Thriving in a multidisciplinary world: technical & communication skills are key

So fun! We got to learn how to say good morning in Portuguese to start off the talk on Managing Scientific Data in a Multidisciplinary world by Claudia Bauzer Medeiros. Halfway through we also had a Tolkien reference! If 10AM seems early to you, this was a talk you could enjoy despite sleepiness.

As Claudia says, we are drowning in data and need to be able to help scientists in other areas cope with that problem. This includes models and visualization as well, which is right up my area of interest! Of course, all of this also means that you need not only technical skills but great communication skills, which seems to be the theme of the conference thus far.

Some of the points made really relates to a lot of what came up in the Santa Fe Institute Complex Systems summer school this summer: you need to learn how to work with people from other disciplines and learn to speak their scientific language to have success. Of course, she calls this Data-driven science, which is different from what we call it in complex systems, but that's because she is coming from the data analysis side. I love that she had an image combining experiments, archives, literature, and simulations as facts used to help answer questions but that it's not enough: you also need data and information from collaborators not just your own set of it.

What I really love about all of this was the point that we are in an interdisciplinary world, and this should inform how we teach computer science. The implication, at least from my perspective, seems to be that the future is in collaboration between computer science and other fields and we need to prepare new computer scientists to succeed in this environment. One point here was how to get access to data as often people are not willing to give it up. Thus the Tolkien reference: the data is their "precious!"

I was not expecting, however, a section title slide called "Emotion." But the point was still a great one: you need to get people emotionally involved to have cooperation. This all ties back into the point of communication abilities being key for collaboration, as it is very difficult to have a successful conversation with someone if you can't understand their perspective on the subject.

Grace Hopper was also a funny lady

Last week there was a lot of tweeting, e-mailing, and Facebooking (is that a verb?) about the Grace Hopper interview on Letterman back in the 80s. Apparently the video has been on YouTube for years, but it seems that at least a large number of us were unaware of it. She is quite the funny lady, and even brought a nanosecond to give Letterman. If you don't understand what that means, please watch the clip to hear the explanation straight from Grace Hopper! Since she is the namesake of the conference starting today, it seems only fitting to share the video with anyone who hasn't had the chance to watch it yet. Enjoy, and see you at the conference!

I'm Back!

Well, after a not-so-short hiatus from the Blogger world, I'm back! I decided I needed to take some time away from the blog to work on another time sensitive one, whose time has now passed. I'm looking forward to getting back to chronicling my graduate school adventures, as I am hoping I will graduate within the next year! I have a lot of exciting things happening this year, and I can't wait to write about them!

How to Actually Help the Undergraduate Program

It's interesting what types of cuts our university is willing to make in order to keep the quality of the undergraduate education. Obviously education is the goal, and when you have over 25K undergraduate students they are also the priority to some extent, even in a research focused university. However, one of the cuts the university makes in order to maintain the apparent quality of undergraduate education (i.e., keep from increasing tuition even more or cutting too many classes) is to decrease the funding given to departments for Teaching Assistantships for graduate students.

I can't imagine I'm the first person to consider that this can't possibly keep the same level of education, when students suddenly have fewer TA office hours to attend, and fewer TAs (and therefore less TA time) to answer e-mail or grade homeworks. This is why I say it is keeping an "apparent" quality of undergraduate education, even if it's not really keeping actual quality. As the sole TA for a class of 40+ students, I can say from just a few months of experience that this definitely impacts the undergraduate students. Luckily we were able to hire another graduate student as a grader for 10 hrs/wk, so I have more time to tend to the students' questions, design their homeworks, and put together the lectures I'll be teaching. But I still don't have enough hours in the week to help them. I can't imagine what TAs do when they aren't lucky enough to have just one other person to help with grading! Even spending the 20 hrs/wk I'm technically paid for isn't really enough. I enjoy what I'm doing, and don't want fewer responsibilities; but I'm really overworked, which is NOT good for my students!!

I wish this cycle was more apparent to the upper level "management" of the university. If you want good undergraduate education, pay your graduate students! We do more than research, and we're often a great resource for students. When there is 1 TA to 40 students, the number of minutes they can spend helping each student is minuscule. How does that keep the quality of undergraduate education?

