Temat: Where is that big programming language ?

Where is that big programming language ?

The last 8 years not much has changed in the top 10 Languages except for...
read more...

Temat: Where is that big programming language ?

My 3 cents from Java perspective.

It seems there isn't any (big programming lanugage). As for current affairs it looks like programming lanugage 'market' (that 's not really a good word, but lets stick with it for now) has become fragmented. If you look on JVM landscapes there are couple of interesting languages that are competing with one another but none of them had really achieved wide adoption.

The one thing here is that for wide adoption (and by that I mean corporations / large companies adopting technology for their less or more critical projects) there is always need for:
* relatively low risk (which in the end means sufficient maturity),
* tools (and this is something that is lacking in most of modern languages)
* stability (takes time, at least several years)
* ease of finding skilled people (and this won't ever work to large extend with fragmented market).

On the other hand what is really important on the large scale is platform maturity - and for now it seems that JVM is here to stay - it meets business demends and lots of tooling grew around it through all these years. Switching to other language that is based on JVM is not really that beneficial. Even if we assume 50% (which I don't think is really possible, as huge part of programming still involves conceptual thinking) productivity increase caused only by switching to new super-duper language on a serious project (where we have lots of costs associated with planning, testing, documentation, globalization, etc etc) development costs may actually be around 20% - 25% of the total costs involved. So, if we are lucky, we got 10%-12% overall cost decrase. And that's after all the staff becomes as productive with new language, as with old plain Java. Taking into account all the risks involed - that does not sound like a sure bargain from the business perspective.

If I would to take a shot at guesing what is going to happen for couple more years I'd say - nothing serious. There will be more languages (seems that we are moving more and more into functional land) but fragmentation will increase, ergo C / Java will rule on such rankings for some time.

Temat: Where is that big programming language ?

What about frameworks? Article says that 'changes are slow' , but I am not sure about it. Java is only a core language, which is useless without a help of the other technologies (JMS, WebServices, Hibernate, Spring/EJB, Security), and I am not able to imagine enterprise project based only on Java. If you take that into account, you will see that new programming language is not so simple. Despite of it, how will somebody convince me that its solution is better than that tested by million devices and projects ? Maybe it's faster to improve current technologies with frameworks than write everything from scratch.

Temat: Where is that big programming language ?

Piotr Skoczek:
What about frameworks? Article says that 'changes are slow' , but I am not sure about it. Java is only a core language, which is useless without a help of the other technologies (JMS, WebServices, Hibernate, Spring/EJB, Security), and I am not able to imagine enterprise project based only on Java. If you take that into account, you will see that new programming language is not so simple. Despite of it, how will somebody convince me that its solution is better than that tested by million devices and projects ? Maybe it's faster to improve current technologies with frameworks than write everything from scratch.

This does not really affect Scala, Groovy or Kotlin (although it makes the point for Fantom or Ceylon - which IMHO seems to generate much more hype that it is worth - or, obviously, Clojure). That said, most of languages based on JVM allows at least partial Java interoperability (it is even the case with Jython or JRuby).

From my point of view - and that is, believing that although languages landscape is fragmented, runtimes will eventually converge to JVM (same history happened to processors, at the end of the day most of us will run on x86 - happened to Apple, HP, Sun/Oracle) - lack of frameworks / libraries will not be that critical.

Last but not least, there are lots of companies that suffer from NIH syndrom - for them lack of frameworks won't be that painful (libraries would still be useful though).

PS. That said - I agree that lack of frameworks (or inapplicability of existing frameworks) is one of valid reasons of slow language transitions.Pawel Dolega edytował(a) ten post dnia 14.05.12 o godzinie 21:45