Show HN: Crowdsourcing academic profiles for 3,600 computer science professors | Hacker News Reader

Show HN: Crowdsourcing academic profiles for 3,600 computer science professors

drafty.cs.brown.edu

46 pointslazyjeff13 comments

So, what's the point? I'm genuinely asking. Is that dataset interesting by itself or is it the crowdsourcing that is studied? I got this impression from a message I got when the page finally loaded, just before it crashed with an out of memory error, it was saying something like "you are free to edit, your edit actions are recorded".

Sorry, was out all day and so was the graduate student working on this. It's back up and might be more scalable now...

The dataset could be interesting for anyone who wants to analyze it for things like hiring trends, or for someone who wants to apply to grad school in a certain topic or faculty positions at a department.

The platform itself is a way to keep dynamic data up to date by asking visitors for help with fixing data. In some cases, it tries to figure out what your interests are (universities or research areas you look at most), and ask you about those.

The smart up to date part sounds pretty cool. Can you elaborate on how you do that?

I could see some interesting time lapse visualizations coming out of this. For example: Have CS departments changed over time to have more / less full professors vs associates? Or adjuncts vs full-time faculty? Is it different for public vs private schools?

I have some hunches but it would be cool to actually execute these queries on data.

Exactly. Like which research areas have universities been hiring for recently, is the male/female ratio of hiring improving for some universities, etc.

Ended up with a 502 bad gateway. As a part time adjunct, I'm looking forward to being able to see this.

Can't even access using good ol' archive.org and Google Cache :D

Great work @lazyjeff, fellow brown alum here, your other projects* look awesome too, thanks for sharing :)

* http://hci.cs.brown.edu/

Works for me (Sun Aug 28 1:52 Eastern US). Looks like a spreadsheet with names, universities, where they got their PhD, .. , up to photo url.

I got this:

  Failed to load the widgetset:
  ./VAADIN/widgetsets/drafty.MyAppWidgetset/drafty.MyAppWidgetset.nocache.js?1472407401168

This is neat. Is the data exploration part something that you've built or off-the-shelf?

We use a java grid library for the table. The part we've built is logging and computing interests based on user interactions, and then prompting the best matching items for visitors to fix.

We'll be open sourcing it when we publish the work, so that anyone can load any updatable data into the platform.

man I can't get in either. this looks fascinating though. can someone describe it?

Reply on news.ycombinator.com