. Lately, this has become less popular. I’ve noticed a sharp trend towards using real names, whether on Twitter (where I’d estimate well more than half of those I follow use their real name or some version of it) and Facebook, where many I know who chose a vanity URL chose their own name, myself included. The concept of handles and nicks seems to be dying off. I’m glad I didn’t keep using handles from the IRC and chat days of my youth, but blisspix was not much of an improvement and finally, it is time for it to go.  I would be interested to hear if anyone has researched this change.

Identity crisis issues aside, trying to change a name on existing services is a complete pain. Some services offer this with a single click, like Twitter. Others require you to send a form to request the change, like Dopplr. Flickr doesn’t seem to allow it at all. Facebook was very upfront in saying that creating a vanity name was a one time deal. 

But if you can change your name – there’s fallout. Your existence has changed, and all those nice URLs are going to break. There’s no redirects, just breaking. Presumably, if you are giving up a name it should be available to others to take, but to have links and RSS feeds just die is pretty painful. Twitter does do an ok job of this, if you change your name you can still track older mentions of you, although clicking on what was once your old username will still lead to a broken link. 

There’s no real easy solution to this, but it does bear saying that the way we all need to think about how we represent ourselves online. What may have worked at 20 as a uni student may not work as an older professional. This isn’t about what photos you put on Facebook, or whether you rant about an employer online, but the very act of giving yourself a name, whatever it is, forms an identity that can follow you for a long time so it pays to consider it carefully.

A while ago now, Jodi Schneider floated the idea of setting up a Planet for Semantic Web and Libraries. I had some trouble with various flavours of the Planet and Venus software, so instead I put something I’m calling a ‘Planet Lite’ together in Yahoo Pipes. Let me know if you like it! And feel free to suggest additional blogs, I have only listed a few to start.

Semantic Web & Libraries ‘Planet Lite’: JSON and RSS also available.

Last week, I attended Mash Oop North, an one-day unconference about mashups in Huddersfield, UK. It was a great event, which I wrote up in much more detail over at Libraries Interact. One of the best parts of the day was being able to spend some more time with Yahoo! Pipes, and the Planet Lite is the result after much tinkering and experimenting after I got home. Hooray for action-based outcomes!

Today I find myself perched in the foyer of the British Library, without enough ID to get a reader’s pass so instead of working, I am making the most of the free wifi here. Or at least I would, but for the service not working on my, ahem, brand new iPod Touch.

Those who follow my Twitter feed would know that I have gone back on forth on the why or why nots to get a Touch or an iPhone. An iPhone didn’t work for me because I don’t use cellular service very much at all, and the price plans didn’t seem logical. I had been holding out for a device that would function as a replacement for a laptop at conferences – eg ability to make videos, take photos, blog on the go etc.

So why did I get a iPod Touch, which doesn’t do most of those things? Apple in the UK are offering the Education bundle which includes a nearly-free iPod Touch, and so I got one yesterday. So far, I’ve loaded it up with RSS readers and social networking apps, but I am hoping to use location aware services that tap into semantic search, like a more accurate Urban Spoon that picks a restaurant for you based on where you are now or the quickest Underground route to get somewhere else.

A barrier to location aware services though is the availability of wifi. At home I can share my (somewhat poor) mobile broadband connection with my iPod using my MacBook. But when out and about, I rely on being able to pick up a connection. Connections with have an authentication service behind them, like that here at the British Library, are problemmatic. The SSID does’t show up (BL), or security apps need to be installed (UTS, Athens airport, EduRoam), or accounts need to be created in a browser (eg The Cloud). While organisations have good reasons for requiring authentication and acceptance of terms and conditions, it can make getting online with a Linux or Mac based system hard, let alone trying to do so with a mobile device. My mobile phone has wifi that I hardly ever use because of authentication layers.

It is very difficult to support platforms like mobile devices which have enormous diversity in operating systems, features and usage in different markets. I have noticed that Blackberries seem to be more popular in London than iPhones were in Sydney, and Push to Talk really does seem to be a feature that only those in the US use. Phones are completely different again in Japan (though I would buy a Japanese phone if I could) and serve very different needs in the African market. How can an organisation possibly provide a service like wifi that will enable all devices to connect easily and securely? How can we steer away from designing apps and authentication just for PCs, Android, or Symbian, or Palm and design universally? Is it even possible?

Beyond authentication, the dream of city-wide wifi seems to be long dead, with many announced project never coming to fruition and existing services mostly being run by a variety of subscription services that greatly restrict where you can get online. For example, I can use The Cloud with my mobile broadband account, but there are no hotspots near me. Yesterday, Ofcom in the UK released a report about Internet take-up across the UK. Turning the figures around, there are a lot of people who don’t have Internet at home (I currently am one of them – there is no phone line in my flat and cable Internet is not wired to my floor). There is a need to provide Internet in and out of the home other than private subscription, and wifi is one way to do it.

With all the distractions of my move across the world, it wasn’t until now (thanks, Jenna!) that I saw “Discovering Linked Data” was published in Library Journal last month. I aimed to cover a few different areas that Linked Data may be useful in, especially research support and discovery layers for catalogues, areas that I was working in at the time of writing the article. I’m really happy with how it turned out.

Open catalogue data is an an issue that arose when I first started working in libraries. I worked in a multicultural music library. I catalogued in more than 60 languages, all original, because our systems didn’t interface with cooperative cataloguing systems (and we had a non-standard cataloguing record, to boot), and the uniqueness of our catalogue in Australia did not make it worthwhile. But surely, there would have been other libraries across the world that we could and should have shared our data with. And I would have liked to share my work, knowing that we were duplicating work with our sister library in another city. We would have been able to identify collection strengths across the world, share resources, and perhaps even find partners to share our collections instead of having to do extensive legwork in obtaining materials in the first place.

I think Linked Data will have a big impact on larger institutions, but smaller ones also have a lot to gain by sharing in ways that were not possible or practical before.