stefw

Dia for Class Diagrams

2010-03-03T03:09:31Z

In my class at NCSU, a UML class diagram is a required part of the design document for each project. I attempted to use Violet UML for this effort as introduced by our instructor, but I had a difficult time creating complex diagrams with its limited functions. So, having extensive experience with Dia for diagrams and figures for documentation, I decided to check out its UML editing features.

I was delighted at how much easier it was to construct the class diagram, especially if I forget a symbol I don't use much. The trick is lining up the terminology in the UML with the terminology in Java so I can use all the features effectively.

For more information about Dia and to download it search for "dia for _____" and fill in the blank with "mac" or "windows" or "linux" (I've used all three at different times).

Setting up my Android SDK and Emulator

2010-02-18T04:55:54Z

Sometimes it's hard to find a "quick start" guide to set up a development environment. Here's a quick checklist for my own (and anyone else) to use as a future reference when setting up an Eclipse-driven Android development environment:

(1) Install Eclipse from http://www.eclipse.org/downloads/
I was already using Eclipse IDE for Java Developers, the Galileo release for Mac, build 20090619-0625.

(2) Download the Android SDK from http://developer.android.com/intl/de/sdk/index.html
I'm using the one for Mac (android-sdk_r04-mac_86.zip).

(3) Unpack the SDK and put it in a local directory you're comfortable referencing it from.
For the Mac install, I just dropped the android-sdk-mac_86 directory into /Applications/.

(4) Install the Android Developer Tools for Eclipse. Basically, open Eclipse, go to Help --> Install New Software, and add the following as an available site: https://dl-ssl.google.com/android/eclipse

(5) Still in Eclipse, go to Preferences from the menus, select Android, and set the SDK location to point to the android-sdk-<something> directory you unpacked to your preferred location.

(6) Also working in Eclipse, install one or more of the SDK platform images from Windows --> Android SDK and AVD Manager using the directions here: http://developer.android.com/intl/de/sdk/adding-components.html
I installed SDK Platform Android 1.6, API 4, revision 2, which is what I'm running on my G1 and what I'm programming for in my class at NCSU. I also installed the corresponding Google APIs; not yet sure what I might do with those.

(7) To test you have a working test environment, create an AVD, and then launch an emulator. The emulator plus AVD creates a virtual machine running the Android OS. Here's more detail for each step:

(7.1) Create the AVD using either using the same Android SDK and AVD Manager in Eclipse, or using instructions here: http://developer.android.com/intl/de/guide/developing/tools/avd.html
In the Android SDK and AVD Manager, I went to Virtual Devices, clicked Add, gave my new configuration a name, selected the Android 1.6 SDK from the targets drop-down list, and saved the configuration. I didn't worry about tweaking any other configuration for the first launch.

(7.2) Launch the emulator with the new AVD either using the same Android ADK and AVD Manager in Eclipse, or using instructions here: http://developer.android.com/intl/de/guide/developing/tools/emulator.html
In the Android ADK and AVD Manager, while I was still in Virtual Devices, I selected my new configuration and clicked Start. The emulator launched in a separate window...

... and it was really kinda cool.

From there, I used the Subversive plugin in Eclipse to connect to the project code I'm working on, check it out, and start working on it. My next step is to figure out just how to pull together my Java classes and manifest into an Android package, add it as part of the AVD, and test it in the emulator.

stefw @ 2010-02-08T10:11:00

2010-02-08T15:11:40Z

Originally submitted at Timbuk2

A mini-messenger for urban adventure

Same Top Quality, But Different Features

By Stef from Raleigh, NC on 2/8/2010

Pros: Attractive, Great Color, Good Strap Length, High Quality, Good Organization, Enough Compartments, Easy Access

Best Uses: Shopping, Traveling, Everyday, Carrying a small netbook

Describe Yourself: Career, Comfort-Oriented, Practical

I got the Click on sale at a great price compared to the classic small messenger bag. It is about the same size, but it has a sewed instead of a clipped strap, and it uses only the Velcro tabs for the flap instead of having the additional clips like the messenger bag. For me, it was a great bargain: all the quality of a Timbuk2 bag, but on a smaller scale that was perfect for carrying my EeePC, a book, and a small notebook for class. I love it, and I recommend it to anyone needing the smaller messenger-type bag without the messenger bag cost.

(legalese)

Turning my EeePC into a Java Programmer System

2010-01-10T23:11:42Z

Technically, I haven't owned a portable computer of my own until now. Sure, I've had work-issued laptops for the last few years, and I've borrowed laptops from friends from time to time. When I bought my mom's Asus EeePC, though, it was the first time I'd had one I could call mine. Everything else I own is a desktop system. :-)

For my Computer Science programming coursework at NCSU, I can easily do assignments at my desktop systems or from one of the campus computer labs. Once in a while, though, I'd like to grab a portable system and go somewhere else. I could use my work-issued MacBook, but if anything happens to it while I'm using it for something other than work, I could be held responsible for its replacement. Plus, even though it isn't but a 13" MacBook, it's still a little bigger than I want to haul around with me.

The EeePC had Xandros pre-installed, and I really like it. However, I didn't want the launcher. When I tried installing the advanced-mode Xandros, I was excited because it had the look and feel I wanted. Unfortunately, any minor change in software bricked the system. I gave up on Xandros after about half a dozen repair installs.

Having participated in the Foresight Linux project for some time, my next instinct was to try installing Foresight Linux Mobile Edition. After all, I have friends and colleagues that have used the EeePC and other netbooks with FL Mobile with much success.

I loved it... but I needed to customize it to my needs as a Java programmer.

My first instinct was the fall back on my Conary system building skills and use rBuilder Online to build a custom one-off of Foresight Mobile. However, the package structure was significantly different in some ways, and I didn't want to spend a lot of time figuring out how to make the system into a Java programmer's toolkit. So, I decided just to use my Conary system administration experience and add and remove packages until I had the system I wanted. Later, if I wanted to, I could capture this list of packages and create a "backup" system image of sorts using rBuilder Online.

For the following process, I used a combination of the Alt+F1 tty and runlevel 3 to drop to a command line as needed. I also made sure sudo was configured (in /etc/sudoers) so I could use it without providing a password.

