The web has been a bottom up, grass roots phenomena, that is why so many people have web servers and there have been so many startups. That is why Linux and Apache have been able to compete - because things have been bottom up, something that performs well as singleton, individual machines have been competitive.
But the sorts of things that people now expect from the internet has increased dramatically. This is good and necessary, because it means that the internet is a richer, more complex and more interesting place. It means that the ways that we can interact with each other are also becoming richer and more complex. This is part of the vision of the internet that many of us hope for.
But this requires that the open source packages become richer and more complex as well, in order to support the richer and more complex internet.
Folks can complain about how complex things break, and are unreliable, etc... However that is what engineers are supposed to do - deal with complexity, deal with reliability - good engineering is about creating a good solution to a complex problem, not insisting that the problem be dumbed down.
The problem space of conquering the internet now needs a more top down, architectural approach - not just well designed bottom up solutions. This is what Microsoft is doing, but providing a top down architecture in.NET
I think that the article hits the nail right on the head - but the problem for open source is more general than just apache and web services. Open source is being attacked by Microsoft with a top down strategy - Microsoft has an architecture that ties everything together, and has a vision for scalable, complex systems. In the enterprise, this is Active Directory and Exchange Server. In the web business it is.NET.
Meanwhile, the open source world consists of little islands of folks working on small projects that fit into a philosophical/ideological "architecture", but no technical architecture. This is the nature of Open Source's distributed, somewhat anarchistic development model - and it is also the achilles heel that Microsoft has correctly identified and is actively attacking.
Microsoft plans to bury us by _properly_ raising the requirements for complexity and richness, and this making balkanized open source tools irrelevant - and in fact, they don't make us irrelevant, it is the open source community's own hubris, and lack of global vision that will make us irrelevant.
Is there a real open source alternative to Exchange Server? Is there an open source alternative to Active Directory. No and no.
Is it necessary? YES!!! Because in any complex environment, it is control of the core services that glues everything that gives you the "high ground". Microsoft is marginalizing Unix/Linux by taking this high ground in the enterprise, and also taking it over in the web sevices world.
The open source world has lots of programmers and engineers, but not enough architects. And that is what Microsoft is betting on - and every time that somebody says "we don't need web services, we have PHP!", or "Perl is way better than Java!", it just validates that strategy of playing for the technological high ground, and leaving the open source hacks to their efficient, but ultimately MICKEY MOUSE tools.
The world needs more complex, more serious tools. For open source to remain relevant, we need to prove that we aren't just an anarchic bunch of punk coders, but people who actually have the talent, experience and rigor to be architects working for a common technical vision.
Sun may be an evil empire wannabe, but if we sign up with this wannabe, at least we'll be supporting someone who _needs_ us around to fight Microsoft, instead of idiotically making clones of tools from a company that actively wants to squash us. The only reason anyone would want to do this is hubris: to prove we can make a better.NET than Microsoft.
But every enemy that Microsoft has BURIED, has died because they tried to fight a battle on Microsoft's terms. And that is _exactly_ what cloning.NET amounts to. The world needs more complex web services, but not on Microsoft's terms.
Once again, the open source movement is playing bottom up, and they've let Microsoft define the terms of "up".
Slashdot Mirror
The web has been a bottom up, grass roots phenomena, that is why so many people have web servers and there have been so many startups. That is why Linux and Apache have been able to compete - because things have been bottom up, something that performs well as singleton, individual machines have been competitive.
But the sorts of things that people now expect from the internet has increased dramatically. This is good and necessary, because it means that the internet is a richer, more complex and more interesting place. It means that the ways that we can interact with each other are also becoming richer and more complex. This is part of the vision of the internet that many of us hope for.
But this requires that the open source packages become richer and more complex as well, in order to support the richer and more complex internet.
Folks can complain about how complex things break, and are unreliable, etc... However that is what engineers are supposed to do - deal with complexity, deal with reliability - good engineering is about creating a good solution to a complex problem, not insisting that the problem be dumbed down.
The problem space of conquering the internet now needs a more top down, architectural approach - not just well designed bottom up solutions. This is what Microsoft is doing, but providing a top down architecture in .NET
I think that the article hits the nail right on the head - but the problem for open source is more general than just apache and web services. Open source is being attacked by Microsoft with a top down strategy - Microsoft has an architecture that ties everything together, and has a vision for scalable, complex systems. In the enterprise, this is Active Directory and Exchange Server. In the web business it is .NET.
Meanwhile, the open source world consists of little islands of folks working on small projects that fit into a philosophical/ideological "architecture", but no technical architecture. This is the nature of Open Source's distributed, somewhat anarchistic development model - and it is also the achilles heel that Microsoft has correctly identified and is actively attacking.
Microsoft plans to bury us by _properly_ raising the requirements for complexity and richness, and this making balkanized open source tools irrelevant - and in fact, they don't make us irrelevant, it is the open source community's own hubris, and lack of global vision that will make us irrelevant.
Is there a real open source alternative to Exchange Server? Is there an open source alternative to Active Directory. No and no.
Is it necessary? YES!!! Because in any complex environment, it is control of the core services that glues everything that gives you the "high ground". Microsoft is marginalizing Unix/Linux by taking this high ground in the enterprise, and also taking it over in the web sevices world.
The open source world has lots of programmers and engineers, but not enough architects. And that is what Microsoft is betting on - and every time that somebody says "we don't need web services, we have PHP!", or "Perl is way better than Java!", it just validates that strategy of playing for the technological high ground, and leaving the open source hacks to their efficient, but ultimately MICKEY MOUSE tools.
The world needs more complex, more serious tools. For open source to remain relevant, we need to prove that we aren't just an anarchic bunch of punk coders, but people who actually have the talent, experience and rigor to be architects working for a common technical vision.
Sun may be an evil empire wannabe, but if we sign up with this wannabe, at least we'll be supporting someone who _needs_ us around to fight Microsoft, instead of idiotically making clones of tools from a company that actively wants to squash us. The only reason anyone would want to do this is hubris: to prove we can make a better .NET than Microsoft.
But every enemy that Microsoft has BURIED, has died because they tried to fight a battle on Microsoft's terms. And that is _exactly_ what cloning .NET amounts to. The world needs more complex web services, but not on Microsoft's terms.
Once again, the open source movement is playing bottom up, and they've let Microsoft define the terms of "up".
Wake up folks!