Between 1987 and 2003, Fugazi played over 1000 concerts in all 50 states and all over the world. Over 800 of these shows were recorded by the band’s sound engineers.

You should already be well aware of this given that it reached the New York Times frontpage, but as this has been a long time coming (in various forms) – I figured I’d mention it here as well.  The NYT article points out another interesting aspect:

 [...] the project also tells a story about musical technology from the 1980s into the 2000s. The earliest recordings were made on cassettes, then came digital DAT tapes, then CD-R’s and a few hard drives.

Go give them some of your money.

Countless times I’ve looked on Knockhill’s website and wondered (often out loud) about driving the short distance from Edinburgh to check out some of the events there.  And so finally, me and my Dad made the trip up there (a whole 25 minutes!) to watch the second day of arguably the biggest weekend on the calendar – the Dunlop MSA British Touring Car Championship.

The day included support from the Porsche Carrera Cup as well as SMRC and the Ginetta GT series.  Our luck was in with the weather as the sun shined all day long – we’ve somehow managed to come home with sunburn.  Not what you expect early September in Scotland!

The racing was fantastic and lived up to all the hype surrounding BTCC;  Knockhill is a short and aggressive circuit which seems to lend itself well to the touring car style.  We witnessed numerous shunts, dodgy overtake manouvres, spinoffs, and so on – the sort of thing you’d almost never get to see at a level like Formula One.  All in all it was a fantastic day out, hopefully the photos I captured reflect that.  Here’s the slideshow of 93 photos on Flickr, otherwise some of my favourites are below….

Me and the missus just spent a fantastic 10 days sunning it up on the cheap in Malta.  I came away qualified as a Scuba Diver, and as if that wasn’t productive enough, I even took some photos!  Favourite thing to shoot was definitely the random selection of cars we came across, although it’s a shame I didn’t decide to start collecting them a little bit sooner.  Still grabbed a few decent ones though.  Some of my favourites are below, the rest are up on Flickr.

This past weekend, RAF Fairford once again played host to the 40th Royal International Air Tattoo.  It’s become something of an annual (more or less) trip for me and my old man, I’m not much of a plane nerd by any stretch of the imagination but I do love the noise and the excuse to shoot a few photos and sit around outside drinking beer.

Speaking of photography, each successive year sees me trying to upgrade my equipment in some way in order to get closer to the action.  I first shot wih a D70 and a 70-200, then a D200, then a D700 + 1.7x Teleconverter (which got me back to square one…) and then this year I borrowed a friend’s D300s which, in conjunction with the TC and the 70-200, had me as close as I’ve gotten so far.  Still need that 200-400 though really…

Annoyingly, the weather and the selection of planes was far worse than it’s been on previous occasions but I still managed to bag quite a few keepers (~80 or so out of about 400), half of which I’ve stuck up online.  Here’s a few of my favourites below, otherwise check them out on Flickr here!

Backups have started to become something of a concern now that my ageing Netgear ReadyNAS NV+ is sporting 8TB of disk and a ~ 5.4TB filesystem.  Only a small fraction of that is media that I really, really care about (i.e photos, documents, content I’ve created) but the time taken to rip all of my CDs and DVD collection isn’t exactly insignificant at this point, either.  Sure it could be recreated from scratch if push came to shove, but I’d rather not go through that again.

The cheapest way to backup your data these days of course seems to be to just buy more disks, but with that comes the problem of what to plug them into. Enter the HP N36L MicroServer.  It might only have a relatively meagre 1.3GHz Athlon II Neo CPU (dual core though!) but it’ll support up to 8GB of RAM and most importantly, it has 4 x SATA drive bays as well as an additional SATA header on board.  A few clicks on eBuyer.com led to this little lot:

• 1 x HP N36L MicroServer;
• 2 x 4GB DIMMs;
• 4 x 2TB Samsung HDDs

All for the princely and scarcely-believable sum of £434, once you factor in HP’s £100 cashback offer.  With a nicely compatible set of hardware onboard the opportunity to double this up as my new vSphere ‘lab’ was too much to resist so I installed and booted ESXi off a USB stick plonked into the header on the motherboard.  I had a spare 750GB drive knocking around which I stashed in the 5.25″ bay at the top and connected it to the spare SATA header, but what about all those other disks?  That’s where I started to scratch my head.

