Home Blog Page 10531

Netscape programmer was there for the “wild ride” of Mozilla.org’s birth

Author: JT Smith

by Tina Gasperson

Open up Netscape 4.x and type “about:akkana” in the location bar. You’ll get a page full of information about Akkana Peck: she’s a Mozilla developer who’s actually employed by Netscape and paid to work on the open source stuff. She’s been employed as a programmer since 1983 and has been with Netscape since 1996. She was there when the Open Source Mozilla project began. She loves her work, calls Netscape “the promised land,” but what really gets her going is watching the moon.Peck, 37, lives in California with her new husband, and works at Netscape’s Mountain View, California offices. She recently took time from her busy schedule and newlywed status to share some of the reasons behind her unique status as one of a very few female Open Source programmers.

Newsforge: What were some of your earliest experiences with computers? How did you get hooked?

Akkana Peck: My first exposure to computers was when I checked some books out of the
local library when I was about 12 or 13. One of the books was a manual
on programming that covered three languages — Algol, Fortran, Cobol —
with a short appendix on BASIC. Partway through the Algol chapter, I
found out that the local Radio Shack had demo models of their new TRS-80
computers sitting out where you could type on them. So I skipped to the
BASIC chapter and learned how to do some little “Hello World” type
programs, then bicycled down to Radio Shack and typed them in.

Most of my programming was “on paper”, though, just for the fun of it
since I didn’t have access to a computer. The next summer, I got
involved in a program helping in a chemistry lab; the professor in
charge asked me whether I knew how to program, and I said sure!,
because I’d been writing all these programs on paper. It didn’t
occur to me until later that I had no idea whether any of them worked.
Fortunately it worked out well and I spent the summer writing data
reduction programs on a Tektronix BASIC workstation.

A year or two later, I was taking science courses at Cal State LA,
and the physics department let me have an account on their PDP-11,
and I fell in love with Unix and C right away.

NF: Describe the path you followed to get that first programming job back in 1983 at Beckman Research Institute.

AP: It started as a summer program when I was in high school — they matched
up science students with local labs, and the student would help on a
summer project, then write it up for science fairs the next year. I
worked at the City of Hope writing data analysis programs in BASIC for a
cell sorting machine. I had a blast, coming up with different ways to
display the data which would give you different insights into what was
going on in the cell population. Then a couple years later I
took a year off of college, and spent the year working in a cell
biology lab at Beckman (a branch of the City of Hope), this time
doing image processing in Fortran.

NF: Did you already know then that programming would become your career?

AP: No, I was originally planning on physics. But I sort of burned out on
science while in college, and wasn’t sure I wanted to spend my life
worrying about grant proposals and tenure, and just sort of fell into
programming.

I still think now and then about going back and getting a physics or
astronomy degree. Or maybe geology.

NF: Why did you decide to get a math degree?

AP: I spent a year at Harvard on a scholarship, then the money ran out and I
ended up transferring to UC Santa Cruz. I was going to double-major in
CIS (Computer and Information Science) and either Physics or Math, but
the CIS degree required that I take a bunch of intro courses like Intro
to Programming in Pascal. I asked whether I could maybe take a
placement test instead, since I’d already been working as a programmer
in Pascal as well as Fortran and BASIC, and their attitude was, “No, it
says right here that course is required.” I thought that was pretty
silly. The math and physics departments were much more reasonable.
So I flipped a coin and ended up in math.

NF: Do you consider yourself a geek?

AP: Sure, I guess so. I don’t get too worried about teminology.

NF: Tell us how you got involved with Netscape. You’ve been there six years;
they must be keeping you busy.

AP: Just over five years now. (ed: Tina’s not a math major, obviously.)
I was contracting at SGI, on their software installation tools, when
the first Netscape betas started making the rounds. I got really
interested in the web — set up project pages for the group, started
building my own home page, and eventually wrote some code to make SGI’s
installer work over the web, so you could just go to a web page and
click on a link and the software would magically download and install on
your machine. Anyway, after the SGI contract I worked at Sun for a
while, but Netscape was really where I wanted to be, sort of the
promised land, so when I got a call from a recruiter at Netscape
I jumped at it.

I worked in the mail group for a couple years, then got traded to the
editor team in one of the reorgs. Just when things seemed like
they were slowing down and the place wasn’t that much fun any more,
they came up with the idea of opening the source and forming
mozilla.org. What a wild ride that was! Most of the Unix developers
in the company were very much in favor of open source, but had never
dreamed that the company would consider a model like that.
I feel lucky I was able to be a part of it. And then it was a lot
of work (but fun) learning the techniques you need for coordinating
a big project like that and helping new people get involved.
We didn’t do too well at first, but we learned.

