I recently installed Ubuntu on a desktop computer at work and do like it but as Ubuntu uses Gnome as its Window Manager and I'm a KDE fan (unsure whether this is due to my familiarity with the desktop environment or its extensive feature set), I was still seeking a version of Ubuntu that used KDE instead of Gnome. I was already familiar with Kubuntu but had heard from a few friends that it had bugs so I wasn't that keen to try it out. Then one of my company's suppliers told me about Mepis and that it uses the same repositories as Ubuntu and best of all it uses KDE!
I installed it on an Acer notebook and it installed with no problems. I thought this was quite impressive considering the recentness of the notebook. I am currently trying to get my Linksys WPC54G Version 3 going using ndiswrapper but haven't been successful yet.
My only complaints are that in my opinion it installs a little too much software by default. For example, it installs Kaquarium, Ksensors, and a whole lot of other utilities that I possibly won't ever use. I suppose, though, the maintainers of any Linux distribution have to balance ease of installation and the number of applications they will include by default (it was nice,though, to see Skype included in the default install). The base install was about 2 gigabytes, though, and Mepis installs this all to the root partition so ensure you make the root partition large enough.

In New Zealand, due to our older telecommunications infrastructure, many people are using ADSL for their broadband internet connections, which uses the older copper-based telephone lines for transmission. This means slower speeds and in my case problematic internet configurations.
Specifically, I was trying to setup a web server that could be accessed over the internet. Our network configuration at work is as follows: we have an ADSL router with 2 network interfaces, a WAN interface (which we'll call r-eth0) and a LAN interface (which we'll call r-eth1). We also have a m0n0wall firewall with 2 interfaces, which we'll call fw-eth0 and fw-eth1. r-eth0 has a public internet address (assigned by the ISP), both r-eth1 and fw-eth0 are on the same subnet, and fw-eth1 is connected to our internal company network (a separate subnet). The web server is on the same subnet as fw-eth1 (our internal company network), and NAT is enabled on the ADSL router and is port-forwarding web traffic (TCP Port 80) through to the IP address of the web server.
The problem was that when people would try and open a website on the web server, it would just sit there and eventually timeout. The m0n0wall firewall logs indicated that the web traffic was successfully getting to the web server but traffic from the web server back out to the internet was being blocked. This was when I worked out that the cause of the problem was the firewall. A big thanks to Kent Brown Lee of Connect Logic here as he spent a lot of time helping me resolve this problem and eventually determined that as the m0n0wall firewall has inbuilt NAT, the outbound connections from the web server were being 'double-NATed'. To stop this occuring we turned on 'Advanced Outbound NAT' on the m0n0wall firewall and now people can access the website over the internet.

Following on from my previous entry - the homebrew brewed out in about 4 days. My friend said that you can tell when it has finished brewing because the carbon dioxide stops bubbling out the fermentation lock. Another perhaps more accurate way to determine that the brew is ready for bottling is to use a hydrometer.

From here we performed the following steps:

1. Check all the bottles for mould by pearing inside the neck of the bottles. It is important to check not only the bottom of the bottle but also the sides as mould can ruin your beer.
2. Use a ratio of 1 teaspoon of sterilisation powder to 1 litre of water and pour the liquid via a funnel into each bottle. Then rinse the bottles out with water.
   
3. To initiate the second fermentation process, put 1 level teaspoon of sugar in each bottle. This process is where the beer gets its fizz.
4. Cap the bottles. I was recommended to use a bottle-capper but I do know of others who successfully use the screw tops that came with their beer bottles.
5. Store the bottles in a stable, warm enviroment, such as a hot water cupboard for 2-3 months. I initially stored mine for 2 months but found that when I opened them, a lot of the beer fizzed out the top and was wasted. If this happens, the beer needs to be left longer, perhaps for an additional 2 weeks, and then tested again until they don't fizz out any longer.
6. Now your beer is ready to enjoy!

N.B. You will need to get a beer glass in order to truly enjoy your homebrew beer and you will also need to learn how to pour your beer properly. This is because there is sediment that settles at the bottom of the bottle and which you don't want to drink (it tastes awful!).

As some of you may be aware, this recipe is licensed under a Creative Commons License, which means you can copy and use it any way you like, as long as you attribute me as the author. As an side, these generous folk have also opensourced their beer making process.

Although I'd rather be  developing in Plone, it really is just 'back to my roots' as I originally trained as a network/systems engineer. It is great, though, being able to apply the skills I've gained over the last 2-2.5 years (a big thanks to Paul at TheVirtual Ltd for his help here) to a role I am already familiar with. Whereas in the past I was only familiar with Microsoft tools and methodologies, now I am aware of a whole raft of opensource tools and approaches.

I can only speak from personal experience but I have found I am so much more efficient and productive with opensource tools. For example, I just implemented an instance of PloneCollectorNG to track IT support jobs. The result has been that staff are happier as they know their job is being cared for and they also receive notifications via email when their job is updated, the IT department is more organised as IT staff can be assigned jobs and all information related to the job can be stored inside PloneCollectorNG enabling IT personnel to be more autonomous. And I suppose the thing to top it all off is the software didn't cost Management a thing!