The MicroServer doesn’t do “proper” RAID, the BIOS essentially fronts a software-based solution which isn’t supported by VMware.  You could import and provision them with ESXi for use by all the virtual machines, but eventually I decided to go down the route of raw-device mapping the disks directly to one VM, and that machine would run FreeBSD so that I could take advantage of ZFS.  The other upside to this is that if I decide for some reason to take ESX out of the equation, I can boot and run FreeBSD natively and just import my zpool.  I don’t have to worry about the messiness of VMFS or anything else like that.

With that decision made and ESXi installed, the first step was to RDM the locally-attached drives to my VM.  This isn’t supported via the VMware client, it has to be done from the commandline as an option to ‘vmkfstools’.  The procedure is well document elsewhere so I won’t bother with the details here, but once that was done FreeBSD saw all four disks on boot and I created a new RAIDZ array with those four drives.  So now onto some performance testing – just how much does my setup suffer when it’s virtualised?  Here’s a real-world example, the time taken to rsync (-av) a ~30GB directory over NFS:

• Native

sent 30211839333 bytes  received 19108 bytes  4125330.57 bytes/sec
total size is 30208073684  speedup is 1.00

real    122m3.212s
user    6m11.601s
sys     5m4.499s

• Virtualised

sent 30211839333 bytes  received 19108 bytes  3530453.81 bytes/sec
total size is 30208073684  speedup is 1.00

real    142m37.140s
user    6m33.874s
sys     5m6.392s

So an extra 20 minutes to do the full 30GB, which isn’t insignificant but I can live with that given its purpose.  The native install had the run of the whole 8GB of RAM, but the virtualised instance had only 2GB allocated – and everyone knows that ZFS loves RAM.  Aside from enabling ZFS prefetch for the VM, there hasn’t been much in the way of performance tuning here either – Jumbo Frames are disabled, NFS is out-of-the box on both client and server in terms of configuration, and so on.  Best do something about that then…

After three and a half years of ownership and > 40,000 miles driven across almost all of Europe, last Friday we finally said goodbye to YK03 DPV.  It was an extremely tough decision, mostly because there’s few cars that can touch the E39 M5 for what it is.  Practical, refined and yet still capable of thoroughly upsetting 911 owners, it took some beating.  And then there was the sound of that V8…

Nostalgia aside, there were a number of big-ticket items on the horizon that would have wanted doing should I have decided to keep the car, and so instead it made more sense to look at something newer.  I did briefly debate a Cayman S and even perhaps an Elise, but after looking around eventually decided on an E46 M3 (ideally a CS) instead.  Largely because I know the marque, but the Cayman doesn’t really do it for me and the Elise is just a tad too extreme.  A Megane R26 was also briefly considered but come on, that would have been pretty depressing coming from the M5.

So here it is – the new motor.  I’d been passively searching around for a little while but then stumbled across this E46 M3 CS at a main dealer down in Carlisle, thoroughly do-able from Edinburgh.  A trip down there to discuss trading my M5 in as a part-exchange resulted in an excellent deal, mainly due to some serious bartering (thanks to Rox!).  And holy shit, what a car it is.  The M5 was comprehensively rapid, but its weight meant that the momentum generated was a bit much through the twisties to say the least.  And then there were the woefully inadequate brakes, typical for M-cars, and which completely wilted on the way down the Stelvio Pass.

The drive back from the dealership put to rest any reservations I had – the M3 CS is an utterly magnificent car to drive.  We headed back along the A702 in the sunshine and the difference on these types of roads between the two cars is like night and day.  The M3 feels like a surgically sharp precision instrument compared to the M5′s blunt sledgehammer approach.  While I miss the sound of the V8, the straight six’s cacophonic howl above 4k RPM is enough to prick the hairs on anyone’s neck.

Enough of my self-affirmation though at spunking a load of money.  Here’s a few pics of the M5 to celebrate!

With VMware’s latest update to vSphere they’ve started to include their own copy of the sky2 driver and some support for various Marvell-Yukon NICs.  However, out of the box the onboard NIC on my Asus P5K-Deluxe – an 88E8056 – doesn’t appear to be recognized.  At first glance it looked  to be as straightforward as creating an updated oem.tgz with an amended /etc/vmware/simple.map containing the relevant hardware ID but alas, that doesn’t work.  Loading VMware’s sky2.o module with vmkload_mod just throws an error regarding unresolved symbols, which suggests that the scope of supported hardware by the driver isn’t quite as broad as you’d hope.