NF: You are involved in a lot of recreational endeavors: photography, sketching,
astronomy, kayaking, biking, motorcycling, etc., etc. How do you make
time for all these things?

AP: Finding time for all the things I’d like to do is always a problem.
There’s too much interesting stuff to do! Generally I do a lot of one
thing for a while, then cut back on it to do something else. And some
can be combined: like I do most of my photography while out hiking, most
of my sketching is of astronomical objects, and so forth.

What I really wish I had is more coding time outside of work — there
are so many projects I’d like to work on (especially kernel and graphics
projects) that I have a hard time finding time for. Sometimes there’s
a temptation to spend all weekend sitting in front of a monitor, but
I need to get outside now and then, go out on trails and do other
non-electronic things, or I go crazy.

NF: Is coding a recreational activity for you (even though you get paid for
it)?

AP: It definitely can be. I really enjoy working on a tough programming
problem and finally getting the program to work well. And I find that
one of the best ways to understand something is to simulate it — I
thought I knew how airplanes flew, but when I went to write a flight
simulator I found out I had completely misunderstood it, and I ended
up with a much better understanding of the physics. And I’ve gotten a
better feel for celestial mechanics by writing astronomy programs.

NF: Of all the software you’ve written, what’s your favorite piece?

AP: That’s a hard one. I was fairly proud of the http installation tools I
wrote at SGI — they were fun to write, and awfully useful. And I wrote
a little Java starchart applet, back when comet Hale-Bopp was on the way
in — there weren’t any web pages where you could get a chart of where
to look for the comet, and I wanted something like that and figured a
lot of other people would too, and I’d been wanting an excuse to learn
Java, so I wrote an applet.

Then there was a little project I worked up a few weeks ago — I’d taken
a series of photos of the Jura mountains on the moon, spread out over
an hour or two, and I wanted to combine them all into an animation
showing how the light changed, but the images had to be rotated and
shifted to match them all up. I thought about the problem and
thought gee, why can’t you just have a program where you click on a
couple of stationary points in each image, and then the program figures
out how to rotate and translate the images to match them all up?
I looked around and couldn’t find any program that did that, but I
thought, how hard can that be? Why not just write it? So I did.
The math turned out to be pretty easy; the hard part is finding a
library that helps with the image transformations. I want to turn
it into a gimp plugin eventually.

NF: Anything you’ve written that you now think, why did I do
that?

AP: Oh, sure, but mostly little stuff (why did I use that
algorithm,
why didn’t I think of this shortcut). I’m more likely to wonder why
I didn’t do something — why didn’t I rewrite that piece of code early
on rather than spending six months trying to maintain it. There are
some parts of mozilla that could seriously use a rewrite, but that’s
true in any big project.

NF: Why did you create the hitchhiker’s guide to the moon?

AP: I had just gotten interested in lunar observing, and there
weren’t any
good books available on the subject like there were for deep-sky
observing. I couldn’t find any good web pages, either. But I knew
there were a bunch of lunar observers out there who posted on AOL but
didn’t come to the high-traffic groups like sci.astro.amateur, and I
thought, what if I set up a web page where people could send in their
observations, and I collected and organized them? Then I could learn
and so could everybody else.

It worked out wonderfully — I’ve learned a huge amount about lunar
observing, and I think the guide has been helpful to a lot of other
people getting started observing the moon.
I was very fortunate that Jay Freeman, a very experienced local
observer
and excellent writer, had written a collection of notes on lunar
features and let me use them to get started. So even from the
beginning, I had enough there to make it worthwhile for people to
visit the page and consider contributing their own observations.

Ironically, lunar observing is getting popular now, and quite a few
good books have come out on the subject recently. So the need isn’t
as great for something like my guide, but it’s still useful. I wish
I had more free time to keep it more up to date.

NF: It’s a fact that a very low percentage of women become
programmers. Is
it completely a result of environmental pressures, or is something
else at
work?

AP: I wish I knew. I think a lot of it is that girls just don’t
think about
the possibility — people around them don’t think of girls as liking
computers, so they give them other presents for christmas and steer
them toward other courses in school. I don’t see any inherent reason
that girls would be less interested in programming than boys, but
it’s pretty clear that adults have different expectations of them.
And there aren’t many role models for girls interested in programming
— maybe groups like LinuxChix can help there.

