CodeMiller

Katie Miller Journalist turned software engineer

[codemiller]

PyCon Au particulars: 10 elements of a quality conference

It's been about two years since I made my entrance into the wonderful world of software development conferences by attending PyCon Australia Sydney 2011.

Over the past few days I have been enjoying my second PyCon Au, this time in Hobart, and taken some time to reflect on what makes the conference unique and how I've grown as a developer since the Sydney event.

At my first PyCon Au I had no other conference experiences with which to compare but I recognised I had stumbled on something worthwhile and blogged about the five major benefits of attending a coding conference.

Two years on, I have attended quite a few conferences and even been privileged to speak at a couple.

In fact, a couple of people I ran into at PyCon Au this year were surprised to see me there as they had met me previously in a different context.

I have stayed true to the intention I expressed in my very first post on this blog, to expand my coding horizons by extending the limits of my language, and that has led me to become a part of multiple programming language communities. I believe each community has something to offer and -- especially as someone seeking to foster women in tech -- I think there is great benefit in being involved in many of them.

Looking through the lens of my experiences over the past couple of years at conferences and other events focused on a variety of languages, I can now discern that there are many things PyCon Au does particularly well.

Here are my 10 favourite features of PyCon Australia:

Inclusion of enthusiasts. I met a lot of software engineers at this year's PyCon Au, but I also met folks from completely different professions who just enjoy dabbling with Python. There were also many students and people from a variety of vocations wanting to learn to program to make their jobs easier. The conference schedule included talks suitable for beginners, so there was something for everyone. It was wonderful to see that people with a variety of professional backgrounds felt comfortable to attend and were warmly welcomed.

Commitment to diversity. The IT industry is struggling with gender (and other) diversity issues, but I've never felt out of place as a woman at PyCon Au. It really is an inclusive event, as outlined in its Code of Conduct, and attracts quite a number of women (enough that you don't feel like the odd one out in the crowd). Special effort is made to make sure women feel welcome with the inclusion a PyLadies breakfast on the program, which men can also attend provided they are invited by a female attendee.

Financial aid program. When it comes to the aforementioned two points, the conference puts its money where its mouth is. Its generous financial aid program helps students, teachers and members of groups that encourage diversity to attend the conference, as well as funding presenters and other community representatives.

Open-source sprints. The main conference event is followed by two days of sprints, where delegates come together to hack on open source code. Core developers from some of the Python ecosystem's key projects are on hand to help out and give the support and encouragement required for attendees to be able to submit a patch with confidence. This event helps to upskill developers and documenters, and garners valuable contributions to the community.

Prioritisation of lightning talks. Many conferences have lightning talks, but I don't think any I have attended have given them as much space as PyCon Australia. There was an hour set aside for lightning talks on each day of PyCon Au this year, which was one of my favourite parts of the day. The delegates responded to the call for talks with gusto, delivering informative and entertaining five-minute presentations. It was nice to have so many different voices on stage.

CodeWars. This is an event held ahead of the conference that isn't quite like anything I've experienced elsewhere. Teams compete in a coding tournament described as "Seriously technical, seriously irreverent, and seriously fun". The challenges aren't your standard coding problems; much work goes into creating puzzles that stretch the imagination and engage the audience.

Sit-down conference dinner. There's something really lovely about actually sitting down and having a proper meal with your fellow delegates. The PyCon Au organisers always make sure there is plenty of good food and beverages to go along with the good company. This year there was also a brilliant keynote by MooresCloud CEO Mark Pesce during the dinner.

Job board. PyCon Au hosts a large whiteboard where attendees can record if their companies are hiring, or if they are looking for work. It's a simple idea but a very effective means of facilitating this valuable delegate communication. 

Next day video. All of the conference tracks were recorded this year and the videos are all expected to go live this week, with some already released. Having this assurance that recordings will be available so quickly makes it much easier when you're agonising over which presentation to attend.

Miniconfs. One-day Django and OpenStack events were held ahead of the main conference. The miniconfs were a new addition this year, but by all accounts they made a significant contribution to the conference experience. 

Although it won't be a regular fixture of PyCon Au, special mention should also go to the amazing setting this year's conference had in Hobart, Tasmania.

The conference team organised a phenomenal venue, with stunning views of the harbour and Mount Wellington.

The 2013 conference was the biggest yet, attracting 310 delegates from as far afield as the United States, Europe and Asia.

Coordinator Christopher Neugebauer announced at the conference close that PyCon Au 2014 and 2015 will be held in Brisbane, so the challenge is there for us Queenslanders to make sure the next couple of years hold even bigger and better things for this excellent event.

  View of Mount Wellington from inside the conference venue.

Sunrise over the harbour, as seen from the Hobart hotel hosting PyCon Au.

[codemiller]

Functional frock: A dress fit for a Lambda Lady

Do I want to look geeky or girly today?

It's the question I face every morning when I peer into my wardrobe. I pick out a cute dress or a programming-themed T-shirt -- and in the latter case perhaps slip on my binary watch -- and then get on with the rest of my day.

The trouble is, sometimes I don't want to be forced to choose; I want to express both my geekiness and my femininity in the one outfit.

It was this desire that prompted me to start scouring the web for geeky dresses. However, to my chagrin, I didn't find much that appealed to me. I spotted a few phenomenal Cosplay outfits, but they weren't really my style, or the sort of thing I could wear to work without raising eyebrows.

Frustrated, I started searching for custom dress manufacturers. Again, the results were pretty disappointing. While dozens of vendors will let you create a custom T-shirt, websites offering dresses or skirts for which you can select the style and material are few and far between. I couldn't find any offering a choice of style and the option to supply your own fabric design.

I did, however, find several websites offering custom material at a reasonable price. Determined to get my hands on a work-appropriate girl geek dress, I started dreaming up plans to learn to sew. A few minutes later I came to my senses and decided the easiest way to get what I wanted this decade was to find a dressmaker on Etsy who would happily use custom fabric.

My time spent swing dancing has given me a deep appreciation for all things vintage (1940s/1950s), so I searched for someone creating vintage-inspired couture and found the woman who would finally make my girl geek dress dream a reality, Tracy McElfresh of the aptly named Custom Dress Shop.

I designed my material print in Inkscape and uploaded the SVG to Spoonflower; it features six spirals of Haskell code expressing sorting algorithms (quick, merge, selection, bubble, insertion and cocktail) encircling a lambda, on a purple background, of course (I may be a little obsessed with purple, as anyone who has seen my hair recently will know).

I figured out the best way to size and print the design (a half-drop makes a layout that is not perfectly symmetrical look much better!) and had the fabric shipped to my wonderful dressmaker, Tracy. Within a few weeks she had created my functional frock, as well as a skirt made from the same material.

I'm not sure how unusual I am to have wanted this dress -- software engineering women are a minority and of course everyone has different preferences on what they like to wear -- but there may well be a market for this sort of custom girl-geekwear. At the very least, I think there would be demand for a website offering completely custom dresses (not necessarily geeky ones -- any custom print) at a reasonable price. It wasn't cheap to produce my dress, so if there are any entrepreneurs out there who would like to streamline the manufacturing process, I'd be delighted to be a product tester.

Here are some pictures of my dress and skirt from when they were being made, as well as the final results. I'm sure they're not to everyone's taste but I'm pretty happy with how they turned out. I don't think having this new girl-geekwear will really make my morning wardrobe decision-making all that much easier (taking a long time to get ready is my prerogative as a woman, right :) ?), but it's nice to have them there. Now I have an option for those days when I want to show the world I'm proud to be both a vintage-dress-loving woman and a computer geek. 