Some digging around revealed the steps necessary to create this homebrewed version.  The author provides a link to a precompiled version, but this doesn’t (yet) have the necessary ID in place in order to pick up my 88E8056.  One quick amendment to the aforementioned simple.map and we’re in business though, and I’m happy to report that this particular NIC now works quite happily under 4.1.  To save anyone else the effort and headscratching here’s a copy to drop in place on your installation: oem.tgz

A month or so ago I stumbled across a post on some forum regarding the replacement of the OEM LCD in an X200 with that of an AFFS model, and on the back of what people were saying I decided to give it a go.  As it was a bit more of a ballache than I was initially led to believe, but less than what the IBM service manual will tell you, I figured I’d document it here.

Firstly, let me just say that yes – it’s 100% worthwhile.  The difference in quality really is almost like night and day.  There’s quite a few problems with the OEM part, not least of which is the backlight leakage and the shitty viewing angle:

There’s plenty more before-and-after pictures in the forum link above, mine’s a weak comparison as I didn’t even shoot on the same angle but you get the idea.  They can be found on eBay, and I highly recommend you ordering from this seller as their service is amazing – China to Edinburgh in five days, and this was over Christmas!  Not only that but the part they sell doesn’t require any modification, it’s as close to plug-and-play as you can get.

So onto replacing the screen itself.  It’s a fairly fiddly job, and you’ll note that the instructions from Lenovo say to remove the keyboard, remove keyboard bezel, remove palm rest, and so on but this isn’t necessary at all.  To begin with you need to remove the three little plastic covers at the bottom of the screen and then undo each screw.  The bezel itself is a bit tricky to take off, requiring a touch more force than you might be comfortable with, but prize it open along the edges working your way around the whole screen until it snaps clear:

There are then six screws along each side, plus two for the webcam at the top.  Unscrew each of these, and the LCD should now be loose in the housing.  Now for the trick that saves you having to take half the laptop apart: If you carefully lift the screen up along the edges, you can also unscrew the horizontally positioned screws that hold the screen itself in its bracket.  Once you’ve done that, unscrew the two topmost silver screws that hold the camera in place, flick that out, and the LCD itself should lift off the back just enough for you to have room to unplug the cable from the old one…

This approach means that you don’t have to remove all the additional bezels to unplug the LCD cabling, as this and the backlight power connector (at the bottom) are the only two you need to worry about for the screen itself.  Plug those two back into your new display, then connect the keyboard and power to do a test before you screw it all together again.  Easy – far less work than following IBM’s official service guide!

How’s this for a milestone in your life:  A few weeks ago, I bought my first TV.  Yes, that’s right – 32 years of living and I’ve only just gotten around to actually owning one.  Personally I think that’s a good thing.  I’d like to say that my life is packed to the gills with extraordinary social demands, much like my cousin, but most of you know that I’ve been something of an itinerant and so the reality is that I’ve basically just watched other people’s all this time.

Anyway, yes.  That wasn’t my point.  This new telly, which is really nice but thoroughly underwhelming, is ‘Internet-ready’.  I hooked it up with a piece of RJ-45, logged into my Youtube account, and for some reason searched for Atari ST (honestly, it was a mindless search).  I then noticed one video in particular, chuckled to myself, and hit play:

Rox walked into the room, snort-laughed, and asked me what on earth I was playing.  It’s just a shitty low-resolution scroller with a digitised (mono!) version of a chart-topping 80′s dance hit.  But the thing is, this demo is one of the first examples that I can remember of hearing ‘proper’ music (i.e not synthesised) on a computer.  Think about it.  This is long before the advent of Fraunhofer’s cash-cow (the MP3 format) and its subsequent ubiquity on almost every device known to man.  We take all that so much for granted now, but once upon a time it was a genuinely amazing thing to behold, at least to my then 14-year-old ears.

