dinopello wrote:Simon_R wrote:The cloud? We really ought to clarify what that means
Putting your data on a server managed by someone you dont know, in a place you dont know.
What could possibly go wrong
Just curious how old you are ? Because I feel the same way (I'm 48) but most of the younger people I know seem to have a different attitude or at least don't think about "where" their data is.
Gosh, I'm glad you guys cottoned onto my question. I asked it open endedly because it touches on a lot of things, one of which is this thing about being concerned about where your data actually is. There are other things as well.
I can use an example to illustrate one, one which doesn't have to do with theft, but access. Some people I know started up a business. It's a type of up-scale fast food. Anyway, their register system is 'cloudy', so all of their accounting is done online. That hasn't been a problem for them. What has been a problem is that the they can't find a connection fast enough for the system to run on and not slow to a crawl. When the place is really busy they are slowed by the glitchy system, which is slowed by not having a fast enough connection. It is certainly possible to write software that would work much better with what many consider a standard connection speed. It is, but where is it?
IBM has always been about intranet, or mainframe, or more local storage of data. As such, to go cloudy is a big venture for them. I mentioned open source because going cloudy these days almost always means some kind of integration of open source. Most of what's going on is somebody's forking of what was an open source project used in their proprietary way in order to do things for their customers. In terms of the cloud this has heretofore mostly been in what is called server virtualization. Going forward that's looking passe and will be replaced by something called Docker. I'm not certain that IBM is at the forefront of either of these. To compete it doesn't mean they have to be, but they do have to show an understanding of them. The challenges, as we've seen, are not all about security.
To compete they can make their systems easier to use, providing a better and more readily approachable interface. Oracle has been trying to do that for years. They've made great strides, but their stuff is still too complicated to adapt it to a MySQL database, instead of the flagship Oracle database, and throw at your average small business owner and expect them to sort it out and produce something that will work over the net. IBM could also address the kinds of problems the people I know are having, by making a smaller information packet footprint more secure over the net. It doesn't appear however that anybody is too concerned with how oldies like us feel about our data being somewhere other than on our own hard drive.
I don't like the big analytic approach that IBM does seem to be going for. They've proven that Watson has all kinds of applications this way. Watson learns, so it can be used in a lot of places. The trouble is that people tend to act differently when they face uncertain expectations. People who don't know what the rules are tend to make up much tougher and stricter rules, especially for behavior. We invent prudishness when it isn't necessary. Watson only reinforces that. I don't like it, but I don't seem to count.