jon torresdal

Bug in Hotmail (update)

Jon Arild Torresdal — Fri, 08 Aug 2008 19:30:42 GMT

A couple of days ago I discovered a bug in Hotmail that you can read about here. As it turns out it was both a bug in Hotmail and in my code (sort of), and a type error in my blog post causing some confusion. Thanks to Rune Grothaug for getting me in touch with the Hotmail team.

The bug in Hotmail is related to how they query for content id's in the email. It's too expensive to load the email into a html parser to get the id's, so they do a string match instead. Here's how I understand it: When they try to match nnug10 they actually get nnug1 back (as the first match). One solution suggested by them was to reverse the order of the searches allowing nnug10 to be found first. I'm not sure if this is what they're going to do. (Update: I got confirmation that this is the solution they're going for)

As for my bug it was that I forgot to add the file extension to the ContentType.Name (see code below). This is the name that actually gets displayed for the attachment image in the mail. In Hotmail (and Gmail for that matter) it shows all linked resources as attachments, and the name of these attachments had no extensions (e.g. jpg). A workaround for the Hotmail bug could have been to do the same with the ContentId. When the id is nnug10.jpg it actually works! In my blog post I said that the cid's where nnug0.jpg, nnug1.jpg etc. which was not the case. It did not have the .jpg extensions.

Taking the above into account this makes the Hotmail bug far less serious than I originally thought.

But..., when you see the property ContentId you're thinking it can be whatever only it's unique right? This is why I didn't think much of what the ContentId was as long as it was unique. However, to be absolutely sure that something like this will not happen again I changed the cid's to be guid's instead (this was also suggested by the Hotmail team). Better safe than sorry :-)

Here's the code after my modifications:


          LinkedResource img
= new LinkedResource(stream,
mediaType); img.ContentId = guid; img.ContentType.MediaType
= mediaType; img.TransferEncoding = TransferEncoding.Base64;
img.ContentType.Name = img.ContentId + fileExtension;
img.ContentLink = new Uri("cid:" +
img.ContentId); resources.Add(img);

Bug in Hotmail forces me to apologize

Jon Arild Torresdal — Wed, 06 Aug 2008 23:43:23 GMT

Update: More on this story here.

I've been playing around with an email tool that I've created to send out emails to NNUG members. It's just a simple command line tool sending an email to each person in a list to avoid using bcc which sometimes get hooked in a spam filter. Another thing this tool does is to take an html page (an html formatted email), scan it for images and embed the images using the AlternateView class and its LinkedResources collection for sending as an html email.

The emails sent by this tool looked ok in Outlook, Gmail and Hotmail on every test I did. Except for when I decided to use the tool to send out an official NNUG invitation to 300 members! The only change I had done was adding one extra image. I noticed that the email I got on my Hotmail account had a wrong image at the end. The image used was the same as the second image in the email. The reason was not hard to find.

When I add images to the LinkedResources collection, I name the images/files nnug[someNumber].jpg like this:

nnug0.jpg, nnug1.jpg, nnug2.jpg etc.

This worked fine for all images up to nnug10.jpg. Can you see why nnug10.jpg did not work and what the bug is? The last image that was supposed to be nnug10 was showing up as the second image instead (nnug1). E.g. nnug10 = nnug1! Hmmm.... Come on!

Then my apology. To all members of NNUG with a registered Hotmail account and to Christian Weyer:

I'm sorry that I did not discover this bug before, so I could have changed my algorithm. Sorry to you Christian for being nnug1.jpg! You replaced the NNUG logo, resulting in the email ending with:

Kind regards [image of Christian]

instead of

Kind regards (image of NNUG logo)

...giving the email a bit different meaning than I intended :-)

Btw. don't forget to sign up for the NNUG events by Christian Weyer and Dan North 25th and 27th August!

Why are we doing Scrum again?

Jon Arild Torresdal — Mon, 04 Aug 2008 20:04:44 GMT

After finishing the book "The Art of Agile Development" which I reviewed here, I was asking myself a few questions:

Why are we doing Scrum?
Why did we start out with Scrum on not XP?

