First off, let me get something straight. I won't go into the ever-recurring theme of "certification value" and why one should or shouldn't not be certified, blablabla... All I want to show is that sometimes pragmatism and realism take over logics :)

In this post and this one, Francis is bragging about the merits of EJB 3.0 and telling people to stop wasting their time with the SCBCD certification because EJB 2.0 will soon be overhauled, and thus, the hardly gained certification will have no value at all.

To a certain extent, he is right. But what Francis is forgetting is that when a new technology comes out, it takes a subtantial amount of time to take off. Industry acceptance is not always straightforward, and even if it is, internal processes and custom practices usually have the last word. People don't use a technology because it is good or because it brings plenty of added value, people use a technology because it is useful to them for their particular needs. Times where we were using a technology just because it was trendy are (fortunately) over. People are now much more reluctant when it comes to make major changes to their systems. Remember the old saying: "If the thing ain't broke, don't fix it". If the whole stuff works with EJB 1.0 or 2.0 for that matter, why on earth would I turn everything upside down to use the latest tech that hasn't made its proofs yet and hasn't even been accepted by a majority of industry actors.

Moreover, as I said to Francis and as far as the certification is concerned, my opinion is that in order for engineers to solve problems efficiently, it is not only important for them to know the technology as it is today, it is also important that they know how it was yesterday!! This is what I call technological culture. It is all about the "Know your enemy" thing, where I compare enemies to the issues of the old specs that you have to master in order to comprehend why and how the new spec gets over them.