An afternoon of more presentations, which were a bit more technical in terms of data indexing, storage, and use. As a result, there are no detailed posts, but here are a few notes and takeaways.
- Be careful when you try to parse a bunch of files you download from the web. Parse and store, distribute up front, and build a test index first.
- Making Software Work – read it
- The results of one study is not the truth.
- It’s hard to make a study repeatable.
- Does agile work? That’s the wrong questions. Really, when does bug fixing has the highest cost?
- High-risk bugs should be done as early as possible, instead of the easy bugs.
- What language? Depends on the problem.
- Make developer happiness hours. (block off time with no meetings)
- Give people open sight lines instead of high cubicle walls.
- Be as transparent as possible (e.g. JIRA) including progress.
- Put projects into short iteration cycles.
- No code without passing tests!
- Slides (PDF) for the last talk: Practical Agile: What’s Working for Stanford, Blacklight, and Hydra by Naomi Dushay
In-browser Data Storage and Me
by Jason Casden, North Carolina State University
- Suma: data collection application using in-browser storage.
- Indexed database API (aka IndexedDB, WebSimple DB) is where things seem to be going, but limited browser support.
- Web (DOM) Storage is basically universally supported.
- Web SQ DB still viable option.
- lawnchair: object storer, but have adapters for a long list of DBs/APIs.
I didn’t take full notes on all the presentations. I like to just sit back and listen to some of the presentations, especially if there are a lot of visuals, but I do have a few notes.
Full Notes for the following sessions:
Building Research Applications with Mendeley
by William Gunn, Mendeley
- Number of tweets a PLoS article gets is a better predictor of number of citations than impact factor.
- Mendeley makes science more collaborative and transparent. Great to organize papers and then extract and aggregate research data in the cloud.
- Can use impact factor as a relevance ranking tool.
- Linked Data right now by citation, but now have tag co-occurrences, etc.
- Link to slides.
NoSQL Bibliographic Records: Implementing a Native FRBR Datasotre with Redis
No notes. Instead, have the link to the presentation complete with what looks like speaker notes.
- Things not taught in library school: all the important things, social skills, go talk to the professor directly if you want to get into CS classes.
- Momento project and UK Archives inserting content for their 404s.
- In response to librarians lamenting loss of physical books, talk to faculty in digital humanities to present data mining etc., look at ‘train based’ circulations, look at ebook stats.
- Take a look at libcatcode.org for library cataloguers learning to code as well as codeyear hosted by codeacademy.
by Annie Cain, Harvard Library Innovation Lab
What’s the point?
Why recreate the physical stack? Why not just use a list? There are advantages to display books like books.
- If you have multiple branches, you can put all the materials into one shelf.
- More visual, such as using page numbers to create thinner or wider images.
- Add more information – color can represent such as frequency of checkouts.
How to Get Started
- build HTML objects and draw using CSS
- works as jQuery plugin
- Start with book data – wrestle data out of your catalog, extract pieces you want and feed it into Solr, or use API, such as WorldCat
- pump out stack view using JSON
Code available on github.
- Usability testing gives mixed reviews. Librarians and those used to browsing the library are happy with it, but others don’t seem to care.
- Experimentation with items other than books is to come.
- No plans yet for catalogue integration as not much use case, but working on it.
- Also take a look at Chrome WebGL bookcase.
More information on available on the Harvard Innovative Lab website, and below is a demo.
by Cory Lown, North Carolina State University
While there is only one search box, typically there are multiple tabs, which is especially true of academic libraries.
- 73% of searches from the home page start from the default tab
- which was actually opposite of usability tests
Home grown federated search includes:
- best bets (60 hand crafted links based on most frequent queries e.g. Web of Science)
- spelling suggestions
- loaded links
- smart subjects
Show top 3-4 results with link to full interface.
From Fall 2010 and Spring 2011, ~739k searches 655k click-throughs
- 7.8% best bets (sounds very little, but actually a lot for 60 links)
- 41.5% articles, 35.2% books and media, 5.5% journals, ~10% everything else
- 23% looking for other things, e.g. library website
- for articles: 70% first 3 results, other 30% see all results
- trends of catalogue use is fairly stable, but articles peaks at the end of term
How to you make use of these results?
Top search terms are fairly stable over time. You can make the top queries work well for people (~37k) by using the best bets.
Single/default search signals that our search tools will just work.
It’s important to consider what the default search box doesn’t do, and doubly important to rescue people when they hit that point.
Dynamic results drive traffic. When putting few actual results, the use of the catalogue for books went up a lot compared to suggesting to use the catalogue.