I started by removing the launcher, which was included as a dependency of the desktop switcher:

$> sudo conary erase desktop-switcher
$> sudo conary erase netbook-launcher

Then, because I wanted to control my software at a command line instead of using the GNOME Package Kit tool, I removed the package kit:

$> sudo conary erase packagekit

Because I wanted to conserve space, I also removed some applications that I knew I would not use on this system. Most were modestly sized, so I probably could have done without removing them. (Browse the application launcher in the default GNOME install to see what GUI applications are installed, and determine which you want to keep and remove.) Here are the erase commands for the ones I removed:

$> sudo conary erase f-spot
$> sudo conary erase banshee
$> sudo conary erase pidgin

At this point, I was still using 58% of the 4 GB solid state drive. I considered replacing GNOME with xfce to save even more space and further reduce processing overhead. For now, though, I'm sticking with GNOME for familiarity's sake, and it seems to be running efficiently so far.

Next, I had to add my programming tools for Java. First, I needed the JDK for Java 6 (aka 1.6), which also brings in the JRE as a dependency:

$> sudo conary update sun-jdk

The install took a few minutes on the Eee PC. Stopping to check my disk space after the update, I was up to 68% used of my 4 GB. I wasn't concerned about room for my data, since I had an 8 GB SD card mounted at /data, ready for my programming work. However, the percentage made me doubtful I'd be able to install and use Eclipse as my IDE. I decided to give it a try, anyway.

I first tried installing just the runtime component (eclipse-sdk:runtime), but when I launched the application, it said it was missing some necessary jar files. Then I installed the eclipse-sdk package as shown here, which also brought in its dependencies: an earlier build of OpenSSL's library component, the Java components for Xalan and Xerces, and Python's runtime and library components:

$> sudo conary update eclipse-sdk:runtime

This install took less time, and it bumped me to 75% of the solid state drive used. I was happy I had the tools I needed installed, and decided the next item of business was to evaluate performance of Eclipse as my Java IDE on the Eee PC with my customized Foresight Mobile.

Before I did anything, I rebooted the system and confirmed everything was still in working order. Everything was fine. Yay!

Before I launched Eclipse, I wanted to set up a workspace directory on the 8 GB SD card. I wanted to be sure my Eclipse work was stored there instead of in my home directory on the solid state drive. I changed to root, changed the owner on /data was mounted, and made sure it was mounted correctly in /etc/fstab. Then, I created a directory there named eclipse-workspace-java6.

When I launched Eclipse, in the "Select a workspace" dialog, I pointed to my new /data/eclipse-workspace-java6 directory. The splash screen since everything is set to be full screen by default (it tiled instead of expanding). The view was the same as using Eclipse on any system, but it seemed to take up a lot of space. I'm sure I'll continue customizing the look and feel from here until it's like I want it.

As I've done a little bit of sample coding, I've noticed that Eclipse seems to work great on my EeePC. I haven't pushed the limited to see how it runs alongside Firefox and other programs, but I'm confident I at least have the Java Programmer System I was looking for.

Installing the SOAPpy Python module

2009-12-21T21:44:31Z

Having not posted since September, I thought it interesting that my new post today is related to my previous post. Basically, I had to get really comfortable finding and installing Python modules, and fixing issues with the installs, so I could make the script run on my local system.

One big thing I had to have (to run JIRAnemo) was SOAPpy, which was apparently called "SOAPy" (with one "p") until about 2003, and which hasn't had a new release since 2004 (stable) and 2005 (release candidate). Also, the dated documentation for SOAPpy is accurate in what you need, but out-of-date in its links to where to get it.

In hopes of helping those of you out there who are trying to get SOAPpy running, especially on Mac or Windows, here's the information about what I did on my Mac OS X Leopard install:

(1) Update your Python to 2.6.x (optional):
    (1.1) Download the newest Mac installer disk image for 2.6.x from:
        http://www.python.org
    (1.2) After installing the package, verify the command line the
        Python Launcher (under Applications/Python 2.6/ i=on the Mac)
        wants to use by default, such as: /usr/local/bin/pythonw
(2) Install the Python module pyXML:
    (2.1) Download it from:
        http://pyxml.sourceforge.net/topics/download.html
    (2.2) Unpack it, change to its directory, and install it to your 2.6.x version of Python:
        $> /usr/local/bin/pythonw setup.py install
(3) Install the Python module fpconst:
    (3.1) Download it from:
        http://pypi.python.org/pypi/fpconst/
    (3.2) Unpack it, cchange to its directory, and install it to your 2.6.x version of Python:
        $> /usr/local/bin/pythonw setup.py install
(4) Install the Python module Web Services for Python (ZSI)... may not need this, but I installed it anyway:
    (4.1) Download it from:
        http://sourceforge.net/projects/pywebsvcs/files/
    (4.2) Unpack it, change to its directory, and install it to your 2.6.x version of Python;
        use "sudo" to pass permissions to install to /usr/local/bin/:
        $> sudo usr/local/bin/pythonw setup.py install
(5) Install SOAPpy:
    (5.1) Download it from:
        http://sourceforge.net/projects/pywebsvcs/files/
    (5.2) Unpack it, cd into it, and install it to your 2.6.x version of Python:
        $> usr/local/bin/pythonw setup.py install
    (5.3) For each file that Python reports that the "__future__" import line needs to
        be at the beginning of the file, make that adjustment, and rerun the install.
        Repeat this process until the install completes successfully.
(6) Install lxml:
    (6.1) Download it from:
        http://codespeak.net/lxml/index.html#download
    (6.2) Unpack it, cd into it, and install it to your 2.6.x version of Python:
        $> usr/local/bin/pythonw setup.py install

My First Python Thing

2009-09-10T03:05:39Z

Okay, so thanks to my experiences writing recipes and working with appliance agent code at rPath, I have already done a lot of Python coding. The difference there, though, is that it is Python syntax within a specific API. Outside of that, though, I had yet to write or contribute to original Python code.

