On 13 June 2011 10:21, Fabrizio Giudici <fabrizio.giud...@tidalwave.it>wrote:
> On 06/13/2011 02:40 AM, Ricky Clarkson wrote: > >> While if I look at my situation >>> I'd say that I won't find a single customer to whom I could sell my >>> hypothetical skills in Scala >>> >> To be fair, as a consultant you're probably generally working with >> people who are slightly behind the curve. I bet Java 5 and earlier >> come up for you too. >> > Declaring Java 5 EOL has been extraordinary effective :-) At least for the > customers I've dealt with in the past year (of course, your point has a > general validity) everybody I work with is aligned to Java 6. This also has > to do with the fact that - thanks God - I'm still able to pick my customers > at least partially. I can partition my customer in two classes, according to > a certain perspective: the short-term and long-term ones. The former class > is made of up to one-two weeks of consultancy and then bye-bye (typically a > course or a mentoring). The latter class is made by customers with whom I've > a continuous relationship - including the "main customer", a single one that > I change after a few years. Currently I'm working with him since three > years. Here my involvement is deeper and related to the products/services > under development. Once every a few months we make some exercise to "inject" > into the development new technologies and practices that we think can be > useful. Most are proposed by me, some by the customer himself. Some are ways > to drop legacy and most aren't beyond "slightly behind the curve", as Ricky > said, others could even be more to the edge - our decisions. Can't go into > details, but there's at least one "leading edge" technology involved in a > prototype. In all of this world, changing language is not an option, desired > or feasible. > > I'm wondering why Scala is seen as being so "scary" from a risk perspective. After all, at the very core it's just another way of generating bytecode that comes with a support library. We've been through all this before, Spring does it, Guice does it, XDoclet does it, IDEs can replace coding with menu options to generate code (and in the case of IntelliJ can then hide some of that generated boilerplate), most of the Jakarta projects replace large chunks of the standard library. Then we also have other projects that replace a focussed subset of the Java library known to be broken (JodaTime, for example). Or move from JUnit to TestNG and you're again writing some of your logic in XML. Why, then, should Scala be perceived any differently? Anyone who has used Hibernate or Spring is *already* using an alternate language to talk to the JVM. In many cases it's XML, sometimes it's string arguments in annotations - sacrificing type safety, easy refactoring, and a great deal of tool support. Scala at least has the advantage of generating that code at compile time and with debug symbols, so it already has a big advantage of Spring's runtime-wiring when it comes to coverage testing or static analysis (for example). Being statically-typed, it also means that you're only ever going to see the refactoring support get better and better over time. So what is it that people fear about Scala, when they're already so willing to move large chunks of logic into XML? -- You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to javaposse@googlegroups.com. To unsubscribe from this group, send email to javaposse+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/javaposse?hl=en.
