Open Repositories 2008, part 1.
Apr. 1st, 2008 12:15 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
deborahAll these papers will eventually be available in the Open Repositories 2008 conference repository. I'm linking to all of the placeholders; papers should be up soon.
This will be very limited liveblogging, because I'm typing in the conference and dictating betwen sessions, so I can't say much. Hopefully I'll get some good fodder for my upcoming sustainability post.
Keynote:
"Repositories for Scientific Data", Peter Murray-Rust
This talk proved its point in perhaps unintentional ways. Peter Murray-Rust's argument was that scientists will not choose to change the ways in which they work, and they won't understand technology -- and he certainly didn't understand the technology. He discussed Active Directory and Samba as storage mechanisms, Subversion as a concept rather than as a specific tool, Word's XML as useful semantic metada. He didn't value preservation and data-interoperability, instead valuing tools to extract and manipulate data now. He saw no utility in PDF because to him, a tool which does not allow manipulation of raw data is useless.
He's quite right that we need to make our data more usable now, but he places too little value on preservation, standards, and interoperability. He also misunderstands the tools he currently uses and what they do. I also disagree with him that we need to accept that scientists will never cooperate with the preservation of data. At some level, we need to attain cooperation from the data producers. Yes, we should adjust to their workflows, but we can't take data without their help.
Sadly, he's probably right that we'll never get useful data from them unless we get inolved in data management at point of creation. Yuck. What a mess of legal, data management, and human spaghetti.
Session 1 – Web 2.0
"Adding Discovery to Scholarly Search: Enhancing Institutional Repositories with OpenID and Connotea", Ian Mulvany, David Kane
This talk didn't cover much I didn't know, but was clearly very interesting to me, because it discussed the strengths and weaknesses of OpenID. Essentially it asked, "how do you have interoperability among a variety of online identities?"
"The margins of scholarship: repositories, Web 2.0 and scholarly practice", Richard Davis
Pretty straightforward stuff. The researchers added standard social networking integrations such as an image viewer, commenting, bookmarking to an eprints instance.
Rich Tags: Cross-Repository Browsing, Daniel Smith, Joe Lambert, mc schraefel
Hmm. I hurt way too much to keep liveblogging. So, nutshell: Scraping papers, available metadata, whois records, text mining, etc to generate non-human-generated metadata which can be discovered in their faceted browser. Cool.
Ow. I'm not doing this for the next session. I can blog at the breaks.
This will be very limited liveblogging, because I'm typing in the conference and dictating betwen sessions, so I can't say much. Hopefully I'll get some good fodder for my upcoming sustainability post.
Keynote:
"Repositories for Scientific Data", Peter Murray-Rust
This talk proved its point in perhaps unintentional ways. Peter Murray-Rust's argument was that scientists will not choose to change the ways in which they work, and they won't understand technology -- and he certainly didn't understand the technology. He discussed Active Directory and Samba as storage mechanisms, Subversion as a concept rather than as a specific tool, Word's XML as useful semantic metada. He didn't value preservation and data-interoperability, instead valuing tools to extract and manipulate data now. He saw no utility in PDF because to him, a tool which does not allow manipulation of raw data is useless.
He's quite right that we need to make our data more usable now, but he places too little value on preservation, standards, and interoperability. He also misunderstands the tools he currently uses and what they do. I also disagree with him that we need to accept that scientists will never cooperate with the preservation of data. At some level, we need to attain cooperation from the data producers. Yes, we should adjust to their workflows, but we can't take data without their help.
Sadly, he's probably right that we'll never get useful data from them unless we get inolved in data management at point of creation. Yuck. What a mess of legal, data management, and human spaghetti.
Session 1 – Web 2.0
"Adding Discovery to Scholarly Search: Enhancing Institutional Repositories with OpenID and Connotea", Ian Mulvany, David Kane
This talk didn't cover much I didn't know, but was clearly very interesting to me, because it discussed the strengths and weaknesses of OpenID. Essentially it asked, "how do you have interoperability among a variety of online identities?"
- Requiring multiple logins is (1) a pain, and (2) might mean you're giving login information for one host to another host
- OpenID is getting widely adopted and is a great way of sharing authentication keys, but provides a large risk of phishing; you need to trust the servers on both ends.
- OAuth provides access to services one service without sharing authentication information.
"The margins of scholarship: repositories, Web 2.0 and scholarly practice", Richard Davis
Pretty straightforward stuff. The researchers added standard social networking integrations such as an image viewer, commenting, bookmarking to an eprints instance.
Rich Tags: Cross-Repository Browsing, Daniel Smith, Joe Lambert, mc schraefel
Hmm. I hurt way too much to keep liveblogging. So, nutshell: Scraping papers, available metadata, whois records, text mining, etc to generate non-human-generated metadata which can be discovered in their faceted browser. Cool.
Ow. I'm not doing this for the next session. I can blog at the breaks.