Now I have! I cannot express thanks enough to Elliot for coming over after work hours and spending 15-20 minutes showing me some essential things to add to my code so I could call it with arguments at the command line. He also advised me on some best practices I had not really picked up on before. There are a lot of helpful resources online, but wading through to find what I needed was going to take more time than I wanted to put into this first effort.

So, what's the program do?

The problem I needed to solve was automating the retrieval of certain information from an installed JIRA (in this case, the rPath Issue Tracking System at issues.rpath.com). I needed to be able to provide a product and a fix version (equal to the release version) for that product, and then to retrieve issue IDs and "README" fields for every issue associated with that product and fix version. This would be a way of assembling the rough draft of the release notes for that product release.

My desired functionality was to run a single command, providing three arguments to that command, and to get the desired results printed to stdout. The arguments are the product ID and fix version, plus the desired type of output. The three types of output are:

(1) from the queried issues, the list of issue IDs for issues with blank "README" fields, needed to determine where we need to add information in rITS;

(2) from the queried issues, the contents of the "README" field plus the issue ID value (or a note that an issue still has no README field), needed to make a first editing pass with changes made directly in rITS as needed;

(3) from the queried issues, the DocBook XML markup needed to generate the release notes for publishing.

One additional need is to ensure that when addressing #2 and #3, identify whether or not an issue should be publicly visible, and only print the issue ID if it is. Markup will include linking these issues as appropriate, too.

The first code attempt created a class called "generel" (get it? GENErate RELease notes?). The class had four methods and allowed me to accomplish output types #1 and #2 (though not yet with the check for publicity).

I can't believe how simple the code is for this, thanks mostly to work already done by Dave Christian's Jiranemo that he programmed back in late 2006, and Andy Grimm's contributions working with pyjira. Thanks guys!

I'll blog more about this one again later when it is complete.

In other news... my first Java programming experiences at NCSU have been very smooth and easy. I look around the class, though, and I see people struggling to understand basic programming logic and syntax the same way I did my first time through it 17 years ago. Without any prior exposure to coding, it is difficult to connect some concepts for people. While I am relieved not to be in those shoes any longer, I am also glad to help those others when I can. Even though there is a bit of boredom to the class, I'm too busy with other things in life to worry about it at the moment... I'll be cherishing these simpler times soon enough, I'm sure. :-)

Introducing Charlie

2009-08-12T20:03:28Z

Charlie is the name of my new Mac Mini, a celebratory purchase made over tax-free weekend. Charlie is my new media center, though I will continue to use the DVD player and AV receiver as part of the overall system. Charlie looks cute next to the Wii.

Charlie plays Netflix, Hulu, and YouTube beautifully, especially with its 2.0 GHz processor and 4 GB of RAM.

Charlie has not yet met Holly (custom-built PC) and Eric the Half-a-TB (network storage device), but I have a feeling they will all get along fine. :-)

Charlie had a little difficulty early on accepting the new RF keyboard/touchpad I purchased. Of course, most Macs are sheltered in their Apple environments for a while before they get out into the world, so I knew there might be a bit of an adjustment period for Charlie. They seem to be getting along well now.

Despite his social challenges, Charlie was delighted to meet the Logitech remote! He is happy to go to stand-by or to waken, following suit with the TV and AV receiver, all from a push of a button on the remote. I was glad to see this ready cooperation.

What does Monty think of all this? Well, I think he was a bit jealous; this morning he was squawking quite dramatically from his cage, trying to make a scene while I watched an episode of the Colbert Report. In time, though, I think he will get used to it, particularly if I play some of his favorite YouTube videos. :-)

Back at rPath

2009-08-12T18:26:40Z

I just realized my last post in this blog was May 20th. Since then, I worked a six-week contract at rPath to update some documentation. Afterward, they extended me an offer to come back to my original position with the company: technical writer. I accepted that offer and started this week as a full-time permanent "docs person."

It is good to be working again. The contract work with rPath, along with my ongoing freelance writing for HowStuffWorks.com (under the pen name Stephanie Crawford), has been an enjoyable way to rebuild my professional confidence. With the resuming of a salaries position, I have been given the opportunity to put things back together again, and a second chance to plan my finances better knowing the type of job market that's out there for people with my experience.

I was also readmitted to N. C. State so that I can finish my Computer Science undergraduate degree. I had almost three years of CSC coursework completed when I changed majors to Mathematics Education. After I got the Math Ed degree in 1996, I didn't think I would ever return to do university coursework. Now, I am looking forward to getting back into the geeky things I loved about CSC, with a renewed perspective to motivate me.

Now, off to do the stuff I get paid for. :-)

Still looking, but certainly not idle!

2009-05-20T19:15:04Z

Wow, I can't believe I haven't posted since March. Not much as changed except that I have had some really good interviews and some small bits of freelance work, but I'm still generally unemployed. I would love to continue embracing the freelance lifestyle, I just need more work in the pipeline to make it work out. If you need any writing, course development, instructional design, or PC repair, let me know!

In the meantime, I have readjusted to using Windows XP for the sake of using certain software: Flash, Fireworks, and Captivate. Because I still need dependable access to my Linux systems, today I learned how to use two PuTTY supplements, PuTTYgen and Pageant, to adapt my OpenSSH keys and use them from PuTTY and FileZilla. BTW, FileZilla rocks the frickin' house... I LOVE this tool!

When I'm not working on projects, I'm looking for jobs and developing my skills. I'm thinking of doing some creative ways of presenting my resume and putting them online. The main page could be a portal to download the Word or PDF copies, or to view a Flash or Captivate version.

Lots of little things to accomplish right now, including these things to enhance my hunt for work.

The Security of Employment versus Controling my Future with Freelance

2009-03-11T12:50:37Z

There is good news on the horizon, but I feel like I'm walking on a treadmill toward that point, not quite reaching it. I have a modest short-term agreement to return to rPath and write some items due in April, but I'm still waiting for other pieces to be in place before that starts. Also, I have an interview today with a firm that may be able to keep me busy with other short-term assignments when the rPath gig ends. These are helping to launch my career shift into being a freelance writer. There are also some other potential assignments that are longer term, some with and without "benefits," but now that I have considered the advantages of freelance, I am looking at those opportunities in a different light than I would have three years ago.

