Trial and error might be acceptable in the development stage but it shouldn't have gone live until it worked properly. Whatever the final benefits might be we will just have to wait for, in the meantime we are having to make do with a non friendly system where we cannot follow threads as easily as we did on the 'inferior' version and this default typewriter font is vile and difficult to read. I guess we should be grateful you don't work in the aviation industry.
Posts 7 | Views 433
Now I need to shower!
PMSL @ STS comment
I thought thats what beta testing was for? Get some real users in to come and play with the system. and break it by doing stupid unexpected things with it.
As for replicating 20,000 users yep thats simple there is software out there to carry out such testing.
I do know a bit about launching websites - i've been doing it for ten years, and that included the last MCN site which was up and down like a tart's undies when it first launched - and it took us a further six months after the core site launch to add the community bits. The site before that crashed continuously for the first month and that was built on IBM Websphere which was regarded as the very best technology available at the time. The truth is that the amount of time you have to load test after functional development is complete is driven by a variety of pressues and you cannot truly anticipate both load and usability issues until you launch publicly. That is what the internet is about these days - launching apps and finishing them with the help of the community that will use them. Of course, it relies on the goodwill and patience of the community to do that effectively.
See if by posting
The invisible posts appear???
Surely you only needed about ermm 1 to see where most of the problems laid?