Acronyms have always filled me with fear ever since my university Software Engineering class where we'd read case studies but with the names changed in a humourous way, like SNAPPERS for an automated processed through a photography lab. What it stands for now escapes me, but leaves me frightened nether-the-less. The telco industry is filled with them, the software industry has plenty go spare too and I'm stuck in the middle.
After reading JPs recent entry about the word 'platform' now being an overloaded word, I remembered what the telco industry was currently going through. I get the impression sometimes that in the industry if it's not worth putting a committee together to solve a problem, then there is no problem.
Take your typical Service Delivery Platform (SDP), talk about a problem that doesn't need solving! SDPs are specific to the telco industry according to Wikipedia. But, why are they specific? If SDPs are so central to (and I cringe as I type this) 'Telco 2.0' why isn't anyone else doing it? Why can't I just buy one? For that matter, why isn't there an open source SDP I can run?
I really like Urmy's "complex is lots of simple" approach. I look at some of the newish companies coming out and they seem to have something in common. David James said something which stuck with me about making your product x, the best damm product x anyone has seen. Take Twitter for example, what they do is really good, and it's simple to use and code against.
When building a generic framework, I'd like to see lots of loosely connected, easy to use web APIs to make it so easy to plug an application together, you could build it before you had time to put a committee together. Yes, those newish start ups do have the luxury of having no legacy systems like telco's do, but that doesn't give anyone permission to shroud those legacy systems in the mysteries of acronyms.
2 comments:
There are, roughly speaking, open source SDPs; "Project Firebird" isn't far off as far as the app server component goes, and there's the SEMS-NG open source SIP router/media server project to go underneath.
There are several commercial ones; IBM springs to mind.
Thank Alex, yes you certainly could cobble together an SDP from several open source projects. I'm not aware of those you mention, I'll check those out.
Robbie
Post a Comment