Yesterday, a mentor and friend took time to talk with me as I worked through an internal debate on what it is that I want out of my career just now. Because I may have a couple of routes before me, I needed to think ahead on which road was going to take me where I want to go. Our discussion, though purely just a hashing out of the facts, helped me to get my thoughts in the proper focus. Here's my thinking list:

Path 1: Secure longer-term positions

ADVANTAGES
* Sense of security in having a job from one day/week/month to the next
* A consistency in routine over that extended amount of time
* A predictable monthly income
* Employer-paid benefits (when offered)
* Sense of camaraderie and common corporate goals with a team of colleagues
* Building trust and good relationships with colleagues over time to strengthen future references

DISADVANTAGES
* Pay rate lock-in for an unpredictable amount of time
* Dependence on opportunities for upward mobility aided by the employer in order to advance professionally
* Less opportunity for taking on overlapping work for extra income
* No guarantee of having the job forever; the possibility of starting over at any given time, disrupting the security and consistency

Path 2: Short-term freelance work

ADVANTAGES
* Control over the selection of work and, as a result, the overall work calendar
* More opportunity to overlap work, with flexibility to adjust the workload based on other life events
* Adjustable pay rate between assignments/contracts
* Choice of a healthcare plan and other items often included as employer benefits
* Professional organizations provide a support system for freelancers if there is a need to replace the camaraderie of long-term colleagues
* Opportunities to deepen the portfolio with a wide variety of experience and increase the number of professional relationships
* Avoiding the uncertainties about whether and when the job will end

DISADVANTAGES
* Unpredictable monthly income
* Time management challenges and the possibility of ongoing changes to the daily/weekly/monthly routine
* Need to take time to self-sell to ensure the work calendar is full [may be able to counter this by working with a firm that does this sort of thing, such as the one I am speaking with today]
* Requirement to charge more per hour to cover health insurance and other expenses often covered by an employer
* Managing my own income tax witholding, 401K contributions, and health and dental insurance.

(I know not all the bullet points here are formed in the best consistency of voice and tense, but the words are meant to convey the idea, not necessarily to meet technical writing standards. ;-)

There are risks with either of these career paths. In time, though, I think I would be more comfortable with the freelance path: it puts me in control of my career instead of depending on an employer to control my future. However, with power comes responsibility, and I know that I alone am responsible for my success or failure in freelance work.

Now, I'm off to bake pies at the coffeeshop... a venture which isn't really making me any money, but thankfully has kept me busy while I continue on this treadmill toward the promising horizon.

Itching to Research and Write

2009-03-03T21:36:44Z

In an attempt to make a little extra cash, especially because the NC Employment Security Commission is overflowing and incredibly slow for processing unemployment claims right now (made mine in January, still not finished processing), I have been baking pies and cookies at a local coffee shop. It has kept me busy and prevented my sinking too deep in frustration about the slow job market and lack of call-backs for interviews. However, today and tomorrow are out; no baking until Thursday while the coffee shop owners work on getting some renovations complete.

So... I got the aching to write, so I got a library card and camped out at one of the Wake County Public Library branches to do writing and research.

First up... Adobe Flex 3.0. It looks simple enough on the surface. I found a book to get started and I'm currently setting up an Eclipse working environment on my Foresight Linux system to give it a whirl. This should help in my professional development, too, as I can create a sample site and put my experience on my resume. I would definitely love to get into web front-end development work.

Next up... history of politics in NC and the US. I want to read about significant events in our state's political history, compare it to events happening nationwide during the same time, and write some blog posts for GoLiberty.net filled with both facts and personal commentary.

I'm sure that is enough to fill my need to study, research, and write for a couple of weeks at least... including reading material for standing in the unemployment line.

Progress or not? Definitely busy!

2009-02-19T03:12:08Z

It is hard to tell whether I'm getting traction on any job opportunities. I have been really thankful for a lot of good referrals and introductions, and I have been excited about some of the listings, but I haven't yet started down a successful road to interview. It is hard to convince people that I am very much qualified for these positions when they have other equally qualified resumes in front of them and, probably, equally as enthusiastic referrals. It is a tough and competitive market, especially in the areas of writing and training where companies are freezing or cutting back when they can.

Even though I am trying not to get too excited about it, there is a possibility I could follow a former manager to Seattle, Washington. I would be working for a different division in the same company, and in an excellent position that puts "senior" in my title for the first time in my career. I feel like I have worked for it, and I can more than handle the job responsibilities, but will the hiring manager be convinced of that enough to want to interview me? Even though I would love to stay in Raleigh, this is a great opportunity for something I really want, and I just can't afford to pass it up! Wish me luck!

There is a local opportunity brewing thanks to another referral, but the individual working on the hiring is on vacation until next week. I had a really good email exchange on that opportunity, though, so I feel good about it so far.

Besides these two possibilities, a lot of resumes and cover letters have been sent out, and more and more hours have been spent searching and filling out online forms and exchanging introductions by email. Beyond that, though, I have filed my taxes, restarted my project at ConaryUncorked.org, and even started baking my pies and cookies at (and for) a local up-and-coming coffeehouse.

That last item also has me a phone call away from putting together a business plan and looking for investors, and I would move on it if I though I could risk little or no income for a couple of months while I get started. Perhaps it is an idea to put on the shelf for later, or perhaps starting an incubator kitchen in Raleigh is a possibility... I wonder how many other aspiring culinary artists would want a timeshare in a state-inspected commercial-line kitchen to help incubate their skills.

My goal when I am employed again is to ramp up to a 6-month runway in terms of income so that if I find myself in this position again, I can go for it with starting my own business if I'm so inspired. ;-)

Speaking of busy... woah! How did my to-do list get so big?

IBM and AWS

2009-02-12T13:52:59Z

Early on in Amazon Web Services (AWS) history when its Elastic Compute Cloud (EC2) was still a limited-access beta, I knew that any any large company with reputable software could really drive people to try this new could computing model. This is especially true for software that required dedicated hardware resources that were often out of consideration for small and medium business. The subscription model for using an application, especially those primarily driven by a web front-end, puts those previously out-of-reach applications into better reach. In addition, the cloud model provides easier and cheaper scalability options for the users when their needs grow or shrink.

