2021-10-14: Joint Conference on Digital Libraries (JCDL) 2021 Trip Report
Day 1
Tutorial 1
First day of #JCDL2021 is started with Tutorial and Doctoral Consortium.@edwardafox from @virginia_tech is presenting Tutorial 1: Introduction to Digital Libraries at @JCDLConf #JCDL2021.@WebSciDL pic.twitter.com/EIPNw9rLll
— Muntabir Hasan Choudhury (@TasinChoudhury) September 27, 2021
#JCDL2021 has just kicked off with an introduction to Digital Libraries (DL) and the life cycle DL covers by Edward Fox @VT_CS. @JCDLConf @WebSciDL @oducs pic.twitter.com/CFDUGgjwRD
— Ajayi Kehinde Peter (@AjayiKehindep) September 27, 2021
Day 2 (Main Conference)
Conference Opening
* Each paper will be presented in 2 sessions to help those across multiple time zones.
— Shawn M. Jones, PhD (@shawnmjones) September 28, 2021
* Conference proceedings are available! 👏🏻 -- I've already imported them into Zotero!
* The papers will be on the IEEE & ACM sites soon.
* Not as many events as if we were "in 3D".#JCDL2021 pic.twitter.com/qt9zLM5rW5
Paper Sessions
@kritika_garg is now presenting "Replaying archived Twitter: When your bird is broken, will it bring you down?" at #JCDL2021 Paper session 1
— Shawn M. Jones, PhD (@shawnmjones) September 28, 2021
Preprint: https://t.co/y4Sqbu62gv pic.twitter.com/qbON8BuoTF
.@kritika_garg(Ph.D. student) a @oducs and @WebSciDL member, is presenting their paper titled "Replaying archived Twitter: When your bird is broken, will it bring you down?" at @JCDLConf #jcdl2021.
— Muntabir Hasan Choudhury (@TasinChoudhury) September 28, 2021
pre-print: https://t.co/IrWqzNYfgp pic.twitter.com/YBCDsH0YUC
#JCDL2021 Paper session 1: @kritika_garg analyzed the loss of info in web archives using @readDonaldTrum3
— Ajayi Kehinde Peter (@AjayiKehindep) September 28, 2021
Preprint: https://t.co/2sMilxvCOs@JCDLConf @WebSciDL pic.twitter.com/flUVeVDMbk
Muntabir Hasan Choudhury, a Ph.D. student from Old Dominion University and a member of the WSDL research group, presented their paper titled "Automatic Metadata Extraction Incorporating Visual Features from Scanned Electronic Theses and Dissertations." Muntabir proposed a CRF-based model which incorporates text-based features and visual features (e.g., the bounding box of a text span) to extract metadata from scanned ETDs. Muntabir emphasized various challenges during the OCR process and showed their method achieved 81.3% to 96% F1 measure on different metadata fields.
.@TasinChoudhury talks about several challenges faced in the OCR process for scanned documents and the need for text alignment. pic.twitter.com/2injcO4Ze0
— Himarsha R. Jayanetti (@HimarshaJ) September 29, 2021
Dr. Shawn Jones from Los Alamos National Laboratory and an alumnus of Old Dominion University and WSDL research group presented their paper titled "It's All About The Cards: Sharing on Social Media Probably Encouraged HTML Metadata Growth." Their research study evaluates the evolution of metadata usage over time. To understand how authors used their metadata budget, they analyzed 227,724 mementos of news articles (e.g., NEWSROOM). They broke down the usage by individual fields and showed that authors favor fields associated with social cards.
Another great and engaging talk by @shawnmjones at #JCDL2021 on the #Twitter Card standard and his research paper: ''It's All About The Cards: Sharing on Social Media Probably Encouraged HTML Metadata Growth''.
— Corinna Breitinger (@BreitingerC) September 29, 2021
Preprint available here: https://t.co/hfhqA0Co0J pic.twitter.com/4ilu8tzQct
Interesting work from @ShawnMJones from @WebSciDL at #JCDL2021 regarding using social cards with web archives.
— Mat Kelly (@machawk1) September 29, 2021
However, one standout aspect is how he formatted his presentation -- not just a floating head over slides but embedded branding.