I know, I know – you can go back way further than that to before computers made any sound at all, to when it was all black and white and so on, but what I’m driving at is this:  When was the last time you were genuinely amazed by a piece of technology?  Like you actually stood back and thought “holy shit, this is amazing!”.  Maybe I’m just old and jaded, but that never happens these days.  Sure, Apple’s iPad is awesome and I understand its potential implications but it didn’t give me a boner in the same way that playing IK+ on the Amiga for the first time did, or seeing the 3Dfx technology demos did, or hearing that demo.

The irony of all this isn’t lost on me: I’m sat here on my Apple laptop, milled out a solid billet of aluminium, with a dual-core processor and 4GB of RAM, effectively running a portable datacentre inside virtual machines.  18 years ago I’d have never imagined such things would be possible (mind you, I don’t think I’d have come up with something so mundane), and it SHOULD be amazing but somehow it isn’t – the magic of these devices and the advances in technology just don’t seem to be what they used to be.

I remember when it was all fields around here…

It’s fairly common to want to run SSH on port 443 these days, especially if you need to connect to your server from behind a restrictive company firewall that only allows egress connections to port 80 and 443.  This is all well and good, but what happens when you want to use SSL/TLS on your webserver as well?

This is where sslh comes in.  It’s a really simple tool that wraps incoming connections to these ports and then depending on protocol redirects it onto sshd back on port 22, or to your httpd on localhost:443.  With me so far?  Good – it’s easier to configure than it is to explain ;)

In this example, I compiled sslh from source and configured it manually in conjunction with Lighttpd.  First up, grab the latest version from sslh’s homepage:

$ wget http://www.rutschle.net/tech/sslh-1.7a.tar.gz
$ tar zxvf sslh-1.7a.tar.gz
$ cd sslh-1.7a

Take a quick look at the Makefile and amend the option for libwrap if you don’t want to use it.  Then it’s simple a case of running:

$ make && make install

This should drop the binary in /usr/local/sbin (unless you changed this in the Makefile!), the default configuration in /etc/default/sslh, and an appropriate init script in /etc/init.d/sslh.  Now all we have to do is amend the configuration and update it to reflect our host’s settings.  In my case, I replaced the ‘ifname’ with dischord.org’s IP address:

LISTEN=212.13.197.13:443
SSH=localhost:22
SSL=localhost:443

Now we have to amend our webserver’s configuration.  In my example, I’m using Lightttpd but it’s just as straightforward for Apache’s httpd.  Firstly, we’ll want to quickly create a self-signed certificate:

$ sudo mkdir /etc/lighttpd/certs ; cd /etc/lighttpd/certs
$ sudo openssl req -new -x509 -keyout lighttpd.pem -out lighttpd.pem -days 3650 -nodes
$ sudo chmod 400 /etc/lighttpd/certs/lighttpd.pem

In Debian’s case, they split out Lighttpd’s configuration into a /etc/lighttpd/conf-available directory.  You then just symlink in the relevant bits that you need into /etc/lighttpd/conf-enabled.  So here we need to edit the 10-ssl.conf and then enable it:

$ sudo vi /etc/lighttpd/conf-available/10-ssl.conf
$SERVER["socket"] == "127.0.0.1:443" {
                  ssl.engine                  = "enable"
                  ssl.pemfile                 = "/etc/lighttpd/certs/lighttpd.pem"
}
$ sudo ln -s /etc/lighttpd/conf-available/10-ssl.conf \
/etc/lighttpd/conf-enabled/10-ssl.conf

Here we’ve specified our newly-created and self-signed certificate, and told Lighttpd’s SSL engine that it should only bind to the localhost loopback address on port 443.  Now we can start sslh and restart Lighttpd:

$ sudo /etc/init.d/sslh start
$ sudo /etc/init.d/lighttpd restart

If all’s well, you should now be able to connect via SSH to port 443 on your server, but also still use HTTPS in your web-browser when hitting your site.

Our mate* Rob has recently qualified, after many many years studying and hard work, as a bona-fide doctor.  Many’s the time along the way that we’ve questioned his chosen career path (not that it’s our place to, but such is the Internet), as his attitude towards people in general isn’t really in line with what you’d expect from someone who is going to have the life of others in his hands.  Odds are that this is just general Internet bravado, but you never know.