Extending from my enthusiasm for the future of cloud technologies and how it will change corporate IT in the next decade, I was thrilled to see the news online that IBM had partnered with Amazon.com to offer DB2, Websphere Portal Server, and other widely-used products to customers wishing to deploy and run them in EC2. Not only does this include product deployment for the software, but it also includes a "Development" AMI that software developers can use when developing their applications that use Websphere, DB2, or Informix.

I hope to see software vendors (ISVs) quickly jump on this opportunity. Our current economy in the US is certainly one thing that could drive a transition to EC2. I think IBM was smart to be proactive in this growing area. After all, if your key hardware products aren't going out the door as much, you had better have a share of the market to which your customers and prospects are moving. ;-)

For more information about this partnership, check out the following article:
http://aws.amazon.com/solutions/featured-partners/ibm/

Also, keep an eye on Jeff Barr's blog or tweets as he is the AWS evangelist and will provide even more ongoing news links:
http://www.jeff-barr.com/
http://www.twitter.com/jeffbarr

Job Hunt 2009, the first month passes

2009-02-09T20:19:37Z

I cannot believe how much of my time it takes to complete and submit job applications, even online!

Some of you know that I parted with rPath a month ago due to some resource changes on their part. I had a great emotional connection to the company and my colleagues there, and I am glad that many have become good friends and lasting professional connections. It is tough for everyone to go through restructuring, and even tougher to deal with it in a down economy. I wish all my former colleagues prosperous careers and hope we never lose touch. I'm glad to have shared this experience with you all!

Were I a software engineer or an experienced web developer, I would have an easier time in this current job market. Because my primary background has been in technical writing and training, I am not only working in an area in which companies are cutting back, but also I cannot compete with the flood of individuals in the job market with more first-hand experience than I have in web development, Linux administration, network administration, scripting, testing, QA, release management, and other engineering-driven tasks. As a result of these difficulties, I have to leverage more than just a resume and cover letter when applying for positions; I have to rely also on personal contacts and the possibility of a major relocation. Here's a breakdown of how things have been over the last month:

* First week: Shock. I knew I might have some limited time at a venture-backed company in the down economy, but I thought I had at least a good year ahead of me before something like this. I had to give things time to settle and work on updating my LinkedIn profile and my resume. LinkedIn connections led to my rPath job and to my IBM job before that, and I have a lot more confidence in my portfolio and references this time back in the job market.

* Second week: Diving in. There aren't a lot of listings for training and writing out there, but there is a promising connection with a certain primary-colored-head-covering-named company that could prove very useful, especially considering my experience with rPath. Unfortunately, I apparently stepped through their online system out of order, and I think the hiring staff got a really old copy of my resume instead of my newer one; the system isn't letting me change it, so that makes me nervous. That's okay, my contact passed on the newer version directly to the hiring manager; that should help, right? I really wish I had found the money to pay for my RHCE before now! In the meantime, I'm registering for CareerBuilder.com, updating my profile at other sites, and submitting for some jobs at N.C. State.

* Third week: Persistence. So my first attempt at using a contact to get a position has failed without even a phone call, just a cold HR email thanking me for my interest. I'm sure my lack of RHCE was the big factor in that; they need someone who already has the certification, and I understand that. I wish I had the money to take the test now! Thankfully, some TriLUG contacts and former colleagues are offering some help and recommendations, and I have found some really good listings locally. I also had a really good phone interview with the principal at a local high school for an exciting position in a special engineering-prep program; it was a confidence booster, but unfortunately, I can't lock myself into that salary for the next six months unless it is close enough to home for me to give up my car. I'm going with my gut instinct to hold out for something else.

* Fourth week: Reality check. It is tough out there. Some folks have it tougher than me, certainly, but the emotional roller coaster of unemployment is taking its toll on me. I had a really good phone screening for a position followed by an email saying they couldn't go forward that insinuated a limitation on their part instead of mine. Also, I had to turn down two other possibilities even before an interview because they were just not going to pay enough, and would leave me no time to take a second job to make up the difference. Making it even tougher is watching so many of my fellow layoff victims starting new positions now. I am happy for them, but it does result in that punch-in-the-gut feeling that I haven't even had more than a phone screening up to this point.

The roller coaster continues: a morning filled with possibilities in new job listings; an afternoon of writing cover letters, filling in forms, and clicking submit buttons; an evening of frustration as my email inbox has had little more than spam and my phone has collected a sheen of dust from lack of use. I have had longer periods of unemployment than this, but I have never had quite this feeling of vulnerability, nor quite such dramatic rises and falls in emotion.

Despite the frustration, there is still a ray of hope, and I will cling to it as tightly as possible.

If you know of anyone hiring in any of the following areas, I would *love* to talk to those doing the hiring and screening and have the opportunity to give a personal account on what I can do to contribute to the team:
* Technical writing (mid-level or senior)
* QA/testing (mid-level)
* Web development (junior/entry-level or mid-level)
* Training (mid-level or senior)
* Linux system administration (junior/entry-level or mid-level)
My resume is linked from my my LinkedIn profile, which also includes my history, experience, and recommendations.

In the meantime, I'll be working on some other projects to keep my mind busy:
* Working with a local marketer who wants to collaborate on revamping his DVD training program (which might bring in a little extra income if he likes my prototype)
* Continuing development of ConaryUncorked.org
* Preparing material for a the Raleigh Television Network project
* Learn some C#, and package Boxee for Foresight Linux (or at least give it a concerted effort to see where our limitations lie, and then perhaps contribute code back to Boxee if I can find just the right adjustment to make it work)

As I opened with, though... WHEW! I cannot believe how much of my time is spend on this job search! Now if I could just get paid for THAT, too... ;-)

Holly Upgrades, Jan 2009

2009-01-16T17:05:17Z

Holly Upgrades, Jan 2009, 4 of 7
Originally uploaded by soundof4strings

A couple of weeks ago, my self-built home PC got major upgrades just in time for becoming my primary system again. I went from a 32-bit Pentium 3 system to a 64-bit dual-core AMD system with 2 GB RAM just by updating four parts: motherboard, processor, RAM, and power supply. It runs 64-bit Foresight Linux beautifully. :-)

New QuickStart for the Appliance Creator in rBuilder Online

2008-12-09T16:45:29Z

Some of you already know that the features you see in rBuilder Online (www.rpath.org) are a beta for features targeted for the rBuilder appliance product. Two of the features currently in rBO that have not really been documented are the Package Creator and the Appliance Creator.