I think the answer to these questions rely on many of the misconceptions around XP. If people really knew what XP was about and not that XP=pair programming and/or XP=no documentation or whatever first impression you get from XP, I think the adoption rate would have been higher. If that's true, what makes people adopt Scrum?

First of all Scrum got really hyped a few years back, many books were written and many adopted the process. They also introduced the certification program of ScrumMasters and Product Owners. If certification is good or bad is still being discussed in agile forums today, but one thing is clear; the certification programs gave Scrum much publicity and I will claim that it helped the adoption rate of Scrum. It also might have helped some companies to easier except the process as a "proper process".

One thing to consider is that Scrum is a management process. It has no practical solutions of how to develop software, only how to manage and plan for successful software delivery. My impression is that after a while teams that started out with Scrum start to look for ways to improve their process and they often turn to the practices defined in XP (or adapted by XP). I think that every successful adoption of Scrum is also adapting one or more development practices like Continuous Integration, Test Driven Development, pair programming etc, and they might not even know that they're a core part of XP.

If you look at what XP says about the management part you see that it resembles Scrum in many ways, or was it the other way around? ;-) I would dear to say that the difference is so small that it doesn't really matter what you choose. However, Scrum cover less of the complete development process than XP does, so my preference at the moment is XP though we're still doing Scrum. In essence this only boils down to what you call your process and which community you want to belong to. So maybe I should just be satisfied by saying we're Agile :-)

This makes me wonder what students learn about Scrum and XP today? If at all? The only process I learned at university was RUP. If they learn both Scrum and XP, which one would they choose? Any students/teachers/professors out there that would care to comment?

This also makes me wonder if a Scrum shop has more credibility towards customers than XP? If you where (or are) a boss of a company that were about to develop some software, would you choose an XP team or a Scrum team for the task? Would you care at all?

Were the Vikings agile?

Jon Arild Torresdal — Fri, 25 Jul 2008 18:09:24 GMT

While I was on vacation in the southern part of Norway (Flekkefjord) I found a postcard with the "Viking Laws". Not sure of the authenticity of these laws, but I still found them quite interesting. Surprisingly I also found them quite agile and a set of good rules to follow for any development team!

Here's the laws and how I link them to software and agile thinking:

§1 Be Brave and Aggressive

Be direct
- Hmmm... I think I'll go for N/A for this one.
Grab all opportunities
- And another N/A...(?)
Use varying methods of attack (attack is interpreted as testing)
- Test your code and functionality by doing TDD, BDD, Exploratory Testing and/or similar things
Attack one target at a time
- Don't do multitasking
- Don't have single tests that cover multiple functional areas
Don't plan everything in detail
- Don't do a big plan up front (waterfall)
Use top quality weapons
- Use good software tools
- Use simple planning tools like whiteboard, story cards etc.

§2 Be Prepared

Keep weapons in good conditions
- Keep plans up to date (e.g. sprint burndown, release burndown etc)
- If 3. party components are used, make sure you upgrade when new versions are available to avoid work to add up
Keep in shape
- To be fit will let you be more effective at work and also have a lot of energy left when you get home, which in turn hopefully means you're motivated and energized for work the day after
Find good battle comrades
- A great team of people will do great things
Agree on important points
- Common understanding/common code base
Choose one chief
- Hmmm... Who's the chief on an agile team? The team or the customer? I would go for the team together with the customer and not one single person.

§3 Be a Good Merchant

Find out what the market needs
- Develop the most important functionality first and prioritize accordingly
Don't promise what you can't keep
- Tell the customer the truth and keep them updated on time schedules and in the loop by utilizing review meetings etc.
Don't demand overpayment
- I'll let the sales people cover this one :-)
Arrange things so that you can return
- Have your code and architecture so that you can "easily" change out existing parts of your system. Think separation of concerns, inversion of control and dependency injection.

