Back to Development - Wed, 1 Oct 2008

Today I started back at development of Behold. It felt good to tackle some of the issues users found and reported to me over the last few months. I’m going to work on these first, and get another release out in a few weeks with some corrections.

I’ve had a lot of time over the last 8 months of website redevelopment to think about the right way to take Behold forward. The current version of Behold is fairly solid. It can handle almost all flavors of GEDCOM quite well, but is limited to GEDCOMs up to about 50,000 people because of speed and memory. That is a very important issue.

I was planning to do the Beta Testing, then release Version 1, and follow it with version 1.1 to add technology improvements, optimization and Unicode. But if I’m going to spend Beta testing time to test everything and document everything, I should do what I was planning for version 1.1 first.

There are a few things happening here:

1. The Unicode version of Delphi, Delphi 2009 just became available, and I’ve now purchased it. It makes sense to upgrade now, and test it thoroughly during Beta testing.
2. Virtual Treeview has been updated to work with Delphi 2009, and Virtual Treeview is *very fast*. I should replace the old ElPack treeview with it.
3. TRichView has been updated to work with Delphi 2009, but TRichView is the slowest and most memory intensive part of Behold. I’d like to take a stab at seeing if I could replace it with Virtual Treeview as well. I’m not sure if I can get VT to do everything that TRichView does, but the potential gain makes it worth a shot.
4. I may be able to make another big memory gain by using a real disk-based database to store the GEDCOM data for processing. This could be like a *fast* GEDCOM storage for Behold. I’ve found the likely database candidate to be Accuracer by AidAim. Their Delphi 2009 compatible version should be available very soon.

If I could get all that in before going to Beta and then to Version 1.0, then there really will only be Saving GEDCOMs to do before getting to the ultimate goal of editing and Version 2.0.

So hold onto your hats. We’re on our way!