Open Practices for the Connected Researcher Presentation by Brian Kelly, UKOLN on 25 October 2012 for an Open Access Week event at the University of Exeter 1 Open Practices for the Connected Researcher Presentation by Brian Kelly, UKOLN on 23 October 2012 for an Open Access Week event at the University of Exeter

UKOLN is supported by: Open Practices for the Connected Researcher Brian Kelly UKOLN University of Bath Bath, UK This work is licensed under a Creative Commons Attribution 2.0 licence (but note caveat) Acceptable Use Policy Recording this talk, taking photos, having discussions using Twitter, etc. is encouraged - but try to keep distractions to others minimised. Acceptable Use Policy Recording this talk, taking photos, having discussions using Twitter, etc. is encouraged - but try to keep distractions to others minimised. Blog: Twitter: #OAweek

3 3 You are free to: copy, share, adapt, or re-mix; photograph, film, or broadcast; blog, live-blog, or post video of this presentation provided that: You attribute the work to its author and respect the rights and licences associated with its components. Idea from Cameron Neylon Slide Concept by Cameron Neylon, who has waived all copyright and related or neighbouring rights. This slide only CCZero. Social Media Icons adapted with permission from originals by Christopher Ross. Original images are available under GPL at:

About Me Brian Kelly: UK Web Focus: national advisory post to UK HEIs Long-standing Web evangelist Based at UKOLN at the University of Bath Prolific blogger (1,125+ posts since Nov 2006) User of various devices to support professional (and social) activities Prolific speaker (~390 talks from ) Member of UKOLN’s Innovation Support Centre ISC at UKOLN: Supporting innovation across higher & further education Funded by JISC 4 Introduction

About This Talk Open access: Benefits for researchers understood Arguments will continue This seminar goes beyond open access: Open practices: sharing ideas on blogs Open engagement: the role of Twitter Dissemination: getting your research read Gathering the evidence: social media metrics Making it work: identifying best practices This talk provides 12 tips for the connected researcher 5 Based on evidence gained from personal experiences Introduction

About You Are you a Roundhead or a Cavalier? “In the century, Britain was devastated by a civil war that divided the nation into two tribes – the Roundheads and the Cavaliers. The Cavaliers represent a Britain of panache, pleasure and individuality. They are confronted by the Roundheads, who stand for modesty, discipline, equality and state intervention.” 6 Who do you admire most? Mo Farah for winning the 5,000 and 10,000m? Usain Bolt for partying with Swedish handball team after winning 100m, & before 200m relay?

Are You A Marxist? “Hitherto, philosophers have sought to understand the world; the point, however, is to change it” Do you seek to change the world through your research or simply understand the world: Will you want to market your research? Will you want others to market your research? Will you have a detached view of your research? 7 Introduction

My Papers My papers in the University of Bath Opus repository 8

Largets downloads 9 Open Access enhances access

Least Downloaded Papers Will papers in a repository be seldom seen? What can be learn from approaches taken for the popular and unpopular papers? 10

Learning from Mistakes “Using context to support effective application of web content accessibility guidelines” Submitted in July 2005 Reviewers comments received in April 2006 Published in JWE in December 2006 PDF uploaded to repository in May 2012 Forgotten paper when bulk uploads made 11 Reflections on implications given in “If a Tree Falls in a Forest” post

Learning From Success “Library 2.0: balancing the risks and benefits to maximise the dividends” 12 Sixth most downloaded paper in repository But only recent download statistics available  2012

Reasons For Paper’s Popularity Possible reasons Quality of paper Quality of metadata Importance of co-authors Provision of full-text, rather than just metadata Formats used (HTML as well as PDFs) Role of social media Other suggestions? 13

Evidence How do we find out more? Peak statistics for repository only available for 1 year But: Blog post about availability in Opus published on 11 August Conclusion: Blog post responsible for initial popularity Further investigation (of all my paper downloads) confirms large peak in August 2009

Beyond the Edge Cases Little-downloaded paper: Uploaded to repository 6 years after paper written I was not lead author Only PDF version uploaded Never blogged about; never tweeted Most popular paper: Available in IR on launch of journal issue I was lead author MS Word, PDF & HTML versions uploaded Blog post published on day of launch Link to paper subsequently tweeted & retweeted About Web 2.0, so likely to be read by bloggers 15 But what about the majority of papers?

