Earl said:
neither of which allow exceptions so it is not standard C++ as we know
it.
Many C++ programmers come from pre-exception days anyway. Whether
programming in exception enabled or disabled C++ doesn't make that big
a difference (IMHO). Also, I don't think EC++ will always prohibit
exceptions. Furthermore, e-processors may not require EC++. The other
thing that I can mention, is that the switch from C++ to Java is easier
than visa-versa. This is not because C++ is a bad language, but because
it supports many concepts that Java don't (or at least hasn't when it
started out). From experience, if we have a new Java programmer working
on a C++ project, we usually had a few nasty ones to look for
. All
said, many C++ programmers (who started out using C++) program
comfortably in both languages, after learning use of Javalib - as with
any library.
Regards,
Werner
The job market has changed a lot though in the last few years. There
are very few jobs now for C++ experts unless you also have business
knowledge.
I'm in favor of believing that a programmer doesn't have to be a domain
expert. He gathers knowledge of the required domain on the fly. There
must be some sort of inteface (body) between the domain and programmers
though, and this body should be able to understand enough about the
domain and software. Many programmers on the other hand, are good
communicators too (Growing up with OOP and UML). They slot into the
requirements world naturally. They ask: What do you want? not: Why?,
except if they see obvious logic flaws. I do agree with DB
principles/theory being important. I don't deem it that difficult to
learn (I'm not an expert, but have done normalisation and ER on odd
occation). I hear of people who even find it boring after a while -
something they haven't experienced with C++ yet.
For all the wonderful things boost has done, it has made a
certain type of programmer redundant.
How did boost make certain types of programmers redundant?