Anyway, Rob is a top chap and has recently found himself newly qualified in the field of whatever it is that doctors have after six or eight years of study.  His recent forays into IRC have sounded pretty grim to say the least, but this latest outburst is a real winner:

[19:37] * rob has joined #quake2.uk
[19:37] [rob] hi
[19:39] [nick] hi!
[19:39] [nick] how's work??
[19:40] [nick] killed many people yet?
[19:44] [nick] no answer :/
[19:44] [nick] can't be a good thing :/
[19:48] [rob] ://
[19:48] [rob] i stuck my 1st finger up someones arse today as a doctor :/
[19:48] [rob] came out all covered in shit
[19:49] [rob] even before i stuck it in, there was diarrhoea like shit smeared all over her anus
[19:49] [rob] and no she wasnt hot, nick.
[19:49] [rob] she was 84 :/
[19:49] [rob] kill me now :|
[19:50] [rob] on that note, im gonna go to the gym and work the anger out of me :|
[19:50] * rob Quit (Quit)
[19:50] [nick] jesus :///
[20:35] [Durz|lnx] rob realising its not like off the TV :/
[20:35] [Durz|lnx] i missed the episode of ER where they stuck a finger up an 84 year olds ass

I think I’ll stick to IT.  Which, ironically, Rob already has a degree in (Computer Science).

Sweated my arse off in Darmstadt’s Golden Krone to take a few shots of these guys playing at the ‘Nonstock’ warm-up.  Click on the small ones below to enlarge, or check out the full set on my Flickr account over here.

I’ve owned the domain ‘freebsd.cx’ for a number of years and it’s got a certain amount of history behind it that a few of you may remember.  However, given CoCCA’s recent price hikes (50 USD a year!) and the fact that it’s now only used by a couple of people for email mean that I’m disinclined to fork out for the renewal fee.

If you’re still using an @freebsd.cx address I can offer you @xen.cx instead, otherwise consider it dead.  RIP!

I can’t believe how quickly the last year has gone, but we (me and Rox) have made a decision to move back to the UK towards the end of summer and this week I handed my notice in at work.

To be fair, the motivation wasn’t really the passing of time, instead there’s been a number of personal / family related stuff going on which when they’re all factored together means that somewhere closer to home is the best place to be for us both, for now.

We’ve had an amazing time out here in Darmstadt,  it’s a cool place and we’ve made some great friends.  But if we don’t start making plans now then before we know it another year will have passed and well, you know how that goes.

I have a three month notice period to work, so it’ll be mid to late July before I start seriously job hunting.  If you hear of anything that would suit, let me know!

Following on from Dan’s how-to regarding the configuration of SunCluster 3.2 within VMware Fusion, here’s my additional notes on doing this from a minimal install (i.e core software group) so as to keep the size of the VMs down.

First thing you’ll want to do is install some packages to make your life a bit less painful.  Connect the DVD device with the Solaris 10 ISO and mount it within your VM:

# mount -F hsfs -o ro /dev/dsk/c0d0t0s0 /mnt/cdrom

Then we can install some packages from /mnt/cdrom/Solaris_10/Product. Let’s start off with some basic utilities, bash, and online documentation (man pages):

# yes | pkgadd -d . SUNWxcu4
# yes | pkgadd -d . SUNWadmfr SUNWadmfw
# yes | pkgadd -d . SUNWbash
# yes | pkgadd -d . SUNWdoc SUNWman

SSH is also quite useful:

# yes | pkgadd -d . SUNWsshr SUNWsshu SUNWsshdr SUNWsshdu SUNWsshcu
# /lib/svc/method/sshd -c
# svcadm enable ssh

The second command will create your RSA and DSA host keys, without this you won’t be able to connect.

Then there’s a bunch of additional prerequisite packages you need to install before SunCluster itself. These are as follows:

SUNWtcatu
SUNWj3rt
SUNWj3dev
SUNWmfrun
SUNWxwrtl
SUNWxwice
SUNWxwplt
SUNWctpls
SUNWxwfnt
SUNWxwplr

You also need some Solaris Zones related gubbins otherwise you’ll see various errors pertaining to missing libraries such as libzonecfg.so.1:

SUNWzoneu
SUNWzoner
SUNWpool
SUNWluu
SUNWluzone
SUNWlur
SUNWpoolr
SUNWlucfg

Finally, the core / reduced networking group installation also disables some RPC related services that are necessary for cluster node communication. You’ll find that installation hangs, for example, when the primary node is waiting for the secondary to reboot. To remedy this, do the following:

