Very interesting, now we have to get to the same point city development is, with software, and then lead it. Easier said then done
Heh, so that's what Pat looks like. This is fun.
Insightful. I've been looking into the Globus toolkit and the OGSA spec for a while now, and although Microsoft are a sponser of the technologies, they don't seem to be in anyway vocal about Grid computing benefits, unlike IBM for example. This talk highlights
many of the benefits Grids hope to achieve...
I know Microsoft are pushing the service-oriented Business Framework, which deals with some grid-esque functionality, but is the Globus vision something Microsoft still subscribes to?
Great points, and I agree that interchangability needs to exist in common areas of infrastructure, such as security, Name Resolution, number crunching, but I feel there will always be value in adding the unique capabilities that may break the standard,
but provide distinct value.
This is a quite interesting perspective to look at the present development and advancement in software with comparision to the development of civilization. History always reminds us of what we should look at for and what's ahead.
Even a computer scientist should know history, and reason to its parallels.
the video's on this site are really cool...i wish more bloggers out there would put stuff on their site like this...maybe in the future..gg guys
I know Microsoft are pushing the service-oriented Business Framework...
Ah, the irony. That would ensure long term sustainability, yes. But is that where they *really* want to go?
I think not.
Pat, that is an awesome post.
As Developers, we have been quitely neglecting this issue out of self-interest.
The day of reckoning is coming, and our profession will forever change. I only hope that the change happens first in the USA, and that American developers lead the way.
Remember what happened to those poor people who got left behind in the technological revolution that actually began during the European Renaissance. They became colonies.
There is one issue I would like to raise: the growing shift from Applicative to Declarative programming.
As a business programmer, I would like to focus on turning Requirements (a declaration of what the users want) into a declarative programming language (a declaration of what the computer must do to satisfy the user).
I want business software (as with SQL) that allows me to specify the results I want. I want business software that then goes away and does it, reporting on any problems it encounters.
THere are an infinite number of ways to re-invent the wheel. I just need something that rolls properly.
Herein lies, I believe, the key to the problem of interchangeability.
Of course, I could just be drinking too much caffine and watching too much television. It wouldn't be the first time
I agree that innovations in domain-specific languages could boost productivity & maintainability (as well and AOSD and other techniques that I believe compress + orthogonalize what you have to deal with)
However it is thinking along the lines of today's SOA discussions that will help you distinguish where to draw healthy boundaries between systems & thus establish rules on how to move them around. IMHO it is the effect of good ol' coupling and cohesion principles
playing within and across IT systems.
The cities metaphor works. And if the purpose of the metaphor is to find symmetries that would lead to learn more about systems, we could be looking around in others as well.
I expect not to be many differences with comparing it to evolution of anything that went from 1 to many to organized many. (Eg..single-cell organisms to multi-cell, cities, cultures)...
can we learn from those metaphors about what comes after? What has made systems that can evolve easily?
Hi Perhaps MS can make more of its applications for other plattforms, like its flightsims for mac or linux. Or does the interchangeability only apply when others have to conform to Microsoft. Yes, I have a mac... Kåre