NF: Did you feel those pressures growing up?

AP: I was lucky — my parents always encouraged me in any
direction I wanted
to go, and if other people around me had expectations that I wouldn’t
like science or math or anything else, I was probably just too dense to
notice. I remember one high school counselor steering me away from
calculus, and I sort of shrugged it off, like, “What a silly idea, of
course I’ll take calculus” and never thought much of it. It never
occurred to me to wonder at the time whether he did the same thing
with the boys, and I’ll probably never know.

NF: Free Software or Open Source?

AP: I don’t care what you call it, as long as you make it
available.
I have a lot of respect for the FSF folks (and I’ve been using their
software since forever), but when you’re talking to people outside
the community, “open source” is more likely to get the point across
without having to make a philosophy speech, so I use that term more
often.

NF: Favorite distribution?

AP: I mostly use Redhat, largely because it’s what we use at
work and
I’m familiar with it, but I also have Mandrake and SuSE on other
partitions, and my husband runs SuSE and Debian and just about anything
else he can find. There are good and bad things about all the distros
I’ve seen. I haven’t found the perfect one yet.

NF: GUI or command line?

AP: Command line for most things, definitely.
GUI for things that need graphics — gimp, web browsers.

NF: Favorite coding beverage?

AP: Chai — spicy Indian tea with milk and honey.

NF: What’s something unusual about you that not very many people
know about?

AP: I’m bi — I like both vi and emacs, and use them both.
I’ve gotten over my compulsion to write Lisp interpreters in every new
language I learn (the weirdest was the one for the HP 41CV calculator).

AOL switches from IE to Netscape in beta test

Author: JT Smith

The Register’s take on this story: “AOL has — finally — shipped a beta of its software that uses Netscape technology rather than IE. This suggests, barring major breakages in the interim, that the company could contrive a defection from the Microsoft camp with the rollout of of AOL 8.0 software later this year.

Definitive guide to writing a Linux virus

Author: JT Smith

From vnunet.com: “Following on from this week’s discovery of the Jac Linux virus, an open source hacker from the University of Linz, Austria, has released a definitive guide to writing Linux viruses … Alexander Bartolich released ‘the Linux virus writing howto’ …

Bartolich said that ‘an astonishing number of people think that viruses require secret black magic.’

This may not be entirely the case, but ‘writing a program that inserts code into another program file is one thing. Writing that program so that it can be inserted itself is a very different art.’ ”

Category:

  • Security

Making Free Software pay

Author: JT Smith

The BBC has a column on Red Hat’s success in Europe. “Because Linux is created and maintained by a worldwide army of programmers – essentially just volunteers – it belongs to no-one and is open to all.

Yet Dieter Hoffmann, Managing Director Central Europe of Red Hat Linux, says his company is turning a healthy profit in what to some might seem a doomed endeavour.”

Testing your memory, the Open Source way

Author: JT Smith

By Russell C. Pavlicek

You are probably used to seeing the memory test that occurs when you boot
most PCs. This proves that your machine has good memory, right? Well,
not exactly.

The memory test a PC performs is quite basic: it will turn up
gross failures in memory. Unfortunately, it does not do much to reveal subtle problems. To do that, the PC would need to do an extensive
memory evaluation. Your machine won’t do it, but Memtest86 will.

Memtest86 is a
stand-alone program that allows you to thoroughly exercise the memory
on your machine. It requires no operating system or human intervention and can boot from a floppy or hard drive partition.
And, if that’s not enough to interest you, it is distributed under the GNU General Public License.

Why bother?

We tend to think that if our machine runs, the memory must be
good. Unfortunately, this is not always the case. A flawed section of memory
that does not normally load kernel-level code can be a time bomb, causing intermittent failures of any application which
happens to load in that section of memory. And, on today’s memory-rich
machines, a bad section of memory might only be accessed once in a great
while.

We often expend considerable effort to make certain our machines are
not vulnerable to hacking or (in the case of Windows PCs particularly)
viruses. It is only reasonable that we take a few minutes now and again
to insure our PC’s memory is sound. Any
system that appears to be less stable than it once was should be tested
with Memtest86.

Installation

Installation is simple, even when using the source kit on a machine with the gcc compiler installed. Download the
tarball, extract it, change directory to the source location, and type
“make.” That’s all.

