The Agile Micro ISV Blog

For the last 8 or 9 months or so, I’ve had the good fortune to be developing a new product called Ergo with the folk at Invu.
Ergo is a search aggregation, visualisation and collaboration tool
written for .Net 3.0 and Windows Vista. We’re currently at Beta 2 and a
free trial version is available here. Feedback, as ever, is most welcome.

Search

Ergo currently aggregates search from 8 sources;

• Local Machine
• Live Search
• Yahoo!
• Google
• Wikipedia
• Flickr
• You Tube
• RSS Feeds

The local machine search uses Microsoft’s Windows Desktop Search
technology to perform really fast searches of your machine. It is the
Desktop Search technology that is probably responsible for hard drive
activity when you’re not doing anything. You can configure what parts
of your system WDS indexes using a little known part of Windows, if you
press the Windows button, type “Indexing Options” and run it, you can
add whichever locations you would like to index.  With the arrival
of Vista SP1, Google will almost certainly write their own search
protocol for Windows, so you should be able to use any Google offering in place of WDS.

The RSS feeds search uses the RSS feeds that you have subscribed to
in Windows Explorer. I suspect this might be expanded to use other
methods of RSS aggregation, Google Reader, for example.

Visualisation

Once the search is complete, we use a custom clustering engine to
organise the results, and we offer numerous different ways to visualise
the results. For example, here are some of the ways I can visualise the
search results for the search term “Physics” on my local machine:

The Network View.

The Galaxy View

The Treemap View

I can then view the documents/web pages in any given cluster, along
with information about a particular document. I can then choose to
preview a document in a lightweight viewer/reader, or open the document
in the associated reader, Adobe Reader or Firefox for example.

There
is also support for collaboration and annotation built in to Ergo. Some
of the functionality, like the Ink support is targetted at the Tablet
PC market, but for instance, if I want to annotate a particular web
page, I can  store these annotations and send them to colleagues
using the XPS file format.

There’s a lot more to the application of course, but that’s the top and bottom of it. Feedback welcome.

—–
EXTENDED

.Net Reactor is a comprehensive protection tool for .Net assemblies, it’s web site is here.

Its key features, include:

- Intellectual property protection through obfuscation, IL modification and sophisticated native code wall techniques.

- A powerful and configurable licensing system, including trial-ware settings, trial to full conversion, SaaS etc.

- An SDK to integrate and automate licensing activities with your system.

So, what do I think of it?

I’ve been using .Net Reactor over the last week or two to protect a sizable WPF application that will be hitting public beta in the next couple of days. I’ve found a couple of wrinkles with Reactor;

1) It doesn’t play nicely with WPF applications. Obfuscation seems
to break resource locating, so most of our assemblies have only minimal
protection (decompilation protection). Merging the assemblies into one
also seems to break resource location.

2) Speed is an issue. As soon as you enable an option that requires
the binary (NAT) file to be generated, you’re hitting big startup
delays. We initially found that if all assemblies in the application
had binaries generated by Reactor, it would take around 20 seconds to
start. To get around this, we enabled only minimal protection on all
but the main exe. For the main exe, we enabled the trial lockout, thus
requiring the binary files to be generated.

The application itself is pretty solid, but there are problems:

1) The help file is mediocre at best. Perhaps not a problem, since
developers usually understand what it is happening,  and I
certainly managed to do pretty much everything I wanted to do without
even opening the help file.

2) Technical support is, as far as I can tell, non-existent. Both
myself and colleagues have emailed asking questions and reporting
problems, all appear to have been ignored. Maybe he’s on holiday, maybe
he’s dead, who knows. To compound the problem, there is no user forum.

So all in all, it is a pretty solid, regularly updated and completely unsupported product. Would I use it again? Yes.

The issues surrounding support I’m sure must be harming product
sales. They seem to be missing part of the equation in the Micro ISV
world:

Is it enough that your application does what it says it
does? No. Customers want to believe in you, in what you do. They want
to love your process as well as your product.

It’s been an interesting couple of weeks for the NSpec
project. For readers who don’t know what it is, it’s a behaviour
specification framework for .Net, specifically targeted at 
practitioners of Behaviour Driven Development. A while ago,  myself and some other active members of the BDD .Net community agreed to merge NSpec with NBehave, a cousin in the BDD .Net arena.

Joe Ocampo has spearheaded
the integration, and I’m happy to confirm that the integration is
complete, and NSpec itself will soon cease to be. Long live NBehave.

It’s new home is here.

Microsoft have also offered to sponsor the project, which is mighty nice of them.