So I wanted to share with you a little story that some of you may have heard me tell during presentations.
My father once purchased a very large number of Italian tiles. (Example for Dramatic effect)
(Why he purchased them is a bit of a long story so I will skip that part.) This was in Lebanon during the civil war in the 1980s and so as the political winds swung one way and the other he had to move the tiles that were in boxes from one storage facility to another to keep them from getting destroyed during the war.
He figured that each time he moved all the boxes of tiles about 10% were getting damaged and he moved them quite a few times. So one day he was talking to a friend and telling him about this problem of losing 10% of the tiles during each move. His friend said "Well I guess the good news is that you also have 10% less tile to move and store". That is when my father really got sad. The bad news was not the loss of the 10% it was the fact that he still needed to move and store all 100% because if he opened the boxes to see which tiles had been damaged that could break more tiles and it could compromise more tiles in the next move.
So how is this possibly related to Interoperability? Well you can think of data as the tiles. The more you move the data the more you can compromise it. However, that is not the bad news, the bad news is you may not know which part is compromised so it can put all your data into question.
Think about that for a bit!
Now there are ways to minimize the risk. For example, by working with formats and software you know well and you understand their strengths and weaknesses. You can also avoid it by working with a hub and spoke model so data is converted 2 times at a maximum or point to point so there is only one conversion. The advantage of a hub and spoke is you only need to support a single file format which is less costly for developers and users but the data fidelity is lower. The advantage of point to point is the data fidelity is much higher but it is also much more costly from the development point of view.
I have seen companies use Revit, Navisworks and IFC to try and minimize these issues. Or use things like DWG that has a well-established standard bearer in AutoCAD. However, at the end of the day the reality is you will always loose some tiles and if you are lucky you will know which boxes they are in. That is why trying to stay in the same place and not move those tiles at all is always your best option but like my father, many of us do not have that luxury.
You may remember that in my Usability and the APP blog post I talked about Seamless Integration and Data Access. Well Data Access is the interoperability piece of the puzzle. The simpler it is to access directly the information in the model and also to create and extend the existing information the less need there is to move that data and risk compromising it.
So do you have a tile story of your own?