Startup Boston EdTech Panel

Attended an interesting Panel discussion as part of the StartupBoston week.  An interesting group that clearly all believed in the value of education.  I had a few takeaways:

• There is plenty of scope for advanced technology (AI) in the learnign process but it is not going to replace the teacher any time soon.
• Universities have a poor understanding of what they do.  Award degrees might be the obvious answer but there are some many other components to it.  Prepare a student for a career,  create good citizens, the list could go on.  This leads to ambiguity in requirements for support systems for colleges.
• Education as a vertical is a little hobbled by desire 'to do no harm', that stifles innovation.

Overall I found the session very interesting.  I think that Boston is in a unique position to really make a difference in the education field and I would live to be part of that.

Refactoring is the most abused term

I was directed towards this article today and it reminded me of a personal rant of mine that I need to share.

Refactoring was a term made famous by Martin Fowler's book.  None of the refactorings discussed in that book are very complex.  The whole point is that the changes made can be proven to make no change to the functionality of the program.  These refactorings can be chained together to create significantly different code base but there is a chain of incremental changes rather than a rework.

How many times have you heard people say they wish to refactor the code base to reduce technical debt but what they really mean is to redesign it?  Redesign is not Refactoring!  They are different words and they have different meanings, don't abuse them!

Power Shell in 4 steps

I love the stuff that they put in the simple-talk newsletter.  This series on Power Shell One-liners looks good.

Flexible css layouts

I have been doing a little more html layout recently. I find it incredibly time consuming, mainly because I have not done enough of it. Too much change something and see what happens. This most true with trying to layout content in columns. I found a couple of good references for this today so I am immortalizing them here.

• Holy Grail of page layouts
• One True Layout

Wish List App Part 1

The Visual Studio templates for Windows store apps are pretty deep.  They probably do much more than most people want but they do get something up immediately which is great.  Run the template, rename a few things and you have a running app - I can see Wish Lists!

The big plus for me here is that all the XAML has been written for me.

The next step though is to look into the Share contract.  There is a template for that!  Pretty soon after that though we get to the issue that we need to persist our wish lists in some way.  The obvious way is probably in the cloud but I want the ability to work off the net and want to keep it local for now.  The next obvious solution is to use the serialization capabilities to turn the wish list into XML and store it on the local disk.  I want to wrap that up in an abstraction so that I can sync with the cloud or just save straight to the cloud at a later date.  I like the concepts in DDD here so will use the Repository pattern for my data model.  Note also that XML Serialization is not supporting in store apps but the DataContract serialization is.  I love unit test too so I start by creating a unit test project and start implementing my Repository.

The first noticeable difference is the use of the async methods.  Any operations that might take a while should be asynchronous so that the user does not have to wait.  This is already very simple with the async keyword that propagates around your code.  Just make sure that in your tests you wait for these tasks so that things happen in the order you expect them to.

Wish List Windows 8 App Series Introduction

I am planning to start a series of posts documenting my progress through implementing a Windows 8 app.  There is plenty here that is completely new to me so don't expect anything ground breaking.  This is more about giving some general guidance and helping others over the pitfalls that I fell into.

The high level view of the app is that it allows the user to collect lists of products that they wish that they could buy.  This will be stored in the cloud so there will be an Azure component to this and ultimately a website to provide access from platforms other than Windows 8.  The primary mechanism for adding to their wish lists will be through sharing URI's via the share contract.  Long term there might be some help in finding good deals for those items but one thing at a time.

Logi Analytics package

Spent a day finding out about the Logi analytics framework.  It seems pretty friendly to integrate with, all web based.  We are not talking big data here but probably a good place to start if you need to integrate with some pretty charts.