Today I had an interesting conversation with my ex-Boss and good friend. ���� ���� �� �� ���� ģ���� ����ִ� - ����� ��ģ�� ģ���� ��ȭ���մϴ�. He asked me to summarize the current state of Java world these days as he was busy with databases and scientific computing. �״� �ڹ��� ���� ���¸��޶�� ��Ź ������ ��� �״����� �ٻ� ������ �����ͺ��̽��� �������� ������մϴ�. Here is my short take on it. ���Ⱑ �� ª�� ���翡�� �װ��մϴ�.

Ruby has become a cult like Apple. ��񿡴� ��Ʈó�� �� ����մϴ�.

PHP guys think they are winning over Java. php ���ڵ��� �� �̻� �ڹ� �׵��� �¸��մϴ�.

Java guys are busy writing tons of frameworks. �ٻ� ������� ���� �ڹ� ������ ��ũ �ۼ��մϴ�. I think some clear thoughts are missing. ���̾�� �� ���� �Ϻ� ����մϴ�.

JSF is a good attempt to solve MVC 2 problems with Struts and other Web frameworks. jsf�� ���� �õ� mvc�� ������ �ذ��ϱ� ��Ʈ�� �� ��Ÿ �� ������ ��ũ�մϴ�.

JSF is very verbose to code fast without an IDE. jsf�� �ſ� �ڼ��� �������̴��� �ڵ带 ��� ide�մϴ�.

Sun Java Studio Creator is a good IDE. �¾� �ڹ� ��Ʃ��� ũ�������ʹ� ���� ide�մϴ�. JSF is the future, with Ajax integration (yet to be built) and after lots of enhancements to core functionalities. jsf�� �̷��� ���̰� ajax ���� (���� ������ ����) �� ���� ���� ���� ����� �ٽ� ������մϴ�.

However it doesn't play well with JSP. �׷��� �� jsp ������ �ʴ´�. So I am sticking with JSP for now. �׷��� �� ���� jsp�� �����մϴ�. It gives me everything I need. ���� ���� �ʿ��մϴ�.

Architecturally when I look at all these IoC frameworks like Spring or PicoContainers, they don't solve the same problems we had with Service Locater pattern. �̷��� ��� ������ ��ũ�� �� �� ���������� ioc �� �Ǵ� picocontainersó��, �׵� ������ ������ �ذ���� ������ �츮�� ���ϰ� ���񽺸� ��ġ�մϴ�. So I don't think they add much value other than providing alternatives. �׷��� �׵��� ��ٱ��� ���� ���� ������ �ʾƿ� �̿��� ����� �����մϴ�.

That's my short opinionated summary. �װ� �� ª�� ���� ����մϴ�.

My friend replied: I think it is a good summary. �� ģ���� ���ߴ� : ���� ���� ���� ����մϴ�. I wonder why it has taken so long to get productive tools and frameworks in java. ���� �� �̷��� ���� �ɷȽ��ϴ� �� �� ������ ��ũ�� �ڹٸ� ����մϴ�. Seems that there is too much "not invented here" thinking. �� �ʹ� ���� ������ ���δ� "�� �߸� ��ٴ� ���⿡"���մϴ�. The proliferation of frameworks has just created camps of loyalists that make it more diffcult for developers to make good decisons about tools and design. ��� ���� ������ ��ũ�� ķ���� Ȯ���� ����� �� �� ���� ����� ��� �����ڸ����� ���� �� �����ο� ���� decisons �����մϴ�. It looks like a mess to me. ����� ���� �� �����ϴ�.

angsuman: Yes. angsuman : �׷����ϴ�. People are out to get fame and recognition. ������� �?�� �ν��� ��ġ���ϴ� ��. Building a new OS framework will get more name and recognition than working with an existing one. ���ο� � ü�� ���� ������ ��ũ�� ���ϴ� �ͺ��� �� ���� �̸��� �ν��� ���� 1. I think it is a disease of OSS environments. oss�� ���� ȯ���� ���մϴ�. The cost to start is low, so quality of most projects are low too. �� ���� ����� �����Ϸ���, �׷��� ��κ��� ������Ʈ�� ǰ���� ���� �ʹ��մϴ�. Even struts guys abandoned struts to go after new shiny toys. ��Ʈ�� ����� ���ο� ��¦ �̴� �Ŀ� �� ��Ʈ�� ������ �峭���մϴ�. Unfortunately IBM or Sun failed to assume leadership role in frameworks. �������� ibm�̳� �¾� �����߽��ϴ� ������ ��ũ�� ������ ������ �������մϴ�. JCP is too liberal and allows even competing frameworks. ������ �ʹ� ������ �������� ������ ��ũ�� ����մϴ�. Groovy is yet another overhyped piece. �� �ٸ� ��ǰ�� ���� ���� overhyped.

angsuman: A framework is as good as the software built on top of it. angsuman : �Ǽ��� ����Ʈ���� ������ ��ũ�� �Ȱ��� �����ΰ��ֽ��ϴ�. Companies avoid producing frameworks because frameworks doesn't pay the bills. ��� ��� ������ ��ũ ���� ��� ������ ������ ��ũ�� �������� �ʽ��ϴ�. So the mantle is on OSS guys! �׷��� �� ��Ʋ�� oss��!

That is in short the gist of the conversation. �Ѹ���� ���ϸ�, ������ ��ȭ�Դϴ�.

What is your take on the current state of affairs? ������ ���� ���¸� ���� �����Դϱ� �־��?