Friday, December 27, 2013

Simple Science: Algorithms for Compressive Genomics

This is a post in our Simple Science Descriptions series of pieces written with the Up-Goer Five Text Editor restricting writers to the ten hundred most used words in English. It's harder than you might think! Send your entries (preferably under 350 words) to jeanyang [at] mit [dot] edu.

--

Learning the life words* that make people, and trees, and animals, and all other kinds of life, helps people like doctors find out how to help sick people, and how life works. One thing we do with these life words is to search all the life words we know to find ones that are like other ones. This search is very important.
Since people started reading life words, the words that make humans and animals and trees, we have found more and more life words. We have found so many life words, that computers aren't getting faster fast enough to keep up!

But, some life words are a lot like other life words. In fact, most life words are very much like other life words. The different parts of your life words and a tree's life words are very small.

I work on helping computers learn how to make all the life words smaller, so they can search only the life words that are not like other life words. This way, computers don't need to search nearly as much. This allows us to look through even bigger sets of life words, like those in your stomach, or in the ground, or in the water.  I hope this will help doctors find new ways to help sick people.

- Noah M. Daniels, Post-Doctoral Associate in Computational Biology

* I chose "life words" for DNA.

Thursday, December 26, 2013

Simple Science: Dissertation on Computer Planning

This is a post in our Simple Science Descriptions series of pieces written with the Up-Goer Five Text Editor restricting writers to the ten hundred most used words in English. It's harder than you might think! Send your entries (preferably under 350 words) to jeanyang [at] mit [dot] edu.

--

Computers have a hard time figuring out what to do when there is a change in the problem they are working on.  My work allows computers to think about possible changes that may happen in order to see which changes should be planned for ahead of time.  Then when a change happens, the computer will already know what to do.

This can be used in many situations in which change happens a lot, such as driving a car, flying things, managing power (the 'Smart_Grid'), or controlling things.

- Robert Holder, Senior Computer Scientist

Wednesday, December 25, 2013

Merry Christmas! Simple Science: Programming and HCI Mission Statement

This is a post in our Simple Science Descriptions series of pieces written with the Up-Goer Five Text Editor restricting writers to the ten hundred most used words in English. It's harder than you might think! Send your entries (preferably under 350 words) to jeanyang [at] mit [dot] edu.

--

I looked at it as a translation problem, so I started with something I had already carefully crafted to be simple and short (but full of technical terms), the mission statement on my homepage:
I work on programming and HCI.
  • crowd computing: making people part of the programming system
  • online education: helping crowds teach each other
  • software development tools: making programming more productive for developers
  • end-user programming: making programming easier for everybody

Here's the translation into TenHundred:

I work on better ways to tell computers what to do.  Some of the things I do are:
  • making people part of the computer's brain
  • helping crowds of people learn from each other using computers
  • making it faster to tell computers what to do, for people who do it for their jobs
  • making it easier to tell computers what to do, for everyone
- Rob Miller, Professor

Tuesday, December 24, 2013

Simple Science: Computers That Help People Learn

This is a post in our Simple Science Descriptions series of pieces written with the Up-Goer Five Text Editor restricting writers to the ten hundred most used words in English. It's harder than you might think! Send your entries (preferably under 350 words) to jeanyang [at] mit [dot] edu.

--

In my work, I build things on the computer that make it easier for people to use computers. Why? Because most people don't want to know how a computer works inside; they just want to use it to do interesting and fun things.

One thing that many people of all ages want to do is learn. People like learning all sorts of new ideas, and computers can make that easier. I am now making things on the computer that help people learn better than just using paper and books. To do so, I need to first watch how people learn, see how they struggle with what they now use to learn, make new things on the computer for those people, have people use them, and see if they learn better.

What I am making today is a thing that helps people learn how to build new things on computers. Why learn how to build new things on computers? Because it is very fun, and also because you can get many kinds of good jobs doing so ... some that even offer free food at the office! But it is very hard to learn how to do this if you don't have a teacher always sitting next to you. Most people don't have their own teacher but still want to learn by themselves.

This is why I am making a teacher inside of the computer. Just like a human teacher, this computer-teacher helps people learn to build new things on computers. A big part of this computer-teacher's job is to draw pictures of things inside of the computer that books only say in words, which makes it much easier for people to learn those ideas. It also tries to explain those pictures using simple words and gives students a place to ask questions and talk about those pictures with other students. This computer-teacher has already been used by people in many parts of the world to draw over 500,000 pictures to help them learn! Right now, I am continuing to make it work better for more kinds of students.
 
- Philip Guo, Professor's Assistant

Monday, December 23, 2013

Simple Science: My Programming Languages Research

This post is the first in our Simple Science Descriptions series of pieces written with the Up-Goer Five Text Editor restricting writers to the ten hundred most used words in English. It's harder than you might think! Send your entries (preferably under 350 words) to jeanyang [at] mit [dot] edu.

--

I make it easier for people to tell computers what to do. I think about how people want to use computers and what computers need people to tell them. The people I have in mind are ones trained to use computers. In my perfect world, Ms Programmer doesn't need to tell computers much and the computers can do everything right while still running fast.

Here is how I spend my time. First, I come up with a new way Ms Programmer can tell the computer what to do. Then I think about why this way is the right way. In doing this, I sometimes have to think about what other people have shown is possible to make a computer do. Then I come up with reasons so other people also believe this way is the right way. To find these reasons, I have to both think hard and also build things. It can be hard to change minds!

I focus on ways of telling computers what to do that seem like they might force the computer to do too much work. I like thinking about these because they often let Ms Programmer do less work. The problem is that sometimes the computer can't finish all of the work. To make the new ways possible, I think about how to get the computer to do less work. In my work, I am helping Ms Programmer make sure things do not get shown to people who should not see them. Right now, Ms Programmer has to do a lot of work to guard Mr Privacy, remembering all the time where everything comes from and where everything is going. With my new way, Ms Programmer can say once how she wants Mr Privacy to be guarded and trust the computer to do it for her. I look at how to make this possible in a way that is easy for Ms Programmer and not too hard for the computer.

- Jean Yang, Ph.D. student in computer science

NeuWrite Boston Starts a Blog

NeuWrite is a collaborative workshop for writers and scientists. We aim to improve the quality and accuracy of science journalism by fostering creative partnerships between journalists and academics. We get together once every three weeks to workshop pieces and/or discuss how to improve science communication. During these discussions, we've talked about wanting to collaborate more and experiment with different methods of communicating--so we've created this blog as a playground to do so. You might see pieces in formats you don't usually see and by people who don't usually work together. We're excited to embark on this process, and hope you enjoy the product!

- Amanda and Jean, co-directors of NeuWrite Boston