Just a quick post for those who make strike the error with the Dynamics Connector. There is a issue if you have installed the WIndows Azure AppFabric SDK on the same box. Uninstall the AppFabric SDK and you will be golden.

So I’m a big fan of taking available light photos. This is particularly good fun at Wedding parties and Christmas parties.

I took a bunch of photos at the Intergen Office Christmas party last Friday night and I’ve had a bunch of people ask “what camera etc… were you using”. SO I thought I’d take a few minutes and write up a post with some notes on taking available light party photos.

So the trick to taking nice party photos is not using a flash. To do this you need a couple of things;

1. A camera that has a decent High ISO noise level.
   ISO is an interesting term in that it goes back to the film days. The ISO (International Standards Organization) measure basically set a standard for the sensitivity of film to light. A high ISO film required less light to create an image, but, a high ISO film also typically produced photos with much more noticeable film grain. In the digital age we still use the term ISO but this time it’s basically referring to the amount of light required by the sensor in order to create an image; while the ISO of a roll of film was fixed we can change the camera ISO (sensitivity) at will because it’s basically just turning up the Gain (a camera sensor is effectively an analog-digital interface). If you’ve ever played with other things that have a Gain knob you’ll know if you put a small signal in and try and use the gain knob to boost it things can get a bit ugly- so it is with camera ISO if you turn the gain up too much things get noisy.
   So to get the nice party photos with available light (no flash and as such relying on not much light) you need a camera that lets you turn the gain up with out getting too noisy. The original ‘super amazing’ low light Camera was probably the Canon EOS 5D Mk II which is now about 3 years old. Indeed it is this camera that I used for most of the shots. It’s an expensive ($5k) body, but, over the past few years that super high ISO technology has filtered down such that you can get pretty decent looking shots from most $1000 bodies at ISO of 1600-6400 (to give you an idea of how revolutionary this is it would be very rare for anyone other than a professional to shoot film higher than ISO400 back in the day. Thus the modern digital cameras are several times more sensitive for the same level of noise than film was ever able to achieve)
   I’m a Canon person so that’s where my recommendations will sit and I think there’s not a lot wrong with the Canon 600D body in the $1000 price point. Take a look at these high ISO shots from DPReview. There will be similarly good options from Nikon in the price point which people can point out in the comments. The benefit of getting the $1000 body is you have more moolah to spend on lenses…
2. The second thing you’ll need is a lens with a large maximum aperture.
   You can tell the maximum aperture of your lens by reading it off the lens. It will be expressed as a ratio on the front ring e.g. 1:1.8 but it’s typically shortened in conversation and marketing to just the denominator. For the tech boffins the F-Stop (Aperture) is basically expressed as a ratio of the focal length of the lens. The lower the F stop number the larger the aperture- a bit counter-intuitive I know.
   For a zoom lens about the largest F stop you’ll get is F2.8 and even then you’ll be paying big bikkies and they weight a ton. There are a few shots in the party photos that are taken using an 70-200mm F2.8 lens.
   What you really want is a ‘prime’ lens. This is a lens with no zoom. Most of the shots taken in the party set above were taken with a Sigma 50mm F1.4 lens (this is a mid-range prime lens in terms of cost) but, for the most part I have it ‘stopped down’ to F1.8. This is for two reasons 1) wide open a mid-range lens like this can lose a bit of sharpness particularly around the edges and 2) A side effect of a large aperture is a very shallow depth-of-field; in the case of F1.4 on a 5D MkII with the subject at 2m only 11cm of depth will be in sharp focus. You’ll see this shallow depth of field in most of the low light photos as either a nice blurry background or as an annoying fact that of two people in the photo only one is in focus.
   So. Which lens to get. This is a place where Canon has a real advantage. If you are not ken to spend much money there is a Canon lens, the 50mm F1.8 MkII (plastic fantastic) that for under $200 will let you take nice party photos too. To me this is almost as good a reason as any to prefer Canon over Nikon at the cheap pricepoint cameras.
   If you have a bit more money to spend both the Canon and Sigma 35mm/50mm/85mm primes are great for under $1000.
   And finally if you have plenty of money go buy a Canon 85m F1.2 for shots like this one I took on a borrowed lens.

So this post became a bit of a ramble. To answer the more direct question of what I used for the party photos.

