A suggestion for Ancestry: There needs to be some sort of process to implement corrections in data transcribed from census reports. We all know that many census reports suffer from poor legibility, and the system in place to make corrections is usually quite sufficient. But, take into consideration the following scenario:
On the 1900 census for Auburn Twp, Clark Co., Illinois, you will find one James W Burkett on page 10 of this enumeration. His family is listed below him, however when viewing the summery for this family on this report James is listed as Burkett while the rest of his family is listed as Murphy. Murphy is the surname of the family listed immediately below the Burkett family. This particular census is clearly written and there is no question as to the distinction between the Burkett family and the Murphy family. Clearly in this instance the error lies with whoever transcribed the census, and it is not due to an illegible enumeration. Going through the usual method of reporting corrections merely adds an alternative name to the listing, it does not correct what is an obvious error. Anyone who has been doing serious genealogy work for any amount of time knows how errors are like snowballs. They just keep getting bigger. Sooner or later some inattentive "researcher" will have James Burkett's family attached to the Murphy family in his data and the next thing you know it will be all over the place. What needs to be done in a case like this is not to offer an alternative name or cursory correction, but the data needs to be changed by Ancestry to reflect what is actually written.