[codemiller]

Fixing flashing Fedora: Linux boot issue solution

I installed a bunch of development tools and packages today on a Linux laptop running Fedora 18 and when I restarted the machine I was only able to boot into single-user mode.

It's not the first time that has happened to me, but the fix for this particular issue came from asking one of my kernel hacker colleagues for a bit of direction, so I thought I should document it for anyone else who might get stuck.

Symptoms:

When the machine booted, it got to a point and then the display started flashing on and off. It would show the start-up logging, then go to black, then about two seconds later come back again -- and repeat. The log output was no longer progressing.

Analysis:

I booted into single-user mode and had a look at the logs. I found that /var/log/messages included repeated blocks of output like this:

localhost systemd[1]: lightdm.service: main process exited, code=exited, status=1/FAILURE localhost systemd[1]: Unit lightdm.service entered failed state localhost systemd[1]: lightdm.service holdoff time over, scheduling restart. localhost systemd[1]: Stopping Light Display Manager... localhost systemd[1]: Starting Light Display Manager... localhost systemd[1]: Started Light Display Manager.

This was as much detail as was provided, so I went to /var/log/lightdm to have a look at the LightDM logs.

The lightdm.log file pointed me to a failure with Greeter, but didn't give any further details. The relevant Greeter log file finally gave me the error I was hunting for:

lightdm-gtk-greeter symbol lookup error /lib/libGL.so.1 undefined symbol _glapi_tls_Dispatch

A Google search gave me no results for this error. Before I attempted to delve any deeper myself, I asked a kernel hacker colleague, and he knew off the top of his head the packages involved and what the problem was likely to be. He advised me to check that all my Mesa packages were the same version:

rpm -qa | grep mesa

I'm not sure how or why they got that way, but the mesa-* packages were indeed from two different versions.

Solution:

I updated all the Mesa packages so the versions aligned (I had to start NetworkManager first as by default there is no network access in single-user mode):

yum update mesa* -y

Problem solved. I hope this helps someone else who perhaps isn't fortunate enough to work alongside someone like my knowledgeable colleague.

[codemiller]

The Journalist Who Became a Software Engineer

May I present a play that is performed frequently in the story of my life.

Narrator: Person X knew Katie was a software engineer but they have just discovered that she used to work as...(with shock) a journalist.

Person X (surprised): How on earth did you go from journalism to IT?

Katie (pauses, to decide which variation of the answer to go with this time): Well...

(EDIT: Script changed to reflect recent developments.)

Narrator: InformIT offered Katie the chance to tell her story in an article on its site, which has just been published.

Katie (with relief): ...you can read all about it at InformIT.com.

End of Story

The information technology industry offers a broad range of job opportunities, but if you've started your career in another arena, how do you go about making the switch? In this article, former journalist Katie Miller shares her experience of transitioning into the tech world and offers advice for others aspiring to follow suit.

[codemiller]

I have gained a lot of support and encouragement from various tech groups for women, especially when I was very new to the industry. However, when I became a part of the functional programming community, I was left wondering where all the women were.

That is no longer the case. About a month ago, myself and four other female FP enthusiasts formed Lambda Ladies and I am delighted to say more than 100 women have already become members of our group.

In the linked blog post I explain how the group came about.

If you identify as a woman and are interested in FP, we would love to have you join our community.

[codemiller]

Sir Tim promotes open data, open doors for girls

Although I am no longer employed as a journalist, last week I had good reason to bring that skill set back to the fore.

It was an honour and a privilege to have the opportunity to meet and ask a few questions of the man behind the World Wide Web, Sir Tim Berners-Lee.

My article based on the press conference I attended and Sir Tim's lecture on the Gold Coast has been posted on GC Tech Wire.

Here is a snippet:

One of the technology industry’s most celebrated men has spoken about the importance of encouraging more Australian women to enter computing.

Sir Tim Berners-Lee, who invented the World Wide Web, visited the Gold Coast on Saturday as part of the TBL Down Under Tour, which was his first trip to Australia in more than a decade.

Speaking at a press conference ahead of a lecture at Griffith University, Sir Tim said the gender imbalance in the local information technology industry was reflected on the website thewebindex.org.

“If you bury down and look at all the details for Australia – which is in fact just a little bit below New Zealand, just to get you going – you find that one of the things to be improved is the percentage of IT experts in Australia who are women,” he said. “It’s not very good.”

Women make up less than 25 per cent of the Australian IT workforce, while less than 20 per cent of the nation’s computer science graduates are female. In the open source community, women are believed to constitute as little as 2 to 5 per cent.

Sir Tim said girls were just as bright or brighter than boys at mathematics and rather than just learning how to use a computer, they should have the chance to learn how to make it do something different.

“I think that it’s really important for girls everywhere to learn that computing is exciting,” he said.

“We need more people who know how to program. It’s important that we have as many girls doing that as boys.”

Read the rest of the article, including Sir Tim's suggested strategies for encouraging more women into technology and what he told his Gold Coast audience, at GC Tech Wire.

[codemiller]

Take 2: An improved tech workshop for teen girls

It's easy to list 'what to do differently next time' when you have no inkling that there will be one. The tough part is sticking to it when next time rolls around.

After writing a reflective blog post on my 'Open Source in the Cloud' EXITE Camp workshop for girls, I was asked to run the session again for another group.

Reading through the list of my own suggestions for improvement, I realised I had my work cut out for me; there had to be less content, fewer words, more pictures, better navigation, improved clarity and more setup done in advance.

The workshop website needed to be overhauled. There was nothing for it but to go through the points one by one, which resulted in this new site.

Perhaps the most difficult requirement was communicating more through pictures rather than words. As I learnt last time, many teens just won't read a block of text.

I am no graphic artist, but I do know how to Google. I discovered that you can find online instructions on how to do virtually anything in the open source Inkscape software and there is plenty of brilliant free artwork available at OpenClipArt.org.

I used Inkscape and OpenClipArt to create some admittedly cheesy cat cartoons to express the key ideas of each section. These became part of my new HTML5 slideshows, such as this one, which replaced the large chunks of text on the previous site.

For the new site I also added an abundance of screenshots, reduced the content from five sections to three and improved the navigation.

The effort was worth it, because it led to a much better workshop experience for the girls.

This time of about 25 students, nine teams finished within the hour (some girls worked in pairs) and almost every teen made it to the final section.

The girls passed the technical checkpoints but they were also creative and loved to customise, even when not required to. They came up with blogs such as 'Pop Mah Afro (Poppin' Afros Braaah)', 'Rainbow Polka Llama' and 'The Amazingest Blog in the Known Universe'.

There were still some unforeseen issues, such as students copying and pasting example terminal prompts as well as the accompanying commands and trying to scroll up in the terminal history and change the past (if only we could), but overall things ran more smoothly.

Another positive aspect of the session was that the girls genuinely engaged with the material and wanted a thorough understanding of it; many were asking myself and the other instructors for a deeper explanation of concepts such as 'source code' and 'open source'. Striking the optimal balance between minimising workshop content and providing comprehensive information is difficult!

One of the best changes made for this second workshop was the inclusion of a web survey, to be completed within the session (as we had Buckley's hope of the teens filling it out afterwards). This provided extremely valuable feedback; I would definitely recommend it to others running similar events.

Although the sample size was small, the results were encouraging. Here is an excerpt:

Had you heard of open source before the workshop? Yes: 18% / No: 82%

Had you heard of Linux before the workshop? Yes: 45% / No: 55%

Had you ever used Linux before the workshop? Yes: 0% / No: 100%

What did you like most about the workshop?

