Month: August 2014

TA/GA Grievance Committee

The agreement between the labor union that represents the teaching assistants, graduate assistants, and faculty at Rutgers University stipulates that there needs to be a protocol to follow when teaching assistants and graduate assistants take issue with reappointment procedures. This is outlined in Article XI of the agreement, “TEACHING ASSISTANT/GRADUATE ASSISTANT PERSONNEL GRIEVANCE PROCEDURE”. At any point when a TA or GA is not reappointed, they can complain to a department representative. If they do not feel the representative has been able to sufficiently resolve it (which, I imagine, would only be in the case when there is a simple filing error and not a flat-out rejection of reappointment), then the aggrieved party would submit a formal request in writing to a committee charged for this purpose on the campus. In this instance, this committee is called the “TA/GA Grievance Committee” and is a necessary step toward to resolution of a reappointment dispute.

The committee needs to convene and make a decision about whether or not to support or deny the claim from the spurned TA/GA. If necessary (meaning if someone asks), a meeting will be held with the TA/GA to hear the case in full. Afterwards, the committee has ~20 days to render a decision about whether to reject or accept the TA/GA’s request. At that point, it is handed up to the Dean of the Graduate School to make a final decision based on their opinion of the matter and the opinion of the committee.

The TA/GA committee is selected by the EVPAA in the New Brunswick campus of Rutgers University and consists of of 3 faculty members and 2 TA/GA members. The pool is selected from the nominees of department chairs whom select internally from their own department one faculty and one TA/GA.

My department asked if I would be willing to serve on the committee and I said I would. Honestly, it is something that interests me; it is useful to see how TA/GA disputes might play out in the wild and I am vested in making sure that graduate students get the right treatment. Maybe it isn’t proper to say this, but graduate students are effectively a “vulnerable” population; we are entirely at the whim of our department and, even more so, our academic adviser.

The adviser holds so much power in a relationship that everything is tipped in their favor and they can get basically any outcome they want regarding their student. If for whatever reason the student and them part ways, the adviser can entirely submarine him or her without having to back it up. As such, being on a committee like this would allow me to bear more witness to that. My relationship with my adviser is not like this at all and I don’t expect it to be, but I’ve borne witness to bad situations with fellow graduate students and their advisers enough to know that I would like to help if I can (AND, most importantly, if it is appropriate!) The graduate student is not always in the right, but I am merely saying that they are at a disadvantage >80% of the time). But I suppose a further expose on a subject like this warrants a different blog post so I digress.

So two issues remain, I guess:
1) The Dean doesn’t have to follow the recommendation of the committee. Not knowing anyone else who has served on this, faculty or otherwise, I cannot say what the usual outcome is. I imagine the Dean will select whatever option they want or best serves the university. I honestly feel that a graduate student appealing a reappointment is forever at a disadvantage and I would be surprised if they ever got reappointed. And then, I sincerely doubt they would be appointed again after that. Most likely they are persona non grata in their department after such a kabuki show.

2) I am not yet on the committee, I would need to be selected from the pool of applicants. Maybe if I just contact the EVPAA directly expressing interest to be on it, they would place me on it after keeling over in shock that someone forcefully volunteered to be on a committee that may only exist to placate a scorned graduate student before their final beheading (academically speaking).

It is interesting to be nominated, though! I consider it a great honor and would happily engage in committee activities. It’s a nice outlet from research and good practice for if I ever become faculty somewhere else, where committee membership is a more important credential of keeping your job than research.

SOUPS – Day 1 – WAY Workshop

I need to do more writing on this. I have to cover all 3 days and it is time consuming to do. Soon! Soon! All of SOUPS will be up here…

 

My overall impressions for the WAY Workshop are that it was: 1) very interesting and 2) seemingly relevant. There were presentations on many diverse topics of authentication, so kudos to the workshop chairs Larry Koved and Elizabeth Stobert for doing a good job on that.

