In many cases I have seen misleading blog posts or replies have their origins in some paper or information published by MS itself and not cleared up later. I have myself had two issues with MS on widely accepted problem issues - the problem with GUIDs and fragmentation, and the problem with # of tempdb files being equal to the # of cores. Granted i didnt speak to someone at the top at MS, mostly the CSS engineers and our TAM, but they were not able to give me a MS link that agreed to these two issues as often explained by well known authorities such as Kim and Paul. (Perhaps there is one now, this was a year or so ago).Where i work for example the management is not very well educated on who is an authority on SQL and who is not outside of MS, they go largely by the book. MS needs to update and correct documentation on a much more frequent basis, and perhaps remove what has been proven outdated.

On how the correction should be made, yes I think Paul is right, people can and should be called out when they have a wide audience and post misleading info. But perhaps writing to them personally would be better than doing so in public. Just my 2 cents.