Learning about different everything in the terminal was interesting, as well as creating the blog.

It seemed as though I was very knowledgable, even though I didn't really know what I was doing.

I liked the fact that I was able to learn something new in a fun and entertaining way.

We got to use terminal, which I think is really fun??? idk I think it's mostly just the feeling of being a COMPUTER HACKER hahahahh

Would you be interested in attending a programming workshop outside school time? Yes: 91% / No: 9%

This last response represents the real challenge. Perhaps it is time for a Girls' Programming Network or similar group to be formed in Queensland.

[codemiller]

Teens don't read: Lessons learnt from a tech workshop for girls

You might not think it watching Bieber and his ilk cruising around, but entertaining 14-year-old girls is really hard work.

Teaching them a thing or two along the way is even more challenging.

Last week I had the privilege of running a technology workshop for about two dozen teenage girls at an IBM EXITE (Exploring Interests in Technology and Engineering) Camp, and I suppose as was to be expected, not everything went smoothly.

I believe exposing girls to technology at a young age is key to tackling the gender imbalance in the IT industry, so I have made a habit of grasping these opportunities when they come along.

Last year I ran an Android app workshop at a local EXITE Camp. This year I decided to focus on cloud computing and open source using OpenShift, which seemed appropriate given I am working at the world's first billion-dollar open source software company, Red Hat (opinions here my own, not my employer's...you know the drill).

I learnt through last year's workshop that having instructions on a series of printouts means instructors spend a lot of time handing out pieces of paper, so this year I created a blog site to automate the process and host everything the students would need.

I customised a WordPress plugin or two to ensure the girls could only reach each new set of instructions after they had completed a quiz for the previous set.

You might think having that automation would leave a workshop instructor with plenty of time to twiddle their toes, but alas it wasn't the case -- myself and the three brave Red Hat volunteers I brought along were kept busy the entire hour.

Although there were a few hiccups, the girls seemed to have fun regardless. Some of the more entertaining comments I overheard from them were: "Haha my database is called 'unicorn'!", "Python...isn't that a snake?" and upon seeing the WordPress Administration screen, "It looks just like Facebook!"

In true Agile spirit, here is my retrospective on the experience. I hope it helps others embarking on similar endeavours.

What went well

The students made it through a fair bit of content, which hopefully meant they learnt a few things. Most made it to at least part three of five of the workshop. Every girl at least made some progress and gave some of the quizzes a go.

The girls had the opportunity to experiment with technologies they hadn't seen before. In one instance, I watched a couple of girls trying out a terminal for the first time, entering 'LOL' and 'OMG' as commands and giggling at the results.

A lot of students seemed to really enjoy personalising their blog sites. They spent a lot of time entering their bios and changing the look of their sites (to the detriment of actually making progress through the workshop, but never mind).

I put the workshop URL and each student's login credentials under the keyboard of each lab machine before the girls entered the room. There was a lot of enthusiasm when I kicked things off by telling them this and they retrieved the planted printouts. A small surprise goes a long way.

The students were exposed to a lot of technologies that were new for them, including Linux. At the start of the workshop, none of them said they had used Linux before and only one said she had "heard of it". By the end, some were saying they liked the interface and could imagine using Linux as their OS.

Having a prize for the first person to finish the workshop was a great motivator. It seemed to encourage the girls to make progress, although for some, when they realised they were too far behind to be in the running, they gave up and slacked off.

Having all the content online meant the girls had something to take away with them. They could continue to work on their WordPress sites if they wished, or revisit the workshop instructions and any links they found interesting. I have seen some activity on the instructions website since the workshop, which is encouraging.

The free caps and T-shirts we gave out to the participants at the end of the workshop received a great reaction. I didn't think caps would excite 14-year-old girls all that much but they immediately starting trying them on every which way and taking photos of each other.

Having lots of helpers was fantastic; many hands really do make light work. You have to expect there will be a stream of unexpected questions. Collaborating with others on the content beforehand and doing dry runs was also helpful.

What didn't go so well

Many students became confused simply because they hadn't read the instructions fully. They would start to try to do something and become frustrated when it didn't work, instead of going back to the instructions where the solution was outlined. There was too much text for them to bother reading it all.

The workshop required the girls to have multiple windows or tabs open and quite a few students struggled with this and had trouble switching between them. They also struggled with navigating between different running programs on what was a new operating system for them, Ubuntu; they didn't find this intuitive and unfortunately, the workshop content didn't cover it. Some ended up with windows in different workspaces.

Some students became confused between the WordPress site hosting the workshop instructions and the WordPress site they were creating and tried to use the login credentials from one with the other.

Students would follow the instructions of whatever was on screen, even if these conflicted with the official workshop instructions they were following. For example, some logged into OpenShift and saw it mentioned that an SSH key was required to make code changes, so they tried to put something in that input box, even though the workshop instructions did not say to do this.

No student finished the whole workshop in the hour allocated; the prize winner made it to somewhere around the end of part four of five. This didn't surprise me, although those who had run through the workshop for me ahead of the day had all managed to get through it in an hour. However, none were teenage girls and in general they were technically literate people, who also didn't have the distractions that come with doing something in a group setting.

Some students didn't have a clear picture of what the workshop as a whole was about. One comment I overheard was: "We made one of these in class. Is this a blog?" It wasn't obvious to everyone what the workshop was trying to achieve.

What to do differently next time

I would cut down the text of any workshop as much as possible. Even if all the required instructions and information was there, some students found it too tedious to read. I would add more diagrams, screenshots and other pictures to try to communicate the information.

I would make the navigation clearer. For example, instead of having a link to click on to start the workshop, I would have a big icon as a button. I would make the navigation more obvious so it was clear how to move backwards and forwards through the instructions.

I would include much less content. This would be both to make it easier to finish the workshop in the allotted time and to make the learning more focused. This workshop included information about a lot of different things; it was too much for the time.

If possible I would try to have the email account/OpenShift registration side of things done in advance. It was a positive that the students registered with their own email addresses, as this is more practical if they were to use OpenShift going forward, however, it just took too long and ate into precious workshop time.

I would include an example answer whenever a choice was required. I found instructions such as 'change the title of your blog' and 'choose a name for your application' slowed many girls down as they simply found it hard to make something up.

I would add clear instructions on what not to do on the external sites. For example, I would add a warning saying not to worry about the SSH Key section on OpenShift and not to worry about filling out a WordPress user profile in full.

I would include tips (or even better, pictures) to help the students navigate a foreign operating system.

I would improve the section at the start of the workshop explaining what it's trying to achieve. I would also talk through this section at the front of the group.

[codemiller]

Using a USB to DVI Adapter with Fedora

I bought a Digitech USB2.0 to DVI Adapter from an electronics store and it only came with an out-of-date Windows driver. No surprises there.

My aim was to set up a dual-monitor display for use with my work laptop -- a Lenovo ThinkPad with Intel integrated graphics chipset running Fedora 16 -- but excluding the laptop's display.

Additionally, I was hoping to use the adapter with my Samsung Series 9 notebook, which also runs Fedora 16.

Although it wasn't labelled as such, the Digitech adapter I bought fortunately turned out to be compatible with the Linux driver for DisplayLink devices and I found an excellent blog post describing the necessary configuration.

As my experience was somewhat different, I have recorded the steps I went through to set up my monitors on Fedora Linux, along with some bits and pieces learnt along the way. Hopefully it saves someone else some time.

I used the USB to DVI adapter to connect my laptop to an AOC 210V LCD monitor with an optimal resolution of 1680x1050, positioning it as the left external monitor. I used a VGA cable to connect a Samsung SyncMaster 931c with native resolution 1280x1024, positioning it on the right.

