Systems Integration Philosophy - lessons learned from this class

I want to share some random thoughts I have about Systems Integration while sitting on the couch eating M&Ms.  Here are some things I've learned that will forever change my view on integration.

My first thought is that this is an extremely complex subject.  I never thought about systems integration much before coming into this class, but after taking this class and learning about the myriad of technologies available, it seems to me that no one can know much about all of these technologies, and the more you know the more superficial the knowledge is across the board, as the technologies are so varied and different.  In fact, I think you can make a whole undergrad or grad degree called MS Systems Integration, for example, and that still wouldn't be enough.

Also, as new and newer technologies emerge, you need more and more techniques for integrating systems, which is sort of a rehash of what I just said, but from a different angle.  The ERP is a relatively new advent, and think of all the technologies associated with it.  Think of all the integration technologies involving the web, databases, servers, etc.  Because there are so many new technologies, it's hard for anyone to keep up with them.  So like most technologies, specialization becomes very important.  So you can be a database integration expert, an application integration expert, an ERP implementer, but I doubt you can be an expert in all the kinds on integration, unless you're just an exceptional superstar.

Another thought - where would the world be without integration?  What good would all these technologies be if they cannot interoperate?  I always thought of technologies as more of standalone islands, but the key is getting them all to talk in a meaningful way.

Another - what does the future hold?  I wonder if the trend will continue of the technologies becoming more and more sprawling, or if we'll start to see more standards and fewer ways to have to integrate systems.  Will the integration of systems become more graphical, automated, and intuitive?  Or will they become more flexible and require coding?  Will they rely more on things like APIs where you don't need much back-end knowledge?  Will there more of a push to open source integration?  As Microsoft loses more and more of its grip on future market share, will systems integration become yet even more complex?  Is there going to be a large vendor that specializes in just integration?

Wow, it's enough to make your head spin.  I really enjoyed learning about all these, but the more I learn, I realize I know less and less.  But the first thing to learning is knowing what you don't know, realizing what's out there, and then you can pick and choose what fits your needs and become expert at it - and go from there.

Are there any other philosophical thoughts on systems integration out there? Am I right in assuming the present is extremely complex to warrant high specialization and even its own degree?  Will the future become even more or less complex?  As I finish my bag of M&Ms, I'm curious to know what you think.  I have a free bag of M&Ms to the best response.  Blog on!