As you probably know the archive.clarionmag.com site has been offline for several months now. It broke badly when we moved the site to a new server. My initial efforts to get the archive back online were unsuccessful and eventually I realized that a proper fix was going to take a lot more work.
When I first wrote what is now the archive site I began with Castle Monorail, a Ruby on Rails-inspired framework for .NET web development. But shortly thereafter Microsoft announced its ASP.NET MVC framework, and I eventually switched to that with the idea that it would be a bit more mainstream and better supported. That was the right decision - Castle Monorail has languished, and ASP.NET MVC (which is now part of ASP.NET vNext) is bigger and better than ever.
It wasn't a completely clean break in favor of Microsoft tooling, however. I needed a reliable data layer, and back then Entity Framework was still pretty green. I had previous experience with Hibernate and Java, so I picked up the .NET version (NHibernate) and another toolkit called Sharp Architecture which provided a framework for using NHibernate with ASP.NET MVC.
Like Monorail, Sharp Architecture hasn't gotten much love recently. When the archive site first went down I tried to update its Visual Studio solution with current versions of the many different libraries it uses, but I ran into a brick wall of dependency errors especially around Sharp Architecture. NuGet is a wonderful tool, but it can't work miracles.
So I decided to rip out Sharp Architecture and some of its related tooling and upgrade everything to MVC 5. Although Entity Framework has pulled ahead of NHibernate in recent years, I chose to leave the EF migration for another time. NHibernate is still well supported and a new data layer would mean an unnecessary delay getting the site back online.
I'm through most of the technical hurdles now. I have a working data layer again, with a slightly different implementation of the repository pattern. The old site used Castle Windsor for dependency injection (DI); I yanked that in favor of Autofac. Actually at first I tried to do without DI at all; why, I asked myself, introduce unnecessary complication? But while creating a proof of concept solution that simply read data from the database and dumped it out to a web page, I had problems maintaining an open database session for the lifetime of the page request. That's just the sort of thing that DI containers do quite well, and Autofac does especially well. But that's another, longer story.
There's a major change on the UI side as well. The original archive site uses ASP.NET syntax with a number of HTML helpers, but back in MVC 3 Microsoft introduced the very cool Razor syntax for web pages. So as I bring individual pages back online I'm also rewriting them using Razor.
As you may recall the archive site used a radius-cornered layout, and it did so via a number of formatting tricks as radiused corners were not widely available in CSS at the time. In the event I have to fiddle much (or at all) with the new site to get the appearance right I may also switch over to a mobile-friendly Foundation-based layout (besides which, radiused corners just aren't that cool anymore). I've been using Foundation in a consulting project and I've been quite impressed with it and with the SCSS-based approach to style sheets. That too is a subject for another time.
Meanwhile I'm working away at the rewrite. I'm not prepared to give an ETA yet but I'm pleased with the progress.
I very much appreciate everyone's patience. It's been a challenging and stressful time in the magazine's history, and I hope it won't take too much longer to get the archive site back online. Meanwhile if you already have full rights to the archive I can set you up so you can search the PDFs on this site. Just drop me a note on this page or send me an email.