Showing posts 1 - 2 of 69. View more »
CSE today was the third largest team with 63 participants and second highest funds raised. It was also a team that truly inspires: from Cheryl Hile's unbounded enthusiasm matched only by her thorough planning and team building to Lindsey Fowler's love of life at UCSD. Our team was full of inspirational images: faculty, students, staff and alumni running together, having fun, even dominating in their divisions and all the while setting trailblazing role models for others to follow!! I could fill pages with the inspiring stories behind each one of them, but here are a few pictures to get the message across.
Here is Christine posing with Jennifer Arguello. Christine ran to the #1 place in her division! Among the rest: Cheryl placed #7, Ryan Kastner #7, Vier Kier #4, Dave Wargo #5, Mihir Bellare #8, Dean Tullsen #15 and yours truly at #14. Brian Hile joined us and placed #2!
On the CSE Facebook page, we track statistics on how our posts are doing with respect to readership, what topics have good traction with the larger CSE community. Here is some data:
* The CSE FB page has 564 likes. This compares to 1,798 likes for the JSOE FB page, and 119 likes for UCSD WIC. Here is a list of other FB pages
By comparison, here are the size of the FB Groups:
The reach of a post is the readership: number of individuals to which a post is shown. This reach can be enhanced by paid promotion (usually about $25). Thus, a reach has an organic as well as a paid component to it. Of course, not all posts are advertised.
Here is data on posts published in last 3 months:
Let us take a closer look at the top posts:
As a step towards strengthening and streamlining our instruction across UG courses, Christine started a survey of course expectations and prerequisites. Now, Yoav and Christine have put together this page specifically focusing on what we expect our students to know entering our upper-division courses. I strongly recommend everyone to look through it and enter information as appropriate.
CSE now has it's own online education server. You can access this server here (You need an account to use this server, if you want to try it out without creating an account, see the MAA101 account below).
Webwork is an online homework system focused on Math. It has a large collection of problems in many areas of math, including discrete math and algorithms.
This fall, three courses will be using the system:
If you are interested in learning more about webworks, I recommend reading this pdf, and then going to the MAA101 demo course here. (log in as guest)
I am also interested in making webwork more adapted to the particular students. Here
is video describing the current work in my group in this direction.
Drop me a line if you are interested
While I did not personally anticipate such a prominent coverage of our lead gift, it is good to see Computer Science (at UCSD) to be in the local press. It helps build our connection to the community. I am proud to be its face and mouthpiece (I have to work on my accent a bit more:), the real hero is, of course, the anonymous donor who choose specifically to stay in the background to make sure Computer Science department got the attention both inside and outside of UCSD!
That was by design and it is already working! While the explosive growth of recent years and attendant problems in a "business-as-usual" campus depressed me often, we seem to be getting past over it. Going by where our students are going this year, it is an exciting time for the word to spread about the academic strength of CSE@UCSD! Here is what I know today, and there may be more:
Yuvraj Agarwal -> CMU
Given the currency of the issues brought on by Coursera, Udacity, Udemy and others, I am putting together a few key links from my explorations of the subject so far. We do have an opening in this regard: our MAS program has a provision for online offerings (for the obvious reasons that the local market for professional engineers is likely to tap out in a few years):
Reproduction of my notes from SNOWBIRD 2012 on this subject:
· This came in multiple sessions related to rise of online learning to CS education principles in college and earlier.
· The session with Peter Norvig, Sal Khan and Spector was most illuminating:
o Sal’s KA is looking to create basically a real private tutor via micro-lectures. In the longer term he sees this becoming a general knowledge resources with instructions customized to individual needs
o There is, of course, a broader universe of Udacity, Udemy, Coursera, EdX that is approaching learning in ways beyond books and papers. Peter’s presentation focused on how he found paper limiting in explaining AI concepts and how online made it possible for him to share with the students a learning process that goes beyond traditional books.
o The #1 issue facing such online (asynch/synch) learning is “credentialing”. Sal thinks it will be decoupled from instruction.
§ “While Udacity etc give a certificate of completion, the employers don’t know what a certificate means.”
o The existing universities will be source “physical experience” that will move up the value chain
§ Corollary: there will be winners (presumably tier 1 universities) and losers in this shift.
o There was quite a bit of discussion on Bloom’s 2-sigma observation that 1-1 tutoring improves student performance by 2 sigmas and how MOOC seek to do much better with astute observations such as “learning is what students do” which is affected by a number of factors some of which are related to social dynamics on a campus
o The goal of teaching should be to enable students to be able to make predictions
§ That is given a problem, understanding a problem and then be able to make predictions. Rather than knowing right and wrong answers to given problems, problem solving requires this essential skill which is gained ironically by errors in the predictions.
o Going back to the credentialing need, there is a need for better transcripts: rather than 48 bits devoted for learning outcome in a year (to describe course grades), the transcripts can be much richer that include details on class outcomes
§ Content for online instructions is exploding with varying quality (of content, presentation)
§ Personalization of this content is a hard problem
§ Testing and certification of learning outcome is a hard problem that demands significant infrastructure
§ Online learning is hampered by availability of interesting exercises that promote problem solving
§ Cheating in an (online) course is a function is the importance of grade as a result.
This one is a bit personal: recent 60-minutes caught up with the phenomenon of Peter Thiel bashing university education as a waste of time and money and encouraging kids to drop out. The sixty minute piece captured the debate, including my son Anand presenting at the 20-under-20 forum; and, of course, dropping out of college, nee "stopping" out of Harvard.
My friend Vivek Wadhwa calls this greatest mislead of our times. Mislead or not, it is obviously clear that billionaires with press accolades have a bit more influence on college kids today than parents! Nevertheless, what did brighten my day was the essay my son wrote re overspecialization being pursued in colleges; why should, for instance, students be required to declare their major before entering school?
So, State Bill 259 is under discussion. UCOP opposes it strongly and with some reason. The UC Berkeley Faculty Association has another view. As can be expected, CSE graduate students and faculty have thoughtfully considered this issue and found the whole notion of unionizing as a labor group for what is a temporary employment category to be wanting. I can not reproduce the discussion here, but there are always nuances in such things that go beyond slogans of either side, ultimately the notion of "not having individual choice" in the matter when union comes to town is the greatest sticking point. Keith pointed to some research on this topic by Tom Schenk Jr here. I still have to go through it, but clearly it is a topic that does need action now than passively waiting for whatever to happen.
Since the readers of this page will not let me get away by stating the obvious things without making an opinion, here is my ambivalence:
a. Personally, I am dismayed by the broader trend in our society of interest groups (yes, I have AARP banging on the mailbox constantly, goodluck). Every is part of some interest groups: teachers, nuns, firefighters to even scientists and must make the (lobbying) case to where the $$ money. In my simple mind, we elect leaders and reps to do their job than be referee in lobby-fests; for that reason, one more interest group is not aligned to how I see the world even if there are tangible benefits and protections against abuse of GSRs;
b. OTOH, we must be proactive in looking after GSR students, many of who -- at least in our domain -- make significant personal sacrifices, putting life on hold -- sometimes with growing family and aging parents -- to earn small in pursue of their dreams. They need to be looked after by a community, not just to the goodwill of their research adviser. In particular, they should not be subjected to arbitrary decisions or exploitative salaries. On the last count, our department puts a floor on the level of appointment a GSR can be at, almost certainly that floor will be lowered in any collective bargaining arrangements running across the university.
On balance, I believe GSRs do need collective representation and mechanisms to safeguard their interests in case they cross paths with the faculty member or even university as an institution. But unionization has more downsides than up given the nature of our work and relationships.
1-10 of 69