You would not believe how many times in the past year I've seen people connect their presentation layer to datasets. From books to articles to full on projects. Admittedly its convenient. But I'm not so sure its wise to do that.
Isn't the primary purpose of the DS to function as a cache? (Aside from the use in web services where it might be used as a true data transport via XML). And if the basic purpose of the DS is to be a cache, wouldn't it make sense for people to still follow OO design practices and create their own "custom" object structures which just happen to get data from a DS on occassion so as not to have to hit the database? Of course there is one part in this question that I can't quite figure - you could define your own object that inherits from a dataset. Seems odd to do that though doesn't it? Lastly, considering that I can cache my own "hand-made" objects if need be, wouldn't that mean I'd want to use a DS for some other reason - which gets us right back into the hot water of possibly binding the DS to UI elements? Thanks, Thomas You can read messages from the Advanced DOTNET archive, unsubscribe from Advanced DOTNET, or subscribe to other DevelopMentor lists at http://discuss.develop.com.
