Aggregating UK Library Job Feeds on Twitter

Been meaning to do this for a while. It seemed like a fairly simple task - in fact I can't believe something doesn't already exist (maybe it does and I just can't find it?). The first job was to gather feeds. I already knew about the jinfo (jobs in information), jobs.ac.uk (academic vacancies) and lgjobs.com feeds. Being sure that there were more (and being unable to find a lisjobnet feed) I put the question to Twitter and got some good tips from @jimmy1712, @ekcragg, @libram, and @joeyanne - including a scraped feed from lisjobnet! (also thanks to @darrenmjones for pointing me towards Feedity, a tool for producing feeds from sites that don't already have them. Haven't tested it yet but looks promising!).

Next stage was to look at aggregating tools and create the Twitter account. @UKLibraryJobs was free so I decided to go for that. As I'm already using Hootsuite I decided to give that a try for aggregating the feeds - if it doesn't work out I look at other tools...

The initial list of feeds is...

lgjobs.gov - libraries category (need to check how comprehensive this one is and possibly find feeds for councils that are not listed?)
Jinfo - Latest Jobs Feed
lisjobnet (unofficial) - in the end I used the feed from @darrenmjones as the other one didn't report a timeline.
TFPL - latest jobs feed
Sue Hill Recruitment - although this looks like it mightbe the jinfo feed with a Sue Hill referral? Will need to look into this...
Jobs.ac.uk - Information Management category - actually I couldn't add this one as it doesn't have a timeline either - I'll look at using Feedity to scrape this.

...I'm hoping that if anyone knows of any others they'll let me know or maybe we can open access to the Hootsuite profile to let others add feeds without having to go through me.

So, that'll do for now. Jobs to do are:
1) Tart up the Twitter account a bit.
2) Check that the feeds are working.
3) Look into the scope of the lgjobs and Sue Hill feeds.
4) Scrape a feed from jobs.ac.uk
5) The most Hootsuite will post is 5 posts an hour from each feed so need to check if this is a problem.

Sure there's other things but that's enough for now.

CILIP 2.0 Council Open Session

cilip2_full by Dave and Bry, on Flickr

I've been thinking about the CILIP 2.0 Council Open Session yesterday. First off, it was fascinating for me to attend an event like this on Twitter. Although I've followed events from hashtags before, as a relatively new Twitterer, never anything on this scale. The tools that were recommended (Twitterfall and Tweetdeck) worked well for me, although at times there was a bit of lag from Twitterfall so I found myself reverting to Twitter Search occasionally.

I did get some sense of what was happening in the room and a number of Twitterers were very generous in describing what was going on (see @tomroper or @janeite for running commentaries or @daveyp's transcript of the #cilip2 tag for more info).

I've seen some criticism of the Twitter stream as being hard to follow and the thought's or ideas not being coherent, which is interesting in itself. Yes, it's a challenge to put complex ideas into 140 characters but I beleive there are coherent streams (or conversations) within the tag. Perhaps it has as much to do with the number of participants as anything. The more people you have in a group, the looser the conversations tend to get. At least Twitter allowed a large number of people to attend remotely and the stream is there to be analysed/ picked over now. I'm sure some interesting threads will emerge.

What did the day achieve though? Well, I didn't pick up on any hard action or proposals. I can't see that any actual decisions were made by the CILIP top brass. But then I'm not sure that was what the event was about. I get the sense that it was more about CILIP Council *listening* to it's members. This is something I've never heard of before and I really think CILIP Council deserve a big hats-off for hosting the event.

I'd like to see CILIP provide a strong lead in this space. Personally I think they've some catching up to do, but listening and finding out more with events like yesterday's is a very good start.

The event got me wondering: actually, what do I want from CILIP with regards to Web 2.0? Do I want them to leap on the latest hot Web2.0 app? Well, partly. I'd like to see them take more interest in the tools that members are using and, if there is significant demand, engage with their users on those tools. But I think using the latest/ coolest tools in the Web2.0 toybox is a bit misleading. Web2.0, for me, is not about the specific applications it's more a way of being. Being open, encouraging participation, encouraging user generated content, concepts like creative commons etc. The big question for me really is not whether or not CILIP have an "official" Twitter account (although that would be nice!) but whether they are prepared to use Web2.0 to be more open and to provide a place for librarians to openly share their ideas and information.

By extension that means that we - the information profession - have to be more open in sharing our information and content. We have to contribute and get involved. Can we make CILIP2.0 a reality?

The Potential Impact of Web 2.0 on the Delivery of Higher Education Library Services

Here's my presentation for the job interview and Bournemouth University. The more I look at this the more it seems outdated and naive but, I worked hard on it so here it is...

Record your screencast online

I may never visit Twitter again! HootSuite does that for me.

I've been trying out HootSuite as a Twitter tool for the last couple of days and I have to say I'm really impressed.

A great feature of HootSuite is that you can use it to manage multiple Twitter accounts from a single place. You can also configure it to allow multiple people to be able to Tweet from a single Twitter account. Great if you're managing a group Twitter stream! These are what got me interested in HootSuite but what's kept me there is some of the other features.