# svccfg -s rpc/bind setprop config/local_only = false
# svccfg -s rpc/bind setprop config/enable_tcpwrappers = false
# svccfg -s rpc/bind listprop
# svcadm refresh rpc/bind

One other oddity I noticed was that I was seeing a bunch of RPC timeouts when looking at anything IPMP related. This turned out to be some DNS resolution going on as a result of the IP addresses assigned to the cluster interconnects and my ISP returning bullshit records. It’s easily fixed by editing /etc/nsswitch.conf and amending the ipnodes line as follows:

ipnodes: files

You should then be ready to kick off the install of SunCluster, and everything else is exactly as per Dan’s guide.

My good friend Greg and his cohorts, known collectively as ‘Married to the Sea‘, rolled into Darmstadt late yesterday to play an acoustic set at a locally (in)famous and unusual venue known as the Gute Stube.  It offered an ideal opportunity to bust out the camera and test out some low-light shooting of which I’ve done very little with the D700.  In short, I was blown away – not just by the music, but by the fact that I was snapping away quite happily at ISO 3,200 and coming out with fantastic results.

Anyway, yeah – the gig!  The band were fantastic, winning the crowd over not just with their music but also with their infectious charm and humour in this very intimate venue.  Even if the songs weren’t necessarily your cup of tea you couldn’t help but smile and just enjoy being a part of the event, it was great fun and hopefully I’ve managed to capture some of that.

Below is a few of my favourite shots from the night, and the full set is over on my Flickr here.

I dusted off the D700 last night to take some shots of Rox’s friend’s band, Immergrün, at the Schlosskeller here in Darmstadt.  It’s a pretty cool little venue but awkward to take photos, as while it’s fairly intimate there’s huge columns in the way (well, it is a cellar…) that kind of limit your options.  Anyway, I got a few decent shots – nothing groundbreaking, but it felt good to be at a gig and shooting some pictures.  Checkit:

It’s always kind of hard to take photos of a band whose music you aren’t familiar with, as you don’t know when they’re going to build up and kick off, those being the moments that you want to capture.  Still, these guys were pretty tight and didn’t seem to mind me making a nuisance of myself at the front.

Finally – Mat Hoffman is being honoured by a documentary that covers his legendary status and contributions not just in BMX Freestyle but for extreme sports in general.  I kind of feel like although he’s revered by most people who know about bike riding, he doesn’t always receive the wider recognition that he so badly deserves.  Hopefully this will help set the record straight.

Mat is awesome as shit, so we decided to make a movie about him.
—Spike Jonze, Jeff Tremaine, Johnny Knoxville

I’m super stoked on the trailer, right from the opening with Fugazi and of course Tony Hawk this seems like it’ll be spot-on.  Can’t wait.

Naa, not really.  But most people are only too aware of this whole “skateboarding vs. BMX” bullshit that’s been going on for some time now, brought to a head more recently by both sport’s popularity in the wake of the X-Games as well as irresponsible articles by the likes of Thrasher Magazine.  As a direct result, in recent years BMXers have faced ever growing levels of marginalisation from those who promote skateboarding and those who provide “skateboarding facilities”, in the form of bans or restricted sessions at various skateparks.  This is worse in the US where there are so many good parks that simply do not allow bikes under any circumstance, although I’ve witnessed it in the UK as well.

Anyway, I noticed today on Ride‘s site that Fudger put up an email chain started by an angry parent that ended up with one park that was about to allow bikes in changing their minds on the back of these few emails and keeping them banned instead.  The parent concerned claims to be one of the old guard, someone who has lived in the area all their life, grew up in the whole Dogtown era surfing and skateboarding and so on.  It’s because of this that I’m even more amazed at the level of hypocrisy that’s exhibited and that goes seemingly unchecked by the people responsible for the parks to which BMXers should be entitled.

Do you think the pools that people originally skated and rode around were meant for skateboarding?  Do you think that the street furniture that’s played such a big part of both sports was intended for something so creative?  What would have happened if, back when it was all starting, empty pools had been policed so fiercely that skateboarders never had a chance to ride?