§4 Keep the Camp in Order

Keep things tidy and organized
- Keep code and architecture clean and easy to read and understand
Arrange enjoyable activities which strengthen the group
- Have fun at work, use breaks for fun activities like fußball or other team games and have a beer on Friday night from time to time.
Make sure everybody does useful work
- Have everybody involved and work as a team
Consult all members of the group for advice
- Do agile modeling when you don't know exactly what to do or how to solve a problem
- Involve the team and ask for advice whenever in doubt
- Do pair programming

Book Review: The Art of Agile Development

Jon Arild Torresdal — Sat, 05 Jul 2008 23:32:05 GMT

The title of this book could easily have been The Art of XP Development. My first impression of XP back in the days where two things; pair programming and no documentation. I guess some people still think that(?). Did they choose to use agile to not scare away the people with these type of misconceptions? James Shore (the author, together with co-author Shane Warden) says on his web site:

Our goal for this book was to provide a complete how-to guide and starter kit for beginning and intermediate agile practitioners. To keep the book concrete and practical, we focus on XP.

So that explains that, but I'm not going to review the title of the book am I? :-)

Talking about the authors, James Shore is a well known name in the agile community. The Agile Alliance recognized James with their most significant award, the Gordon Pask Award for Contributions to Agile Practice. He is also an internationally recognized speaker and have done lots of consulting for agile teams. Shane Warden is the co-author and editor of this book. He promotes open source software for O'Reilly's Open Technology Exchange.

To cut strait to the point, this is probably the best book on XP I've ever read and one of the best agile books I've come across as well! Combine it with Mike Cohn's book; Agile estimating and planning, and maybe Agile Modeling by Scott W. Ambler, and you're well on your way to become agile.

My motivation for reading this book was based on my knowledge of Scrum and my willingness to adapt more of the XP practices. One of the questions I've been asking myself is how do "they" approach team management (our should I say self management...) compared to Scrum? That question and many others where answered in this book!

XP is of course a full fledged development process and (in my opinion) have been colored by its some "extreme" (at the time) practices and deserves more attention than it gets. So I would strongly recommend reading it, even if your focus is not XP.

As I've mentioned before, many books are very theoretical when it comes to software processes, but this book is very hands on and the authors clearly have a great extent of experience with the process. This gave me as a reader good value and motivated me to try out different aspects, as well as new things on areas that I thought we were doing ok already. I also rediscovered some things I just forgot about.

The key takeaways I got from the book was:

XP covers a lot more around the management part than I thought.
Scrum and XP is closer than I expected and XP cover more ground.
If I had known what I know about XP today, I might have introduced XP instead of Scrum to our teams.
I got many confirmations that what our teams are doing today is good, but also found aspects already known to us where I knew we could improve, and some aspects I hadn't thought of.
All the provided examples of what to try or do in specific cases allowed this book to be a how-to instead of a theoretical book about XP/Agile, which I valued very much.
The focus on technical debt gave me new insight.
The testing focus. Good article covering exploratory testing by Elisabeth Hendrickson.

The book is structured into three parts:

Getting Started
Practicing XP
Mastering Agility

For more details about the content of each part, have a look at Jame Shore's web site, where he lists the chapters. For me part two and three were most valuable. Part one give a good introduction to agile, XP the Manifesto and the values behind it all. You should defiantly read it if you are new to the process or feel that you could need an update, like I did. In part two every chapter starts with a mini-étude, and every topic within a chapter has an explanation followed by five paragraphs in the end; questions, results, contradictions, alternatives and further reading (if any). Here's what the authors say about the mini-études:

Have you ever heard a musician playing scales? That's an étude (if a boring one). An étude teaches mastery through precise and careful repetition...
...Beginning agile teams can use the études to refine their practice of agile development.

I think these are good, but I found myself skipping most of the mini-étude's and the five paragraphs in the end. In the beginning I read them all (maybe except from the études), but the further I progressed in the book the more I skipped. I did this because much of the material was known to me and I found the main part of each topic most interesting. If I felt something was unclear or had questions, I read the other parts as well. I'm not saying this is the way I recommend reading the book.

Part three of the book was very interesting to me. It covers the following topics:

Improve the Process
Rely on People
Eliminate Waste
Deliver Value
Seek Technical Excellence

The interesting part is that they give you insight to how you can go about to improve the process when you've become comfortable with it. In essence; apply the process on the process itself.