IEEE Pervasive Computing Magazine

My paper  to the IEEE Pervasive Computing Magazine special edition on Security & Privacy has been accepted! This marks the end of a long and arduous journey from the original submission way back in early 2014. The paper is a broad strokes paper on the state of gesture recognition in the context of authentication. Currently the title is Engineering Gesture Recognition for Authentication but the title is subject to change by the IEEE editors, so we’ll see what it ends up being. Something more flowery and fancy, maybe!?

Admittedly, the first iteration of the paper was not something that was up to par for publication. It was written too quickly, overly long, and contained highly-technical discussion of various concepts. The second iteration was better; we had better figures/tables and removed a highly specious section on use cases and replaced it with an evaluation of gestures focusing on usability & security. This final iteration concerned a full-scale attack on unclear statements, unexplained reasoning, and generally form/flow/structure. This final result, I believe, is quite a good effort. This was an interesting learning experience in how to write a good paper. An experience worth having.

Much thanks to the reviewers and colleagues for giving us tons of useful advice to improve the organization and clarity of our paper! It is even better that there were no run-ins with adversarial reviewers.

Don’t know when this will be published but I’m hoping it is at some point in 2015! There’s no date set on the calendar which makes me think it might not be until 2016. Which would be annoying but understood since the editors have quite a job ahead of them reducing computer science and engineering gobbledygook to readable material.

SOUPS 2014 – Day 0

This post is about the SOUPS 2014 Conference I attended that took place July 9-11, 2014, Menlo Park, California. I meant to write a post after the conference but I spent too long getting my website ready.

I had submitted a paper to Who are you?! Adventures in Authentication: WAY Workshop but it got rejected. At the same time, I had applied for the travel grant right after I submitted the paper. One day, at noon, I received notification that the paper had been rejected and I said, “Well, there goes SOUPS.” Fast-forward three hours later, I received an email from Tiffany Todd at Carnegie Mellon University informing me I had received their travel grant! Naturally, I accepted and was on my way to SOUPS as a paid-for tourist.

This trip represented a few firsts for me:

  1. First time going to California.
  2. First time flying Virgin America.
  3. First time attending SOUPS!
  4. First time using a ridesharing service — Lyft.

Naturally, I was gung-ho about the whole deal. I scheduled a 1:00PM departure from Newark Airport to San Francisco with a connection in Los Angeles. I got to the airport at 12:00 since, with online check-in, I figured I was alright with time. To my dismay, the TSA line for Terminal A was exceptionally long. So long, in fact, it stretched outside of the normal zig-zag holding area, past the adjacent restaurants & checking cashing places, landing almost near where you would board the AirTrain. In other words, this line was ABSURDLY long. It got to the point where, at 12:40, I had to ask people in front of me if I could cut in line or else I would miss my flight entirely. I made it to my flight with a few minutes to spare; we took off within 3 minutes of my boarding and getting settled. Needless to say, I will not underestimate Newark Airport again.

The Virgin America flight was relatively comfortable. I preoccupied most of my time by reading the papers for the WAY Workshop and working on stuff for the Machine Learning course I was taking. The most interesting thing that occurred on the flight was, since the World Cup was still going on, the now-infamous Germany vs. Brazil match. This man started yelling and sliding out of his seat in shock over Germany’s non-stop scoring offensive. Honestly, it was like watching a basketball game with a score like that.

The connection in LAX resulted in a delayed flight. This meshes with everything I’ve been told about that airport so I’m not surprised. Also the places to eat in that terminal were terrible. :/

After arriving in SFO, I wandered to the entryway to the airport and requested a Lyft driver. At that time, San Francisco Airport had recently banned ridesharing services from operating inside their property so I was curious to see if anyone would show up. My driver did show up, but his car was bereft of the typical adornments Lyft advertising had told me to expect. Sunny, the driver, informed me that they had to strip that off in order to keep operating in the airport otherwise they would be escorted out by airport security. Additionally, they don’t park outside the pickup lanes anymore but stay right off of airport property. So they were trying to skirt the law. I didn’t complain, since my 26 mile drive only ended up costing me $25! Hearing from others at the conference that their costs were around $80 or higher, I felt I’d really won out.