👇 is his rec. video being played. 🤯 pic.twitter.com/XoTsWOTPTr
.@shawnmjones(Dr. Jones) from @LosAlamosNatLab and a @oducs and @WebSciDL alumnus is presenting his paper titled "It’s All About The Cards: Sharing on Social Media Probably Encouraged HTML Metadata Growth" @JCDLConf #jcdl2021.
— Muntabir Hasan Choudhury (@TasinChoudhury) September 28, 2021
pre-print: https://t.co/L9yN4cMHS1 pic.twitter.com/rRNDTGPLSi
Dr. Alexandar C. Nawla from Observatory on Social Media (OSoME) at Indiana University, Bloomington, and an alumnus of Old Dominion University and WSDL research group presented their paper titled "Garbage, Glitter, or Gold: Assigning Multi-dimensional Quality Scores to Social Media Seeds for Web Archive Collections." Dr. Nawla explained that how reference-rot usually causes significant web resources to disappear. He also emphasized how web archive collections (e.g., seeds) selected by experts or scraped from social media can help mitigate the effects of this reference rot.
#JCDL2021 main conference Paper session 1 with @acnwala on Garbage, Glitter, or Gold.
— Ajayi Kehinde Peter (@AjayiKehindep) September 28, 2021
Preprint: https://t.co/AFgKxyCGi5
I'm honored to present our (@weiglemc @phonedude_mln) paper today (10:30am ET) in the first paper session of @jcdl2021 #JCDL2021
— Alexander C. Nwala (@acnwala) September 28, 2021
cc @WebSciDL https://t.co/fCHNzZ4tZ7
Dr. Sawood Alam from Internet Archive and an alumnus of Old Dominion University and WSDL research group presented their paper titled "Profiling Web Archival Voids for Memento Routing." This work defines and explores Archival Voids to establish a means to represent portions of URI spaces that are not present in a web archive.
Now @WebSciDL alum @ibnesayeed is presenting "Profiling Web Archival Voids for Memento Routing" #jcdl2021
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
Preprint: https://t.co/g6XA5I2vca
Slides: https://t.co/grKLWRwLQq pic.twitter.com/CZ6c9YNkaQ
Another interesting paper titled "ScanBank: A Benchmark Dataset for Figure Extraction from Scanned Electronic Theses and Dissertations," by Sampanna Kahu from Virginia Polytechnic Institute and State University and co-authored by Dr. Jian Wu from Old Dominion University, presented their study and methods on extracting figures and tables from scanned ETDs. Along with WSDL papers, some other exciting papers, including "S2AND: A Benchmark and Evaluation System for Author Name Disambiguation", "Automatic Metadata Generation for Fish Specimen Image Collections", "NoteLink: A Point-and-Shoot Linking Interface between Students' Handwritten Notebooks, and Instructional Videos" were presented by Shivashankar Subramanian, Joel Pepper, and Ranjitha Jaddigadde Srinivasa, respectively.
Sampanna Kahu from @VT_CS is presenting their paper titled: ScanBank: A Benchmark Dataset for Figure Extraction from Scanned Electronic Theses and Dissertations at #jcdl2021 @JCDLConf. https://t.co/DDt14O0f1i pic.twitter.com/fZRys4WYgH
— Muntabir Hasan Choudhury (@TasinChoudhury) September 28, 2021
Shivashankar Subramanian is presenting "S2AND: A Benchmark and Evaluation System for Author Name Disambiguation" #JCDL2021
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
Preprint: https://t.co/Jua4mzgg9W
This is something that those working in bibliometrics (@vneblitt) struggle with when manually evaluating authors' work. pic.twitter.com/atqMUs0NAl
#JCDL2021 Joel Pepper from @DrexelUniv is presenting "Automatic Metadata Generation for Fish Specimen Image Collections"
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
They are working to develop with automated tools to ascertain metadata about images. pic.twitter.com/050DCLLKvN
— Muntabir Hasan Choudhury (@TasinChoudhury) September 28, 2021
Keynote 1
While searching about "Black Girls" on Google, some ubiquitous result returns, an abuse of Google Search Engine. This left her question, how does this search engine algorithm works? pic.twitter.com/bijL3NidaK
— Muntabir Hasan Choudhury (@TasinChoudhury) September 28, 2021
@safiyanoble mentioned a time before search engines where "online libraries" existed. She mentioned how Brin/Page were concerned about abuse of Google. Why do these results exist in her "black girls" search example? What does this mean about the search algorithm? #JCDL2021 pic.twitter.com/kEWfQjMqdu
— Shawn M. Jones, PhD (@shawnmjones) September 28, 2021
#JCDL2021 @safiyanoble is discussing how these racist and sexist search results do not exist in a vacuum. We see that the stereotypes are embedded in our history and culture. Solving the problem is an interdisciplinary challenge.
