The sites I work with also integrate into other pieces of software or custom code. This actually a huge reason to hold off on upgrading, because these integration points are what will cause the real problems. I don't mind updating a style or a few plugins, but if all our custom coding fails because the code it's anchored to has changed or is gone, then we're up the creek if we jump to the latest version too early.
However, thorough testing and a good staging site can catch these problems. Sometimes a simple backup / test on the forums is not enough - you have to test the entire site and its suite of applications. This is another reason I'm a strong advocate of unit testing -- these kind of things can usually be caught at the flip of a switch (or pound of a button).
Also, there's no guarantee that these problems are truly unique to their set up. If the problem caught them, it can catch others with custom code too - it's impossible to say.
Knowing that the site is still down 8? hours later with the team that they have is VERY alarming. Hopefully this doesn't paint the picture for those upgrading a lot of custom code.
This is obviously an extreme view coming from a developer. This should not be an issue for the vast majority of other license holders.
Perhaps I'm being misunderstood or I misunderstood the posters I replied to. I took their comments as they wouldn't upgrade their sites to 4.0 when the software was considered stable (this coming from the fact that nobody but the beta testers have access to the beta version). That's what it looked like to me. Of course I wouldn't upgrade my live site with a beta...but since IB isn't offering the beta to customers right now, that's not even an option anyway.
10-23-2009, 12:12 AM
Threaded Mode