I ended up arriving at the hotel at 10PM, feeling severely tired. The accommodations were nice; ultimately, a soft bed wins me over. What was odd was that I had pictures of pebbles and wheat in my room, which I didn’t quite understand. It reminded me a lot of how SUBWAY puts up pictures of fresh, tasty looking bread and vegetables. Neither of these places had anything approaching what were on these pictures so I didn’t understand the point. Subliminal messaging?

I ordered a bit of food for the room and sat around reading the papers I hadn’t finished on the plane. I didn’t examine them in exacting detail since they’d all be presented tomorrow anyway. I ended up sleeping after finishing my food and set alarm for around 6:00AM.

UIST Student Competition

Our proposal to the UIST Student Competition has been accepted!
The team consists of:
Point-of-contact: Gradeigh D. Clark
Victor Kaiser-Pendergrast
Elie Rosen
Yulong Yang
Faculty Adviser: Janne Lindqvist

This year’s competition focuses on innovations with the Kinoma Create. From what I gathered while doing research for this, the Create is a maker kit designed for software developers. Apparently coders have a lot of trouble doing hardware projects with Arduino and sensors? I didn’t know that was a problem that needed to be filled, but a successful crowdfunding venture says otherwise.

The creators describe the device as a “The JavaScript-Powered Internet of Things Construction Kit”. I look at it as an Arduino with a 3″ display unit on it. They really want web developers involved in maker projects and they are really into using JavaScript. Extremely into the idea. Constantly, the notion that because this thing uses JavaScript it must be easier to use. I’ve never had that experience (though I’ve been told it’s “because of user error”).

Not to mention the Kinoma Platform Runtime has this frightening “plus”: “Created by architects of QuickTime, Apple Media Tool, and iShell”
The architects of QUICKTIME worked on this? Yikes.

Some thoughts when looking at their specifications:

Speaker and microphone – I don’t know why, but the speakers made me think of this

No breadboard required for many breakout-board-based sensors – The video on the website showed 2(I think?) projects that had external breadboards dot dot dot

Custom, lightweight Linux distribution to support the Kinoma platform – This never fails to strike fear in my heart.

PWM (3) – Well, on Arduino, every pin is a PWM … 🙂

After “much” brainstorming, our ideas were:


  1. Ephemeral Locks – Use the touchscreen interface with a companion application to do dynamic locking of any door in the house (or near the Kinoma)
  2. Privacy-preserving occupancy detection – Use the Kinoma with various sensors to determine where occupants are. We wanted to do it do differentiate children from adults, but ultimately this devolves to an IR sensor.
  3. Gesture recognition for activities – This was a cool concept but endlessly difficult to execute in the time frame that we had. The idea would be based on concepts from this, using only WiFi signals to distinguish gestures and translate those into home-based actions depending on user data. Well, maybe it can still be done.
  4. Retrofitting – Use the Kinoma display in conjunction with actuary/haptic/control sensors to automatically control the shower heat. I still like this idea but the rest of the team wasn’t so hot about it (ha ha ha).
  5. Kitchen management – Use a barcode scanner in conjunction with the Kinoma display in the kitchen to manage food inventories.

Ultimately, we went with the last one. Since the Kinoma allows for the creation of easy companion applications for Android/iOS integration we ended up actually eliminating a barcode scanner for the proposal. Instead, we’ll implement a barcode scanner that will talk to our Kinoma via a smartphone. Of course, we’re not that hardworking so it will likely only be an Android application.

Finally, the conference this year is in Honolulu. While that’s great for a vacation spot, it increases the travel expenses significantly for people who are there to just demo a project like this. In previous years, they offered a prize between $1500-$2000 for the winning team, which would be enough to recoup the cost of a single student flying to present the finished product. Last year it was Surface tablets. While the Surface is a pretty nifty device, it doesn’t pay the bills. Here’s hoping they announce some kind of prize soon because right now that page has no details.

Also, we don’t get the Kinoma unless we register for the conference first! This is insurance to force us to go, but it puts a market price on this thing at ~$350. Boo.