SEO or SMO SEO: Helping Google find your papers through: Writing style, document structure, … In-bound links SMO: Helping other people find your papers through: Viral marketing Engaging with one’s peers Sharing on social media services 16 SMO:Good for new papers, but not relevant for popular papers written from SEO:Document structure consistent. Difference appears to be significant nos. of in-bound links

Tip No. 1: Be Pro-active Tip No. 1: Be pro-active 17

W4A 2012 Paper Case study: Paper on “A challenge to web accessibility metrics and guidelines: putting people and processes first” given at W4A 2012 conference in Lyon in Apr 2012 Four co-authors agreed: To collaborate in raising awareness of paper and presentation of the paper How: Writing blog posts on or just before conference To participate in Twitter (e.g. responding to comments while speaker is presenting paper) Benefits: Reaching out to a wider audience based on our 4 professional networks 18

Preparation We: Uploaded paper to repository so URL was known Provided a link to the paper in speaker’s slides Uploaded holding slide to Slideshare so URL was known (slides were finalised shortly before talk) We could then: Prepare blog posts in advance Create short URLs in advance 19 Examples of approaches to follow

Tip No. 2: Identify Key Channels Tip No. 2: Identify the key channels 20

Opus Repository Paper uploaded to Opus repository 21

Metadata provided to give context to slides 22

Final slide provides (active) links to related work 23

Tip No. 3: Monitor What Works Tip No. 3: Monitor what works (for you) 24

Capture Statistics On 18 Apr 12: 1,391 views on Slideshare Other slides had 3 and 311 views By 11 Sep 12: 6,976 views on Slideshare 25 “Lies, damned lies & statistics” – but 3 rd most downloaded 2012 paper in 2012

Topsy and Event Hashtag 26 Buzz around event hashtag captured by Topsy

Topsy & Discussion About Slides 27 Topsy recorded discussions about slides

Topsy & Discussion About Paper 28 Topsy recorded discussions about paper Note tweets about event (25) and slides (20) more popular than paper (7)

Repository Statistics Opus repository stats: Views began in March (before conference). Publish on embargo date didn’t work! 29 Largest downloads took place on 7 March, day blog post published Post about collaborative tools for writing paper, not contents of paper

Other Repositories 30 In some disciplines other repositories may be popular “palaeontologists for some reason very much like Academia.edu. Perhaps your research communities centre around IRs - this will vary from discipline to discipline” Ross Mounce, Bath

Tip No. 4: Don’t Forget the Links! Tip No. 4: Don’t forget the links 31

LinkedIn Links to paper added to LinkedIn Academia.edu My pages on UKOLN Web site and blog … 32

Academia.edu 33 Note: Links to papers in IR (not uploaded) Importance of tags Academia.edu users may find my papers here and LinkedIn users in LinkedIn. Why would I make it difficult for them?

The IR 34 Your papers should be hosted on your institutional repository

The Institutional Web Site 35 You may also wish to provide links on your institutional Web site Note direct links to paper in various formats

The Institutional CRIS 36 Your papers may also be listed on the institutional CRIS

The Blog 37 If you have a blog you can provide links to your papers (again to all formats)

Commentable Pages on Blog 38 Recent development: Commentable pages for papers with links to key resources (IR & publisher’s copy, metrics, citations, …)

Tip No. 5: Don’t Forget the Google Juice! Tip No. 5: Don’t forget the Google juice! 39

Importance of Google Context: Between 50-80% of traffic to IRs are from Google (may be higher if direct links to PDFs not recorded by Google Analytics) What provides ‘Google juice’: On-page SEO techniques (structure, writing style, …) Links to pages, especially from highly-ranking sites 40

Importance of Google Context: Between 50-80% of traffic to IRs are from Google (may be higher if direct links to PDFs not recorded by Google Analytics) What provides ‘Google juice’: On-page SEO techniques (structure, writing style, …) Links to pages, especially from highly-ranking sites What’s different about IRs? Same page structure Therefore importance of links to repository 41

Importance of Google Context: Between 50-80% of traffic to IRs are from Google (may be higher if direct links to PDFs not recorded by Google Analytics) What provides ‘Google juice’: On-page SEO techniques (structure, writing style, …) Links to pages, especially from highly-ranking sites What’s different about IRs? Same page structure Therefore importance of links to repository 42

What Delivers Google Juice? Survey of SEO ranking of 24 Russell Group IRs carried out in August Findings: Google, YouTube, Blogspot, Wikipedia and Microsoft are highest ranking domains with links to IRs 43 Blogspot.com & WordPress.com have significantly larger number of links to IRs Links from institutional domain (e.g. locally-hosted blogs) provide little Google juice! Blogspot.com Wordpress.com

