Yesterday I took delivery of a brand new shiny 3.06Ghz iMac. I’ve been hankering after a decent desktop for a while. MacBooks and MacBook Pros are great for when you’re on the move, but when you want to sit down and do some serious coding / photo touching / gaming, you need a decent desktop.

So, I ordered a custom build iMac, which is now proudly taking pride of place on my desk.

After initial setup through restoring from a Time Machine backup, the iMac worked a treat, but from there, things deteriorated… I installed the 10.5.5 update.

After the update WiFi started playing up. The signal strength would keep varying, and constantly drop and reconnect altogether. Applications relying on internet connectivity would constantly fail.

After scratching my head and tinkering with all of the WiFi related settings I could think of, I was ready to start installing a new wired connection into the office… a damned pity given that this is the only appliance in the house that has a WiFi issue.

Then, i noticed something odd… I was running a little application called iStumbler to get a better look at the WiFi signal strength and characteristics. Strangely enough, whilst this application was left running, my network WiFi issues disappeared!

So now, I have a work around, but am completely at a loss as to how to properly fix this… if anyone has any suggestions, they would be very much appreciated!

The content on this blog over the years has gradually morphed from a discussion of Java technology, software development and software engineering methodologies into a far more personal blog where I tend to have the occasional rant about something on my mind.

I’ve decided, partly inspired by a discussion with Jason of BookShelfApps, that I should split my blog into two… technical content one (this one), and all my random ranting and rambling in the other.

So… if you know me, and have a vague (or even passing) interest in my rantings/ramblings, you will now be able to find these over in the Atonewell blog.

I’ve imported all of the post from The jZone into this blog, and will shortly be pruning both blogs so that they only contain the relevant content.

I’m just in the process of clearing out various odds and sods, including a Blackberry Curve that has been made redundant by my shiny new iPhone 3G.

Before shipping it off once it has been eBayed, I wanted to make sure that it had been completely wiped and reset to factory defaults, but couldn’t work out how.

Google to the rescue.

Google pointed me in the direction of a post on Lockergnome by Greg Hughes that advised exactly how to do this.

Go through the following menus to “nuke” your Blackberry:

• Options menu
• Security Options
• General Settings
• Menu
• Wipe handheld
• Enter password (”Blackberry” or your business-assigned security password)

It takes a couple of minutes, but afterwards, you should have a completely wiped Blackberry.

I’ve got fed up with updating The jZone with new builds of WordPress. I’ve got better things to do with my time.

So, I’ve set up a new hosted blog with WordPress.com, and set up a ReWrite rule to direct traffic from the original site.

It may cause a few broken links, but will allow me (hopefully) to ignore this site, beyond writing the odd blog post or two of course.

I’d just like to thank my colleagues Bill and Tom for defining this new pattern:

Sarcophagus Pattern
Intent

Provide a unified interface to a preserved, unspeakable horror that cannot be purged for sentimental or compatibility reasons. Sarcophagus is similar to Facade, but carries the implication that the box should only be opened in the presence of police, necromancers, or others familiar with the dealing with dead things.

Sarcophagus defines a higher-level interface that prevents the contents of the tomb from eating the brains of living. This can be used to simplify a number of despicable interactions and nightmarish dependencies into a single interface.

See also
Facade Pattern, Mediator Pattern, Adapter Pattern

Origins
Inspired by Tom during discussions of [removed to protect the authors]. Defined by Bill.

From the “Online Etymology Dictionary”: sarcophagus
“stone coffin,” 1601, from L. sarcophagus, from Gk. sarkophagos “limestone used for coffins,” lit. “flesh-eating,” in reference to the supposed action of this type of limestone (quarried near Assos in Troas) in quickly decomposing the body, from sarx (gen. sarkos) “flesh” (see sarcasm) + phagein “to eat” (see -phagous). The stone sense was the earliest in Eng,; meaning “stone coffin, often with inscriptions or decorative carvings” is recorded from 1705. The L. word, shortened in V.L. to *sarcus, is the source of Fr. cercueil, Ger. Sarg “coffin,” Du. zerk “tombstone.”

My Drupal installation was playing up, and its too memory intensive on the host I run it on, so I thought I’d give WordPress another try.

So, I’m just in the process of converting the site, which has been made significantly easier using the post I found on vrypan.net.

Now all I’ve got to figure out is how to rebuild my previous style with the minimum of effort, and work out what to do with posts that contain Textile-like mark-up.

One thing I have immediately noticed is how much more responsive the site now appears to be.

Floyd Marinescu has just published a free short book on InfoQ that I think all software developers, java developers in particular, should read.

It is a short, succinct explanation of Domain-Driven Design, and takes us back to the basics of application design using a domain based approach.

Far too often, Java designers and developers spend too much time worrying about “how” to build an application, with far too much focus being placed on the technical aspects. Floyd’s book re-introduces the domain driven approach, focusing on “what” to build.