The udlfb driver that handles DisplayLink devices is now part of the mainline Linux kernel, so I didn't need to recompile the kernel to include it. Once I plugged in the USB adapter the AOC monitor screen went bright green, indicating it was ready for use. I found I had to plug in the USB adapter once the ThinkPad had booted or the display would not be detected and the monitor would not go green.

I ran the command dmesg in a terminal and in the output saw the message:

udlfb: DisplayLink USB device /dev/fb1 attached. 1680x1050 resolution

This confirmed the device had been detected and was mapped as /dev/fb1.

The next task was to obtain the DisplayLink X.org driver. I cloned the xf-video-udlfb Git repository and changed to the new directory with the commands:

git clone http://git.plugable.com/webdav/xf-video-udlfb cd xf-video-udlfb

Continuing to follow the original blog's instructions, I then ran ./configure. I received the error:

checking for XORG... configure: error: Package requirements (xorg-server >= 1.0.99.901 xproto fontsproto ) were not met: No package 'xorg-server' found Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables XORG_CFLAGS and XORG_LIBS to avoid the need to call pkg-config.

I figured out the package I needed was xorg-x11-server-devel, so I ran sudo yum install xorg-x11-server-devel to install it.

I ran ./configure again and this time it was successful. I then ran sudo make && sudo make install to install. To link the library, I ran the command

sudo ln -s /usr/local/lib/xorg/modules/drivers/displaylink_drv.so /usr/lib64/xorg/modules/drivers/

Alternatively, if on i386 architecture (uname -a will show you this), run

sudo ln -s /usr/local/lib/xorg/modules/drivers/displaylink_drv.so /usr/lib/xorg/modules/drivers/

The final, vital step was to create a suitable xorg.conf file. To build a base file to work from I ran Xorg :1 -configure as root, as suggested here. This created the file /root/xorg.conf.new.

I moved this file to /etc/X11/ and started to try different configuration options. To start an X session with the configuration, I used the command startx -- :1 -config xorg.conf.new (Ctrl+Alt+Fn switches between sessions). Once I was happy with the result, I created an alias to change to the X11 directory and launch an X session with this configuration.

Below is the Xorg.conf I ended up with for my dual-monitor setup, plus another file for use with my Samsung notebook when connected to a single external monitor. Some useful commands and lessons I learnt while drafting these files were:

The monitor connected with the USB adapter must be set as Screen 0 in the ServerLayout section. It is also best if it is positioned on the left. I found I could number the USB-connected monitor as Screen 0 and position it on the right, and the configuration would load, but the Gnome panel on the left screen would stretch a quarter of the way on to the right screen and most of the bottom panel was also off the screen. It looked as if the positioning was calculated based on the assumption the entire Xinerama screen had the resolution of the larger monitor only. I couldn't find any tools that could fix this.

The order in which the Screens are listed in the ServerLayout section is significant, even if the optional screen-num field is included. I had to list the USB-connected screen first, even if screen-num was included and it was num 0. If it wasn't listed first, even if the screen-nums were correct, only the VGA-connected screen would load. If it was listed first but set as screen-num one, the screens would go into clone mode.

The Intel format for configuring outputs in the Device section is 'monitor-<type>', ie: 'monitor-LVDS1' (see the Intel man page for more details). This was important to enable the laptop display to be disabled (the configuration still worked if I didn't disable the laptop display, but I wanted it disabled).

It may be worth adding the DisplaySize property to your Monitor listings, as described here. The command xdpyinfo | grep -B2 resolution will display the DPI. I found when the detection was automatic, my dual-display was at 90x88 DPI. To make this square (90x90 DPI), I calculated the physical screen width and height. For one monitor, these figures were listed in the manual's 'Resolution' section under 'Horizontal scan size(maximum)' and 'Vertical scan size(maximum)'.

Bit depth matters. Changing the bit depth options under Screen could mean my session would not load. I found having depths of 16 and 24 available worked, with 16 the default.

The monitor connected with the USB adapter did not seem to be able to handle anything other than its optimal resolution. When I tried other settings, the display was garbled.

I tried playing with modelines, but ultimately they did not enhance anything. As long as the EDID was being read, there wasn't much point. Suitable modelines can be created with the command gtf [resolution width] [resolution height] [refresh rate].

The XRandR tool was only useful for checking the resolution. I wasn't able to use it to change resolutions. The command printed the error "xrandr: Failed to get size of gamma for output DL" and would give output like "Screen 0: minimum 1680 x 1050, current 3046 x 1050, maximum 1680 x 1050", where the current resolution was greater than the maximum. The logs mentioned that xrandr-DL was "unable to get EDID for ouput", although the EDID of the USB-connected monitor was successfully read by other processes.

I was already using Gnome Fallback mode, but when I tried launching the configuration from Gnome 3, I found the new session went to Fallback mode anyway.

Here is the Xorg.conf for a dual-monitor external display, with a disabled ThinkPad display. Some irrelevant sections have been abridged.

Section "ServerLayout"   Identifier "X.org Configured"   Screen 0 "LeftScreen" 0 0   Screen 1 "RightScreen" RightOf "LeftScreen"   InputDevice "Mouse0" "CorePointer"   InputDevice "Keyboard0" "CoreKeyboard"   Option "Xinerama" "true"   Option "clone" "false" EndSection Section "Extensions"   Option "Composite" "Enable" EndSection Section "Files"   ... EndSection Section "Module"   ... EndSection Section "InputDevice"   Identifier "Keyboard0"   ... EndSection Section "InputDevice"   Identifier "Mouse0"   ... EndSection Section "Device"   Identifier "Card0"   Driver "displaylink"   Option "fbdev" "/dev/fb1" EndSection Section "Device"   Identifier "Card1"   Driver "intel"   BusID "PCI:0:2:0"   Option "monitor-VGA1" "RightMonitor"   Option "monitor-LVDS1" "LaptopMonitor" EndSection Section "Monitor"   Identifier "LaptopMonitor"   VendorName "Lenovo"   Option "Ignore" "true" EndSection Section "Monitor"   Identifier "LeftMonitor"   VendorName "AOC"   ModelName "210V"   DisplaySize 473 296 EndSection Section "Monitor"   Identifier "RightMonitor"   VendorName "Samsung"   ModelName "SyncMaster 931c"   DisplaySize 377 302 EndSection Section "Screen"   Identifier "LeftScreen"   Device "Card0"   Monitor "LeftMonitor"   DefaultDepth 16   SubSection "Display"     Viewport 0 0     Depth 16     Modes "1680x1050"   EndSubSection   SubSection "Display"     Viewport 0 0     Depth 24     Modes "1680x1050"   EndSubSection EndSection Section "Screen"   Identifier "RightScreen"   Device "Card1"   Monitor "RightMonitor"   DefaultDepth 16   SubSection "Display"     Viewport 0 0     Depth 16     Modes "1280x1024"   EndSubSection   SubSection "Display"     Viewport 0 0     Depth 24     Modes "1280x1024"   EndSubSection EndSection

Here is the Xorg.conf to connect a single external monitor to a Samsung Series 9. Again, some irrelevant parts have been condensed to make it easier to read.