First Biology Conference

Last week I went to my first biology conference, Systems Biology of Human Diseases, held at Harvard Medical School. It was definitely different from the computer science conferences I have been to thus far. Of course, for me the biggest difference was that I was surrounded by people that were experts in a completely different field from mine, but yet were still working on some similar problems. We just happen to have very different vocabularies. Overall though, I think it was a worthwhile conference to attend. I learned some new keywords, learned more about how biologists describe cancer, and got some new ideas for how to present and do my research. Not bad for a day's worth of time. Unfortunately I couldn't stay through the poster session thanks to a migraine, so hopefully next time I won't have that problem. Luckily this conference will be in Boston again next year before they merge with a European conference, so I will have easy access for at least that year. After that we will have to decide if it is worth traveling to Europe for it.

Undergrad Class Again

Earlier in the year I wrote about experiencing undergrad class again. Later in the semester I sat in on two lectures of my own department's intro programming class, as part of an observation for my "teaching undergraduates" class. I was in the same horrible classroom/auditorium I was in for the other class, but it was much emptier and thus quieter.

Overall, I was very impressed with the faculty member giving the lectures. Sure, he wasn't perfect, but he had a great energy to him. This is one of the reasons I had chosen to observe him over some other professors; I knew he brought a lot of energy and excitement to teaching at the graduate level, and I really wanted to see that happen at the undergraduate level.

In the second of the two lectures he brought his research area (Computer Vision) into it. He helped the kids think about how they might go about solving problems in the area with what they had learned in the class: arrays, I/O streams, etc. Really, I think if more professors used this type of lecture throughout their intro courses, we would have a better retention rate in Computer Science. It takes a step back from what can often be boring topics, and shows that there is an interesting use for all of it after all. I hope when I start teaching I am able to bring this sort of thing to my lectures as well.

Computer Science and Volleyball

One of my few outside-of-computer-science activities that I actually do every week is volleyball. After our intramural game last Wednesday, I was thinking about all of the ways that the two are similar. For instance:

1. Time Outs - In volleyball, sometimes you just need to call a timeout, mentally regroup, and then get back into it. In our game we were losing 3-9 to a team we had just squarely beaten; I called a timeout, and then we proceeded to win 15-9. What a difference that timeout made! In research it's the same way. Sometimes you just need to take a step back from what you're doing, regroup mentally, and then start off again where you left off.
   
2. Team Effort - In volleyball, you have 6 people on the court at any given time. If you don't trust your teammates, you end up running over them to get balls they were perfectly prepared to hit, usually better than you end up hitting them. To really play well, you should back up other people and trust their abilities. In research you also need to trust your group members, but still encourage and back them up when possible. It's all about sharing information, ideas, and letting each other find his/her own way without stepping on each other's toes.
   
3. Encouragement - When playing a team sport, you want to try to keep everyone energized and feeling good about their playing. If someone starts to doubt their abilities from a few bad hits, they are going to second guess everything else they do that game. A little "that's ok, you'll get it next time" or "great job!" will go a long way. With research, you should also encourage others even when they are struggling. Everyone needs encouragement in everything they do in life, but especially so in "high" stress situations like volleyball and research!
   
4. Practice! - Obviously, if you don't practice volleyball you are not going to retain your skills. Much of it is muscle memory and reactions, which can be learned and refined over time. In computer science research you also need to practice skills such as public speaking, programming, and critical thinking. It may not be muscle memory, but it is still a set of learnable and improvable skills.
   

It's true that the comparison between team sports and teamwork in school and work has been made many times. However, it's nice to be reminded occasionally that our extracurricular activities really do relate to graduate school, and working on good habits and skills in one can easily help improve the other. All the more reason to be sure to get adequate "play" time in addition to work time!

Barbara Liskov wins Turing Award!

What an amazing day! I was just casually checking my e-mail and what do I find out? That another woman has won the Turing Award!

MIT Press Release
ACM Press Release

Although I feel bad to be surprised about the fact a woman has won (why should anyone be surprised about that??), I don't know how I could not be excited. When Fran Allen won it seemed like a barrier had been broken, but it still seemed like it would a long time before the community was willing to honor another woman. Apparently we are better than I thought! From reading through the press releases she sounds like an amazing researcher. I wanted to share a few paragraphs from the ACM Press Release:

The award cites Liskov for her foundational innovations to designing and building the pervasive computer system designs that power daily life. Her achievements in programming language design have made software more reliable and easier to maintain. They are now the basis of every important programming language since 1975, including Ada, C++, Java, and C#. The Turing Award, widely considered the "Nobel Prize in Computing," is named for the British mathematician Alan M. Turing. The award carries a $250,000 prize, with financial support provided by Intel Corporation and Google Inc.

