In a previous blog I asked the rhetorical question, "is leasing software really better than buying", rhetorical because in a real sense there is no choice but to lease commercial software right now, End User License Agreements (EULA) are, for all intents and purposes, more akin to leasing than buying for you rarely really own the software and you are usually forbidden from doing things to the software that you can routinely do with anything else you buy.

For instance, if you want to buy a car and then strip it down to it's component parts to determine how it works, you are completely free to do so. You can even rebuild the car with new and different, possibly custom parts and even resell that new car as your own. This isn't just legally possible, it is commonplace and a lot of automobile innovation has occurred because of it. However, if you do the equivalent of any of that with most of the commercial software you use, you are (probably) breaking the law.

But of course, nobody is forcing you to lease said software to begin with so it is hard to get too upset about it, right? I'm not so sure about that really, but I think I'll leave that bit alone for now.

The truth is that a EULA doesn't legally stop you from using the software under the Terms and Conditions, even if you decide you no longer want to pay maintenance or support. I've never seen a software vendor even try to enforce support payments, though some of the routine conditions is that to qualify for "upgrade" prices you need to ensure you are up to date with "support and maintenance" payments. The EULA doesn't really need to address long-term use, without software maintenance the application will degrade toward the point of uselessness within about five years. The problem isn't the application but the environment around the application that changes to the point that the original application no longer functions in some critical fashion. APIs change, hardware changes, even file formats change. If the changing environment isn't accounted for with software maintenance eventually the application will simply cease to function in a useful manner even if the application would continue to solve the business problem it was designed to.

That is the promise of Cloud Computing, that an application can be written against a non-changing (or at least backward compatible) Cloud API and stay working without worrying about the infrastructure. If the computing environment doesn't change then the application doesn't need to change and the need for application maintenance to deal with environmental changes is practically zero.

Well, I hate to be cynical but I can't see that happening any time soon, the business case to make such software is simply not there for the majority of software vendors. No corporation would willingly give up a lucrative revenue stream unless they have to, there has to be, as an old friend used to say, some "secret sauce" involved so that ownership of the code (the so-called Intellectual Property) remains with the vendor. Using cloud storage doesn't hurt most ECM vendors (with one notable exception of course, EMC² still makes the majority of it's money on storage) but using cloud APIs to perform computing tasks seems an incredibly risky course of action for a software company ... it's pretty hard to claim proprietary code when using public APIs.

One can envision a company attempting to use a Cloud Computing model where they are currently using an open source software (OSS) today, however as far as I can tell there are damn few commercially successful software vendors using OSS (except as the "bait"), it is a very difficult business model. Software development is very time consuming and really good software always starts at the design phase, however that is the stage where there is no revenue available at all and so there is a great commercial pressure to produce demonstrable code ... this rush to code leads to poor design choices that often are a real hindrance to use of the product. The plain fact of the matter is that if the OSS product cannot be used by a reasonably competent IT-savvy user without the assistance of the sponsoring company, it won't be a commercial success. (NOTE WELL, I'm not claiming there isn't good and useful OSS ECM software, only that there are few commercial successes using OSS as the model).

Nuxeo is, in my opinion, an example of a company that has not put enough effort into their OSS product ... I'm a veteran of both IT and ECM and yet once NuxeoDM was installed it was not at all obvious what I was supposed to do with it and it took a very long time just to figure out how to change the administrator password. I think the company is honest in it's attempt to provide the product but has produced something that pretty much forces somebody to purchase hours from a service provider to morph the free product into a useful offering. There may well be no license revenue but there are still significant costs and long-term concerns about the viability of the application and it seems to me that it is as much a proprietary system as non-OSS offerings.

Alfresco is an example of a bait-and-switch offering in my opinion. The free OSS code is old and outdated, it solves some relatively simple problems related to ECM but many of the really interesting problems are not solved in that code base and instead you are directed back toward the more traditional licensing of proprietary software.

Where cloud computing really comes into play are for the non-commercial uses of software, like Wikipedia. There is no concern that Wikipedia will see it's code or data stolen, both are already freely available. Entrepreneurs will find the cloud a great place to try things out but will, I think, quickly find that any commercially viable solution will have to be proprietary, Otherwise it will be the cloud providers themselves that will be pushing the cloud, aggressively pushing it to consumers. I think it would be a mistake for most ECM-programs to consider the cloud as a useful resource, even if the software vendors are offering that option.