Section "ServerLayout"   Identifier "X.org Configured"   Screen 0 "LeftScreen" 0 0   Screen 1 "RightScreen" RightOf "LeftScreen"   InputDevice "Mouse0" "CorePointer"   InputDevice "Keyboard0" "CoreKeyboard"   Option "Xinerama" "true"   Option "clone" "false" EndSection Section "Extensions"   Option "Composite" "Enable" EndSection Section "Files"   ... EndSection Section "Module"   ... EndSection Section "InputDevice"   Identifier "Keyboard0"   ... EndSection Section "InputDevice"   Identifier "Mouse0"   ... EndSection Section "Device"   Identifier "Card0"   Driver "displaylink"   Option "fbdev" "/dev/fb1" EndSection Section "Device"   Identifier "Card1"   Driver "intel"   BusID "PCI:0:2:0"   Option "monitor-LVDS1" "RightMonitor" EndSection Section "Monitor"   Identifier "LeftMonitor"   VendorName "AOC"   ModelName "210V"   DisplaySize 473 296 EndSection Section "Monitor"   Identifier "RightMonitor"   VendorName "Samsung" EndSection Section "Screen"   Identifier "LeftScreen"   Device "Card0"   Monitor "LeftMonitor"   DefaultDepth 16   SubSection "Display"     Viewport 0 0     Depth 16     Modes "1680x1050"   EndSubSection   SubSection "Display"     Viewport 0 0     Depth 24     Modes "1680x1050"   EndSubSection EndSection Section "Screen"   Identifier "RightScreen"   Device "Card1"   Monitor "RightMonitor"   DefaultDepth 16   SubSection "Display"     Viewport 0 0     Depth 16     Modes "1366x768"   EndSubSection   SubSection "Display"     Viewport 0 0     Depth 24     Modes "1366x768"   EndSubSection EndSection

[codemiller]

Installing Linux on a Samsung Series 9

When I bought a Samsung Series 9 (900X3A-A03AU) notebook recently I was intending to install Linux on it, but I wasn't sure which distribution to go for.

I started out with Linux Mint 12 (Gnome + MATE) and while the installation was easy and I didn't have any of the kernel panic issues I'd seen mentioned on various forums, not everything went smoothly.

WiFi did not seem to work on the MATE desktop, even after I'd installed the recommended Broadcom driver, but I eventually managed to get it functioning on Gnome Classic.

I built an optimised kernel based on one blog and tweaked the touchpad settings and key mappings in keeping with some other posts, but I still couldn't get the core touchpad gestures working properly and only half my function keys were responding correctly.

After much fiddling, I managed to get the function key meant to turn on/up the backlit keyboard to respond, but the key to turn it off/down was still inactive.

This wasn't a very useful state of affairs as my focus was to avoid draining my battery with a constant backlight, so I decided to give another Linux distribution a try. Fedora was the distro the Samsung Series 9 key mapping fix posts I'd found were written about, so I gave it a go.

It took less time to get everything up and running with Fedora and the key mapping hacks I'd tried previously now worked a treat, finally giving me control of all my function keys.

As described below, several things that I had to battle with on Linux Mint just worked out of the box with Fedora.

In the few days since, I have noticed it is slower to load and shut down Fedora than it was with Mint and there do seem to be a couple of little bugs to iron out, but overall I’ve been happy with the OS so far.

Here are the steps I went through to install Fedora 16 on my Samsung Series 9, along with some of my observations:

Backed up my system and data using Windows 7's Backup and Restore tool and the Samsung Recovery Solution tool.

Downloaded the Fedora 16 Desktop 64-bit ISO.

Used Universal USB Installer and a spare USB stick to create a Fedora USB installer.

Pressed F2 to get to the BIOS settings and selected Boot > Boot Device Priority. Moved 'USB HDD' to the top, then selected 'Save Configuration and Exit'. For the record, it seems F1 is the shortcut for the Windows Boot Manager and F11 gets you to Edit Boot Options on a Samsung Series 9.

Put the USB stick in the USB port on the right-hand side of the notebook, which can handle bootable media, and booted into Fedora.

Clicked Activities > Install to Hard Drive to kick off the installation and then followed the prompts. I had already reduced the size of my Windows partition and created plenty of free space for Linux with GParted during my Mint installation. During the Fedora installation, I still chose the custom install option to ensure my partitions were the way I wanted (there’s some advice on partitioning for Fedora here).

When I logged in to Fedora, I found the screen brightness, touchpad and volume function keys already worked. The other function keys did not seem to work. WiFi was working out of the box. Regarding the touchpad, the right-hand side was set as a scrollbar and I had to actually press down the touchpad to click, rather than just tapping. No other touchpad gestures seemed to work.

As I am not a fan of Gnome 3, I switched to a Gnome Classic look by installing dconf-editor and using gnome-fallback mode as per the instructions here. 

Ran 'sudo yum update' in a terminal. Found (by typing uname -r) that the kernel had been upgraded to a 3.3 version. 

Went to Applications > System Tools > System Settings > Mouse and Touchpad and selected 'Enable mouse clicks with touchpad', 'Disable touchpad while typing' and 'Two-finger scrolling'. These gestures all then worked.

Opened the '/etc/default/grub' file, added 'elevator=deadline noatime discard i915.i915_enable_rc6=1 i915.lvds_downclock=1' to grub_cmdline and changed grub_default to 0. Then ran 'grub2-mkconfig > /boot/grub2/grub.cfg', as per instructions here (although some of that blog post's suggested options I didn't think were necessary so I have left them out).

Went back into touchpad settings and unticked 'Disable touchpad after typing', instead using a custom script to set the time to wait before enabling the touchpad after typing, as described here. Found 0.2 seconds was about right for me.

Found Bluetooth was always on at start up, so disabled the Bluetooth Manager through Startup Programs (type gnome-session-properties in a terminal) and ran the command 'chkconfig bluetooth off' to stop it starting automatically at boot time, based on advice here.

Set up Git and cloned the 'samsung-laptop' Git repo, continuing to follow instructions from this blog.

Installed 'kernel-devel' and other packages for building a kernel listed here. Ran 'make' to compile the samsung-laptop module and then ran 'insmod drivers/platform/x86/samsung-laptop.ko' to load it. Copied the .ko file to '/lib/modules/$(uname -r)/kernel/drivers/platform/x86/' and ran 'depmod -a' (all with sudo).

Created extra keymaps and scripts as per the instructions here and found all function keys now worked. Note that this blog's custom shortcut commands are missing the 'bin' directory from some filepaths; make sure you use 'sudo /home/username/bin/samctl.sh' if that is where your script is.

Added notifications to each action in the samctl.sh script, for example:

wlan_state_value="$(cat $wlan_state)" if [[ $wlan_state_value -eq 0 ]];   then echo "1" > $wlan_state   su username -c "notify-send --hint=int:transient:1 -t 250 'WiFi:' ON"

The hint option means each notification does not have to be manually dismissed, which would be a pain.

Found there is a bug in the notification-daemon that does not accept boolean values like this to be sent as ints, even though boolean isn't a permitted hint type in notify-send. This causes the daemon to crash when you run notify-send with this hint option.

Could have just taken out the '--hint=int:transient:1' but was not content with manually dismissing function key alerts. Found someone had written a Debian patch for the bug but couldn't find a patched RPM file, so downloaded the latest source, applied the patch to the C code, rebuilt the RPM (using this as a guide) and installed the new RPM (plus emailed the authors so the patch will hopefully end up in the next version). This fixed the alert problem but I found the expire-time flag (-t 250) did not seem to have any impact, which I decided was a problem for another day.

Found the battery duration prediction was 4h10m at full charge. Installed powertop and tuned.

Discovered the keyboard shortcut to open a new terminal was not working. Installed libgnome based on this advice and found it worked after a restart.

Had a look at the kernel config (in '/boot/config-[kernelVersion]' file) and decided against building a custom kernel like I did for Mint. For those who might like to do so, found a wiki on this here.