Java’s complexity (ironic for a language that was striving for simplicity) is the reason why so many Java developers have such poor focus. Using a simpler language (Ruby anyone?) allows the developer to focus on the “what” without having to worry about the “how”.

Unfortunately, Java’s complexity is here to stay. Which is why I think it is so important to stress to developers that they shouldn’t lose focus on the “what”.

I have blogged about Java and caching of DNS once before.

It looks like the peeps over on Java Lobby have run into this common problem too.

There are a couple of ways around this problem that dont involve restarting all of your Java network based applications.

There is a property that can be set via java.security.Security, “networkaddress.cache.ttl”, that sets the poll interval (in seconds) that Java will use before refreshing a cached DNS name.

To set this value, you just need to include the following snippet somewhere in your code:

bq. java.security.Security.setProperty(“networkaddress.cache.ttl” , “60″)

There is a second property, “networkaddress.cache.negative.ttl” which is pertinent too. This property sets the poll interval used for unsucessful DNS lookups.

You can set both of these properties to zero, which will disable DNS caching all together.

In addition to these settings, there are two internal system properties that can also be used, but depending on your VM, these may or may not work.

“sun.net.inetaddr.ttl” & “sun.net.inetaddr.negative.ttl” correspond directly to the two previously mentioned properties, but are System properties rather than Security properties, so they have the advantage of being configurable via the command line, or centralised with other application properties.

There is a useful page over on the Sun Java site that discusses all of the available Networking Properties. It is well worth bookmarking.

This is a great quote, that all project managers should heed:

Once I learned the very hard lessons of how to be a leader instead of a manager, I fully embraced the idea that a successful team was one where people of varying skills had the opportunity to fully demonstrate their talents. At the same time they could count on the different skills of others to help them be more successful than they could have been by themselves. That’s a mouthful, but it served me well.

I’m not moving… the site is. I’m just in the process of moving it to a new host. It’ll take 48 hours or so for the DNS changes to propogate.

On Wednesday this week, for no apparent reason, the battery on my three month old MacBook Pro decided to die. All of a sudden it refused to charge, reported that it had no charge, and threw all of its toys out of the pram.

I searched on the web, and found someone else had experienced the same problem.

So, first thing Thursday morning I gave Apple a call, and they agreed to send out a new battery pronto.

Yesterday, the battery arrived, and everything now seems to be back in order.

Well done Apple for such a prompt response.

Would the comment spammer at IP 195.225.176.160 please stop stealing my bandwidth.

Ah… nmap reports:

8000/tcp  open   http-proxy         nginx http proxy 0.3.34

Let me guess… open proxy…

My new MacBook arrived last Thursday, but I’ve been away over the weekend, and have only just managed to get online with it, so more on the MacBook later.

However, while I was away, I also took the opportunity to have a little play with iWeb, to see what its like. The results can be found here.

I don’t think I’ll be ditching Drupal just yet.

I’ve taken the plunge… rather than repairing my “poorly Powerbook”:http://www.jzone.co.uk/node/136, I’ve decided to keep it limping along until the launch of the new “MacBook Pro”:http://www.apple.com/macbookpro/. I’ve pre-ordered one of the new Macintels with 2Gb of memory, and a nice speedy 100Gb 7200rpm drive.

Various reports have sugegsted that the performance claims made about the new Macintels are a little over-hyped. I’m not too worried about the performance under “Rosetta”:http://developer.apple.com/documentation/MacOSX/Conceptual/universal_binary/universal_binary_exec_a/chapter_7_section_1.html. Sooner or later the apps I run will all be compiled as Universal Binaries (hopefully sooner rather than later). The most painful wait will be for Photoshop, which it has been suggested wont see the light of day before the end of the ‘06.

In many ways, it is against my better judgement buying a first generation product, and the first generation Macintels are bound not to be without problems of their own. As the new MacBook Pro will be my main pc, I will be somewhat dependent on it, so I really hope that Apple manage to pull the rabit out of the hat on this one. They are obviously going to be concerned over the anti-Intel crowd’s fears being realised by a release marred by problems.

The good news is that so far there has been very little bad news concerning the new Macintel iMacs, which are already in the stores. However, it is still early days, and I’m trawling the news feeds daily for any signs of woe.

The other concern over the MacBook Pro is that the models seen at MacWorld were hastily knocked up prototypes, and that the final product is not yet production ready, and that there may yet be delays in getting the product shipped. Hopefully I can keep the old one running if the dates slip too much.

I hate waiting.

At long last “Google”:http://www.google.com/ has released a version of “Google Earth”:http://earth.google.com/ for Mac OS X.

I’ve been playing with a version of the beta for a while now that I managed to find on a dodgy Spanish blog. It looks virtually indestinguishable from the Windows version.

It makes for beautiful eye candy as well as being a very useful application.

The version posted on the Google site is still beta, but its well worth a test drive. You can snag a copy of it “here”:http://earth.google.com/downloads.html.