When project schedule run amuck
One thing I can't really resist getting amusement from the is the development cycle of Helm 4, the Windows hosting control panel suite by Web Host Automation. A while ago, I was looking for some control panel software to run on my server (which has since been parted out) and ended up buying a license of Helm 3. In the process of researching the options, I found out about the history of Helm 4's development cycle and became rather intrigued. In the Windows server market, the choices for control panels are pretty slim, so while I ended up buying Helm over some of the others (Plesk, DotNetPanel), it wasn't exactly because it wasn't the best one, but rather the lesser of other evils (DotNetPanel probably would have been notice, but not as mature).
Version 4 of their product has been in development with continually delayed released dates for over a year and a half now. Initially, they had talked about a beta release mid-last year, with an official release in late 2005. That never happened. Fast forward to October 2006, Helm 4 is still being the Duke Nukem Forever of the webhosting industry. They did finally release a "beta" version in in July, and had posted about a final release of October 2nd. October 2nd came, and literally that day, they pushed it back to the 9th. However, the kicker is they said it was 90% complete and that there would be an announcement about timing on the 9th (not a release). How is a product that has been in development for over a year, in beta for 3 months, and supposed to have RTM'd that day only 90% complete? And if you quantify it as 90% considering all that, how do you expect to make enough progress in a week? As you might guess, the week delay to October 9th became an additional two week delay to October 23rd and then an additional three week delay until November 13th.
They now have released Beta 4, and my opinion is that 4 betas is too many. If it takes you 4 test releases and you still haven't nailed things down, is it really anywhere near being releasable, or for that matter, stable? I've seen a number of product releases (from other programs) that were labeled as a Release Candidate that are anything but that. The Wikipedia definition has always been my opinion of what qualifies as a beta, an more importantly, a release candidate:
"The term release candidate refers to a final product, ready to release unless fatal bugs emerge. In this stage, the product features all designed functionalities and no known showstopper class bugs. At this phase the product is usually code complete."
A release candidate should be capable of turning into the official release of no showstoping, fatal issues arise within a week or so of testing (for smaller products, not big things like Operating Systems... they could take longer).
The thing that struck me most about WHA was how loyal many of their users are despite almost no project management and a poor track record of continually missing deadlines. The users actually defended them! I love our Community Server users, but we would never get away with that kind of track record. There would be chaos in the streets if we set hard dates and continually pushed them back. Especially if we had official release delays of a year in length. The amazing thing is that their customers are usually business owners as well... surely they must understand a missed deadline would affect their own business. I just don't get it.
With the release of the betas, I did try out Helm 4 on their online demo and was anything but impressed. It had the same confusing interface, it seemed slow and bloated, with a seemingly heavy dependence on third-party UI components, and not a lot of "Web 2.0" functionality built into it. This is what people have been waiting for? This is the future of the program I bought? What was all that time spent on?
Throughout the delays, there were a number of mentionings that it was due to features being stolen and such... they had to go back to the drawing board to one-up the competition. In my opinion, they were most likely referring to DotNetPanel, though I find it difficult to believe in stolen features when DNP has been out with a stable release for a while, and Helm has yet to have an official release of those features. So if they aren't first, can you really call them stolen?
I may sound like I am bitter. Perhaps I am. I bought Helm without actually getting to try Helm 4, and then I was disappointed. Either way, I am now no longer using my Helm license, and I'm instead waiting for them to finally release Helm 4 so I can upgrade my license and then sell it (if I sell it now, it won't qualify for an upgrade). Then as a developer, I am just amazed a small business could operate like that. It doesn't seem as though there is a lot of business experience backing the company, and not a good recipe for a long-term future. I was also just in a mood to stur *** up.