I have always liked Graffiti. Even before they had officially announced support for VistaDB I was a fan of the design and liked the fact that it was a commercial product.
I was tired of having to manually patch other blog software all the time, keep track of their latest exploits, patches, etc and really just wanted a product that I could use and it would take care of itself.
Community wants the product
The amazing thing is that after the lack of product development customers still support Graffiti. The community actually wants the product to succeed. In a very well thought out post about the situation Lee Dumond makes an impassioned plea for Telligent to Open Source Graffiti to allow the community to embrace it.
Despite all that’s happened, there are still lots of us out here that want to see Graffiti succeed in one form or another. I strongly believe that if Telligent were to throw the Graffiti code base up on CodePlex tomorrow, there would be dozens, if not hundreds, of developers all over it within hours. Within a week or two we’d be seeing some serious code commits. Inside of a month most of the important but still-missing features will have been implemented. And going forward, Graffiti would be well positioned to take its place as the de facto .NET CMS solution.
I don't know if that many people would be willing to work on the project at this point, but it is a very interesting argument. Perhaps a multiple license version where there is an Open version, and a Closed supported version. Of course that only works when the majority of the work comes from the company, if the community takes off with it they will not tolerate all that work going back into a closed version lightly.
Telligent should be doing everything they can to embrace these passionate developers from their community. They don't seem to understand how lucky they are that people are even taking the time to post and fight for Graffiti. Community is a rare thing to build, and once it goes sour it rarely comes back. I think they are blowing a fantastic opportunity, but that is just my opinion.
VistaDB Moving to BlogEngine
Yes, we are moving to DotNetBlogEngine for our blog. They have a solid release cycle, and a history of listening to their users. Yes, they do support VistaDB (including VistaDB Express). Does this mean they are perfect? No, but I think we are much more likely to see updates and community support.
I expect us to complete our migration over the next week or so.
Getting your data out of Graffiti
I was originally going the hardest route you could possibly take to migrate the data - migrating from one schema to the other. Then I read a blog post about a tool to export your Graffiti data to BlogML. Aha! Why bother with one database schema to another when they are both blogs? Move out from one, and into the other.
Curt has a nice simple tool that will get you to an XML file you can then import into Blog Engine.
Import into BlogEngine
The import was a little strange. It is a click once deployed app that talks to your website over a service interface. That seemed really strange to me since it was all right here on my machine, but I can see the use for people running it against an already installed system. It would have seemed to be easier to me to let the user upload the XML file and run it all on the server end, but it works.
Tags didn't import, and a few other minor things, but it got over the bodies and comments. That is by far the largest chunk of work.
What to do about trackbacks and permalinks?
The biggest problem I am facing right now is what to do with the dozens of folders and files that Graffiti generated for permalinks and trackbacks? We may either do a URL rewrite like we did for YAF, or maybe just update these files with 301 redirects to get Google to flag as permanently moved. I would love to hear from anyone else who is doing the same thing. So far everyone I have talked to has basically told me they just lost the links - ouch. That would hurt bad.
Social Media Links
Then we have the added problem of DotNetShoutOut, DotNetKicks, Delicious, etc. All of them are voted on and tracked per the full url, something that will obviously change with the move to the new engine. I would hate to lose all that valuable data, but right now I don't see an easy fix.
EDIT
If you are reading this now you are looking at BlogEngine.net. The sharing links below all work again as well.
Subscribe