Here's a bit more detail:

• Posting: In-line url shortening makes it much simpler to add links. Delayed posting is also useful. If you've got more that one Twitter account set up you can post the same tweet from multiple Twitter accounts - simply tick each account you want to post from.
  

• Stats: If you use the in-line ow.ly shortening HootSuite can keep stats on how many people follow your links. The ui is really good, you can see a breakdown of clicks on all of your links...

...or click on the little graph icons in the "sent" view to get a breakdown of how many hits a particular link has had. Good for working out what your followers are interested in!

• Searches: HootSuite saves each search you run as an icon in your "Searches" tab. Really handy if you want to keep track of some trending topics or particular people. Some of the search options are quite powerful too.

• User Management: Really simple to give others editing rights. Just add a users email address and name and select whether you want their password to be sent to you or them. You can also delete editors with a single click.

• Twitterfeed-like function: There is an option in "Account Settings" that enables you configure Tweets from RSS or ATOM feeds. It's not quite as fine grained as TwitterFeed but fine for most purposes.

I've really only skimmed the surface here. It does lack follower management and the grouping functionality of say Tweetdeck right now. Another thing to be wary of is that account confirmations seem to end up in your email spam folder! In short though, I'm really impressed. The site is fast and simple and provides a great user interface for becoming a more advanced Twitter user.

Useful Links Pages Using Social Bookmarking - Potential Banana Skins

So the Useful Links page using Social Bookmarking was finally finished. Or so I thought. The idea was fairly simple: use a Social Bookmarking tool to allow users to tag web sites to add to the library's "Useful Links" page. The information flow basically happens like this:

1) User finds useful website.

2) User bookmarks web site in social bookmarking application using library tag.

3) Social bookmarking application adds bookmark to feed (in this case ATOM).

4) Magpie RSS aggregates feeds and parses them to html.

5) Library web site displays links on "Useful Resources" page.

In practice though, I came across a few problems so I thought now might be a good time to highlight some of the pitfalls and document my solution!

Pitfall #1 - The Accidental Tag

In my case, I came up with about 10 different tags to allow the Useful Resources to be sorted into categories (e.g. Databases, Libraries etc). Each category has its own distinctive tag (e.g. hurlibdat, hurliblib). The tags have to be fairly unique as otherwise you'd end up with a load of stuff you don't really want.

Say for example you used a generic tag like 'magazine'. It's not beyond the realms of imagination that sooner or later somebody would tag the Playboy website with 'magazine' and it would end up on your useful resources page. They probably weren't intending to introduce soft porn to your library users but the effect is the same.

Lesson #1 - Library link to Playboy Magazine = Grumpy users and complaints! Use tags that nobody is likely to accidentally tag items with!

But what if it's no accident? This brings me to...

Pitfall #2 - Intentional vandalism

Ok so it's not very 'Web2.0' what with extreme trust and all that, but if you're putting your name to a page you have to be able to ensure that the links on it are of a certain quality. So what can you do to stop people intentionally tagging inappropriate content? Well, nothing really. What you can do is set out guidance on what resources are appropriate and point out the fact that, in social bookmarking, everybody knows who's tagged what - if you vandalise there will be comeback!

Lesson #2 - Give users clear guidance on how to tag and which kinds of resources are appropriate.

I actually found a very similar site on our Intranet and revised the guidance for the library page.

Despite this it occurred to me that, if someone really wants to add inappropriate links, there is really very little I can do to stop them. Or is there? More on that later!

Pitfall #3 - How to fix broken links

This one only occurred to me after I finished the page and discovered, to my horror, that one of the linked sites had already vanished. So. I found a new link for the site and edited my bookmark in the social bookmarking tool to point to fix the link. Then it hit me. How can I fix broken links from users when I can't edit their bookmarks? ACK!

The Solution to ALL My Problems...

...is actually very simple. Moderation. Although I was initially against this I've come to the conclusion that, to avoid a page full of broken links six months down the line I need to own the bookmarks that feed the useful links page.

The mechanism to do this had me stumped for a while but the solution is actually fairly easy.

I need to create one tag for users to bookmark sites they think will be useful to the library. I monitor the feed for that tag and create my own bookmark (using a different tag) any time a user tags a new page. The Useful Resources page aggregates my tags rather than the users', allowing me to moderate and organise the content.

This actually improves all three problems.

For pitfall #1, the tags that I use can be so obscure that nobody could possibly add resources by accident. Also, just having one simple tag for library users is alot easier for them - they don't have to categorize the resource or remember 10 different, obscure tags for different categories.

For pitfall #2 - I'm now able to filter out any sites that aren't appropriate just by not bookmarking them!

For pitfall #3 - I own the bookmarks feeding the Useful Links page so I'll be able to edit or delete broken links. This should be really easy, just schedule in a review of the page every six months or so and edit the bookmarks of any broken links.

Shimple! If only I'd of thought of it sooner. Ah well, we live and learn and at least it shouldn't be too hard to fix.

The Times. Are They A Changing?