So am I not going to criticize (on the negative side) anything in the book? Well, I could because there are some things I don't totally agree with, but all in all it deserves its good credit. However, I'm not going to pick on the few knitty details I personally don't like/agree with.

When I did some research about the authors I found an interview with James Shore over at InfoQ that you should check out: http://www.infoq.com/news/2008/05/Interview-James-Shore

As Brian Marick, I'm considering buying in several copies of this book to distribute among our agile teams. I really think that it could generate great value for anyone reading it, being any type of agile team member.

August - The best NNUG event ever!

Jon Arild Torresdal — Tue, 01 Jul 2008 18:41:34 GMT

If everything goes as planned NNUG Bergen will in August have two great speakers. Dan North from ThoughtWorks and Christian Weyer from Thinktecture! Is that a great lineup or what?! And the best part... it's FREE!

Currently the plan is to have Dan North on stage the 25th of August and Christian Weyer the 27th (Monday and Wednesday). Here's a bit about the two speakers and what they're going to talk about:

Dan is a principal consultant with ThoughtWorks, where he writes software and coaches teams in agile and lean methods. He believes in putting people first and writing simple, pragmatic software. He believes that most problems that teams face are about communication, and all the others are too. This is why he puts so much emphasis on "getting the words right", and why he is so passionate about behaviour-driven development, communication and how people learn. He has been working in the IT industry since he graduated in 1991, and he occasionally blogs at dannorth.net.

At NNUG Dan is going to talk about The relationship between Domain-Driven Design and Behaviour-Driven Development.

Christian is co-founder of ThinkTecture, a European software development support company. He has been modeling and implementing distributed applications with Java, COM, DCOM, COM+, Web Services and other technologies for many many years. Recently his focus has been on the ideas and concepts of service-orientation and their practical translation in customer projects with Windows Communication Foundation (WCF) and Windows Workflow Foundation (WF) being the two main technologies applied. Especially the more than natural marriage of WF and WCF currently has gotten his attention.

Christian's talk will be about WCF, but other than that he's quite open to suggestions. I'm thinking it would be interesting to hear about why we should move from Asmx to WCF and the benefits (and any drawbacks) we get from that move. What do you want to know about WCF? Drop me a comment and we'll see what we can do... Be quick though, we need a decision soon.

More comics

Jon Arild Torresdal — Fri, 20 Jun 2008 19:43:48 GMT

Found the Complexitor comic through PolymorphicPodcast. This looks like it has great potential, so go over and subscribe. He (Scott Fletcher) will try to have a new strip ready every Wednesday. Really looking forward to how this one is going to turn out. Man those graphics looks great!

Contiki comic strip

Jon Arild Torresdal — Wed, 18 Jun 2008 20:45:47 GMT

A great friend and colleague of mine, Torbjørn Marø, have created at comic strip (the contiki strip) with people and stories from the company we work for, Contiki. For us this is great fun, but I suspect we're not alone. Great work Torbjørn!

As of now there are two strips available:

Btw, this is me :-) :

Looking forward to the next one!

Not a valid win32 application on Vista

Jon Arild Torresdal — Sun, 15 Jun 2008 19:58:30 GMT

Recently I've done some work on automating the creation of our deployment packages. In this process I've used some tools on my Vista client to make sure things work before running live on the server. Like yesterday I used mt.exe to embed a manifest file into our exe files, which (after several runs) resulted in:

mt.exe is not a valid win32 application

What!? I looked up the file in Windows Explorer and saw that the file's size was 0KB! What is it that I do so wrong that I'm able to totally clear out everything in an exe file just by running it?

This has happened to me several times. Once with mt.exe, once with mage.exe and 2-3 times with Excel.exe (2007)! Excel was ok after log of - log on, but the other files I had to manually replace to get them to work.

After searching the net I found that many has this issue with Excel 2007, Access 2007, Snippting Tool and Windows Media Player after installing Vista SP1. I guess I can add mt.exe and mage.exe to the list...

