An interesting subject

It is interesting that you mention SNA - I did work in/with IBM (SNA) systems but also Hitachi, Ericsson, Tandem and Memorex (ITT actually) 1280 controllers until the developers did walk out after a mismanagement (what else?). Ericsson was a huge player in SNA world in Europe until - the same story, a mismanagement?

Now, the worst company breaking SNA standards was, you guess it, IBM. Maybe for competitive reasons (definitely not because lack of skills, the architects and developers were - are first class), one day our controllers, terminals, printers just didn't work. I had all the fun showing that the IBM host, whatever was breaking the standard. I still have a lot of IBM friends from that time - it was a friendly competition which we all enjoyed - really!

So - standards are good as long as they don't get mixed with marketing. Is that possible? A real standard looks a lot like some used in multimedia - a container standard where everyone can add their own extension and to compete that way. I personally hate when someone breaks a set standard, you can't reach the developer (as it used to be), you try to reverse engineer what and why happened, and so on. And this seems to be more a rule today - especially in any standard which has anything to do with security or, even worse, a competing product.

Another problem with standards today is, they are just too big (even SNA "bible" was nothing) and full of loopholes. It almost seems that IT is taking the model from lawyers - instead of putting clear and short ground rules to place, current standards (6000+ pages anybody) would take years to clear even the obvious problems and even then have loopholes?

Selecting a product or a vendor based on standards - depends. Standards are not just technical - rules, regulations, laws, country standards, de facto standards, etc are also standards. And "standards" change. So, if the company has resources to deal with all the changes and "small" vendor exceptions, it doesn't really matter but if the company doesn't have the resources (or interest) the old saying "nobody gets fired by buying IBM" has a meaning. It's a cost issue. Right or wrong, another issue!

Click to read the article this is in response to.