Yesterday, I published a "QuickStart" guide to stepping through the Appliance Creator, which includes the Package Creator steps within its wizard. You can literally start with an rpm, deb, tar, or compressed tar archive, and use nothing outside the rBuilder web interface to package the application, build the JeOS appliance, and generate images under your product definition. This even includes the ability to tweak your Conary recipe files for the package and for the appliance right there in the interface!

This is a HUGE step forward to getting people started with rPath products, and I have already planned a similar QuickStart guide to go out this week to introduce the new rBuild command line utility, currently in beta.

Check out the new QuickStart guide at docs.rpath.com under "QuickStart Guides" in the "rBuilder" section. You can view it online or download a PDF. Here's a direct link to the online version:
http://docs.rpath.com/quickstart-appliance_creator.html

Conary Migrate Finally Documented

2008-11-29T05:10:04Z

So, somehow over the course of the last year, the highly useful conary migrate command was never fully documented. Sure, the conary(1) man page defines the basic command and syntax, and the inline help provides a list and short description of options, but it was never really documented clearly what the command does by default and why anyone would want to use it.

So tonight, in an attempt to define it to two people back-to-back in IRC, I decided it was time to stop repeating myself and provide a reference that I could point to when speaking to others, either in Conary and Foresight community discussion, or as part of the professional work with rPath.

So here's the new page: http://wiki.rpath.com/wiki/Conary:Migrate
I welcome feedback that might help me to improve what I wrote there, and any other registered user on the rPath Wiki should be able to contribute modifications, too.

I also added a line and link in the Conary QuickReference for the command.

I still have an ever-growing desire to get Conary documented on its own, separating the valuable community side of Conary from the rPath roots which have driven most of its development. I feel that the separation of branding will invite more distribution development using Conary, which will lead to more people using Conary, contributing feedback, and putting their experience on resumes and weblogs. Conary has the strength and features to be its own brand, which I think would encourage developers to offer their software packaged for Conary just as they would for RPM. :-)

That said, look for me to launch this project full-force in about a month. I look forward to it!

Election Day 2008 with Mike Munger

2008-11-20T21:30:10Z

A little aside from my usual posts here...

Election Day 2008 with Mike Munger
Originally uploaded by soundof4strings

This is me posing with gubernatorial candidate Mike Munger between watching election results on November 4, 2008. I'm so excited about the LP in NC meeting the state's ballot retention requirements, and I think Munger's campaign was the primary reason for that. Many thanks to both Mike, his campaign manager (and LPNC chair) Barbara Howe, and many enthusiastic volunteers who helped make this happen.

I look forward to taking advantage of that continued ballot access alongside fellow Libertarians in 2010 and 2012.

Preparing for Cloud Training and Using Amazon EC2

2008-11-07T19:05:02Z

It has been over a year since rPath provided the feature on rBuilder Online to create an Amazon Machine Image for an appliance which is automatically uploaded to the Amazon Elastic Compute Cloud (Amazon EC2). I played with the Firefox plugin which provided a convenient GUI in place of using the EC2 command line, and I launched and used some images while also providing some of my own rBO product releases as AMIs. Now rBO has the rBuilder Catalog for EC2 which has made instance management even more convenient for those of us using rBuilder to build the AMIs.

Now, cloud computing in general has become a hot topic, and companies are evaluating its features to see what role it has for their infrastructures. Path has responded with training to inform those folks of how they can start the transition process to an infrastructure in the cloud. There are really two audiences for such information, and rPath Training Services is working to address both in its courses:

(1) Infrastructure Planners, Network Administrators, and System Administrators

These are the people responsible for how software gets deployed as part of a corporate infrastructure, and how the network topology is maintained and scaled over time. These people need to know:

How does cloud computing differ from using a virtual hypervisor?
How do public and provide cloud computing solutions compare to each other and to traditional infrastructure solutions?
What are the ongoing costs associated with cloud instances as opposed to hardware and virtual hypervisors?
How does running in the cloud change troubleshooting and maintenance for the systems and software?
How can I launch an existing application in a cloud instance instead of using a local machine or hypervisor?
How do I ensure the infrastructure in the cloud has reliable uptime, minimal points of failure, and efficient backup, restore, and update strategies?
What tasks are involved in launching shared resources and ensuring multiple instances can reliably use those resources?
How can I provide load balancing between cloud resources?
What resources or actions are required to scale the network and storage over time?

I have been learning a lot about how Amazon EC2 addresses these questions, and about how other cloud solutions (including private solutions for the enterprise) compare when addressing the same questions. I am excited about working with the folks at Amazon Web Services(TM) to develop a comprehensive course designed to address these questions and more. The first session of the course is already scheduled for December 10-12, 2008, in Redwood City, California, and I look forward to folks joining me in the Bay Area and stepping through these hands-on lab activities. Check out the following links for more info:

(2) Application Providers

These are the people responsible for delivering software so it can be deployed as part of a corporate infrastructure. This may include independent software vendors delivering directly to their customers, or it could be enterprise software acquisition teams responsible for purchasing software applications and providing the internal mechanisms used to deploy each instance of those applications. These people need to know:

How does the underlying operating system choice impact providing the application in the cloud?
What needs to be done to prepare the application for cloud deployment?
What post-installation configuration tasks are required on a system during a typical deployment?
How can I use the rPath tools to prepare the application as a cloud image?