The first U.S. woman to be awarded a Ph.D. from a computer science department (in 1968 from Stanford University), Liskov revolutionized the programming field with groundbreaking research that underpins virtually every modern computer application for both consumers and businesses. Her contributions have led to fundamental changes in building the computer software programs that form the infrastructure of our information-based society. Her legacy has made software systems more accessible, reliable, and secure 24/7.

I want to go dance around the block! Maybe one day (in many decades) my research will be great enough to receive this type of acknowledgment. Once can only hope! Despite her obviously being a deserving recipient, her success will help highlight what great things women can do (and are already doing) in the field of computer science.

Experiencing Undergrad Class Again

Classes just started up again, and although I am officially done with all course credits for my MS and PhD, I am sitting in on a class. This is not just any class though; it's an undergrad biology class! I know, I know...shocking.

The first class meeting was yesterday morning, in an auditorium. It wasn't a huge auditorium though, probably only about 100 people can sit in there (at least if you are limiting people to sitting on chairs). Of course, my first impression was "oh my goodness, they are all so little and young!" but after that all I could think was "shush!"

Sure, sure, I might not be able to say anything about looking young (I can still be mistaken as an undergrad by people over the age of 30!), but I definitely have no qualms talking about the noise level. It was quieter during class than before or after, but the chatter never completely stopped! I can see people thinking they could get away with it in a 200 or 300 person class, but 100? Maybe it was just because I was in the 3rd to last row that I heard it all, as maybe the first 10 rows know how to be quiet, but I doubt it. There was a girl sitting 2 seats away from me who was only quiet when the professor was explaining how the class would be graded. Otherwise she was chattering away with the people around her the entire time!

On top of the general noise, 5 minutes before class was supposed to end it got very noisy. I was very confused for about 20 seconds, and then the professor said "By my watch it is only 12:00, and class ends at 12:05 unless we end early." Apparently everyone was packing up and chatting and getting ready to go! 5 minutes early! I still can't believe it. We knew better than that in high school, and I'm really not THAT much older than these kids. To be fair, where we were at in the slides at 12 easily could have been mistaken for a stopping point, but the professor hadn't stopped. At least they quieted down rather quickly, we had 5 more minutes of class, and then we left.

I do feel sorry for the kids in the class though. They have FOUR exams this semester, quizzes, and the class is going to be taught differently than normal. Usually it is on 2 topics, half and half, but this semester it is primarily only 1 of those topics. It should be interesting to see how much that affects their later classes. Not to mention that those seats are hard, close together, and not very comfortable! There was lots of dried gum under the pull-out writing desk for my seat as well. Gross!

I have to say that I do not miss taking this type of class, and I think it will definitely be interesting to be teaching one in a few years!

GHC Keynote Fran Allen on Diversity

Both of the keynote speakers at GHC have been very interesting. On Thursday we were honored by hearing Francis Allen, the first woman to win the Turing Award (the highest award in the computing field). I actually had the honor of meeting Fran Allen on the shuttle ride from the Denver Airport to the Keystone resort. She's a great lady, and definitely has strong opinions about where the field of computing is going as well as the state of women in our field. Although she spoke primarily of technical aspects of the field during her talk, she also spoke on her beliefs about the reason the number of women in computing dropped, based on her experiences in the field. Apparently, in the 1960's the field of computing started to become a major at universities and was based out of engineering departments. Well, engineering departments were primarily male, so suddenly women who were entering computing from other fields were blocked from entry as they didn't have the engineering knowledge that was suddenly defined as necessary. So by the 1970's the number of women had dropped in the field, and the glass ceiling started to exist.

This is a view of the problem that I've never heard before, and find very interesting. For everyone who actually thinks that women don't belong in computing or aren't capable of being in computing, this could finally convince them they are wrong. Not only are women now a large number of engineers graduating each year (I believe it's around 50%), which proves that we are capable of being engineers, it also proves that women have always been interested and the reason they left is that they were pushed out by sudden strict requirements. Of course, these are things that most of us believe anyway, but maybe a true history of the changes in computing would help remind everyone that men are not automatically better at it, they just shaped the field in a way that preferred them at the time. We have bigger problems to look at too, like visibility and perception of computing, but if we can work on increasing everyone's knowledge of how our field became defined it may help the rest of our work as well.