That's all I've had the chance to do so far. Hopefully someone finds it helpful. Happy hacking!

[codemiller]

Next transaction: 10 influences developing a dev

I've heard it said (and tweeted) that if you don't hate code you wrote six months ago, then you haven't learnt enough in the past six months.

It didn't even take one month for me to start despising the 'fiscal challenge' code I blogged last year -- I admitted it didn't seem right to me in the original post -- but at the time I wasn't sure how best to refactor it.

Recently I had reason to reattempt the same cash dispenser programming exercise, and I soon realised the best approach was not to refactor my old code at all but rather to start from scratch, taking into account all I had been learning.

The finished product was vastly different from my first effort. While I would still not argue that it is perfect code by any means, I believe the second version has a much cleaner and more maintainable structure with better separation of concerns. Here is a snippet of the new code, representing the same portion of the exercise as what I first posted.

As I mentioned the Haskell programming language in the original post I thought I should try my hand at a more functional solution as well. I came up with a different algorithm for the Haskell version (and was aiming for readability rather than brevity) but this code snippet addresses the same part of the cash dispenser problem.

Looking over the new code I was quite surprised at the degree of difference a few months of experience can make. As this is my first blog for the new year I thought it would be appropriate to spend some time reflecting on what influences or practices have been helping me to improve my programming. I came up with the following:

The right learning environment. If you are exposed to best practices daily, you are likely to start to incorporate them into your coding. 

Talented technical mentors. It's much easier to learn from experienced engineers than to try to figure out everything on your own.

Using test-driven development. You have to be disciplined to work this way but it can produce leaner, more testable code and encourages a focus on requirements.

Pair programming. When you're new in a team and/or to professional coding it really helps to see how others do things and the standards they apply. I have also picked up some nifty keyboard shortcuts this way. 

Exploring functional programming. It makes you think about problems differently and helps you to write better code no matter what language you are using, as I have argued in a previous rant.

Learning new languages. Different programming languages have been designed to solve different problems. If you only know one you're bound to be missing something from your toolkit.

Reading informative books. If you're coding in Java, I can't recommend Effective Java enough. The Pragmatic Programmer is another worthy entry on the must-read list.

Studying code libraries. There's a lot of fantastic stuff both in standard language libraries and beyond them, but you're never going to use it if you don't know it's there.

Practice. It's all very well to code for work or study, but there are many more opportunities than that. Playing with Android has been fun and instructive for me. Coding competitions and open source projects are other avenues I'd like to further explore.

Attending coding meet-ups and conferences. It's a great way to get exposed to new technologies, new networks of people and new ways of doing things. Special mention goes to Code Retreat; you are forced to delete your code after each iteration so you can truly start afresh in the next one, which is a valuable lesson.

That's what came to mind for me, but I would be interested to hear what more experienced software engineers would add to the list. Please let me know if there is anything I've left out that has helped you to improve your coding over the years.

[codemiller]

Haskell will improve your coding, Simon says

THERE were many brilliant talks at the YOW! Australia Developer Conference in Brisbane this week but for me I think the highlight was Simon Peyton Jones' keynote.

The moment the Haskell inventor took to the stage barefoot and full of beans I knew we were in for a treat, but it wasn't just his entertaining speaking style that made the presentation memorable (for the record, I also found Damian Conway, John Hughes and Jim Webber to be consummate showmen).

What set apart Simon's talk for me was the timeliness and relevance of his message.

As the Microsoft researcher (pictured) said, he probably wouldn't have been invited to give a keynote on functional programming five years ago but something has been happening in the software industry that has been increasing the popularity of the paradigm.

It seems Haskell really has escaped from the Ivory Tower and is influencing today's programming styles and languages, and for good reason.

As Simon enthused: "Learning Haskell will make you a better Java programmer ... it will rewire your brain in a way that will make you write better Java programs. What I'm trying to do here is not so much to infect your hard drives but to infect your brains ... these ideas really are important."

He stressed that learning key ideas such as purity and types is more important than learning any particular language.

I totally agree and feel this message needs to be declared as I have found that not everyone appreciates the value of the functional programming approach.

It was only this year that I graduated from my IT studies, but I was unable to take a subject on functional and logic programming at my university as the course had been removed from the curriculum.

Thankfully, one passionate professor taught curious students about FP in the library in his own time. I am grateful for that opportunity and believe without that experience, I probably wouldn't have chosen to delve further into Haskell this year.

Perhaps my university's decision is not indicative of what is happening broadly (at least I certainly hope not!), but I have come across plenty of programmers who did learn functional programming at uni but consider it too difficult or esoteric to be worth putting into practice today. I hope they hear a talk like Simon's and reconsider their position.

I am still learning, but I have already experienced many benefits from exploring functional programming with Haskell. It helps me to look at programming problems differently and also offers many widely applicable techniques to improve my code.

When one of my fellow Java developers at work told me about an Optional type the team had implemented, I could immediately equate it with Maybe in Haskell and understand what he was going on about. I've found concepts such as immutability, list filtering and lazy evaluation to be extremely relevant, even if they don't always make for pretty code when expressed in Java.

If any coders out there still haven't jumped on the bandwagon and started to take functional programming seriously post-university, I encourage you to Learn You A Haskell (or some other FP-oriented language) For Great Good.

You don't have to do it alone. I joined my local functional programming group and have also found a small study group with which to go through the Learn You A Haskell book.

Hopefully when these ideas are well understood across the programming profession, together we can create the nirvana Simon described of a completely safe, useful programming language.

[codemiller]

Don’t lose precision with Hibernate3 Maven Plugin

If you’ve been using the Hibernate3 Maven Plugin to produce table definitions for Java types such as BigDecimal you may have noticed the resulting SQL does not include scale or precision.

In fact, if you specify scale and precision in your *.hbm.xml files or @Column annotations, it seems to be completely ignored.

For example, a definition such as:

@Column(name = "MY_BIG_DECIMAL", scale = 5, precision = 2) private BigDecimal myBigDecimal;

Produces a column definition of: MY_BIG_DECIMAL numeric

I found this to be the case when using what is currently the highest version of the plugin, 2.2, but also discovered an easy fix, which I’d like to share for the benefit of anyone else facing this problem.

If you have a look at the POM.xml for the plugin, you can see the version of Hibernate Core it uses is managed by its parent, which depends on an old version of Hibernate Core, 3.3.1.GA.

To fix this, all you need to do is give the plugin an upgrade by adding an explicit Hibernate Core dependency, specifying version 3.6.1.Final or higher.

For example:

<plugin>   <groupId>org.codehaus.mojo</groupId>   <artifactId>hibernate3-maven-plugin</artifactId>   <version>2.2</version>   <executions> ... </executions>   <configuration> ... </configuration>   <dependencies>     <dependency>       <groupId>org.hibernate</groupId>       <artifactId>hibernate-core</artifactId>       <version>3.6.1.Final</version>     </dependency>   </dependencies> </plugin>

I found doing this ensured scale and precision parameters were added to the column definitions of relevant types, including appropriate UserTypes.

A BigDecimal, for example, for which you do not specify scale and precision will now become MY_BIG_DECIMAL numeric(19,2). The aforementioned example would produce MY_BIG_DECIMAL numeric(5,2).

According to an open bug report found here, the column definition of BigDecimal particularly should actually use ‘decimal’ rather than ‘numeric’.

Updating the Hibernate Core dependency of the plugin does not change the resultant column type, but it does make things much more workable by allowing scale and precision to be defined.

[codemiller]

Back on the Grid: Selenium Grid2 and Maven

I survived three weeks without the internet. True story.

