Category Archives: technology

I just finished watching DimeCast
#15.  It completely blew my mind.  Part of me really likes the elegance
of the syntax of using Linq to query the FS.  The rest of me can’t imagine the
performance of joining files to directories in such a manner would be efficient. 
I need to dig in and do some side by side comparisons and get to the bottom of this. 
Sounds like the makings of an in-depth blog post.

Simply Awesome

Yesterday I posted about
my experience with the Errata submission process at Wrox.  I promised an update
and this is it.

Within an hour of my submission, I received an email reply from someone in the customer
service department.  It turns out that my observation about the missing figure
had already been noticed.  If I would have looked harder, I would have found
the replacement figure in the form of a JPG file in the downloads section for the
book.  I still believe that it should have been listed in the errata section
- after all it IS an erratum.

I also received an email from Jim Minatel. 
It seems that the comment functionality of my blog was down so he emailed instead. 
I was very satisfied ith the service I got from everyone involved and am happy to
say that I have had my faith in Wrox restored.  I’ll continue to buy books from
them when I have an opportunity and I will continue to hold them to a high standard
of excellence – as I do with everyone I recommend.

Let me preface this by saying that I understand people make mistakes.  I make
them all the time.  I expect mistakes in technical books.  This is NOT a
rant about the quality of any particular book or technical books in general.

Today I was reading a particularly interesting section in a technical book and ran
into a figure that was clearly incorrect.  After making sure it wasn’t simply
a case of the figure reference being off by a number or two, I decided to take a browse
over to the Wrox site for the book and check out the Errata page.  This is where
my journey becomes very frustrating.

Problem #1 – When I search for a book using the exact title, I expect it to be the
FIRST hit, not the fourth.

Ok, so I find the title, bring up the book’s page, and click on the Errata link. 
So far so good.  I’m a little troubled by the fact that the book has only a few
items listed – I know I’ve run into more errors than this and I’m only in Chapter
2.  Either I’m the only person reading the book or something is broken with the
errata process.  I click on the “errata form” link.  I expected to get to
a page that clearly indicated that I was submitting an errata for a particular book. 
Instead I’m at a generic “ask a question” page.  I thought this was strange enough
that I went back and searched for a completely different book to see if the link itself
was broken.  Nope – all the books take you to this generic page.

At this point I’m starting to see the issues with the Errata submission process. 
I’m imagining some English major doing an internship getting an email and yelling
at her screen something like “I’m glad that you noticed that ‘obtuse’ was misspelled
on page 578, but IN WHICH F**ING BOOK?”.  So, I try to help the intern out and
provide the book’s title and printing number along with my errata submission in the
“ask your question” box.

I move on to the next *required* section which requests that I enter a category. 
I eventually find an option for “figure error” which seems to fit the bill. 
Selecting this enables a text box where I can put in the book’s title and ISBN number. 
The text boxes are empty.

Problem #2 – Don’t make me enter data which you have no excuse for not knowing. 
I clicked on a link on your website that contained all the information you are requesting. 
Don’t make me do your work for you.

Problem #3 – x1000 when we’re dealing with a web development book.  Perhaps you
should read some of your own books from time to time – you may learn how to make a
website that doesn’t suck.

Deep Breath.

Now, I click the submit button.  Feeling rather pleased with myself that I took
the time to report the errata in spite of all the challenges.  Oh, what is this
- they want me to create an account?  This is optional right?  Nope. 
They want my address and phone number to submit a freaking errata?  Fail.

Problem #4 – When someone is trying to help you out, you should be thankful and appreciative. 
This should not be viewed as an opportunity to pad your marketing database.

Against my better judgment, I filled out the silly little account creation form. 
I didn’t give any real information except for my email address.  Submit. 
We’re finally done, right?  Wrong.

I get redirected to yet another page that gives me some troubleshooting steps that
I can perform myself before submitting my question.  The first on the list -
Have you checked the Errata page for this book?  Congratulations – you just bought
yourself a blog post.  Epic Fail.

I did get an email saying that I’ll hear from someone in “support” within 24 hours. 
I’ll update if/when my submission ever makes it to the errata page.

Updated here.

I just installed the Release Candidate of Firefox 3 on my MacBook and my Vista bootcamp partition.  My first impression is that it looks very shiny, but I’m not sure I can last log without a google toolbar version that is compatible.

Over the past two weeks, I’ve been adding a bunch of new tools to the old toolkit.  First was Ninject,
which I decided to take out for a spin after seeing Nate‘s
excellent talk at the Cleveland Day of .Net.  That started me down a grand path
of goodness.  I had been using xUnit in
yet another attempt to get unit testing off the ground.  I had tried in the past,
but kept failing when it came to doing anything complex.  It turns out I was
trying to do dependency injection by hand – and failing miserably at it.  With
Ninject, a lot of the resistance to developing for testability is removed and writing
testable code and the unit tests that go along with them are easy/fun/addictive -
you pick the adjective.

With my new found passion for unit testing, I looked for other tools that could help.  NCover was
a natural fit – not only am I striving to always have green unit tests, they need
to have a high level of code coverage.  Fast forward a couple days and I discover NDepend -
quite by accident – and fell in love with the extremely detailed and insightful metrics. 
The most recent addition to the toolkit is testdriven.net.  It brings all the
tools together in a nice convenient right-click package.  I’ll have to re-evaluate
things once ReSharper 4.0 goes RTM and xUnit supports it, but for now, my toolbox
overflows with goodness and programming is fun again.

I’m planning on attending, assuming I can get domestic clearance.  My former colleague Nate
Kohari will be speaking.  My former employer is
a sponsor.  It should be interesting.