My favorite quote from her talk is that she is "both honored and concerned" about being the first woman to win the Turing Award. If you want to read about the technical aspects of her talk & her views on awards, the GHC Wiki page on her keynote has lots of details (written by me!).

Girls and Math

In Computer Science we talk a lot about why girls shy away from math, computers, etc. even when they are obviously good at it. We spend a lot of time trying to figure out ways to encourage them to pursue math and science, which includes trying to break stereotypes, talking about the many great careers that exist with computers, and giving them opportunities to try out many computer related activities. Not to mention all of the research on the topic. Still, the numbers of women in CS undergraduate programs in the USA is dropping.

I think the above comic helps illustrate some of the problems that we face. Often, girls and women are judged not as a single person, but as representative of their entire gender. As the comic shows, if a male does something wrong another male will assume it's because he is stupid; if a woman does the same thing wrong, that same male will assume it is because all women are stupid. This happens very often, which seems to not only put unnecessary pressure on women, but also gives people a skewed view of how people perform on a task.

As far as the pressure is concerned, women often feel less confident about their abilities even when they are doing just as well as (or better than) the males in their class. Especially with the low numbers of female students, they can easily feel like they don't belong, or that they are singled out and judged more than the men. Some women do not feel this way, but even one woman who feels this way is too many. With the added pressure of feeling like you represent all of your gender, this can be incredibly daunting.

So, what can we do to fix the problem? I wish I had the answer. I think the best beginning is to make people more aware of these tendencies. Maybe we if we talk about how we judge others more with younger people (college freshmen at the latest), they will be aware of it and try to fix it within themselves. It would be great to help make professors aware of this as they can (and do) make this problem worse as well. Not forcing women to represent all of woman-kind with every move they make, and just letting them be who they are themselves, would be great improvement for women in Computer Science.

Academia: Disillusionment

Academia is an odd animal, at least for Computer Science. To succeed as a graduate student it is known that you must have plenty of publications and do plenty of networking. To network you must attend conferences that your work is accepted at, and go to many talks that you can. All of these requirements are quite reasonable and easy to understand; if it doesn't sound interesting, then academia is not where you should be.

However, the odd part comes with the money. It is hard to get funding for travel, both from your own university and from the conference itself. For an example, I will not be going to my conference in Switzerland because the conference would only waive my registration fee and the department could only give me enough money to cover less than 1/2 of the cheapest plane ticket I could fine. The university may have been able to give me more, but their deadline for applying was before the acceptance date at the conference and required proof of acceptance. However I don't know the probability of getting university funding; it may be slim, or may be high. Of course, if it works the same as with the department where the student must pay up front and then be reimbursed it almost doesn't matter; I do not have the ability to buy a $1200 plane ticket, even with credit cards.

So I am disillusioned at the requirements coupled with reality. We are expected to do things that are near impossible. We certainly cannot afford most conference trips on our student salary, but yet it is difficult to find funding. How ironic that one of the most important things is so far from our grasp. I even know of someone whose advisor has good funding (mine currently does not), but her advisor would not (or could not?) fund her for a good conference her paper was accepted at. What a lost opportunity.

Overall, we are better off than non-science graduate students as we have almost-guaranteed salary funding, but we still have a long way to go before overall funding is realistic. The other part of academia that disillusions me is the fact that I'm being paid less this summer than I pay in rent; but that's an issue I'll have to strike up later, probably when I start wondering why I went to grad school right out of college instead of taking a few years to save up some money.

Cancer Stem Cells

Event: Robert Weinberg
Date: Tuesday, April 18, 2006
Time: 4:30 PM to 6:00 PM

This talk was on cancer stem cells, a topic very pertinent to my research. The beginning of the talk was on basic cancer information, which made it very obvious that it was occurring at an undergraduate venue (Amherst College). He did reaffirm a few numbers for me though, such that a cell can only split a max of about 50 times and that that probability for mutation is very low (at most 1 in 1x10^6 according to him).

