I’ve been pretty hot on Amazon for a while now. No that’s not a stock tip, rather a statement of my amazement at their success and innovation as a company.
If you haven’t realized, Amazon has been branching out beyond being “a bookstore”, in fact they hate it when people contextualize them as such. You knew they sold electronics but did you know you can get groceries from them too?
An even more strange and interesting curiosity is the fact that that you can “subscribe” to discounted deliveries of toilet paper delivered free with Amazon Prime!
So it seems like a natural progression… Amazon’s moved from bookseller, to general retailer, to an online super Wall*Mart who can deliver fiber, suggest an appropriate book and routinely deliver appropriate accessories for the experience.
However, Amazon’s become more then any of that, much more… They’ve moved from being an online presence to enabling others, which is a powerful transition because it leverages other business’s success, avoiding the burden of their failures and acts as a force multiplier for their revenue!
First it was their marketplace, which is to ebay what Facebook is to MySpace. However, they’ve since moved beyond alternate definitions of ‘retailer’. Now, in the techno-sphere, Amazon’s best known for their WebServices which have been used building all kinds of interesting businesses. They’ve grown so influential in the Web2.0 world, that James Governor at RedMonk suggests IBM should consider purchasing Amazon!
So currently, it seems no company can compete with the breadth of Amazon’s offerings. However, Kevin Burton and Nick Carr have been discussing whether or not Google may soon challenge Amazon’s hosted offerings, aligning itself more fully against Amazon.
Now the argument centers on whether or not Google will open its compute resources, and Kevin asserts that neither Amazon or the potential Google offering is sufficient. As a ‘techie’ I understand his perspective, which I would insufficiently summarize as being “vendor lock-in is bad”.
However, having been introduced to the broader business perspectives, I disagree with his conclusion.
- Historically vendor lock-in hasn’t prohibited adoption when it’s strategically and tactically expedient (e.g. Microsoft, Token-Ring, Intel)
- I strongly believe OpenSource’s momentum is inevitable and will destabilize and nullify permenant vendor lock in (e.g. WINE, the recent opensource implementation of java, and numerous MS Office “import/export” techniques)
- Technology changes too fast to waste time “not doing” or “reimplementing the wheel”. This is less about the latest and greatest gadgets and more a statement of business need.
There’s a great phrase by Marc Hedlund which has stuck with me since I read it; “Jedi’s build their own lightsabers”. Marc leverages this pop culture meme to remind us to never outsource critical for your business components (You must read it!).
I believe the correlary suggests that any function, fundamental but incidental to your business value should be a candidate for procurement.
Even at the risk of “vendor lock in” there’s a list of trade-offs including “time to market” and “cost of competency” to consider. In Kevin’s ideal world it sounds like we’d all build from scratch and have unlimited time to implement, tweak and benchmark to our heart’s content.
Unfortunately, business necessitates an alternate reality and if expediency, simplicity and accuracy mean vendor constraint, so be it.
However, Kevin would be right to point out that many companies require “flexible on-demand IT” in order to succeed, even if they don’t realize it, trapped as they are in legacy models and processes.
When companies talk about outsourcing these components, or letting a vendor’s software product dictate their business & IT processes… I always check to make sure my lightsaber is close.