Canon 7D Body (early evening) with a Canon 24-105 F4 zoom
Canon 5D (low light) with a mixture of Sigma 50mm 1.4, Carl Zeiss 85mm 1.4 and Canon 70-200mm 2.8

  \

I’ll start this post by stating that I am generally supportive of Kiwisaver as a scheme. I deeply regret the decision taken by the national government to remove the ‘carrot’ from the scheme and would much sooner have seen them scrap the bat-shit crazy interest free student loans rort to make their numbers.

So I am a little torn with the new Labour policy for compulsory Kiwisaver. I think there are a few things we can all agree on

1. It will suppress wage growth. DPF posts on that here including a Labour party admission of such.
2. Kiwisaver is a "low risk” or conservative investment. Even if you pick the Forsyth Barr build your own scheme and load up on equities you’ll still be buying bluechips and not speccy mining shares.
3. 9% of pre-tax income is a relative SHITLOAD of money for most people and would, I suspect, represent the entirety of investment income for  the 99% (post written about the time the smelly hippies were occupying Aotea Square)
4. Saving and investing (rather than spending) is a good thing and as a country we don’t really do enough of it
5. At 9% of pre-tax income Kiwisaver will impact the common kiwi savings vehicle of the family home. People will be slower at paying off their mortgages (yes even will all the mortgage diversion stuff)
6. Banks aren’t particularly interested in letting you gear against your Kiwisaver assets

So beyond all the usual stuff that’s been discussed today I have one really major concern with compulsory Kiwisaver

It’s going to knock the stuffing out of small business startup capital. Basically the money that many people would have put into a liquid investment such that they could draw down to start a business will instead be locked away in a conservative investment vehicle until they are 65. Now starting your own business is risky to be sure, but, are we really saying that someone on $80k in a solid job should be denied the opportunity to put their money aside to invest in their own venture?

What do others think?

People will point to the success of compulsory super at these kinds of pre-tax levels in Australia, but, the Australian scheme is quite different to Kiwisaver and in particular their Self Managed Super Funds provide some relief to the above concerns.

The argument about whether small business owners can bear the burden of these Kiwisaver payments is somewhat moot if they were never in a position to start their small business in the first place.

So unless you’ve been living in a cave you’ll be aware of how rad helmet cams are for the adventure sports inclined amongst us. I held off buying one for quite a long time because they have been FREAKING expensive. The other issue is that the ‘industry standard’ GoPro HD makes you look like a total mongtard when you are wearing it.

Figure 1: How to look like a gaper mong!

So I really like the look of the cameras from Contour. They have recently announced the Contour ROAM which is their new budget camera at $199. But, this post isn’t about that new camera. It’s nice and all with more waterproofing that the original ContourHD, but, it lacks the 720p 60fps mode which I think is critical to nice footage in fast moving environments.

Instead this post is about the CountourHD; because it’s basically been replaced by these newer models it’s available super cheap. It supports full HD 1080p as well as the important 720p at 60fps mode.

Figure 2: How to not look like a gaper mong!

It can be had at Backcountry at the moment for about US$140 (http://tinyurl.com/contour-backcountry). If you’ve been in the market I think this is the deal to snag! Shipping is a bit to New Zealand but it’s still waaay cheaper than buying locally. Particularly if you get a spare battery and a few other things in the same order and/or order with friends.

In my travels I run into lots of people who are running Microsoft Office 2010 but many of them are also running the 64 bit version. I thought I’d do a quick post to discourage this behaviour!

Yes, we’ve had it drilled into us that 64 bit is the future, and, for operating system installs you do indeed want to be running Windows x64 because you’ve probably got more than 4GB of RAM in your machine. However, when it comes to applications that have been specifically compiled for 64 bit I suggest a much greater degree of caution and reflection before diving in. It is basically a trade-off by way of sacrificing raw performance for the ability to use more RAM in that application process. What do I mean by this; Well, there is an overhead in working with 64 bit pointers and this will mean that you take a performance hit in running the 64 bit application- see this article for some simple benchmarks http://www.osnews.com/story/5768. The flip side, however,is that you can then work with a full 64 bit address space and thus your application can use more than 4GB(ish) of RAM.

So, why do I want to discourage you from installing 64 bit office?  

1. You probably don’t need that much RAM for a individual Office application.
   Unless you are working with extremely large spread sheets in Microsoft Excel it is unlikely you’ll need to use > a couple of GB of ram for a single Office App.
2. There will be a performance hit.
3. There are significant compatibility issues
   If you use Office add-ins and other 3rd party extensions you’ll find that compatibility with x64 Office is patchy at best.