Here's the post mentioned. The interesting part is this:

Microsoft is aware of this issue and is working on a fix. As several people have observed, this is konwn to affect Excel 2007, Access 2007, Snipping Tool, and Windows Media Player on Vista SP1. The fix is to either exit as many apps as possible that you are running (e.g. Outlook), and then load the app (and then you can reload the apps you had running), or a logoff/logon will clear it up for a while (days/weeks). The error messages are confusing, but the problem is neither an invalid application file nor insufficient system resources. It only affects certain applications, and typically only after the user has been logged in for an extended period of time.

and then a response on 14th of May from a MS representative:

We're working on it, it's going through full testing right now and should be available in a few weeks if there are no issues found.

In the meantime: You don't need to reboot, just log off and log back on. It's a per-user issue. That will clear it up if it's the same issue. I realize it can be a pain and I'm not trying to minimze the issue at all, but the workaround is simpler/quicker than a full reboot.

It hits different people in different time frames. It depends on how many apps you are running and how often you have loaded/unloaded them.

I appologize that it takes this long, but we'd rather not give you an update that had other problems/regressions, so we want to make sure it's fully tested, and that can take a few weeks in an environment as complex as this, as I'm sure you know.

This is all good, but now in mid June there's still no fix to be found. The mentioned workaround works for me with Excel, but not the other apps. Please let me know if you see a fix for this, and if you work for Microsoft and has any info about the issue please let me know.

Firefox 3 and a World Record Attempt

Jon Arild Torresdal — Sun, 15 Jun 2008 19:29:19 GMT

Firefox 3 will be released on the 17th of June (Tuesday). At the same time Firefox will try to beat the world record of most software downloads in 24 hours! Interesting marketing stunt :-) Check out the new features of Firefox 3 and Firefox Download Day 2008.

Exposed as agile

Jon Arild Torresdal — Mon, 09 Jun 2008 08:12:41 GMT

On Friday we had some friends over and enjoyed the beautiful weather in Bergen. I and (let's call him Trond) were sitting outside talking a bit bout work, agile, .net and java. I fired up the grill to throw on some burgers and I went in and out quite a few times to get stuff I needed. First I went and got the meet I needed. At the same time I also got some lettuce, ketchup and mustard. I went out again and threw the burgers onto the grill, keeping the conversation going with Trond. Then I went back in, got some plates and some grill equipment I forgot last time. I went out again and starting flipping some of the burgers. I then realized we needed something to eat with, and went to get some knifes and forks, as well as a couple of beers. Back out again, I flipped the burgers, went in again and found something else I needed.

When I got out again Trond said: "I must say you really ARE agile! You even barbeque agile! Only getting what you need there and then and not planning too far ahead."

And he was right. This is how I am. Somebody might say chaotic, others agile :-) Maybe that's why I was so easily convinced about the agile way of working.

But seriously, one of the most important aspects of agile development (from my perspective) is the idea of postponing all decisions to the last responsible moment (note: it's responsible not possible). This will let you be more adoptive to changes and force you to not implement things you don't need (e.g. it's very tempting to implement something that you think you will need in the future). On a traditional project where "everything" is planned up front (a waterfall like project), you often run into problems where you see that the way we thought about the solution back then is not the way it is today. This might effect the architecture in a big way. If you're being told to stick to the planned architecture, you find yourself creating workarounds to fit the architecture, which probably is not the best solution.

Fußballfun @ work

Jon Arild Torresdal — Wed, 04 Jun 2008 18:16:22 GMT

At Contiki we have a lot of fun. Both with programming and other things. One of the activities we do a lot of is fußball (aka. table football). Fußball has become such an integrated part of almost everyone's day that if someone don't play at least once during the day, something got to be wrong. It's a really a nice game for breaks. One game takes from 5-10 minutes.