Even on an old machine, the build process is very fast. The entire kit
takes up less than a megabyte of space unpacked. A modern PC will build
the program in a few seconds. Even most slow machines can build the
software in less than the time it takes to get a fresh cup of coffee.

To load Memtest86 on a floppy, just pop a formatted disk into the drive
and use the “dd” command to copy the memtest.bin file onto the floppy.
That’s it. You’re ready to go. No configuration necessary.

Execution

Running the program is even easier. Simply insert the floppy disk into the
machine you want to test. If the machine in question does not normally
try to boot from the floppy, you will need to enter the CMOS setup and
make the floppy a boot device.

Boot the machine. Memtest86 will load itself from the floppy and begin
execution automatically. Yes, you can pick and choose what memory tests
to run by entering the configuration menu (by pressing the “c” key), but
the default suite does a good job of exercising the machine.

Note that even using the default suite of tests, each pass can take some
time. On my Thinkpad 755CX (75 MHz Pentium with 40 MB of memory), for
example, each pass through the default memory tests takes about 23
minutes. On my 1 GHz Athlon with 128 MB, each pass takes about half that

time. But the tests run fine without intervention, so you can attend
to other matters for a while and come back later to see how things are
going.

It should be noted that the program just runs and runs. Don’t wait for it
to finish, because it won’t. It will show progress as it is running, so
simply decide how many passes you want to endure and then press the escape
key when you are satisfied with the results. On most machines, the escape
key will cause the system to reboot. On my Thinkpad, it caused the
program to halt, but I had to use the halt button to power down the
machine and then manually reboot.

I noted one other small glitch while running the program on my Athlon box.
Entering the configuration menu and instructing Memtest86 to restart the
tests caused the program to hang. A simple reboot both cleared the
problem and had the originally desired effect.

The software comes with an in-depth README file. Look at this file to
decode the errors that may occur. The README also provides technical
details regarding the various tests that are employed, as well as tips for
troubleshooting memory problems. If problems occur during testing, you
will definitely want to consult the README to determine your best course
of action.

Conclusion

Memtest86 is both useful and simple to use. It can run from any x86 PC
that can boot from a floppy. And, by testing memory thoroughly, the
program can save you untold agony down the road by identifying subtle
memory errors before they have a chance to corrupt your data. It is well
worth the time to install and run.

Curious employees are biggest security risk

Author: JT Smith

Register: “Forget about Internet crackers, employees are the biggest security problem for most businesses. That’s the main conclusion of a survey of UK IT managers which suggests that most firms are prepared for the threats posed by viruses and hackers, but are still struggling to secure data on their own networks.”

Category:

  • Security

AbiWord: Open Source’s answer to Microsoft Word

Author: JT Smith

O’Reilly: “Tired of putting up with Microsoft Word’s bloated file size and price, but still need to deal with documents in Word format? Then you
should take a serious look at AbiWord. This open source word processor is able to read and write most documents in Word’s *.doc
file format. (AbiWord does this by incorporating the wv library into its code.)”

Category:

  • Open Source

OS X and Me, perfect together

Author: JT Smith

Anonymous Reader writes: “I have to admit that I haven’t been the most ideal Mac user out there in fact I totally left the Mac party all together in 1998, not for enemy (Windows) but for Linux.
For the last few years I’ve used nothing but Linux, (even started a Linux website, ReactiveLinux.com). It wasn’t until OSX Public Beta came out that I thought “hey, I gotta check this out” after all it is BSD isn’t it? plus the UI looked “oh so sweet”! Read More

License fees for SuSE Linux?

Author: JT Smith

PDAJames writes “SuSE says it will have Linux ported to Hammer by November. This interview says that they also might be converting the entire German parliament over to Linux, and that they’re going to start charging ‘license fees’ for SuSE Linux.”

Category:

  • Linux

Linux advisory watch – March 15th 2002

Author: JT Smith

From LinuxSecurity.com: “This week, advisories were released for zlib, mod_ssl, xtel, pam_pgsql, cyrus-sasl, netscape, mod_frontpage, openssh, rsync, gzip, NetBSD kernel, php,
fileutils, and cvs. The vendors include Conectiva, Debian, EnGarde, FreeBSD, Immunix, Mandrake, NetBSD, Red Hat, Slackware, SuSE, Trustix, and Yellow
Dog Linux. Many serious advisories affecting nearly all Linux vendors were released this week, it is advisable that you patch your systems
immediately.”

Category:

  • Security