One Runtime To Bind Them All

Sowbug writes "Here's some interesting Saturday night reading: a critical examination of many of the advertised benefits of .NET's CLR (Common Language Runtime) and the other technologies (MSIL, CLS, CTS) that make it possible. It's written from the perspective of a Java advocate, Osvaldo Pinali Doederlein. "

Comments

Re:Unbiased Articles?

[joto]

I believe this article is actually quite unbiased, considering the source it came from. It admits that CLR is probably better than JVM for implementing other languages than C# or Java, but that it's far from the holy grail MS makes it sound like.

Then it goes on to say that surely the JVM can be extended if this proves to be a selling point for .NET (which is probably true, but it kind of makes you wonder why sun haven't already done it).

Very nice article.

[DaveWood]

What a pleasure to see such a balanced, well-written and thorough analysis of the situation. I didn't see any great evidence of Java "advocacy" - this person appears extremely well-versed in langauge design and familiar with a good variety of languages, as well as more than willing to point out Java's flaws.

The author is saying pretty much what I figured, which is that .NET is much better than what MS has been doing in the past, however it's still just a sugared-up clone of J2EE, whose "cross-langauge" benefits are ultimately dubious and primarily a marketing invention.

I would also make the case that "unsafe" mode/pointer arithmetic is a flaw, but that's not the matter at hand. The high point of the article were these two paragraphs in the conclusion:

"Playing with the .NET SDK, the cross-language support looks impressive, but the illusion holds true only until realizing that all languages in the mix are virtually identical. Microsoft has actually invented the concept of skinnable language: changing a language's most superficial aspects, and claiming the result to be a new language. There is only One True Language that is C#, and "skins" offered by Microsoft and third parties. Just like in GUIs, these skins will alter the system's look and feel, add a few features, but never compete with a fully new toolkit."

For those quick to make an ignorant response, he's not saying more radical structural departures are impossible, though many are - but more often that diverging "client languages" suffer in performance and, in many cases, have been "embraced and extended" in order to become compatible. He goes on:

"There are, actually, many successful "common language runtimes", with names like Pentium, SPARC and others. Mainstream CPUs are equally fitted to very different languages as they only do the most fundamental, low-level operations, so they cannot be biased towards particular languages. There aren't many different ways to perform a conditional branch. However, there are radically different ways to support methods and functions, or most constructs found in high-level languages. The consequence is that every language needs different compilers and runtimes to implement their features, and different libraries to support their vision of software development."