— Shawn M. Jones, PhD (@shawnmjones) September 28, 2021
Source of image in slide: https://t.co/33xMv8zLXQ pic.twitter.com/cowy6jgUx4
In her keynote, Dr. Noble mentioned a time before search engines where "online libraries" existed. However, a lot of things are concerning especially abusing of Google search engine. While searching "black girls" on Google, it returns racist words. Dr. Noble mentioned, "Why do these results exist in her "black girls" search example?", and "What does this mean about the search algorithm?" Dr. Noble also discussed how racist and sexist search results do not exist in a vacuum. We see that the stereotypes are embedded in our history and culture. Thus, solving this problem became an interdisciplinary challenge.
Day 3
Keynote 2
Per Twidale: "We are always constructing metaphors to make sense of things in order to learn."
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
Per Colburn & Shute: "the language of computer science is laced with metaphor"
What metaphors do we use? When do we use bad metaphors? Dead metaphors? Zombie metaphors?#JCDL2021 pic.twitter.com/iUbhPqqYZZ
Twidale cautions against having tight metaphorical thinking. In computer science you can mix metaphors to understand what is going on. Metaphors help us learn and communicate ideas.#JCDL2021 pic.twitter.com/CUKzMUIK1T
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
#JCDL2021 Per Twidale: Digital Libraries are like open stack libraries.
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
Is the narrow slot from the physical library represented by the text box in our search engines?
What other methods could we use other than a narrow slot? pic.twitter.com/5Vuh23eWYC
#JCDL2021 Per Twidale: think of the physical library as a super high resolution data visualization -- the bandwidth and refresh rate of reality is amazing. pic.twitter.com/98BZSrndO4
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
Per Twidale: How long should a Digital Library last? Some libraries are centuries old. Linnaeus' metadata standards still influence metadata today. #JCDL2021 pic.twitter.com/zqsD3ifZod
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
Twidale is now suggesting that we apply design thinking to research ideas. Ideas can be designed and constructed. There is a design process of how you look for a problem area, empathize, ideate, prototype, and test it. #JCDL2021 pic.twitter.com/J8zyNFuI3j
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
#JCDL2021 Twidale: It is important for a software tester, UX designer, and researchers to stray from the "happy path". Asking awkward questions? What are the unintended consequences? How will your system integrate with others in a software ecology? pic.twitter.com/sfbWuBhwCq
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
Twidale wraps up by mentioning that metaphors can be a mirror to help us see what we are doing and how we might do things better. They can help us design better research and better systems. They can help us spot problems. #JCDL2021 pic.twitter.com/4Trl2dvZ81
— Shawn M. Jones, PhD (@shawnmjones) September 29, 2021
In conclusion, Dr. Twidale wrapped up that metaphors are not just for end-users, and they can help us design better research, better systems, and spot problems.
Posters and Demo Sessions
.@JCDLConf #JCDL2021 minute madness of Poster/Demo sessions. @WSDL members @mahanama94, @Gavindya2, and @OpenMaze will be presenting today from @NirdsLab.
