This was only a few months ago. I was offering support for a particular reply (out of 3 different ones) and said that you can "do this, if this_result=that, etc" and was told off for not supplying working, tested code - even though the next paragraph began "But don't do it, it is too complicated and can be achieved simply by using Fred's response" then went into further detail about why I supported that response.

I was sent a private email by AK telling me off for this so I don't bother any more.

>Hi Nigel,
>
>>The problem I had on CompuServe was that the requirements for answering queries were too strict. You MUST put valid code in any response, you MUST check the code works first, etc. Basically, you MUST tell the enquirer EXACTLY how to solve the problem. I think people from UT will find that difficult. I prefer to point somebody in the direction where they can find the solution, not to tell them. That way, they learn MUCH more.
>
>That was many years ago... The requirements for technical replies nowadays are much more like the one here on the UT. It's perfectly OK to simply point people to a web site or a knowledgebase article, and if you have an idea, you can always start a reply with "Try..." or "I think that". You don't have to offer the perfect and final solution, any step towards a solution helps.
>
>Like here of course we prefer valid code, or at least a comment that you haven't tested the code, but it's no requirement. IF you can accept that you get corrected in a reply. *g*
>
>Christof