Custom log is being used right now by tracking searches (timestamp, action, query, referrer URL) and tracking click-throughs. An alternative might be to use Google Analytics.
For more, see the slides below or read the C&RL Article Preprint.
by Kirk Hess, University of Illinois Urbana-Champaign
Why Google Analytics?
- small tracking image (visible via Firebug) = mostly users not bots
- works across domains
- easy to integrate with existing system
Some useful things in the interface:
- heat map
- content drill down – click on page and see where users went from there
- visitor flow
Export Data Using API
- Analytics API
- export any field into a database for further analysis (in this case MySQL db)
- Which items are popular?
- How many time was an item viewed?
- Effective collection size – see if people seeing/using
- found typically, many things are not popular
- discover a lot of other things about users
- found, need to change site design
- change search weighting
- allow users to sort by popularity (based on previous data)
- recommender system – think Amazon
- add new tracking/new repositories
- analyze webstats – hard to look at direct access
The event analysis code has been posted on github and adding events to link code will be added later to his Github account.
- technology developed and maintained by California Digital Library
- supports the search/display of digital collections (images, PDFs, etc)
- fully open source platform, based on Apache Lucene search toolkit
- Java framework, runs in Tomcat or Jetty servlet engine
- extensive customization possible through XSLT programming
- user and developer group communication through Google Groups
- search interface running on Solr with facets
- can output in RSS
- has a debug mode
- Aid activities for the Great East Japan Earthquake through collaboration via wiki
- input from museum, library, archive, kominkan = MLAK
- 20,000 data of damaged area
- Information about places, damages, and relief support
- Key Lessons
- build synergy with twitter
- have offline meet ups & training
Andrew Nagy – Vendors Suck
- vendors aren’t really that bad
- used to think vendors suck, and that they don’t know how to solve libraries’ problems
- but working for a vendor allows to make a greater impact on higher education, more so than from one university (he started to work for SerialsSolution)
- libraries’ problems aren’t really that unique
- together with the vendor, a difference can be made
- call your vendors and talk to the product managers
- if they blow you off, you’ve selected the wrong vendor
- sometimes vendor solutions can provide a better fit
Andreas Orphanides – Heat maps
The library needed grad students to teach instructional sessions, but how to set schedule when classes have a very inflexible schedule? So, he used the data of 2 semesters of instructional sessions using date and start time, but there were inconsistent start times and duration. The question is how best to visualize the data.
- heatmap package from clickheat
- time of day – x-dimension
- day of the week – y-dimension
- could see patterns in way that you can’t in histogram or bar graph
- heat map needn’t be spatial
- heat maps can compare histogram-like data along a single dimension or scatter-like plot data to look for high density areas
Nettie Lagace from NISO
- National Information Standards Organization (NISO)
- work internationally
- want to know: What environment or conditions are needed to identify and solve the problem of interoperability problems?
Eric Larson – Finding images in book page images
A lot of free books exist out there, but you can’t have the time to read them all. What if you just wanted to look at the images? Because a lot of books have great images.
He used curl to pull all those images out, then use imagemagick to manage the images. The processing steps:
- Convert to greyscale
- Contrast boost x8
- Covert image to 1px by height
- Sharpen image
- Heavy-handed grayscaling
- Convert to text
- Look for long continuous line of black to pull pages with images
Code is on github
Adam Wead – Blacklight at the Rock Hall
- went live, soft launch about a month ago
- broken down to the item level
- find bugs he doesn’t know about for a beer!
Kelley McGrath – Finding Movies with FRBR & Facets
- users are looking for movies, either particular movie or genre/topic
- libraries describe publications e.g. date by DVD, not by movie
- users care about versions e.g. Blu-Ray, language
- Try the prototyped catalog
- Hit list provides one result per movie, can filter by different facets
- don’t over rely on the context
- but context is still necessary for understanding e.g. “mobile” – means on the go, what they want on the go
- sometimes there is no better term e.g. “Interlibrary Loan”
- brevity will cost you “tour” vs. “online tour”
- Time ran out, but check out the rest of the slides
Simon Spero – Restriction Classes, Bitches
- lets you define properties
- control what the property can apply to
- control the values the property can take
- provides an easy way to do this
- provides a really confusing way to do this
The easy way is usually wrong!
When defining what can apply to and the range, this applies to every use of the property. An alternative is Attempto.
- Processing: open source visual programming language
- Processing.js: related project to make processing available through web browsers without plugins
- While both tend to focus on data visualizations, digital art, and (in the case of PJS) games, there are educational oriented applications.
- Kanji Compositing – allows visual breakdown of Japanese kanji characters, interact with parts, and see children.
