A proposed solution of the different planets problem


First we should think about the problems that arise from the dilemma stated in my last blog entry (by the way read that one first to get the train of thought). When the end user is owner of the data this comes with the freedom to twist, change, mangle and calculate with the data like she or he wants to. But it also comes with the responsibility about it. The owner must take care about backup, about accuracy somehow, about privacy… Since the end user is most of the times busy of doing his own job he does not remember doing the IT department’s job too.

The consequences are well known and I think we all suffered from it occasionally.

So what solutions do exist? I think there are at least three …

Police State:
The IT department is named owner of the data. All PCs are taken away and replaced by what is called Thin Clients (beside the PCs of the executives who normally don’t accept being patronized that way ;-). All functions are only accessible via a web front end which in the end means that we are back to the good old host times.
Whenever pressure is forced onto a system, the system reacts and moves into a position with the least amount of pressure remaining. This is true in Physics and for the end user / IT department scenario. So the end user will react: Possibly they will buy their own PCs (I’ve heard from companies where employees bought equipment on the travel budget with the accomplishment of their direct manager. So if there is a need there will be a way, too) or do what ever they can to regain the old situation. Remember: They will do that not to cheat the IT department (OK, maybe a bit)  but to keep on being productive with the same luxury they had before.

Forget about the IT department and the bus loads of consultants they usually hire … just kidding. This model is the other extreme regarding the police state. The IT department is reduced to some guys using coal shovels to throw new harddisks into the arrays. The users are free to store any document on the server what so ever. IT provides a search engine to harvest what is on there.
This will not address all problems stated above (at least the backup thing), will lead to redundancy and out of date data… So in the end (and I think this is the destiny of all anarchic systems) nobody is happy with it.

In the middle:
Nothing is really black and nothing is really white so the solution lies in the middle (which is grey… not very attractive model but let’s accept this for the moment).
Next time I will lift the curtain how the best solutionat least in my point of view should look like...