OK so that's not completely accurate...I managed to connect to free WiFi a few times...but the reason I haven't posted anything in a while is because I have been off indulging another passion -- cruising.

I spent 18 days sailing from Hawaii to French Polynesia, New Zealand and finally Sydney aboard Royal Caribbean's Radiance of the Seas.

For any keen cruisers, I thought the Radiance was a beautiful ship and the entertainment was superb, but surprisingly the service was markedly inferior to what I've experienced on other, less grandiose ships. I have a new appreciation for P&O!

Anyway, to return to regular programming, I was playing around with Selenium Grid during the week and realised there weren't any instructions for integrating it into a build with a tool such as Maven.

Below is an example of what you could put in your POM to get some WebDriver integration tests to run on a Selenium Grid(2).

Here are the general instructions for setting up Grid2 with WebDriver and some examples demonstrating the Maven AntRun Plugin I've used (yes, it's a bit hacky but it worked). The use of 'maven.compile.classpath' is a shortcut you can read about here.

I realise the 'tasks' tag is deprecated but I had no joy with 'target'. I also found 'fork="true"' did not work for me, even though this seems to be the standard syntax used in the docs, but realised 'fork="1"' did. Lastly, arguments supplied with the <arg value=""/> syntax seemed to be ignored, but <arg line=""/> worked for me.

You'll probably need to add a bunch of dependencies that would otherwise be included in the standalone JAR, such as: selenium-server, selenium-remote-driver, commons-logging, servlet-api, maven-antrun-plugin, jasper-compiler, jasper-runtime, jsp-api and possibly the maven-surefire-plugin if you want to run tests in parallel.

Here is a POM excerpt for setting up a Grid2 hub and a single grid node that can support five instances of Firefox.

<build>   <plugins>     <plugin>       <groupId>org.apache.maven.plugins</groupId>       <artifactId>maven-antrun-plugin</artifactId>       <executions>         <execution>           <phase>pre-integration-test</phase>           <configuration>             <tasks>               <java classname="org.openqa.grid.selenium.GridLauncher"                classpathref="maven.compile.classpath"                failonerror="true"                fork="1">                 <arg line="-role hub"/>               </java>               <java classname="org.openqa.grid.selenium.GridLauncher"                classpathref="maven.compile.classpath"                failonerror="true"                fork="1">                 <arg line="-role webdriver -hub http://localhost:4444/grid/register -port 5556 -browser 'browserName=firefox,maxInstances=5'"/>               </java>             </tasks>           </configuration>           <goals>             <goal>run</goal>           </goals>         </execution>       </executions>     </plugin>   </plugins> </build>

To try running test classes in parallel, add something like: 

<plugins> <plugin>   <groupId>org.apache.maven.plugins</groupId>     <artifactId>maven-surefire-plugin</artifactId>     <configuration>     <parallel>classes</parallel>     </configuration>   </plugin> </plugins>

The Grid2 instructions explain how to set up a RemoteWebDriver to use the grid. Happy testing!

[codemiller]

Calling Pythonista sisters: Five reasons to attend a coding conference

If you have just started building a machine that goes BING, pledged to steer clear of Python’s dark corners and seen code combat end in a dance off, chances are you were at PyCon Au 2011.

The two-day Sydney conference – which was preceded by a CodeWar evening and followed by two days of coding sprints – attracted about 300 delegates this year. This included 35 women, which was a healthy increase on the 10 female delegates among about 200 attendees at the inaugural event last year.

I was privileged to be one of this year’s delegates, thanks to one of several Gender Diversity Grants generously funded by Google. In my six months or so in the IT industry I’ve attended many coding group meet-ups, bar camps and developer days, but PyCon Au was my first real geek conference and I was stoked to be there.

However, after speaking to some of the other women at the event, I realised a few had been quite nervous about attending and at least one very nearly did not come because of this.

So, in the interests of encouraging more women and other reticent delegates to attend future conferences, here are some of the benefits I found in attending PyCon Au:

1. Gaining knowledge

Okay so this is obvious, but there is no question it was a fantastic opportunity to learn more about the Python programming language and related tools and frameworks.

The PyCon sessions ranged from beginner-level workshops, such as Python 101+, through to advanced lectures pitched at experienced engineers and scientists, so there was something for everyone.

I particularly appreciated Peter Lovett’s workshops, Katie Bell, Georgina Wilcox and Tim Dawborn’s talks on the National Computer Science School and Girls’ Programming Network, Dr Graeme Cross’s presentation on Python-controlled machines and Mary Gardiner’s keynote.

As an industry newcomer, I have noticed the more of these tech talks I attend the more conscious I become of just how much I have to learn, which can be daunting. If this fazes you, I think it’s worth keeping in mind where you started; when I stop and reflect on the fact that two years ago I was a journalist/sub editor with no post-school IT training or experience, I appreciate how far I’ve come rather than being overawed by how far I have to go.

2. Encountering challenges

Gaining knowledge at a conference has an associated benefit in that learning about new or improved practices can challenge your thinking and motivate action.

When you’re caught up in the world of your own workplace or projects, it can be difficult to find the time to investigate new ways of doing things. There is a risk of continuing to be set in your ways while the industry has moved on.

A conference offers an avenue for exploring a variety of techniques and technologies that you might not be familiar with that may well challenge your status quo.

Your response may be to adopt something new, or it may simply prompt you to evaluate the benefits of your current approach and justify why an alternative method would not be beneficial.

Even as a newbie with few set ideas, I found PyCon challenged me to try out some different ways of using Python, as well as of course to improve my skills.

3. Networking

Another massive benefit of attending PyCon Au was that I met all sorts of interesting people from many parts of Australia and even the US.

There was a fantastic blend of coders with different ideas, perspectives and programming language backgrounds, which made for many rousing conversations and a few passionate debates.

There was also a whiteboard for job/hiring opportunities that become so full it had to be extended, which was an excellent tool for those looking for work or employees.

4. Finding community

There was a tremendous sense of community at PyCon Au as the common interest of Python brought a diverse group of people together.

Within this broader community, there were also many informal subgroups based around occupations, framework choices, locations, hobbies or any number of other characteristics.

As I have already mentioned, one of these subgroups was the female contingent. PyCon Au did a fantastic job of catering for women by providing diversity grants, imposing a Code of Conduct and holding a PyLadies breakfast that facilitated women meeting up before the conference had even begun.

While the percentage of women at PyCon Au was still far short of being representative of the female population generally, the numbers were strong enough that we were not an oddity and that was refreshing (as Brianna Laugher has eloquently blogged).

I am typically the only woman at most of the coding groups I attend and I have become used to guys looking at me curiously as if they think I might have wandered into the wrong room. I don't recall getting any of those looks at PyCon.

The female attendees were able to discuss some of the difficulties we encounter as coders and receive validation and encouragement from each other in discovering many of these experiences are shared.

A group of us were able to talk about some of the barriers to getting more women programming and share tips and techniques for overcoming these.

Whether you’re a woman, identify with some other subgroup or both, PyCon and other conferences like it offer the opportunity to connect with like-minded or accepting individuals with whom you already share an interest, which in my experience can be hugely beneficial.

5. Discovering ways to contribute

Last but certainly not least, I found a benefit of PyCon Au was that it gave attendees both the opportunity to contribute to the Python community and guidance on how best to do so.

Aside from the chance to conduct workshops or lectures, there was provision for delegates to share brief snippets of information through the lightning talk sessions.

Some conference presentations focused on specific opportunities for delegates to become involved, such as Mary Gardiner's keynote and Nick Coghlan's presentation.