I can understand the whole ‘concerned parent’ point of view, but at the end of the day your kid is involving themselves in something where danger is a huge part of the fun.  If it’s not a bike in a skatepark, it’s a car in the street.  Understand what you’re involving yourself in, and don’t complain when you get bitten every once in a while.  Oh, and teaching kids some basic park etiquette really, really helps.  Respect your surroundings.

On the flip side, the attitude that’s prevalent these days is doing BMX no favours.  There’s a level of arrogance that’s been layered on in both over the years, kids don’t really appreciate their roots nor were they around to witness the dark ages where there was literally fucking nothing to ride in terms of skateparks or whatever.  It’s that complacency that breeds the attitude we’re seeing now and mentioned in the email chain, and it’s a real shame as it’s inevitably going to widen the rift between the two and ensure that they’re kept out for good.

I don’t have a solution for this, but I do think it’s something that both sides need to work on.  I grew up riding with skateboarders, some of my best friends in fact, and I’d hate to imagine what it’d be like if the animosity grew to the point that both were kept essentially separate and at arm’s length from one another.  Let’s cut the bullshit and work together.

Hackintoshes have always seemed like way more effort than they’re worth to me.  Custom kernels, problematic updates, random crashes, it all flies in the face of why you use an Apple operating system (and computer) in the first place.  While I can appreciate the hacker aesthetic, I’ve never really had the inclination or patience to try it for myself – until now.

Dan‘s been running Mac OS X on his Dell for a couple of years with some success, but it was hearing him talk about Snow Leopard and how well that works pretty much out-of-the-box, including updates from Apple, that made me rethink my position.  Fast forward a couple of weeks and here I am in front of my own PC which now runs an (almost) flawless installation of Cupertino’s finest operating system.  The installation was laughably straightforward, but I’ve outlined it below just for shits and giggles all the same:

Hardware

• Intel E6600 2.4GHz Core2Duo Processor
• Asus P5K Deluxe WiFi mainboard.  SATA configured as AHCI in BIOS, no other changes of note.
• 4GB DDR2
• EVGA GTX260-216 896MB VGA
• 250GB Seagate HDD connected to SATA5

Installation was made with Empire EFI v1.085.  I almost gave up at the first hurdle I wasn’t able to boot from my SATA DVD drive as it echoed some random EBIOS error, to which anecdotal evidence on various forums suggested dodgy media.  I was using a disc which I knew to work so this wasn’t the problem in my case, and in the end I made an OS X USB boot disk by following some instructions out there on the Interwebs, and then booted the installation from that.  The only thing of note here was that I had to add the flags ‘GraphicsEnabler=Y PciRoot=0′ in order to get graphics to work, as my nVidia 260-based card relies on that injector for compatibility.

Anyway, the 250GB disk was dedicated to OS X and so partitioned accordingly during install.  The remainder of the process went smoothly, and after rebooting OS X came up pretty much as clean as a whistle.  All that was then left was to run myHack (on the Empire EFI CD for convenience!) which installs the ‘standard’ set of compatibility kexts as well as the Chameleon EFI bootloader onto your disk along with a really fucking ugly set of boot screens.  Everything was working from this point except for wireless and my NIC, the latter of which was easily fixed by a simple modification to one of the Apple kexts for MarvellYukon devices.  Incidentally, the wifi I don’t actually care for as my machine runs into a GbE switch that also has a wireless bridge attached, and so at this point I was happy.

The final litmus test for me was to install 10.6.2 via SU (as well as the recent security updates), migrate everything from my MacBook, and then sit back in amazement as it all worked flawlessly.  So now I have a solid, fully functional install of SL on my PC.  As I’m sat in front of an Apple LCD and using an Apple keyboard, as long as you don’t look under the desk who is to know that it’s not really a Mac Pro!? :)

The only thing that’s broken at this point is sleep mode, but as that’s something I’d never use on my workstation I’m disinclined to spend time hacking around with it any more.  To be fair the installation does seem very fragile and prone to basic errors, such as if the permissions are out of whack on any of your kexts it simply won’t boot with no clear indication as to what the root cause is, and quite often recovery can take a little while so it’s really more hassle than it’s worth.  In practice it’s extremely solid with everything working at full speed (including graphics and I/O) and it seems super snappy overall.  I don’t have to worry about updates or it breaking on a whim and so that’ll do me.