On Wednesday last, Niels stood up in front of 380 Umbraco developers and announced that v5 was being dropped. It was quite a bombshell. You can read more about the details elsewhere (read a good description of what's going on, watch the keynote, or read the comments on the original blog post for a flavour of people's varied reactions to the announcement), so I'm not going to into that here. I'm just trying to share my thoughts as a freelancer whose business is based on Umbraco.

Firstly, an alternative description of what is happening was suggested to me at Codegarden: versions 4 and 5 of Umbraco are going to be merged over the next number of months. Perhaps Umbraco HQ's aversion to "spin" meant a brutally honest approach, but maybe the reactions wouldn't have been so extreme if the decision had been presented in a slightly different light.

Umbraco is returning to its roots by returning to the community, which has been sidelined over the past while due to the development of Umbraco version 5. Developers who had been waiting for version 5 to mature before starting on their next package or idea can now focus on version 4 (myself included). And all of the pending work with version 4 which had been put aside because it was due to be dropped can now be picked back up. Everyone can focus on version 4 and take it forward.

For a lot of developers, myself included, Umbraco version 5 was to be the next step in our professional development. We were going to get to use the latest technologies like razor and (especially) MVC. The idea of a ground-up rewrite with a beautiful architecture was certainly a seductive one. But it was just too hard to use compared to version 4. (You can also read about some of the technical issues with the codebase).

Shannon has already demo-ed a working version of MVC with Umbraco version 4. This isn't in the core just yet, but I'm expecting to see it there very soon. Razor is already a part of version 4, thanks mostly to lots of hard work by Gareth Evans on the razor engine. The XML cache is not going away, so XSLT is still available for those who use it (again, myself included). This means that there is even more choice for developers when creating websites with Umbraco - use the tools that you and your team are most comfortable with. Personally, I plan on getting into razor over the next few months, but I'm not dropping XSLT - for me it's the perfect tool for translating XML into HTML.

So my conclusions?

While I feel bad for those who put a lot of work into Umbraco version 5 only to have the rug pulled from under them, I am mostly relieved. Having built hive providers for version 5 and built a basic site using it also, I know how complex it is. Version 4 is much easier to use and simpler to understand. It's a proven product, and one that is only going to improve. The team have learned a lot from their experiences with developing version 5 and this is not lost - it will eventually lead to improvements in version 4. The best bits from version 5 will all end up in version 4. And meanwhile I can get back to doing what I love to do - create great websites and web applications using Umbraco.

While this may not be a popular thing to say in the Umbraco community: this is business. I have built my business on Umbraco, and while I contribute to the community where I can, my continued use of Umbraco boils down to it being the best and most profitable way I know of building great websites.

Umbraco makes my day job fun, it makes me money, and it keeps my clients happy. While that's true I'll continue to support it and use it as best I can.

Related links

Never rewrite the code from scratch

the single most important quality of code is maintainability