Programmers (and others too I guess) often tend to work for long periods of time in front of the computer, without taking a break. Research has showed that taking frequent breaks improves productivity and lessens the risk of injuries. There's probably a lot of writing about this topic, but one interesting article I found was this: Take a Break and Refresh Your Productivity. Fußball is a great way of having this small break once in a while. So that was the excuses, now to the fun stuff! :-)

Last week we finished a 10 month long league. Torbjørn (our Web lead) has created a great app for entering and keeping track of scores. During these 10 months I've played 350 games, some played more others less. Take a look at Torbjørn's blog for more about the app (in Norwegian). Now however, he's extended the app to support a Contiki league and I'm looking forward to a great season of fußball! :-)

I don't know how big fußball is in Bergen or Norway in general, but I would really like to know. Do your company have a fußball table? Let me know! I like to think we've been quite good at it and it would be fun to challenge another company in a game. Anyone up for it? I also have a plan to do some fußball stuff through NNUG, but I'll blog about that later.

Our table	THE table we would kill to get
Our glass fußball room in the middle of the working place	Worlds biggest fußball table

And to end this off, here are some inspiring fußball videos:

Sysinternals tools on file share

Jon Arild Torresdal — Tue, 03 Jun 2008 18:36:07 GMT

Don't want to download and install all the tools from Sysinternals? Want to have access from anywhere on any computer? Just hook up to the UNC path: \\live.sysinternals.com\tools and you can run them directly. Thanks to John St. Clair for letting me know. Is that cool or what?

For more information about the different tools go to http://technet.microsoft.com/sysinternals.

PDC2008

Jon Arild Torresdal — Thu, 29 May 2008 08:41:21 GMT

The PDC web site is up! At the time of writing it's 151 days, 3 hours and 25 minutes to the conference start :-) Anyway, if you live in the US you've probably bought a ticket already. If you don't and are lucky enough to have a budget big enough to travel to the US, PDC is something to consider. My impression of PDC compared to TechEd is that you'll get to know what's coming, as apposed to TechEd where you get to know what's already there. I would love to go, but I'm afraid I've spent most of my budget and are left with buying books only :-) I'm going to NDC2008 though, which will be a great event as well.

New features in TFS2008 SP1

Jon Arild Torresdal — Tue, 27 May 2008 05:47:50 GMT

Brian Harry over at the TFS team has a great post about the new features that will be available in the coming SP1 for TFS 2008.

FIT

Jon Arild Torresdal — Sat, 24 May 2008 21:33:35 GMT

Lately I've been trying to focus on how to make testing work better at our company. We've fully integrated our testers into the Scrum teams, but there's still some things I feel is missing. Especially related to the tools/frameworks we use for testing. One of the things I'm looking into is Fit and FitNesse (Framework for Integrated Test) created by Ward Cunningham in 2002. The first time I got an introduction to Fit was in Nils Christian Haugen's presentation at JavaBin back in March. This got me very excited, but I've hadn't had time to look enough into it, but now I think I will.

In essence Fit is a framework that lets your user stories's story tests (or acceptance tests) to be automatically tested/verified. The way you do this is by using a table structure (as showed on the left) to give in values and expected outcome. This is a very nice way of working with tests from a customer perspective. Everyone can understand this by having a short introduction to how it works.

Much like you do with unit tests, this process is automated. The preferred way of authoring unit tests is by using Test Driven Development (TDD). Similarly, working with Fit you can use Story Test-Driven Development (STDD). I really find this way of working to be very interesting and I hope to try this out live soon. Hopefully I can post some more articles on this later when I have some actual experience with it :-)

David Hussman does a great job describing Fit in his presentation.

Pex: Automated Exploratory Testing for .Net

Jon Arild Torresdal — Sat, 24 May 2008 21:31:09 GMT

Earlier this month Fredrik Normén wrote a post about argument validation. Yesterday Rune Grothaug from MS Norway posted a link to a new tool MS has developed in their research lab called Pex (Program EXploration). Combining these two (argument validation and Pex) have to be a perfect match.

Real wireless broadband

Jon Arild Torresdal — Tue, 20 May 2008 21:02:19 GMT

PS! This article is about Norwegian wireless broadband's, but you might find similar solutions in your country.