— Muntabir Hasan Choudhury (@TasinChoudhury) September 29, 2021
Youtube demo: https://t.co/eU6TfIynkO pic.twitter.com/vHa4ZqncSA
.@shawnmjones from @LosAlamosNatLab, a @WebSciDL and @oducs alumnus presenting his poster titled "Hypercane: Intelligent Sampling for Web Archive Collections" @jcdl20201 #JCDL2021. pic.twitter.com/p43tlwwS8U
— Muntabir Hasan Choudhury (@TasinChoudhury) September 29, 2021
some other interesting posters and demo sessions @JCDLConf #JCDL2021 pic.twitter.com/QyOfkmd5Ey
— Muntabir Hasan Choudhury (@TasinChoudhury) September 29, 2021
Very cool UI of @JCDLConf #JCDL2021 poster sessions. pic.twitter.com/8PmWtvXSO8
— Muntabir Hasan Choudhury (@TasinChoudhury) September 30, 2021
.@WebSciDL gathering @JCDLConf poster session. :) #JCDL2021 pic.twitter.com/NY8pMw65fN
— Bhanuka Mahanama (@mahanama94) September 30, 2021
Our annual @WebSciDL members + alumni group pic at the #JCDL2021 poster session!
— Michael L. Nelson (@phonedude_mln) September 30, 2021
Also at @JCDLConf but not pictured: @OpenMaze @mart1nkle1n @kritika_garg @HimarshaJ pic.twitter.com/DxW9yhVJVF
Closing Ceremony
- Vannevar Bush Best Paper Award: "Towards Novel Information Access to the Past: Estimating Contemporary Relatedness Degree of Archival News Articles" by Mari Sato, Adam Jatowt, Yijun Duan, and Masatoshi Yoshikawa
- Best Student Paper Award: "Automatic Metadata Generation for Fish Specimen Image Collections" by Joel Pepper, Jane Greenberg, Yasin Bakis, Xiaojun Wang, Henry Bart, and David Breen
- Best Poster Award: "Analyzing Unconstrained Reading Patterns of Digital Documents Using Eye Tracking" by Bhanuka Mahanama, Gavindya Jayawardena, and Sampath Jayarathna
Congratulations to Mari Sato, Adam Jatowt, Yijun Duan, and Masatoshi Yoshikawa for winning the #JCDL2021 Vannevar Bush Best Paper Award for "Towards Novel Information Access to the Past: Estimating Contemporary Relatedness Degree of Archival News Articles" pic.twitter.com/WPkl0sJ93f
— Shawn M. Jones, PhD (@shawnmjones) September 30, 2021
Congratulations to @WebSciDL members @mahanama94 @Gavindya2 @OpenMaze for winning the #JCDL2021 best poster: "Analyzing Unconstrained Reading Patterns of Digital Documents Using Eye Tracking" pic.twitter.com/doi4iBVU45
— Shawn M. Jones, PhD (@shawnmjones) September 30, 2021
#JCDL2021 Congratulations to the winners of this year's Best Poster Award! pic.twitter.com/cmOvHQO4RX
— JCDL Conference (@JCDLConf) September 30, 2021
— Gavindya (@Gavindya2) September 30, 2021
Later, Dr. Stephen Downie thanked the two keynote speakers Dr. Safiya Noble and Dr. Michael Twidale, at JCDL 2021, for giving an excellent and insightful talk. He also thanked the committee chairs and student volunteers, and sponsors.
Thanks to #JCDL2021 committee: @profdownie, Hussein Suleman, @DanaChatter, David Nichols,@Faryane, @OpenMaze, @FLoizides, claire timpany, @acnwala, Dion Goh, Xiao Hu, @GeorgeRBuchanan, Edie Rasmussen, @machawk1, and Dan Wu!
— Shawn M. Jones, PhD (@shawnmjones) September 30, 2021
Also thanks to our student volunteers! pic.twitter.com/wF68X9D3f9
#JCDL2021 is not quite over, but #JCDL2022 @jcdl2022 is being planned. It will be in Cologne, Germany and will take place June 20 - 22, 2022.
— Shawn M. Jones, PhD (@shawnmjones) September 30, 2021
Papers due: 10 January 2022
Site: https://t.co/CQx64stLWd
No longer a distinction between long and short papers. pic.twitter.com/HldwzmecI5
Finally, JCDL 2021 ended by announcing its next conference date (June 20-24, 2022), and it will be held in Cologne, Germany.
-- Muntabir Choudhury (@TasinChoudhury), Ajayi Kehinde Peter (@AjayiKehindep)
Comments
Post a Comment