The rPTS-101 course from rPath is the door-opener to providing applications in virtual environments, including cloud computing environments. This course is the primer those developing offerings that are pre-packaged for cloud computing solutions, introducing rPath tools and stepping through hands-on labs that go from application to complete, customized virtual machine. In addition, both rPTS-101 and the self-paced starter Application to Appliance: A Hands-on Guide (rPTS-099) are about to go through some updates to incorporate even more new strategies for providing applications for virtual environments:

Course objectives for rPTS-101 Providing Applications on the rPath Platform

This is such an exciting time for virtual and cloud computing! I look forward to equipping people for the journey!

Much love for rbuild

2008-11-06T22:33:57Z

With the new product model in rBuilder, featured as part of rBuilder Online, a single XML file can make for a quick step through the appliance creation and building process. As a result, rbuild has been in development for a few months and recently released for beta testing with rBuilder Online, wraps the cvc and rmake development tools together with replacement functions originally performed by the rbuilder client. This triple-threat tool does not preempt a developer's option to use Conary and rMake tools, but it adds the following key features:

(*) The option to skip initial Conary and rMake configuration and dive straight into recipe development
INSTEAD: Run rbuild config and step through a quick questionnaire that completes an rbuild configuration for you, so you can go straight into checking out and modifying your source code.

(*) Eliminates the need to produce exact flavor specifications in the build command to match the desired images
INSTEAD: Run rbuild init with the label on which your pre-defined product definition XML source exists (product-definition:source). Then, after performing the checkouts and editing the package and recipe sources, you only need rbuild build packages or rbuild build groups to build the given package or group. The rbuild tool uses the product definition to line up the appropriate flavor specifications required for building the images indicated when you set up that product version in rBuilder.

(*) Eliminates the need to commit the binary build of the package or group as an extra step after the build is complete
INSTEAD: The rbuild tool will automatically commit the source and the binary build as its default behavior. According to the inline help, the --no-commit option can be added to override this option, which should keep the build local and allow for other operations such as creating a changeset from the rMake job for testing.

(*) Eliminates the need to jump back to the rBuilder interface to build images
INSTEAD: Run rbuild build images to launch the image builds of the latest committed source.

I'm excited about where this will allow us to take users in training. It provides a more introductory-level set of steps before delving deeper into Conary and rMake configuration and advanced options.

For now, I don't have time to provide a detailed set of instructions, but this should help you get started using it on your rBuilder Online products:

(1) If you have an existing rBuilder Online product, be sure you have created a new product version that creates a product definition committed to a given label. See the latest rBuilder documentation for more information about the product definition:
http://docs.rpath.com/rBuilder_4.1_User_Guide/ch-create_a_product.html

(2) In your local development environment, such as one launched using the Application to Appliance Development Image (app2app.rpath.org), run the following command to install the still-beta-level rbuild tool and accompanying application stack from Michael K. Johnson's product on rBO:

#> conary update {rmake,decorator,rpath-{common,xmllib,product-definition},rbuild}=mkj.rpath.org@rpl:2 --resolve

(3) Restart rMake after the update is complete:

#> service rmake restart

(4) Step through rbuild configuration and initialization for your desired label:

$> rbuild config

To see more details and information about each question in the questionnaire, see mkj's original post about rbuild here:
http://blogs.conary.com/index.php/mkj/2008/08/29/simplifying_assumptions

(5) Pick a directory within your home directory to start the checkout tree associated with your product version. I just picked ~/conary. You can use the following commands, but be sure to replace rpts.rpath.org@rpath:rpts-2008.12.1 with the label on which your product-definition:source component resides:

$> cd ~/conary/
$> rbuild init rpts.rpath.org@rpath:rpts-2008.12.1

(6) Change to your Development stage subdirectory under the checkout to check out any additional package and group source components for continued development work:

$> cd ~/conary/rpts-2008.12.1/Development/
$> rbuild checkout group-rpts-appliance
$> cd group-rpts-appliance

(7) Build packages for the appliance. Flavors are automatically selected based on your product definition. I do not have examples for this; I was just using it to build groups so far.

(8) Build groups. Flavors are automatically selected based on your product definition. Again, the binary build and the source (if it has been modified since checkout) are committed after the build completes successfully:

$> rbuild build groups

(9) Build images. This launches the image builds in rBuilder, so if you Control-C out of the feedback at your command line, you can always continue watching the status in the associated rBuilder page for that image:

$> rbuild build images

Now, I don't really have more to say at this time. I literally started this less than an hour and a half ago, and about half of that was spent on this journal entry while another fourth was spent just talking with mkj about the features. Don't worry, though, I'll be testing the heck out rbuild in the coming weeks and will be sure to post information if I have any more to report... including the link to documentation when rPath can get something published (probably about the time the tool is released on the main rPath labels).

I'd love to hear what folks think about rbuild, including if you have any issues or questions. You can start with questions at the rPath Forum, and file issues in the rPath Issue Tracking System if necessary:
* http://forum.rpath.com (I'd recommend using the rBuilder, rBuilder Online, or Appliance Development boards)
* http://issues.rpath.com (The rbuild tool has its own project, "RBLD")

Dedicated Training Staff at rPath

2008-09-16T22:55:29Z

It is official: I'm now 100% course development and training for rPath. I will still continue to write and contribute documentation when it is directly supportive of training. With the new role seems to have come a wave of new training engagements which is likely to put me traveling over a third of the remainder of the year.

Though I had looked forward to the switch from one role to the next, the news was bittersweet as I watched internal responsibilities shift away from having a dedicated product documentation resource. However, this was just my emotional attachment after two years of lobbying to raise the priority for documentation. In fact, because rPath now has a dedicated effort toward "D3" (documentation-driven design), what customers and community enthusiasts should see is better communication of new features and a more user-friendly front-end for products that significantly reduces the amount of things that product documentation should address.

After having spent time reworking rPTS-101 and rPTS-201, I'm now ready to get re-energized for the fourth quarter (did I mention there are sales figures attached to my job now?) and tackle the development of two new course offerings.