A couple of years ago I considered buying a subscription for a wireless broadband from ICE to use when I was out traveling or at our lodge. ICE uses the old mobile network in Norway (NMT 450) at 450 MHz which had and still has a great mobile coverage in Norway (though not used for mobile phones anymore). This network has a max down speed of 3,1 Mit/s and upload of 1,8 Mbit/s. However, at the time the subscription was too expensive (around 400kr/$80/£50) per month. It was not an option to have both a fixed broadband at home and the wireless subscription on top of that. Recently though they have started to offer a different subscription based on usage. Or actually it's a subscription where you pay for one week at the time, which is more attractive but still quite expensive if you're just using it in the weekends. However, having this as an option to get on the Internet when you absolutely have to is of great value. DinSide.no has an article about ICE.

Another company offering a similar concept is Telenor. Telenor offer wireless through normal GSM network (EDGE/3G/Turbo-3G). They has several subscription options from monthly, weekend to just usage. You can find more info here. There's also a test by DinSide.no here.

A third company (NetCom) also provide this service. Their subscription services is very similar to Telenor's, except they don't offer usage only.

Common for all of these is that you'll need a USB modem that your provider will provide as part of your subscription. Some providers let you use your mobile phone or your PC modem with a SIM card instead.

Are you deploying applications to Windows Vista, and doing ok?

Jon Arild Torresdal — Tue, 13 May 2008 20:08:42 GMT

Update: This article is not limited to deploying applications to Vista, but also when changing files that are only writeable by administrators (e.g. files in the Program Files directory).

If you have not made specific changes to your installation to support UAC in Vista, this is for you. I hope this will help others to not spend many days of confusion, desperation and sleepless nights like I have.

Lately I've been getting reports from some of my coworkers running Windows Vista that the app.config file in our application was not updated after installation. I've also experienced this myself. We've solved them without knowing exactly how (which I hate). Typically "I uninstalled the application and installed to a different location, and it worked!" Well, this is a workaround that are not particulary popular to any of our customers, so I dived into the problem to find the exact reason.

To save you some reading I'll just say that the reason and solution is to provide your application with a manifest file to handle UAC in Vista as described in this article: Create and Embed an Application Manifest (UAC) I haven't actually tried this out yet, but I trust the article to be correct :-)

Virtualization
But that's not the interesting part. The interesting part is what happens (or what can happen) if you don't have an application manifest for this purpose? If your superstitious you might think that something weird is going on with your computer. If you're like me you KNOW that something weird is going on with your computer, and you need to find it and fix it. The weird thing is Vista virtualization.

"File virtualization addresses the situation where an application relies on the ability to store a file, such as a configuration file, in a system location typically writeable only by administrators"

"Virtualization is implemented to improve application compatibility problems for applications running as a standard user on Windows Vista."

As I write this I have exactly this problem on my computer. I've installed our application (Contiki ECM) to this location: C:\Program Files\CMA Contiki AS\Contiki ECM\. This is where I've always install our app. So today I installed and checked that the app.config was updated with all my changes done during installation. And it hadn't. Actually it was updated, but not with my changes. It looked like an old file that I had some time ago. How could this happen?

Textpad .vs. Notepad
To make this even more confusing, I discovered this reading the file in TextPad, but when opening the file in Notepad it was fine. All my changes was there! So I thought that this was a TextPad problem. I started our application, but it turns out that when .Net tries to read the config file, it reads the same old stuff as TextPad did, resulting in old config to be loaded into our app. Why does Notepad get the right file? I don't know. Please tell me if you do.

The solution
So after Googling for a while I found this article on MSDN, and things started to make sense. But first let me show you what it looks like on my computer. The file causing my headaches are Contiki.Windows.Application.exe.config.