- Primer on Bezier Curves – scroll down to see interactive (i.e. if you move points, replots on the fly) and animated graphs.
- Obvious use might be instructional materials, but how might we apply it in this context? What other applications might we think of in the information organization world?
Since doing the presentation, I have already gotten one response by Dan Chudnov who did a quick re-rendering of newspaper data from OCR data. Still thinking on (best) use in libraries and other information organizations.
It’s over for today, but if you’d like more, do remember that there is a livestream and you can follow on twitter, #c4l12 or IRC.
Once again, I didn’t take full notes on all the sessions, but some takeaways below.
- Non-English searches should not suck.
- Favour precision over recall on large-scale searching.
- Develop measures of assessment in order to measure success.
- Leverage the correlation between academic degree and type of materials used, and focus on discipline-related materials and authors in case of ambiguity.
- If a user built-in interface doesn’t work, you can always put something on top.
Many of these sound like common sense, but not enough people do it.
See my other posts for notes on the presentations I wrote more on:
by Lisa Kurt, University of Nevada
If you can get three things down, you can get a good design:
- Typography – simple
- Composition – a lot of white space, conventions
- Colour – minimal
Study the designs that you love and those that you hate. What works and what doesn’t?
On Photos: If you use clip art, don’t use clip art that looks like clip art.
Look at designs with fresh eyes. Make sure it’s balanced.
Have fun too!
Really know your audience. Beware of decorative typeface: it can become hokey, very quickly because they look more like illustrations.
Designing for Mobile: Sans serif and white background with dark text is easier to read on mobile.
While you need to be careful of branding, you can use it to link different elements together.
Design by committee does not work! Provide three design and be firm that you will not combine them, etc. Usability can help support your design.
For more, check out Lisa’s website and the presentation slides below.
by Michael Poltorak Nielsen, Statsbiblioteket/State and University Library, Aarhus, Denmark
Current Mobile Interaction Paradigm
You do a lot with your hands, everyday. Our hands are a really good tool, but currently, the handheld interaction is based on glass. That is you do functions by sliding your fingers, which means there is no feedback on what it does, i.e. it’s not intuitive.
Take a look at Pictures Under Glass: Transitional Paradigm dictated by technology, not human capabilities by Bret Victor.
- direct manipulation
- gesture driven
The near future may mean combining something like the Wiimote and the iPhone.
The idea was to build an HTML5 app that searches library data, favourites, view own items, renew, and request. Currently in beta, but to be published soon.
The search app can be augmented with gestures, gestures combined with multi-touch interactions.
Possible interactions with focus on
- keyboard – typing
- microphone speech
- screen – touch, visuals
- camera – pattern, movement
- accelerometer – acceleration
- gyroscope – rotation
- compass – direction
- GPS – movement, position
Might include simple ones using accelerometer data, including
The problem is that gestures are only really supported by Firefox, and partially supported by Chrome. Thus, it was decided that development would move to the native iPhone app environment with gestures, and HTML5 web app without gestures (but possibly later when supported). Features that are implemented include:
- Restart search – face down
- Scroll – tilt up and down
- Switch views – tilt
- Request items – touch and tilt left
- Favourites – touch and tilt right
Check out the demo:
- no standard mobile gestures
- gesture maybe individual
- gesture may not be appropriate at all
- sophisticated gestures are hard to code
I did not take notes on everything in part because some of it was very technical and it can be hard to do notes, but here are some takeaways from the morning:
- Versioning Control: Use it, Git or Mercurial. Doesn’t need to be code, can be data too. – Description and Slides
- Take library data and make it available to users, can’t expect them to search for it.
- Linked Data doesn’t need to be a huge project. Start small.
- Why RDF? It’s flexible with easy addition of new attributes or classes, and works cleanly with an iterative approach.
HTML5 Microdata and Schema.org
by Jason Ronallo
Other than getting good ranking, we need to provide rich results, i.e. rich snippets. Some digital collection have been providing rich snippets already, such as NCSU Libraries.
How do we get this?
- embedded semantic markup
- HTML5 Semantics include nav, header, article, section, footer
- HTML5 Microdata is a syntax for annotating content to communicate meaning of data to machines
- similar to RDFA, other microdata
- Microdata comes back as tree based JSON and allows for DOM API
<div itemscope itemtype=”http://schema.org/Organization” itemref=”logo”>
<a itemprop=”url” href=”http://code4lib.org/”>
where: scope = about something
type = type of item
prop = properties
For the user, there is no difference as display is the same. This provides a complete data model.
Schema.org is a one-stop shop for vocabulary in describing items on the web.
Apologies, I did not take extensive notes on it, but to read more, check out the slides below or the Code4lib article he wrote.