So there you have it. The typical person I se running 64 bit office is the ‘IT Guy’ or enthusiast for whom more bits == more power. This just ain’t the case.

Here is the official Microsoft article on the situation: http://technet.microsoft.com/en-us/library/ee681792.aspx

So Deb and I were in Vegas at the Venetian this week. Just for shits and giggles we tried to recreate our photo of real Venice... How did we do for authenticity?

The Grand Canal in ‘Real Venice’

The Grand Canal in Fake Venice (Venetian Hotel, Las Vegas)

Trust Deb to ruin it by not wearing the same top!!! Me, ever reliable in my yellow cool-dry.

Frankly I’m torn. The chlorinated canals in Fake Venice smell much nicer than those in real Venice…

Hi All,

Saw this rather interesting snippet come through on the Twitter feed this morning.

http://mobilitydigest.com/icloud-brought-to-you-by-microsoft-and-amazon/

Clicking through shows you some partial HTTP messages (some stuff is blurred out)

Certainly looks like iCloud is using both Azure Blob storage and S3.

They are probably using Blob storage for the ability to use Shared Access Signatures for file upload, but, without seeing the full URL in the HTTP request that’s a bit of a guess.

My guess is that the call to the iCloud servers for authorizePut will be fetching a SAS and then this is being used in the PUT request to the Blob storage endpoint.

There is a header in there called AuthorizationBSharedKey. I certainly hope that’s not the storage account shared key for the Azure storage account! But again, without seeing the full messages I can’t really tell. It’s certainly not a standard Azure header, but, it does have a somewhat worrying name.

Anyone able to pull the headers in full for some analysis? chris(at)syringe.net.nz

Chris

So one of the kayaking items on my bucket list was a trip to the Ottawa River. Just ticked that one off with a couple of runs over the last couple of days.

Didn’t take all that many photos but got a few decent ones of me running The Elevator Shaft chute at Garvin’s rapid on the Middle Channel. The flow yesterday was 13.0

Here is a MASSIVE panorama of the rapid. Thought I’d upload in all it’s glory for people who may want to examine it later to plan their lines. Clicky Clicky for the monster.

Elevator Shaft is hard River Right. It was a pretty simple drop but it’s got a heap of rocks in the bottom which kinda freaked me out at first. Someone on the trip who had run it a heap of times said you don’t hit em, just go for it. So I did.

And finally a view looking back up the river. With the Chute we ran on lookers left.

I awoke this morning to another great Ray Maker review, this time of the Polar RCX5. Now I have to admit a bit of a sports watch fetish, certainly nothing approach Ray, but I’ve still owned a bunch of Polar, Suunto and Garmin devices. I’ve pretty much settled on Garmin as they support the open(ish) ANT+ sensor platform, but, it’s not without challenges and as such I’ve recently been combining data from two watches.

Towards the end of Rays piece* he notes the issue with the RCX5 in that the desktop client basically spits the data as two separate files: a *.gpx for the Geo data and a *.hrm for the traditional polar biometrics and sensor data.

Ray says this

“For example, if you try and import into Sport Tracks, after selecting both files it simply sees them as two separate training entries – one using GPS data, and one using footpod data:”

Which is strange, because I do this all the time. Reason being is that the heart rate monitor that Garmin ship (the soft one) is a piece of turd. I ordered an older style strap from Garmin but they sent me the wrong one (an old FR301strap) which doesn’t bloody work. So, I am stuck using my Garmin 310XT for GPS + Footpod and my Polar s625x with the ever reliable analogue HRM and wearlink strap for HR.

Basically at the end of each workout I let the Garmin 310XT sync with Garmin Connect and I open Polar Personal Trainer and sync the s625x into that.

Then I can begin import into SportTracks. The trick is to import them in a two step process… I’ll use a worked example here which was my run/hike up Lynn Peak near Vancouver this past weekend.

I typically start by importing the Garmin *.fit file direct from the file system (it gets dumped out by the Garmin ANT Agent)

This should feel like a standard import.

Next I change to the Polar Folder and import the HRM file.

