I’m not sure if this has caused a misunderstanding. In any case, most people would say that Java has strong typing i.e variables are bound to specific data types. Java also has static typing. So it has a static, strong type system.

Now, to answer your questions:

“Are you saying that developers should write tests when they use a language that *doesn’t* have a strong, static type system?”

Not only then. I am saying that it would nice if more developers wrote more tests generally. And if they don’t write tests, they may tend to have more bugs in their code when programming in, say, dynamically typed languages.

“Are you saying that programmers should do strong testing when they program in Java?”

See above. It would be nice if they did. It’s not the end of the world if the don’t - as long as they are capable of writing code that isn’t riddled with bugs.

“What if they don’t? Should they be fired?”

I’m not sure if I understand the tone of your question here. But getting into disciplinary procedures is probably a bit off-topic IMHO.

“Or are you arguing that programmers that don’t write tests should only use languages that *do* have a strong static type system?”

I’m saying that if you’re not going to write unit tests, then you may catch at least some bugs if you use a language which allows you to find some type errors at compile time.

I’m not sure I can imagine a developer so incompetent that they couldn’t get to grips with JUnit. It’s hardly rocket science, is it? But you’re right, functional programming languages like Haskell are interesting from the point of view of being able to check code “correctness” - and very different from imperative programming languages in that respect.

There aren’t many languages that do have a strong static type system and certainly none that do are widely used. Java, for example, has a very simplistic, weak static type system and detects most type errors at run time. Are you saying that programmers should do strong testing when they program in Java? What if they don’t? Should they be fired?

Or are you arguing that programmers that don’t write tests should only use languages that *do* have a strong static type system. I don’t think that will work — if a programmer is not competent enough to get to grips with JUnit I don’t think they’ll get very far with Haskell, the only languages I can think of that has a static type system strong enough to even come close to unit testing.

Where credential’s best role comes into play, I think, is in credibility of factual information. I’d love to be able to trace every fact in every discussion or publication back to primary sources, but I just don’t have the time. So I’ll have to look at how that fact fits with my current knowledge, and I’ll also have to assess the credibility of the source of the fact. For some kinds of facts, this is fairly easy. For others, less so, particularly when we get into softer topics where more judgement is called for.

A secondary factor is the tone and style of the communication. For example, for me, if a given post or essay seems to be a rant, that’ll affect how much credibility I extend to the facts in the rant.

Balanced against this is the subtle issue of accessibility of writing. The point of any piece of writing is to communicate something. While it may be proper to qualify and clarify every single aspect of a topic, that may also turn what might a short, concise, readable, even enjoyable essay into a long, boring, flat, unaccessible white paper. Sometimes appropriate, sometimes counterproductive. I tend to use this quote for my sig file, from the intro to one of my own writings:

“I’m going to make broad, sweeping generalizations and strong, declarative statements, because otherwise I’ll be here all night and this document will be four times longer and much less fun to read. Take it all with a grain of salt.”

And Mr. Obie Fernandez, I know Hani was out of line, but your comment hasn’t aided you in any way.

I believe he has a personal vendetta against me due to the way I called him out on his hypocrisy after he biled me earlier this year. He won’t dare bile me again, so he has been throwing darts via weblog comments. I think his comments above are lible and character assassination of the highest order. Luckily, he is incredibly sloppy when it comes to actual facts. I find it remarkable he can’t even get my age right! I am in my 30s, not early 20s, and verifiably so.

The other is significant lie is that I am “desperate for ThoughtWorks to assign me more Ruby work”. In fact we have as much Ruby work available as we can handle, probably more. In my role as a senior consultant and technologist I have been evangelizing Ruby *and* billing for Ruby on Rails related consulting fulltime for the last six months.

As recently as last weekend I was sitting in a lounge with well-known ThoughtWorker Paul Hammant, and he assured me that Hani is actually likable in person. I argued that no matter how likable he might be when you are speaking to him face-to-face, Hani is one of the most vile, despicable hypocrites of our extended social circle. His penchant for character assassination goes beyond satire into the realm of malicious, sociopathic behavior.