44 UK Web Focus blog has links to papers

45 UK Web Focus blog has a rotating Featured Paper link

46 UK Web Focus blog has a rotating Featured Paper with links to funders & organisation on all pages > 5,000?

But Isn’t Google Clever? Google will address cheats aimed at artificially boosting search results (“black hat SEO”) 47 But: This approach is “white hat SEO” SEO analysis of Bath IR shows top 5 highly ranked page are to my papers 2 Accessibility 2.0: next steps for web accessibility, citations 3 Can LinkedIn and Academia.edu Enhance Access to Open Repositories?, Open Metrics for Open Repositories, 2012 Evidence: Largest number of downloads in IR Largest number of links from highly-ranked trusted domains Success in dissemination. Need to correlate with nos. of citations

Google Scholar Citations 48 Downloads (IR) Most cited papers according to Google Scholar Citations Nos. Current Graph

Tip No. 6: Encourage Feedback Tip No. 6: Encourage feedback and discussion 49

Comments On Your Blog A blog post about a paper provides opportunity for feedback & discussion 50 Feedback on an embarrassing typo! Legitimate question about research assumptions

Blog Comments Increasing tendency for discussion and comments to be distributed (e.g. on other people’s blogs or on Twitter) Some topics do attract large numbers of comments (evidence-based surveys in my case). 51

Distributed Discussions Comments may take place on other blogs. You should allow trackbacks so you see links to your posts 52 You can see links from blogs which reference yours. You can see links from other services which reference yours. You can see the chain if you refer to previous posts in your blog.

Referring Blog Post 53 Citation, but no additional comment

Scoop.it 54

Scoop.it Seeing no. of links from Scoop.it growing Implications: Others highlighting your ideas Scoop.it giving your Google juice Helping you to grow your network 55 This is of interest to me. I must follow the Dual Impact Scoop.it account

Tip No. 7: Develop Your Network Tip No. 7: Develop your network 56

“It’s About Nodes and Connections” Cameron Neylon keynote at OR 2012: “Networks qualitatively change our capacity” With only 20% of a community connected only limited interaction can take place This increases drastically as numbers of connected nodes grows Examples: Phone networks (no use with only 1 user!) Tweeting at this seminar Galaxy Zoo 57 “Filters block. Filters cause friction” Need for client-side, not supply-side filters.

Tweetchat Tweetchats: Discussions on Twitter Specific topic covered at specified time Use hashtags e.g. #PhDchat, #ECRchat Summary at Survey findings: “give a community & shared space to explore ideas” “regular opportunity to network with a wide range of people I wouldn’t otherwise meet” “have very interesting and thought-provoking discussions/debate” 58

Tip No. 8: Understand Your Network Tip No. 8: Understand your network 59

Social Bros 60 SocialBro: people in my network typically follow Twitter users

Social Bros 61 SocialBro: people in my network typically tweet 2-5 times every day

Traffic To My Blog People arrive at the UK Web Focus blog via: Google Twitter Web site (nb client visits not included) Facebook! 62

Tip No. 9: Know Your Limits Tip No. 9: Know when to say “No!” 63

It May Not Be For You Your working style; your discipline “The Social Web & the Belbin Model” post suggested “Plants” & “Resource investigators” may find Social Web fits their working style You may have concerns about privacy You may work in different area of research with different practices; … 64 Your time: “Blogs, Twitter, LinkedIn, Academia.edu, scoop.it, bit.ly, YouTube… - I haven’t got the time!” Need to prioritise, based on awareness of possibilities, relevance, … But remember there can be light-weight approaches you can use

Tip No. 10: Seek Improvement Tip No. 10: Seek gradual improvement 65

Use of Slideshare Improvements for next time. 66

Use of Slideshare Improvements for next time. 67

Tip No. 11: Be Ethical Tip No. 11: Be ethical 68

Fake Reviews Don’t be tempted to create a “sock puppet”: an online identity used for purposes of deception. Note technological developments are likely to spot fraud / inappropriate patterns of use. 69

Personal Statement First public draft of a personal approach to use of social media to support my research based on my working practices: I will promote my research in order to maximise awareness and take-up of ideas I will use social media to support this objective I will regard outputs which do not have a sufficiently large audience as a failing I will not use fake accounts to promote my work I will be aware of potential concerns regarding spamming I will gather evidence of the effectiveness of my use of social media 70

Tip No. 12: Participate Tip No. 12: Participate! 71

Participate 72

Health Warning! Suggestions given can help to enhance the visibility of one’s research. Highly visible and popular research is not necessarily an indication of quality! 73

Conclusions 1Be pro-active 2Identify the key channels 3Monitor what works for you 4Don’t forget the links 5Don’t forget the Google juice 6Encourage feedback and discussion 7Develop your network 8Understand your network 9Know your limits 10Seek improvements 11Be ethical 12Participate 74

Questions Any questions or comments? 75