Skip to Content

Language Surprises

One of the places I go to, when I’m trying to understand what is happening in the IT World is the O’Reilly Radar.  Tim often puts up charts containing data about his book sales, as a means of gauging the market, but this one via this, took me by surprise.

What’s Up?

Javascript is clearly the big winner, which I completely missed –  this I can only put down to me just considering it part of the “fabric” of the (IT) universe, but it must be surging on the back of AJAX.  The next surprise was SQL – SQL, to me, is generally, uncooperative, unexciting, and generally a “means to an end”, so what is making this one resurgent (anyone?)?  Finally Ruby – this is no surprise to me.  Ruby has been hanging in the wings for a number of years now, and is finally coming into it’s own on the back of frameworks such as Rails, and the fact that it is the only significant language out there that has its roots firmly in the South East Asian sector – it wouldn’t surprise me in the least if it goes all the way.

What’s Down?

The big clanger here is Java.  As far as book sales is any indication, then Java took a hammering last year, closely followed by C++, Visual Basic, and sadly – Perl.  Perl is still wallowing in it’s own self-destructive juices of the never-to-be-quite-ready-for-release Perl 6.

Watch this space

The title of Tims’ original blog is a little unfortunate in this context as I’m not interested in “Language Wars”, but  I most certainly am interested in understanding where things are going outside of the SAP micro climate, because you can bet – sooner or later, kicking and screaming even – those of us living in the micro climate are going to be affected by these trends.

With this in mind – the most striking macro-trend, to me, is the total market share, and trajectory of the Dynamic Languages.

You must be Logged on to comment or reply to a post.
  • Piers:

    Thanx for that info and links…I watch O’Reilly stuff just when someone send me the links -:)

    My thoughts…I have never used AJAX and have used just a little bit of JavaScript…But I don’t see it a Language…I see both more like a tool…Ruby is great I love it and I’m happy that is getting bigger…About Perl, used it for one of my books, but just to make some little examples…Python…Same as Perl -:)

    I agree with you that is not Languages War…I think it’s just a matter of taste…Some people like Visual Basic and some people like Python…It’s the programming language that makes a product great…It’s the programmer -:)



  • It took me by surprise too but if you see it from the point of view, that JS is the native language of every bodies favorite client it makes complete sense. More and more of our daily apps move into the browser and JS is the glue code to make this experience become interactive. IMHO for a web developer this is the _one_ language you have to know (beside HTML/CSS but we don’t count these as languages, right?).

    IIRC ActiveScript from Adobe is also very close to JS and there are plans to integrate it into Mozilla/Firefox.

    On the other hand take a look at GWT and its approach to get rid of this step child through code generators. Maybe this is the right way to go.

    Could you please remember to do this again next year for a comparison? 😉


    • Oliver,

      Small correction: there are no plans to integrate ActionScript in Mozilla. Simply, ActionScript and future version of JavaScript are based on the same standard ECMAScript 4.

      Adobe donated relevant sources of its Flash VM to help Mozilla community create new high performant JavaScript interpreter. See


  • Piers,

    The collapse of Java is not surprising. Here I see 3 major reasons.

    The first cause is Java 5 itself with all its new “features”. The factor of time works against Sun here: it was long religious resistance to alter language, then the inevitability of changes became obvious, finally to catch-up with C# Sun rolls-out the worst possible implementation of generics and the most buggy release of JDK ever. The new languages features come at very high price – in fact developers (at least, libraries/framework developers) have to re-learn language once again. “Generics by erasure” makes things even worse. Literally, developer must mentally perform work of compiler to understand cause of compilation errors. Wildcards is plain pain – I feel myself uncomfortable when I read someone’s code with intensive usage of generic’s wildcards (also I have several years of C++ background including STL and ATL). To the mix, the syntax becomes utterly verbose and redundant.

    In short, Sun created completely new language that is binary backward-compatible with previous versions, but logically is incompatible with my brains, both backward- and forward incompatible. I fear that I’m not alone here. So let me say this straight: Java 5 is a failure.

    To turn the tragedy into farce, Sun plans to add new pack of language features in upcoming (7?) release like closures and properties. 5 years ago I would take my hat off, but currently I can say for sure – this is suicidal act from Sun.  It’s impossible to collect features from C# 3.0 / Ruby / Python and keep language backward-compatible or the price would be too high to keep language useful at all. The arms race with US destroys USSR, the “language features race” with C# / Ruby will destroy Java. But the race is already on, so Java 7 will be even more dramatic failure.

    Secondly, J2EE 5 is too late to keep Java domination on server-side. J2EE before version 5 is/was unacceptably complex for majority of web applications (mid- and small size, at least). Sun/JCP failed to roll-out new specification before Ruby/RoR phenomena and .NET attacks J2EE steadily.

    Finally, Java on desktop… Should I comment here?

    P.S. Btw, totally agree with you that JavaScript growth was caused by Ajax hype, also I guess that JavaScript will have to share some points with ActionScript in near future due to growing popularity of Adobe Flex.

    • nice perspective. a little dramatic but your points make sense. I’m wondering how Abap fits into your view.

      The success of a language depends on more factors than its technical superiority like established developer community, customer adoption, mass market deployment…

      Maybe the success of platforms like Jboss or Websphere or Weblogic will offset the problems related to Sun approach to Java.

      • Andre,

        Regarding “dramatic” part — I understand that there are thousand of billions invested in Java applications, so obviously Java will stay alive for long time. But it will be “COBOL-like lifecycle phase”.

        “established developer community, customer adoption, mass market deploy” — all of this will keep Java (semi-)alive for very long time (I guess more then 10 years). But how many _new_ applications will be developed in Java? And, most importantly, in _new_ Java? This is a question…

        Every language has an idea behind it, when the value of idea is over the language start to collapse. Also if language creator itself destroys an idea then language collapsing even quicker. One of the main points of Java as language were simplicity and readability. And Java has these attributes… till version 5. Btw, idea of compile-time type checking is also questionable in modern world. At least, such languages need type inference — non existing Java feature. It keeps me wondering how whole proposal(s) for Java closures started without considering type inference at first hand…