One throat to choke

Isn't it better to have one throat to choke?

The explosions of the personal computer saw two hardware and operating system vendor models emerge. The first was the IBM compatible or “PC” camp where hardware vendors could produce components to standards and other vendors package these bits into a saleable product. With any luck the operating system ran on top of it fairly well. The other model was the Apple way and, at the time, it was the lone proponent of building the hardware and software under one roof.

So, 20 some odd years later which is better? And what can we learn form this?

While there is a lot to be said for “lowest dollar cost” I think the “overall hassle cost” is often under estimate in the first model. In this model there are a significant number of “seams”. These “seams” are the boundaries of responsibilities and interactions between different companies providing the full solution to the customer. In the “PC” model this means that if my printer stopped working because the driver for my webcam did something this problem is all but unsolvable. No party is responsible and ultimately I have to be the IT expert to get anything done. A very high hassle cost.

In the second model, the Apple Way, the vendor is responsible for supplying a full solution to the consumer. Great effort is undertaken to eliminate and reduce the “seams”. In the case of many Apple products this may appear to be a closed system but there are many advantages in terms of quality and experience. Whilst the trade off is sometime seen as proprietary the other option is trying to synchronise your Nokia phone with Microsoft Outlook on a Dell laptop. A feat yet to be achieved by mankind or at least anyone in my household.

I have oversimplified these to illustrate the issue and do agree that neither model is perfect in the extreme. However as we might see the Apple market share in double figures and a stock price with triple digit growth, we might ask ourselves what we can learn from this?

If we apply this to a more tangible problem we might see a better way of doing things.

Let’s buy some enterprise software.

In the ideal world the best application would be run by the best IT provider on the best hardware and software so the end user gets what they need. This is the best business solution. Off we go and traditionally select a vendor, hardware, software, database, etc... In a perfect world our IT department would bid against outsourcing providers for this as well.  At some distant point in the future the solution is deployed and the world is a better place. Right? Well, not always, or not even most of the time.

In the same way as you cannot phone Microsoft to fix your printer when the drivers go wrong, the end business user becomes the ultimate problem resolution coordinator.  IT blames the vendor, the vendor blames the network and the finger pointing continues.  Selecting a solution with this mixed bag of offerings is going to get snagged on “seams” that make themselves evident in the day to day operations.But what is the alternative?

In the Apple model everything is under one roof. The quality of the hardware and software are provided in a package that is designed to meet the end users need. This oversight of the solutions components means delivery can take place in a better way than the transitional, lowest cost assembly of standard components, route has.  I would say that this model is what is behind the SaaS (software as a service) model.

In the non consumer world the best equivalent solution would be when you have everything  under one roof (software, hardware and most importantly the company the wrote the software) and consequently if something were to go wrong there would be “one throat to choke” and a minimum of finger pointing.  Problems get resolved faster and the hassle cost is kept to a minimum.

In these instances it is truly providing a solution to the market rather than a jumble of bits cobbled together with significant solution seams.  The Apple way is relevant to even the largest enterprise software purchase.