Posts

Some notes on final grading…

As mentioned in Discussion Feedback and Grading, there is no simple way to get grading for the online parts of class (Online Discussions and Seek and Finds).

There are the “quantitative score” (the 1-5 score you got each week) - these are uploaded to Canvas.

Then there is the qualitative part. Where we group things together into 5 week chunks and give an overall assessment. We had given you the statistics for your posts for the first weeks, but the actual grades were done manually.

The “qualitative scores” are posted as a Canvas comment to the grading column “Qualitative Discussion” (please do not reply to Canvas Comments!).

We will explain how we will use these scores later.

For the past 15 weeks, you would be expecting the “Week in Vis” posting around now.

To help you transition to the world without the weekly Friday posting, here is a last one. (probably not the last posting)

There is a survey Survey 15: End of Semester (due Fri, Dec 11) - technically, it’s due today, but I won’t fault anyone for doing it after DC3 is handed in. In fact, you get my sincere appreciation for doing it at all. The feedback will really help me plan future classes. The survey will stay open through next Wednesday (the 16th).

As you know, the DC3 Deadline is here. The deadline is officially Sunday Monday, 11:59pm (yes, there was a typo that said the 14th was Saturday). Please turn everything in as promptly as possible.

The project description is a little vague on turning things in late. It says “this deadline is firm” in bold, but it also says you can turn things in late (on the 15th) for a penalty. Let me pre-empt the wave of Piazza posts and emails…

On Monday Tuesday morning (December 15th), Young and I will meet and start grading. If you turn things in before then, we won’t notice. And no, I can’t tell you when we will start (OK, it won’t be before 9am).

Once we start grading… we will look through all the assignments and put them into batches so we can group like assignments together. If your assignment isn’t there, we can’t determine which batch to put it in. Also, we will use Canvas to download everything in “batch” - so if yours isn’t there, we will have to make a special trip to Canvas. It’s a hassle for everyone. Also, when we do our initial check on Monday Tuesday, if there’s a problem, we will contact you with a chance to fix it. If you try to turn things in late, you will lose that opportunity. So, there’s no specific penalty - but it affects how we evaluate assignments.

So, just please turn everything in before Monday Tuesday morning.

A Proper Ending for Class

Having BBCU die in the middle of lecture was not the way I had hoped to end the semester. I am not sure what to do after that. I’ll probably make a posting at some point over the weekend. There will also be a final posting explaining the details of how grading worked out. And maybe something about how to pursue Vis beyond class for those who are interested.

And about grading details… We have a lot of grading information, but we might not be able to get it all uploaded to Canvas.

Instead of Online Discussion 14, I am going to ask you to do a design exercise.

You will turn it in as Online Discussion 14: Dear Data Design Exercise - the initial posting is due on Friday. I do hope that you’ll try to give feedback and critique to others after that!

The exercise is based on the book Dear Data. I’ve done variants of the exercise in class for a few years - it’s always fun and inspiring. This year, we don’t have time in class, so we’ll try parts of it as an “at-home” design exercise.

Understanding the nested model is really important. So, we’ll have an exercise that will force you to use it and think about how it applies to DC3: The Tree of Stuff (which you should be working on).

This is an exercise I usually do in class. We’ll do it individually as “homework”. You must turn it in. We will grade it for “effort” (check/no check, and maybe a notation for something exceptional).

You must draw out all of the parts. Draw it by hand - do not use software to do it (that defeats the purpose). Turn in a PDF or image file with your pictures. Graph Exercise (due Mon, Nov 23)

One caveat: once you see “the answer” everything else is obvious. Try not to cheat.

This is an alternate option for DC3. I expect most people will want to do the regular DC3: The Tree of Stuff assignment. This assignment may be more work, and is more exploratory, but if it connects to other work that you would want to do anyway, then it may be attractive to you.

Visualization for machine learning is a hot topic that many people are interested in (from the Vis side and the ML side). Given the interest in it, I wanted to let students have the option of doing their final “project” in this domain. But, the problem is that in order to do visualization on machine learning things, you have to have done the machine learning work to have things to visualize.

If you are willing to do some machine learning work to generate your own data read on…

Here is some information on how DC1 was graded.

I know you were expecting Design Challenge 3 (since we skipped Design Challenge 2) - but first, we’re bringing back DC1.

In this assignment, we will give you a chance to revisit DC1, to allow you to learn from the feedback that you have gotten. Note that this counts separately from DC1 - we will grade it separately, and give you a letter grade. Since it’s a 1 week assignment, it will count 1/4 as much as DC1 and DC3.

For Monday, November 9th, you may turn in another design that presents the same story as one of your designs.

For online discussions and seek and finds, we are giving you two different evaluations. One is the per-assignment “quantitative” evaluation - this is posted for each assignment, and is on a 1-5 scale. The other is the more “qualitative” feedback, where we look at your work over the course of a few weeks and give you a grade.

This posting explains these two, so you can interpret the scores you are getting.

Here are the examples for the 2nd interaction lecture - I wanted to put the links in one place so you can try them yourself! (and I had a place to copy/paste them into the chat)

As part of class, you are required to “attend” some of IEEE Vis2020 (VisWeek). This posting explains what you need to do for class, and provides some advice on how to find things at the conference.

A warning: this is the first virtual IEEE Vis conference. It’s all an experiment. I think they are figuring things out as they go. My advice is based on older conferences.

The virtual conference thing is an experiment. It is an experiment for the conference. It is an experiment for this class. I have no idea how well it will work out. But I think it’s an exciting opportunity. Even if you never look at visualization work again, learning about online conferences and presentations will be useful.