Chris Boos on Automation

I promised to give some free advice on how to select your Application Management provider. We’re talking about outsourcing maintenance and continuous improvement of custom applications here. Of course, selecting an appropriate provider is crucial if you want to achieve the desired ROI. And unfortunately it will take time to measure the actual improvement. So I tried so compile the known criteria that make a provider a match for you. Meaning: the criteria you can check before you make the deal.

Here’s the list:

• Select an Application Management Provider. Not just any outsourcer or software development supplier. Even if you know them and trust them. The reason: It takes a very specific approach to be successful at Application Management. One that is different from the economies of scale known by typical outsourcers. And it’s different from managing software development projects, too. Go ask for what you need and don’t stop short of it: an Application Management provider.
• Find a provider that understands your business. Of course, that still doesn’t mean the provider knows your application, which is unique. But it will eliminate a considerable amount of communication overhead and possible misunderstandings. After all, the provider will have to make changes to the application as required by your business departments. So he/she’d better known what they business talks about.
• Technology matters as well: there are different technological ecosystems (AJAX/Java/Unix vs. DotNet vs. SAP vs. Host …) and not all providers know how to optimize all kinds of applications. Development tools may be important, too, in some ecosystems where they are not standardized.
• Make sure the provider’s operating processes can be matched to your own. Nowadays, nine times out of ten that comes down to: you both speak and live ITIL. That will help to fit the provider in nicely.
• Check the provider’s methodology for setting up application management for your app. At the very least,  check that there is a methodology. As with any project, the success of application management is determined largely by how it begins.
• Check the contract offered to you:
  • Define the application and its interfaces. It should be obvious that a clear definition of the scope is necessary. However, sometimes boundaries of an application become fuzzy and their definition non-trivial.
  • Define the service levels you need (no less but also not more: high quality service is expensive – buy it only where it is needed)
  • How do you participate in the efficiency gains achieved by your provider? There should be a perspective for reduced costs for you.
• For those of you who still wonder how this blog entry got its headline, here’s the answer: before you outsource an application to an Application Management provider (“marry”), make sure you will always be able to source it back in (“divorce”). That requires an up to date documentation and the use of common development environments. Among other things. Make sure your Application Management provider offers you that – without you having to ask for it. Because if you have to ask for it, it is not common practise with this Application Management provider. And it will not become common practise just because of you. Instead, when you need it, you will find out that it never really worked. Insourcing will still be possible, but tiresome then.

This is my list of general rules for selecting an Application Management provider. Plus, there always are your specific circumstances to take into account.

Got more points to add to the list? Let me know! (info@syngenio.de)

Despite the fact that the new Google competitor Cuil is considered to be more energy efficient (because the black page background seems to be a big energy saver - if they get as big as Google of course <LINK>), Google is cutting edge when talking about the energy revolution. Sites like http://www.blackle.com or http://www.ecofree.org/ which claim to provide a more energy efficient Google search interface (at least on CRTs), by having had the black background idea ahead of cuil, are a minor issue to this revolution.

Google committet to be carbon-neutral by end of 2007 <READ HERE>, a goal, which they just missed, but as you might guess, they are working on that issue. Google needs to do somethink, as long as some of us believe that one Google search consumes as much as a 11-watt light bulb <READ HERE>.

There are rumours/news that Google buys large chunks of land (800 acres in Oklahoma <LINK> and 1000 acres in Iowa <LINK>). So what to build on a site being 4 kilometer long and 4 kilometer wide? A fancy new datacenter with some Hi-Tech, sophisticated cooling facility? Housings and facilities for an army of human taggers to annotate the whole earth? A theme park?

Nope, my I guess is the big Google has other plans for using these site. A hint could be found on a page Googles homepage titles “Powering a clean energy revolution“, where Google describes their effort toward reaching the goal of beeing carbon neutral. I like especially the part, where the author name some companies Google and Google.org are cooperating with. Do you see the point? It’s obvious.

So in the near future Google will not only provide grid-like computing services with GAE, but they will provide carbon-free computing services and this might be putting some pressure on some people investing billions of dollars at the moment in build hi-tech datacenters. Sounds like Google not only wants to index, map and annotate the earth, but they want to save it.

Good luck.