Now here is the trick. If you have the clocks on your watches correctly sync’d then SportTracks will typically pickup that the workouts overlap and automagically offer to ‘update’. In my case I didn’t because the olar Dual time doesn’t change the date. So instead I choose the radio button option to Update Existing Activity

Note that I have
1. Selected the radio button
2. Chosen my previously imported activity
3. Chose to import heart rate and elevation (the s625x has a *proper* altimeter which is a heap better than relying on GPS data from the 310XT)

And that’s really all there is to it folks- GPS data and running cadence from the Garmin and Heart Rate data and elevation off the Polar. While I don’t have an RCX5 to test with (feel free to send me one) this process should work just fine with a GPX file and an HRM file.

*Ray: As a quick ask would love to see the sections of your reviews with #Anchor tags so we can link direct to them.

So there’s not a whole lots of stuff on the interweb that gets you started with drives and SQL Compact 4 is only just out into the open. My need was for a lightweight database for a sample app I’m building with Chris Klug for our session at Tech Ed New Zealand.

I thought I’d run quickly through how I got things going.

Getting the Bits
I grabbed the bits using the Microsoft Web Platform installer. I wanted two things. 1) SQL Compact 4 CTP1 and; 2) The Web Matrix tool.
It’s hopefully fairly obvious why I needed SQL Compact. The reason I wanted Web Matrix is because it’s currently the only tool I could find that will allow me to create and edit SQL Compact databases.

I’ll be using Visual Studio for much of this post; I’m guessing y’all know how to get your mittens on that now.

Creating a Database
To build our database we’ll be using Web Matrix. It doesn’t have a specific database designer that is independent of the site editor so we’ll need to create a site and then create it in the context of our site.

We’ll start from a template.

…and create an empty site.

Choose Databases from the left hand nav.

Choose to create a new database from the ribbon

Double click to select tables. And right click to add a new table

Create a basic table definition with a ContactID and ContactName. Use the Ribbon controls to create new columns. Use a BigInt identity column for the ID and make it the Primary Key.

Save the table as Contact

C:\Users\Chris Auld\Documents\My Web Sites\Empty Site\App_Data

We need some data in our database. We could ahve used Web Matrix to do this but I wanted to show another tool that allows us to execute raw SQL against the DB as it may be more suited to some developers.

The tool in question is the SQL CE Toolbox (you’ll need the latest version which is a standalone application)
Grab it from here: http://sqlcetoolbox.codeplex.com/
Don’t just install the VIsual Studio add-in version as at the time of writing it doesn’t support the v4 sdf format.

Open the tool and right click to add a new connection

Open the database file and test the connection

Expand the tree to show our Contact table. Right click and choose ‘Script as Insert’. This will create a template insert script for us.
Enter some sensible (or stupid if you prefer) values and then execute the query. I didn’t attempt to insert the identity column and so removed that from the query.

Before we leave this tool click the Estimated Plan toolbar button. As long as you have SQL Server Management Studio installed it’ll open that tool and hand over the query plan text.

That’s right folks. That is a graphical query plan graph of the query for my embedded database. That’s pretty damn cool!

Getting Your Database Into Windows Azure Storage

Being a database we want our *.sdf file sitting in a WIndows Azure drive. There are a bunch of ways we could do this:

- We could put it into our service package, create a drive on startup, copy to the drive and then do a lazy check each time we start so it’s only done once.
-We could create the whole VHD on the client machine and upload that to storage. This removes any of the once only init code from our application. THis is the approach I chose to use

We need to create a VHD. Hooray for Windows 7- this is trivial. If you are a hard-ass like Steve Marx then you can use a batch command to do this but you all know that I’m not really a hard ass so I’ll do the draggy droppy clicky clicky thing.

Open Disk Management. I’m a fan of the search box in the start menu- type ‘Disk Management’ and then select ‘Create and format hard disk partitions’

Use the Action menu to Create VHD. Stick it somewhere useful like the desktop as we’ll need to grab it later. You’ll want to choose a fixed size- I chose to make it 100MB. You’ll only pay for the actual data you store on disk (not the empty stuff) but more on that shortly.

Once created you’ll need to Initialize the disk with an MBR

.. and Format as NTFS

\

The drive will now be available in Windows Explorer and we can copy our *.sdf file across from the My Web Sites directory to our new VHD

Once we’ve copied it over we can go back and Detach the VHD

Now we can upload the VHD to Blob storage. Before you crack out your favourite Azure storage client (I like Cloud Studio from Cerebrata) you’ll need to want to check that it supports sparse Page Blob uploads. Cloud Studio doesn’t so we’ll use the VHDUpload Tool from here: http://blogs.msdn.com/b/windowsazurestorage/archive/2010/04/11/using-windows-azure-page-blobs-and-how-to-efficiently-upload-and-download-page-blobs.aspx I took the liberty of building the tool: VHDUpload

More in the Part 2 of this post tomorrow.

