Marian Dörkin 1996, @[email protected] gave us the task by data type taxonomy (with overview first etc), but when did the very prolific, yet slightly dated triad user–data–tasks become a thing in visualization? asking for a friend
Carlos Scheidegger
- 18 Followers
- 21 Following
- 35 Posts
https://cscheid.net assistant prof at arizona computer science. data analysis, data vis. humans, data, scale
@aparrish if you take "Adaptation." 's script to be a piece of written work, it is presumably partly about itself. Does that count as an infinite chain? :)
I literally wrote that section of the paper 45 minutes ago. If you ask me what is the design of the study we used, I can tell you "yes, we showed all independent factors to each participant". But I cannot tell you if that's "within" or "between" or "across" or whatever-they-feel-like-calling-it-today
(Add another one to my growing suspicion that I have some weird form of dyslexia-spectrum thing) I can never remember if it's across-subjects or within-subjects or between-subjects. I know the _meaning_, but the word association never sticks to my head. Might as well keep that google tab open forever
I'm hacking on some JSX+React stuff my student wrote. I feel so modern.
@rpgove my point is that the "standard" datasets play much the same role as synthetic datasets, and in this sense they _are_ important.
@rpgove well yeah, you don't only show things on synthetic datasets. But you also don't only show things on "real" datasets.
Well, that's one more thing Kosara and I disagree about.
Standard datasets _are_ important. You literally don't want to be surprised by the dataset when you're trying to understand a technique.
Synthetic examples serve a similar purpose: you _know_ what's in them, so you can check that your technique is doing what you expect.
If you design a new technique and show it on a new dataset, readers can't separate one from the other.
@nrchtct do you have a link available?
@aparrish I don't think anyone has done exactly what you're suggesting, but I remember this: https://erikbern.com/2016/01/21/analyzing-50k-fonts-using-deep-neural-networks.html