The later part of the talk was actually on stem cells, and was interesting after we got past the description of what the different types were (stem, progenitor, differentiated). Apparently it is possible to separate stem and non-stem cells from a group by the proteins they each have on their surface. The stem cell will have a different type from the other ones. So, you can bind a die to an antigen that will bond with only the stem cell protein, and then bind dies to antigens that bond with the other cells' proteins, and use light refraction to see the different colors and separate them out. That is very interesting for both cancer and regular medical/biological research.

Also interesting was a discussion on whether or not tumors are organized hierarchically like regular cells are. In regular sets of cells, we have stem cells that are a sort of "super parent" in the fact that they can create any type of cell (most people would know this just from the regular news). These cells can create progenitor cells (he called them "transit-amplifying cells" and I like to call them "blast cells") that are capable of differentiating into a specific type of cell. The children of blast cells eventually become differentiated cells, which are basically a mature cell that will not change its function during its lifetime. Blast cells can also create more blast cells, and when stem cells split they create a new copy of themselves (so that there is one blast cells and one stem cell). Originally it was assumed that tumors were only made of mature cells, and that stem cells did not exist in a tumor as stem cells are less likely to mutate. There was speculation that a tumor cell might incur a change that turned it into a tumor stem cell, however.

So, at this talk he described a study showing that tumor stem cells did indeed exist. A group apparently split apart tumor cells the same way that regular cells were split apart, using antigens. Tumor initiating cells appear to be a small minority, which is a property of stem cells. This makes sense, as other cells would be unable to split enough times to form a proper tumor, although personally I'm not sure that limit would still be in place in a tumor cell. When an experiment was done with the separated cells, however, it was found that the cells they thought were stem cells could create a tumor in a mouse if only 200 of them were implanted whereas 20,000 of the others failed to form a tumor when implanted. Usually researchers implant 100,000's of cells to get a tumor growing, so it is interesting that only 200 of a specific type would be necessary. This experiment is why many agree that tumors do have stem cells.

One implication from the existence and ability of tumor stem cells involves metastasis, which is when a tumor migrates to another part of the body or tissue to become cancer. If only tumor stem cells are capable of growing a tumor, would only tumor stem cells be involved in metastasis? If so, could we stop metastasis by killing the stem cells?

There are still many things that are not known though. For instance, no one is sure if regular stem cells are likely to become tumor stem cells. Since stem cells don't replicate often and the majority of mutations are going to occur due to replication (or at least be spread by it), it seems unlikely. Most people currently assume that progenitor cells are the ones that become tumor stem cells, as they are only slightly different from stem cells to begin with.

There are treatments out there (he named "Gleevac") that can kill off all of the tumor except the stem cells. So once you stop with the treatment, the tumor is just replenished from the stem cells. Companies would like to make a treatment to kill the tumor stem cells, but no one can figure out how to do that without killing the healthy stem cells as well (which would probably kill you in the end, as the body is constantly killing and replenishing cells). The reason the current treatment doesn't work is that it ruins a cell's DNA, and cells that split frequently end up dying because they can't fix their genes in time to split. However, stem cells don't replicate frequently, so they are able to repair the damage before splitting.

So if we can research tumors in a similar way to regular cell growths, and figure out how to get rid of those stem cells, we've cured cancer! Which of course is easier said than done.

Global Climate Crisis

I saw Al Gore give a talk at Google on Friday, April 7th, and even got to shake his hand afterwards. It was a great talk about the state of the climate changes and global warming. I was incredibly impressed with the fact that his talk concentrated on the facts, and was not political. There were of course political jokes here and there, and slight pokes at the current administration, but I do not feel that they detracted from the point he was trying to get across. The Q&A session afterwards had more of a political agenda due to the questions that were asked, but overall I think it was very well done.

Below is a summary of what he talked about, including quick facts. All of this information will be in the movie coming out this year entitled An Inconvenient Truth, which I recommend seeing and plan to see myself.

• He calls it the "climate crisis" because "global warming" doesn't truly encompass the issues, and makes it sound less like a local problem
  
• The word "crisis" is written as a combination of 2 characters in Chinese, one for "danger" and one for "opportunity" -- this tidbit of information is a good summary of the talk
  
• There are a lot of global places that are showing signs of global warming: Mt. Kilamanjaro is out of glaciers, and Glacier National Park is out of glaciers, for 2 examples.
  