The Windows Azure Management API uses x509 certificates to authenticate callers. In order to make a call to the API you need to have a certificate with both public and private key at the client and and the public key uploaded into the Azure portal. But, if you then want to call the management API from your Windows Azure VMs then you’ll also need to install the cert into the instances by defining them in the service definition This post will show you how.

I found it a bit of a pain to get going so here’s my simple guide. I used this to setup the certs for my favourite open source Azure toolkit Lokad-Cloud. We’ll be creating a self signed certificate, then uploading that certificate into the Windows Azure management portal. Finally we’ll add the certificate to our service model to ensure that Windows Azure installs the certificate into our VM instance when it is started.

Here’s the approach in pictures so you can follow along.

1. Create a self signed certificate in the IIS7 Manager
   Open IIS7 Manager
    
2. Expand the node for your local machine
    
3. Double Click Server Certificates
   
4. Choose Create Self-Signed Certificate
   
5. Give it a friendly name
   

You’re all done in IIS7 Manager. It’s just created a new Certificate and added this into the trusted root certification authorities on your machine. This is a highly trusted location for the cert so do be careful if you ever export it with the private key included.

We need to export it with the public key included so that we can upload it to the Windows Azure Portal.

1. Run CertMgr.msc - just type it in full into the Start box.
   
2. Expand 'Trusted Root Certificates Authorities –> Certificates"’
   
3. Find your cert using the friendly name column and right click and choose Export
   
4. Choose the option to NOT export the public key. You should generally avoid exporting the public key if at all possible*.
5. Choose to export as a DER Encoded binary
6. Save it somewhere useful – we stick ours into source control
7. Choose Export again, but, this time choose to export the private key
   Leave the PFX options all unchecked
   
8. Give it a sensible name.

You’re all done exporting. Now we need to upload it to Azure.

1. Go to the Windows Azure Portal
2. Choose the appropriate project
   
3. Click the Account tab across the top of the page
   
4. Click Manage My API Certificates
   
5. Browse to and upload your *.cer file you created earlier
   
6. You should now see the certificate listed against the account.
   Note that it’ll be listed using the Subject rather than the friendly name. So you’ll need to identify it by the thumbprint if you have multiple uploaded from the same machine. Azure Team: Can we please have the friendly name listed here?
   

Now we’re ready to use our certificate from the Windows Azure Tools for Visual Studio

1. Open your Windows Azure Project – We’re using Lokad-Cloud here
2. Expand the Cloud Project to show the Roles.
   
3. Right click the Web Role and choose properties
   
   This will open the Snazzy Windows Azure graphical UI. This is much nice than editing the Model XML by hand.
4. Choose the Certificates Tab
   
5. In our case there is already a certificate entry defined by the default Lokad Model definition.
   Click the elipses (…) at the end of the thumbnail column to open the certificate chooser dialog.
   
6. Choose the cert from the dialog
   
7. Repeat for the worker role

The last thing we need to do is upload the certificate into our cloud service before we can upload our packages.

1. Create a new Cloud Service
2. Browse to the bottom of the page to find certificates and choose Manage
   
3. Browse for your certificate (including the private key)
4. Punch in your password and hit upload
5. Confirm that you’re all uploaded.
   
   

You’re all done. Now you can happily deploy the app to Windows Azure along with having your certificate deployed into your Azure instances as well. This means that your Azure roles can now call the management API themselves.

*There will be times when you need to export the public key. We’ve had to do that in this scenario as we actually want to install the private key into our Azure instances. Another good example here is in a shared development environment. You have the option of either sharing a certificate among everyone or each uploading your own public key. Because we’re using the lokad tool and this explicitly supports just a single cert we’ve actually exported our cert (password protected) into our source control system so all developers can install the same private key.

If you have done an export and want to install the private cert onto a new machine then you’ll need to add it into the Local Computer –> Personal store.

1. Run MMC by typing MMC in the start run box
2. Add the Certificates Snapin
   
   
3. Choose Computer Account
   
4. Choose local computer
5. Open the Personal Store and choose More Actions>Import
   
6. Browse to and import your certificate.

The certificate should now be visible in the Windows Azure Tools for VS.NET cert selector dialog.