Do you see when it's modified? 08:56 this morning, but the file that is being opened is from 13th of April! How do I know? Well, if you have a look at the virtualized files which you'll find in C:\Users\Username\AppData\Local\VirtualStore\[Your path], it might help out. So I looked at C:\Users\1jontor\AppData\Local\VirtualStore\Program Files\CMA Contiki AS\Contiki ECM and found this:

Modified back in April. I opened this file and it was exactly what I saw in TextPad. But this file is the same both in TextPad and Notepad! :-)

Explorer gives you a shortcut to the Virtual Store folder if any files are virtualized, and will show this button in Explorer:

The end
A couple of notes from the MSDN article:

" When you enumerate resources in folders and in the registry, Windows Vista will merge global file/folder and registry keys into a single list. In this merged view, the global (protected) resource is listed along with the virtualized resource."

"The virtual copy will always be present to the application first.... even if [some file] is updated"

Even if the file is updated!? Who though of that? Was that really such a good idea? I can see what they where thinking, but it would have saved me a lot of work getting an error message saying "Access denied"...

To end this off I picked another good quote from the article:

"Microsoft intends to remove virtualization from future versions of the Windows operating system as more applications are migrated to Windows Vista."

I appreciate that... :-)

Problem with Reg-Free COM and ClickOnce

Jon Arild Torresdal — Tue, 06 May 2008 22:05:41 GMT

Today I was struggling to figure out why I kept getting this error message when downloading a ClickOnce app:

The file 'C:\Documents and Settings\[user]\Local Settings\Temp\2\Deployment\RCAX3K4W.T8E\GRY27V4Q.GBE\SomeDll.dll' already exists.

First, let me just say that I don't use Visual Studio to create the ClickOnce deployment, but the Mage.exe utility. If you use Visual Studio this problem will be solved on Publish in Visual Studio.

I knew there was only one of this dll present on the server where ClickOnce downloads its files, but still it insisted that it was already downloaded. I checked my application manifest and there where only one entry for my dll. The dll causing this problem is a COM dll with generated manifest files (Reg-Free COM) as described here.

I started to inspect the Reg-Free COM manifest file generated by Visual Studio. It turns out that this manifest file use the same syntax as ClickOnce, or actually ClickOnce use the existing manifest schema that was introduced with Windows XP and Reg-Free COM. So in my ClickOnce manifest I had this:

...
<file name="My.dll" size="1121280">
<hash>
    <dsig:Transforms>
      <dsig:Transform Algorithm="urn:schemas-microsoft-com:HashTransforms.Identity" />
    </dsig:Transforms>
    <dsig:DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1" />
    <dsig:DigestValue>UwpS1EpkD7aJk3uWdOtX+3BHiOA=</dsig:DigestValue>
</hash>
</file>
...

...and in my Reg-Free COM manifest I had this:

...
<file name="My.dll" asmv2:size="1121280">
    <hash xmlns="urn:schemas-microsoft-com:asm.v2">
      <dsig:Transforms>
        <dsig:Transform Algorithm="urn:schemas-microsoft-com:HashTransforms.Identity" />
      </dsig:Transforms>
      <dsig:DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1" />
      <dsig:DigestValue>UwpS1EpkD7aJk3uWdOtX+3BHiOA=</dsig:DigestValue>
    </hash>
    <typelib tlbid="{2d5e2d34-bed5-4b9f-9793-a31e26e6806e}" version="4.1" helpdir="" resourceid="0" flags="HASDISKIMAGE" />
    ...COM Registry Entries...
</file>
...

It's exactly the same as ClickOnce except from the <typelib> part which is used by Reg-Free COM. ClickOnce treated this as any other ClickOnce manifest, and correctly stated that the file is about to be downloaded twice. The surprising part for me though was that the Reg-Free COM manifest was read at all. Since the deployment manifest is only linked to the application manifest...

So what's the solution? Just delete the file entry in the ClickOnce manifest and re-sign it! Deleting can be done like this:

ApplicationManifest manifest = (ApplicationManifest) ManifestReader.ReadManifest("ApplicationManifest", @"My.exe.manifest", false);
FileReference fileRef = manifest.FileReferences.FindTargetPath("My.dll");
manifest.FileReferences.Remove(fileRef);
ManifestWriter.WriteManifest(manifest);

Curious about how Visual Studio solves this? It merges all the manifest files into the application manifest and use the imported manifest entries instead (or something like that). At least that happened when I created a sample program to test this.