Change is good and I think that we are starting to see some change in the way that publishers are thinking about eBooks. I've been thinking alot about O'Reilly's adoption of Bookworm and promotion of the ePub format. O'Reilly have also started to make eBooks available in a wider range of formats with their eBook bundles. This has led me to think that *possibly* some of the barriers to libraries offering their users eBooks may soon come down.

Price...

I saw a recent blog post from @ijclark on eBook pricing. Ian found that, for individuals at least, there was no real price benefit to getting an eBook. For libraries, this point is even more salient as you often have to pay platform or hosting fees on top of the cover price. An interesting aside here, contrary to eBooks in general O'Reilly eBook bundles are significantly cheaper than the print versions (try clicking on any with a green E logo from this page to compare). It seems that O'Reilly have recognized the price barrier and opted to pass production savings on to eBook readers.

The Dreaded DRM and proprietary standards...

DRM only really works on closed standards and O'Reilly have correctly (just my opinion) identified that closed standards are preventing eBook uptake. Amazon are particularly bad at this - using a proprietary file format for the Kindle which makes it difficult to read books that you have bought for the Kindle on any other device (and vice versa).

My library patrons use a number of different operating systems, mobile devices and eBook readers. Buying eBooks that all of my library users can read (regardless of their chosen platform) is very important to me. At the moment I don't have this ability. However, there are signs that more publishers and eBook manufacturers are getting behind the open ePub standard.

As a librarian I want to decide which books I buy...

Until recently, most of the best computing publishers (O'Reilly, Addison-Wesley, Pearson etc.) have offered their eBooks exclusively through a distributer - namely Safari Books.

Safari works on a subscription basis. Getting all of this content for $45 per month may seem like a bargain but actually, how many computer books can you read a month? When you look at this on an institutional level (for example providing library users with access) the pricing becomes prohibitively high - especially when you know that your users don't need alot of the content that you're paying for.

Whilst some distributors (such as Netlibrary) will let you pick and choose which titles you buy they are hamstrung in that they can't sell books that are exclusive to other distributors. To build the collection you'd like to as a library, you'd need to have deals with a number of distributors. This would really break the bank as well as confusing your library users. In practice it's Hobson's choice between paying through the nose for content that you don't need and offering more selective but lower quality books. The solution, for me, is for publishers to offer their content directly to users in a common format. Cut out the distributors.

Light at the end of the tunnel?

All this brings me back to Bookworm. When I first saw Bookworm I got very exited as it is a free platform that allows individual users to build bespoke collections that are hosted on the Internet. It's actually not far off being a platform that libraries could use to provide access to collections that aren't hampered by subscription/ exclusivity deals or limited content that we get with existing platforms.

Ironically, the lack of DRM is what makes this a no-no at the moment. Obviously, hosting a load of eBooks on Bookworm and giving users access is not an option. Everytime a user downloaded an eBook they'd be creating a copy which would be a massive infringement of copyright. Cleveland Public Libraries have overcome this problem using OverDrive's platform to host their ePub books (they are the only library that I can find who are loaning ePub books). The way this works is that when a user downloads the file it becomes unavailable to other library users. After a time the user's copy of the file expires. At the same time the copy on the library system becomes available to the masses again. However, OverDrive are a distributor so, although the ePub file format is open, you'd soon come up against the exclusivity/ limited content problem.

Soooo, it's not quite there yet. However there is hope. With publisher's like O'Reilly recognising that overpricing, proprietary file formats & DRM, and the limitations of distributors are hampering eBook uptake by individuals, hopefully it won't be long before some of these issues are resolved for libraries?

Well played HCC!!!

I'm referring of course to Hampshire County Council's live Twittering of the Local Authorities Transforming Public Libraries conference. Regardless of the content of the conference I think we all have to stand up and give a round of applause to HCC for making it accessible to the broader community. As well as a Twitter stream there is also a conference website and slides from the speaker's presentations are available. Video clips of the presentations should be added by the end of the week.

Here are some things I liked about the Twitter coverage:

• The Twitter account was set up in advance of the conference and by this morning had almost 50 followers. Posting about the purpose of the event gave folks enough info to follow it if they wanted. I probably wouldn't have ever heard of it if it wasn't on Twitter.
  
• The stream now exists as a persistent record of the event - almost like minutes. The links to the speaker's presentations were neat as well and by looking up @libconf (and reply's to @libconf) we can reflect on the speakers, revisit the topics and add our own responses. I for one will be thinking hard about Tim Coates pitch (possibly another post on that later!).
• Hampshire R&HIT web team (behind the Twittering) explained that it was experimental and a new venture. Which set the expectation. That said, they did very well in reporting what was going on and even managed to respond to some replies. There were about 100 tweets from libconf on the day - no mean feat I'm sure with everything else going on.
  

I really hope they do this for future events and that other organisations follow their lead (mentioning no names *cough*CILIP*cough*). It shows a willingness to engage with the broader profession and allowed about 50 information professionals to get the inside scoop on something which, otherwise, may have appearred to be a very top-down conference. Yes, it's a shame that the speakers didn't field questions from Twitter and also that the backchannel from delegates appeared to be pretty poor but it's definitely a step in the right direction.

Well done Hampshire!