Ideal Workday
|
| |
|
 |
You're a materials engineer. Standing in the morning shower,
you suddenly have a "eureka" moment with a new idea about how to achieve
the properties you want for the material you are designing. |
| |
|
|
|
You rush into your office (preferably dressed), start up your
computer, and start to do some preliminary calculations to make sure your
idea actually has some potential. You don't have any numbers to work with
yet, but you can get a rough handle on things--such as the manner of interaction
of various parameters--by expressing everything symbolically. |
| |
|
 |
Now you need to know something about, say, the resistive behavior
of certain kinds of thin films. You check out your company intranet, and
you find a couple of papers from colleagues--one in Santa Fe, the other
from a team in Osaka--that look like they might be helpful. But they're
not just papers; each also includes a set of computer simulations all ready
to run. Better yet, they're ready to run on your platform. You spend 10
or so minutes looking over the simulation source code and modifying it for
your new case. Since the code is written in a clear, self-documenting style,
you're getting your first results shortly. By noon, you've collected a number
of interesting calculations and a few 2D and 3D plots showing how the parameters
interact. |
| |
|
 |
Your modified code, your results and graphs, and your notes
are all collected in a single, convenient document serving as a project
notebook. Your progress since morning gives you an opportunity to do something
unusual: eat your lunch away from your desk. On the way out of your cubicle
you encounter your supervisor, citing a sudden large sales opportunity and
insisting that you make your preliminary findings available to the entire
company ASAP--even if it takes you late into the evening |
| |
|
|
|
Grimacing, you try not to let on that it's only a three-minute
task to convert your project notebook into HTML ready for immediate upload
to the intranet. You save your project notebook in HTML form. The graphics
and even the mathematical formulas convert themselves automatically, and
all you have to do is upload it to the company intranet server. You promptly
do this--after lunch. |
| |
|
 |
More simulation studies follow in the afternoon. Over lunch,
perhaps inspired by the thin-film behavior of your pastrami, you think of
some refinements to your original plan. After some additional simple modifications
to the simulation, you're ready to run the revised model with a full set
of parameters. You're leaving some of the parameters symbolic, while putting
numbers into the others. (You've imported the numbers from the output of
some software that collects lab data.) You run more tests, and everything
still looks good. Since you've carefully specified the precision of the
numerical inputs, you'll see directly from the precision of your output
the degree to which you can rely on your results. |
| |
|
 |
On the way out the door as you head home, your supervisor appears again
and suggests that you prepare a presentation for your workgroup to get
them up to speed on your new idea. You know that you'll be able to do
that in the morning. The entire task of preparing the presentation consists
of moving your project notebook onto a computer attached to an overhead
projector, and you'll be able to punctuate your talk by demonstrating
your simulation from within your presentation software.
|
| |
|
| |
It's been a busy but productive day, and you didn't struggle
through it all, trying to get an assortment of barely compatible programs
to cooperate with each other. |
| |  | | | |
| |
|
![[ja]](/common/images2003/lang_bottom_ja.gif)