• You can compare the amount of CO2 in the atmosphere by checking the amount in glaciers (they save the gas in pockets of air) and comparing it to numbers taken over the past 40-some years. The amount of CO2 has never gone over 300 ppmv (parts per million), but is now at 400 ppmv. 45 more years of following the same increase trend as the past 40 (there was a graph) would be catastrophic due to the large amout that would then be in the air. We are talking exponential growth here.
  
• The actual temperature of the ocean is higher than predicted by an actual computer monitor taking only natural causes in to account. A computer model that predicted the increase based on human interference from 30 years ago has accurately predicted the temperature up until now. That model shows it to continue to increase at a rate that will probably kill off most of the sea life in our lifetimes.
  
• In March 2004, Brazil had the first typhoon in South America; a very big sign of changing global climates and ocean temperature.
  
• We are having more flooding and more drought: the heavy rains cause flooding, but due to the rains all coming down at once the aquifers do not have time to refill, causing drought in the non-rain time.
  
• Glaciers:
  
  • We have lost 40% of the thickness of hte arctic glaciers
    
  • There is currently a higher temperature increase at the north pole than at the equator. If these glaciers melt, the ocean will be warmed a great deal because these glaciers currently reflect a majority of sunshine in the winter months. If this warming occurs, the therma-haline pump between Canada and Europe could stop working, which currently helps control the ocean temperature and warms Europe. Europe could be thrust in to an ice age if this happens, with high consequences to the rest of the planet as well.
    
  • Larsen Ice Shelf in Antartica disappeared. It is just completely gone, this huge piece of ice. The small melting areas tunneled to the bottom of the glacier, heating the water and causing the land based ice to fall.
    
  • Land based ice melting causes the ocean height to rise, because that water was not part of the ocean previously. Water based ice melting does not cause the ocean height to rise, because it was part of the ocean to begin with.
    
  • Greenland melting (which it has already begun to do visibly) will raise sea levels worldwide by 20 feet, as all of those glaciers are currently land based. Currently 50 cubed miles are lost a year from the Greenland glacier. If this melting is not stopped, all of the Netherlands (not to mention most of Florida and many other regions) will be entirely under water in our lifetime.
    
• Due to this warming, there is less frost on the ground during winter, and therefore invasive speacies come sooner. This is causing devastation to whole forests and other natural habitats.
  
• Coral has died in the Caribbean because of warming ocean temperatures.
  
• There will be more infectious diseases and mosquitos due to this warming, as they will be able to survive easier.
  

The year to fear is 2050...yes, if things keep going as they are, we will have a flooded and possibly uninhabitable planet by that time.

If you take nothing out of these bullet points, there are 2 things I want you to remember. First, 53% of the news puts global warming in doubt, whereas 0% of actual scientists doubt it. Big oil companies and others are literally paying people to act as specialists and claim that global warming is fake. Two, we are basically the one big country that has not ratified the Kyoto Protocol, and it cannot succeed without us because we output much more CO2 than the other countries, even per person. The rest of the world understands the problem, and wants to fix it but for some reason we are too worried about what it will do to the economy. But by not following the Kyoto Protocol and decreasing our emissions, we are losing some of our economy. For instance, our car makers are suffering right now; if they made more fuel efficient cars they could sell them to other countries, but right now they can't because their cars are not up to the standards required!

There's no reason not to try to fix our mistakes and save the Earth. Otherwise, our children might be the last generation before we plunge back to the dark ages or become extinct. It really is that imminent, and something really does need to be done about it.

If you are unsure of terms or why I think something matters, just ask. I did a semester project on global warming when I was an undergraduate, I took a geology class where we discussed much of this, and I have a full year of college level chemistry behind me. I could write a post on each of those bullet points I think, but I will refrain from doing so for now. However, it really is a very important topic, and we really need to take care of our planet because IT'S THE ONLY ONE WE HAVE!

Please see his movie or research more on this topic (preferably not in the american news, as there are many lies there) if you are not already familiar with the key points. If we do what we are capable of doing with the technology we have now, we actually CAN get our CO2 levels down to that of 1970! It's only a matter of trying and making the government know that it should be a priority! Because really, what's the point of fighting terrorism and finding cures to diseases if we aren't even going to have a planet in 50 years?