In the meantime, I would like to thank all of my colleagues in rPath Engineering for being so helpful and supportive while I was part of the department (June 2006-September 2008). Though you are all a "boys club" once again (for a while), don't forget to come out sometimes and visit those of us outside the "short cubes." :D

Artoo can stay on the webcam... for now. ;-)

Using the new image set feature in rBuilder Online

2008-06-27T22:44:00Z

One of the painful things about getting out a new release of the Simple Machines Forum Appliance has been stepping through generating each image needed at the time of release. Fortunately, rBuilder Online has added the new image set feature so I can predefine all the images I need, then use a three-step questionnaire to generate all of those images at once. I was excited about the new feature, but then suddenly a bit disappointed as I realized its early limitations. The primary one was this: existing projects from rBO have an existing Conary label structure and appliance group name, and image sets require establishing a product version (also called a "definition") which can only use a set label structure. I imagine in time this will be more configurable, but for now it means:

Promoting all the custom packages to the expected development label for that defined version

Creating the new appliance group on that expected development label with the expected name for the product definition, but with code copied from the current appliance group

Note that any new rBO products (created since the 4.1 update) have a template appliance group recipe at the expected label, and developers can start fresh by checking out that starter recipe. For those of us with existing appliance projects (now called "products") in rBO, we can either choose not to use the image sets for now, or we can choose to move things to what rBO expects for these image sets. I chose the latter for the SMF appliance, so I am currently working through the following steps. Use these steps as a guide in updating your own appliances to use this new image set feature:

Create a new product version from the main page of the product (again, formerly known as "project").
- Be cautious of what you use for the "namespace" and the "major version" value because these make up part of your new Conary label for the product.
- Be aware that the advanced options for each image can be specified by expanding the section using the little chalk icon on the right side of the row for that image in the set list.

Use Manage Images --> Create a set of images from a product definition, stepping through selecting the product version and initial ("Development") stage, to see what group name and label are expected to have a binary build of the appliance in order to generate images from the set. This has an expected format of group-shortname-appliance=shortname.rpath.org@namespace:shortname-version-devel (replacing shortname, namespace, and version as appropriate for your product and version)

Set up your local development environment with an appropriate Conary context for the expected label.

Promote custom packages and non-appliance groups from their existing label to the new expected label. If necessary, make modifications to accommodate their new home.

Use cvc newpkg to create a new group with the expected group name, and copy the recipe from the existing label to the new group. At this step, be sure to rename the recipe file to match the new group name, and open the recipe file to make adjustments throughout to match its new home.

Continue development work as usual on this new label.

Use the new rBO interface features to generate images when necessary, and use cvc promote as always to promote from one stage to another in a release management process (though be sure to conform to the expected values for that product definition: -devel suffix for devel, -qa suffix for test/qa, and no suffix for release).

After releasing the updated appliance on the new label, change all the custom packages, non-appliance groups, and appliance groups on the old label to redirect to their new locations on the new label (see this wiki page for more info on changing these to use a redirect recipe).

Because I'm also doing other updates that will delay the "continue development" step for SMF, I cannot testify to how all this worked out for that appliance. Hopefully I can post an announcement soon, though, about the release of version 2.1.1.5 of the appliance (with SMF 1.1.5) on the new version 2 label.

rBuilder Online... Beta!

2008-06-24T03:42:18Z

So... what does this "beta" mean? Well, check it out: after a weekend-long maintenance window, rBuilder Online has been launched with several new features and other changes as described in the rBO page at the rPath Wiki: http://wiki.rpath.com/wiki/rBuilder_Online

It's good to see this new stuff in action, and I should get motivated now to update my Simple Machines Forum appliance with 1.1.5 and offer an appliance version of the 2.0 beta. I hope I can find some time to get that accomplished very soon.

SFO, the first 12 hours

2008-06-17T01:52:37Z

I am currently on the road for rPath Training, delivering the rPTS-101 course in Redwood City, CA, and staying in a hotel in nearby San Carlos. Though I hope to be able to swing down to San Jose to visit friends later this week (if they'll email me back already :-), I am currently focused on getting things ready to deliver. Thanks to the release of rBuilder 4.1.0, I was excited to update our rPTS-101 course to correspond to the new rBuilder features, but it has taken some time to get the course materials up-to-date.

Yesterday, the chaos of airline travel did result in my having time to concentrate on some course preparations, but it was utterly exhausting to have lost so much sleep. My 2:40 PM flight from Raleigh took off on time, but was in an indefinite weather-prompted hold at JFK so long that we diverted to Atlantic City, NJ, to refuel. Thanks to that diversion, adding a couple of hours to the normal flight time between RDU and JFK, I completely missed my connection to SFO, which left on time without me. The JetBlue staff was prompt about meeting passengers at the gate, though, with an updated boarding pass in hand to the next SFO flight. The new flight was supposed to board around 8:00 PM. However, after two delays and four gate changes, we didn't board until after 11 PM and didn't get off the ground until almost an hour later.

I did get a lot of time to hang out in the larger JetBlue terminal at JFK, and I was able to plug in my MacBook Pro and work for a while. I lost a lot of sleep though: I don't sleep well on planes as I can't really get into a comfortable position for myself, so I might have had at most an hour of sleep on board when you add all the snoozes together. After arriving, picking up the car, and getting to the hotel, I only got three and a half hours of good sleep before I was up to start preparations for this week's training.

In the meantime, though, I'm having a good rental car experience so far. Avis was out of subcompact and compact cars as I had requested, so they offered me a choice between a Ford Escape and a Ford Mustang, either at the same price I had reserved for the subcompact. I love the Escape from previous rental experiences, but I didn't want either that or the Mustang because of fuel costs. I took the Mustang, though, as well as a very helpful Garmin navigation tool on their new "where2" system. The nav has been VERY helpful so far, especially because I didn't feel like fumbling with navigation tools on my phone while driving the unfamiliar roads.

I am going to eat and work a while now while I'm still motivated after a very successful day of getting things set up for class tomorrow. I am looking forward to my first delivery for rPath Professional Training Services!

rPath Wall of Test Statuses

2008-04-25T22:08:46Z

rPath Wall of Test Statuses
Originally uploaded by soundof4strings

rPath Engineering uses Bamboo in combination with a web power switch to track how new code tests against its test suites. Each lava lamp in this picture corresponds to a product's test suite and is "on" if the tests are breaking. When everything is working, though, the "cold beer" light is on (see the smaller light above the mounted monitor).