1. P. Boyce 1 Use of Astronomy’s Info System : The Highly Productive User Peter B. Boyce Maria Mitchell Association and Past Executive Officer American Astronomical Society pboyce@aas.org

2. P. Boyce 2 Branding for Authors Adding to, solidifying author’s reputation Adding to, solidifying author’s reputation –Peer review (independent of publisher?) –Perceived quality of journal –Perceived quality of publisher Author wants to be identified with most prestigious journal Author wants to be identified with most prestigious journal

3. P. Boyce 3 Branding for users Want to ensure reliability of information Want to ensure reliability of information For journals -- depend on For journals -- depend on –peer review –journal reputation –reputation of author –Reputation of delivery system (?) e-prints depend only on author reputation e-prints depend only on author reputation

4. P. Boyce 4 Branding is Complex Combination of many factors Combination of many factors Depends on sophistication of user Depends on sophistication of user Broad range of users Broad range of users –Electronic information reaches broad audience –Old definitions no longer apply Users confused about source Users confused about source –Library vs. Journal vs. Aggregator –All linked

5. P. Boyce 5 Ease of Access is Critical Ease of access is most important factor Ease of access is most important factor –Working academics will preferentially use what is available Usage data may be distorted by factors other than quality and branding Usage data may be distorted by factors other than quality and branding “If it is not on the Web, it does not exist at all.” -- Sarah Stevens Rayburn, STScI “If it is not on the Web, it does not exist at all.” -- Sarah Stevens Rayburn, STScI Yet, users, like cats, will go to great lengths to get what they want… Yet, users, like cats, will go to great lengths to get what they want…

6. P. Boyce 6

7. 7 Make things simple Our job, as information providers is to make information simple to find and access… Our job, as information providers is to make information simple to find and access…

8. P. Boyce 8

9. 9 What Does a User Want an Information System To Do? Consider the tasks … Consider the tasks … Look at the system components Look at the system components

10. P. Boyce 10 Tasks and Components of Electronic Information System TasksComponents Find - Searchable Abstract Database Access - Linked Accessible Collection Read - Journal Article Use - Tabular and Auxiliary Data Publish - Journal Store - Archival Quality Collection

11. P. Boyce 11 Astronomy – A Good Testbed Small literature – 11 core journals Small literature – 11 core journals Self referential – 90% references to core Self referential – 90% references to core NASA and NSF funding for development NASA and NSF funding for development Internationally connected discipline Internationally connected discipline Most journals owned by societies Most journals owned by societies AAS publishes 40% of core literature AAS publishes 40% of core literature –25,000 pages per year Strong, active library community Strong, active library community

12. P. Boyce 12 Astronomy’s Interlinked System Searchable abstract database (ADS) Searchable abstract database (ADS) Linked electronic journals (since 1995) Linked electronic journals (since 1995) Scanned full text backfiles Scanned full text backfiles Links to astronomical databases Links to astronomical databases Machine readable data tables Machine readable data tables Nearly complete coverage (90%) Nearly complete coverage (90%) Design input from end users and librarians Design input from end users and librarians

13. P. Boyce 13 Astrophysics Data System (ADS) Searchable abstract database – Linked  Searchable abstract database – Linked  –One click away from full text  Journals or backfiles –Covers core journals, conference proceedings Forward and backward citations Forward and backward citations Interoperates with publisher Interoperates with publisher –Check references Supported by NASA -- freely accessible Supported by NASA -- freely accessible

14. P. Boyce 14 Awareness and Use of ADS by Astronomers

15. P. Boyce 15 Awareness and Use of e-print ArXiV by Astronomers

16. P. Boyce 16 Levels of Productivity (Astronomers) LabelDefinition Non-Productive < 1 refereed articles per year Average 1-2 refereed articles per year Productive >2 refereed articles per year 46% of respondents 78% of refereed articles

17. P. Boyce 17 Perceived value of Resource Percent Rating Resource as Critical or Very Useful for Keeping Up with Recent Developments

18. P. Boyce 18 Perceived value of Resource Percent Rating Resource as Critical or Very Useful for Obtaining Definitive Information

19. P. Boyce 19 Journals Preferred Journals (as a brand) are overwhelming choice Journals (as a brand) are overwhelming choice –Peer reviewed –High reputation –Familiar e-prints used for current information e-prints used for current information

20. P. Boyce 20 Articles Read per Month vs. Productivity

21. P. Boyce 21 How Articles Are Located vs. Productivity

22. P. Boyce 22 To Find Articles Productive authors Productive authors –Prefer to search - mostly on-line –Less likely to get reference for colleague Non-productive authors Non-productive authors –More likely to find article by browsing –Less likely to follow citation

23. P. Boyce 23 When are e-prints submitted? Before or after acceptance by Journal

24. P. Boyce 24 Branding for readers Journals preferred as info source Journals preferred as info source –For current information –Overwhelmingly for definitive information e-prints used for rapid communication e-prints used for rapid communication –Before journals become available Most productive users value info more Most productive users value info more –They read twice as much

25. P. Boyce 25 Branding for Authors Authors depend on journal peer review Authors depend on journal peer review –Peer review is important confirmation Submit to e-print server after acceptance by journal Submit to e-print server after acceptance by journal Young and productive authors more likely to wait for journal acceptance Young and productive authors more likely to wait for journal acceptance Non-productive authors more likely to publish only an e-print Non-productive authors more likely to publish only an e-print

26. P. Boyce 26 Final Thought Productive authors Productive authors –Produce 4/5 of articles –Do 2/3 of reading –Find ½ their articles by searching Shouldn’t we tailor our products and services to the needs of productive users? Shouldn’t we tailor our products and services to the needs of productive users?