In addition, after the conference there were two days of guided coding sprints that were open to programmers of all levels.

I was only able to attend the first of these days but enjoyed the collaborative spirit and the chance to give something back.

It was certainly useful to have experts on hand to ensure we were hacking away at the right thing.

So there you have it. That’s my take on the major benefits of attending PyCon. Feel free to leave a comment with anything I left out.

Hopefully we will see an even bigger crowd attend PyCon Au in Hobart next year!

[codemiller]

Androiducation: Teaching Android to schoolgirls

There is a stereotype that the young people of today are technology experts, seemingly born to text, tap and type on tiny touch screens.

According to the media hype, these tech-savvy teens are joined at the hip to their mobile phones and can do all sorts of weird and wonderful things with them.

Given this prevalent image, I was a little surprised when I asked a group of teenage girls attending an IBM camp if they had smartphones and received the reply “what’s a smartphone?”

I had been asked to run a workshop for about 30 schoolgirls as part of an Exploring Interests in Technology and Engineering (EXITE) event aiming to foster interest in technology careers.

I had decided to show the students how to create Android apps with App Inventor, which I thought would be right up their tech alley.

While the majority of these bright young women completed the workshop material in record time and did an excellent job, I was taken aback by some of the gaps in their knowledge.

While it was evident that these teens were adept consumers of technology, I found many didn’t know even the basic technical details of the products they were using.

For starters, a high percentage didn’t know what Android was.

I was asked to inspect a string of phones, including Nokias and iPhones,­ to tell their owners if they were running Android and could install the app they were making. Despite this discussion, I still saw one girl trying to download the Android app to her iPhone.

I was also surprised to discover many of the students didn’t know how to unzip a file, one of the few workshop instructions I didn’t give explicit steps for.

To my chagrin only one girl put up her hand when I asked if anyone had learnt to program, and when I asked what language she’d learnt I was told “scripts and things”. I was extremely glad I’d chosen to use App Inventor rather than trying to teach Java.

Despite these hiccups, once I filled in the gaps most of the girls did brilliantly in the workshop, a version of App Inventor’s HelloPurr tutorial. Some even filled in the gaps for each other; a student piped up and gave a definition of ‘smartphone’ to her peers after my initial question and the response.

I broke the HelloPurr activity up into four worksheets, with an extra sheet of bonus challenges, which seemed to work well. I’d say about 30 per cent of the group made it to the bonus challenges inside an hour, which was excellent to see.

One girl with an Android phone managed to download her version of the app to her mobile and went around the group excitedly showing the others.

I was delighted to see many of the girls going beyond the workshop instructions and experimenting with App Inventor, putting in their own pictures and sound files and adding new functionality to the basic app.

There was no need to worry about any bad behaviour from these students; most were fully engaged as they saw the creative power of technology unleashed.

Although as a programmer I was a bit disappointed (though not surprised) that there were no keen coders among them, I do hope the experience inspires the girls to explore further.

At least if they know what operating systems their phones are running they can hunt down details of what they can do with them. I did see one girl doing a Google search on how to create iPhone apps at the end of the workshop, which was encouraging.

I applaud IBM for running the EXITE camps and am delighted to see bright students coming through the school system. However, I don’t think these events alone are sufficient to address the apparent technical disconnect.

We know we need more women in technology. I hope more can be done to ensure young women -- and men -- are aware of the myriad career options technology offers and the true potential of those tiny touch screens they carry in their pockets.

[codemiller]

Android emulator internet access behind a corporate proxy

I have been experimenting with Android web and native app testing at work recently and have found operating behind a corporate proxy a bit of a pain.

I’m no networking expert but the questions I have seen on Stack Overflow and Google Code sites suggest I am not the only one to have experienced Android Virtual Device (AVD) proxy challenges, so I’m going to share the approach that worked for me.

How to set the proxy on an Android Virtual Device

When you launch an AVD you should find you are able to use its browser to access websites within your internal network without doing any extra configuration at all (make sure you use complete internal addresses, not just hostnames).

However, if you want to access external sites in the emulator, you will need to give the device some proxy details.

There are apparently three ways to do this: on the command line, in a particular settings database, or in the emulator itself. The only one of these methods that worked consistently for me in a corporate Windows environment was setting the proxy through the AVD itself.

To do this, load the emulator and go to Settings > Wireless & networks/Wireless controls > Mobile networks > Access Point Names > TelKila (or whatever APN is there; it differs between platforms) and you will then be able to enter your proxy, port, username and password.

The only platform I tried that this did not seem to work on was Android 2.1; it appeared to ignore the proxy settings altogether.

Unfortunately, you can’t specify proxy exclusions, so while this approach should get you external internet access in the emulator, you'll likely find your internal sites no longer load.

How to enable access to internal and external sites on an AVD

In order to be able to access both internal and external sites without any proxy setting changes in between, you can set up an intermediate proxy to sit between the emulator and your corporate proxy.

If I were able to use Linux I would have probably used TinyProxy but as I was limited to Windows, I chose Squid for this purpose.

I set my intermediate proxy to send external requests to its parent, the corporate proxy, but handle internal site requests directly.

After ensuring my intermediate proxy had localhost access enabled and was running, I was able to make requests through it in the Android emulator by setting the proxy to 10.0.2.2 (this is an alias used to refer to localhost (127.0.0.1) on your machine) and the port to my intermediate proxy’s port number.

Once this was set up, I could view both internal and external sites in the emulator browser.

Setting up Squid as an intermediate proxy on Windows 7

One software solution you could use for your intermediate proxy is Squid.

You could use Cygwin to run Squid on Windows, however, I chose the native port, Squid for Windows, instead.

There are clear installation instructions with each release. The only advice I would add is:

Make sure you run command line as an administrator during the install

It is a hassle to install Squid anywhere other than ‘C:\squid’, so try to avoid this if possible

Don’t forget to add 'C:\squid\sbin' to your PATH environment variable

Once you have Squid installed, you can launch it by typing 'squid net start' on the command line.

However, you’ll probably need to edit the squid.conf file to get it running and behaving as you'd like.

Every environment is different, but as a guide, here are the main changes to the squid.conf file I would suggest you try based on my experiences (you may not need to implement all of them).

Once you have Squid running, it is safest to make these changes one at a time so you can easily tell what the problem is if you get errors.

Set http_port to the port you want to run the proxy on (or leave as the default if desired)

In the http_access section, add the line 'http_access allow localhost'

Set up your parent proxy in the cache_peer section by adding a line such as:

cache_peer  myproxy.my.internal.network  parent  [PARENT PROXY PORT]  3130  default no-query login=[USERNAME]:[PASSWORD]

Add an always_direct command with appropriate rules for the internal sites you want to access. For example: 

acl local-servers url_regex ^http://.*my.internal.network.* always_direct allow local-servers

Add the line 'never_direct allow all' in the never_direct section

Add your local nameservers under dns_nameservers (type 'ipconfig /all' in command line to find out the IPs)

Comment out the acl localnet entries and any other mentions of localnet; you don’t need them

If you have any problems, check the logs in C:\squid\var\logs to see what is going on. 

If you find Squid 'terminates abnormally' every time you start it, check the cache.log file. If you see the line 'FATAL: ipcache_init: DNS name lookup tests failed', try adding the -D command line option when you launch Squid (ie. 'squid net start -D'), which is meant to disable these internal DNS tests.

If you still get this error, try setting dns_testnames in your squid.conf file to an internal site you know can be accessed locally. You have to have at least one site on the list.

For more help, try websites such as this Squid FAQ, this other Squid FAQ or this Squid configuration guide.

Good luck!