[erlang-questions] your first choice?

David Welton davidnwelton@REDACTED
Mon Feb 23 10:10:17 CET 2015

>>> Never pick a "framework", since they always limit you in the long run. If I
>>> *had* to, I would research nitrogen or n2o.
> A framework has technical merits, flaws, and limitations.
> It also has a community of people who (puts on cynic's hat)
> have a strong self-interest in seeing other people succeed
> with it in order to validate their self-investment in it.

In economics, the concept is called "positive network externalities"
or "network effects": http://en.wikipedia.org/wiki/Network_effect -
and it's not just about 'self validation'.  It's something that's a
factor in most programming languages and other tools we use.  The more
people use Erlang (or PHP, or Ruby or Visual Forth ++ or whatever),
the more libraries there are likely to be.  There will be more books,
and more jobs, and more people to hire if you create a company.  It's
a real and tangible benefit to have other people to ask for help, or
that write packages that you can use easily.  So trying to get other
people on board with whatever is probably in your own self-interest,
to some degree, long term, even though the actual benefits at the
margin are not large.

I'd highly recommend this book for a more thorough discussion of the
economics of 'information goods', which of course includes programming
languages: http://www.amazon.com/Information-Rules-Strategic-Network-Economy/dp/087584863X
- one of the authors is now the chief economist at Google.

> So it's not just what _you_ can do in a framework,
> it's what you can be _helped_ to do by the proponents of it.
> (And with some frameworks, it's what you can stealXXXXXcopy.)

Most good frameworks are just a nice collection of pieces without too
much glue to hold them together, so, yes, you can certainly take what
you like from them, and perhaps replace it with other bits and pieces
more to your liking.  This presumes, of course, that you're proficient
with the language and its libraries, and with the problem you're
trying to solve.  You might consider the purpose of a framework as
helping to get you to that point in the first place.

David N. Welton



More information about the erlang-questions mailing list