New ways to version everywhere

The big problem with Configuration Management, I think, is that everybody thinks it’s an IT thing and nothing to do with the business.

This isn’t true, of course: sending the internal version of a document to an external customer or working to an outdated version of a contract can be as fatal – or more fatal – to the business as releasing a piece of software without the latest bug fixes. Perhaps the problem is that evangelists for configuration management often come from IT, and who wants to accept advice from IT even if it’s right? And, of course, configuration management needs a bit of forethought about what’s important to delivering a business service, what’s related to it, and what might go wrong – and no-one likes thinking about what might go wrong (your boss saying “I think you’re thinking a bit negatively there” is usually the first death-knell of a promising career).

To its credit, Perforce, with its “version everything” campaign (I was at the Perforce Roadtrip 2012 in London), is trying to do something about this. Whether “versioning” is the right term is a pub discussion for configuration management professionals; what’s important, for everyone else, is recognising that there is seldom just one version of a business asset and that managing the versions – so the right people always get given the right version – is an important aspect of business governance.

Leaving aside Perforce’s rock-solid underlying technology (it makes a change to hear conference delegates with only externals to grumble about), used by people like Google because it scales so well, Perforce has introduced 2 new products that perhaps take it into the general business space. Well, perhaps one new product, as Commons is really just configuration management made more accessible, with a simplified “drag and drop” interface and some open source components on top of the Perforce engine (although Mark Warren of Perforce points out that the open source struff is “really just framework and enablement for the web browser client rather than something users would be aware of”). It’s “Drag-and-Drop Content Sharing Meets Enterprise Control”, Perforce says. However, Chronicle (to quote a Perforce support technician) “really feels like a new product for the first time”. Chronicle is a sophisticated web content management tool (still built on established Perforce configuration management technology) with version management built in – “Manage the Flow of Website Change”, according to Perforce. Which doesn’t mean that it does things you can’t do with established web CMS (content management system) tools; it just means that you can do them yourself, cheaply, without having to employ expensive CMS or configuration management consultants who’ll give you a version management system that works but is probably a continuing maintenance overhead.

I don’t think there’s much point in working through the use cases for Chronicle and Commons here – they’re both on the web for you to have a look at and play with, as time limited hosted services. Hosted options mean that you can knock up a “proof of concept” without committing your employer. The message from Perforce is that one person spending just 30 mins a day trying to find stuff, or the right version of stuff, can cost the business some £4k pa – so there may well be some interest in a proof of concept…

I think that the “success metric” for versioning in the business is the CEO’s PA using something like Commons to find the appropriate current version of his/her master’s “message to the troops” and publishing it to the company via Chronicle. That, to me, seems entirely feasible.

So, what are the issues, if any? Well, culture. If supporting this stuff is in an IT silo, then the business won’t buy into it. If it isn’t, the business has to come to terms with some of the disciplines around managing assets and different versions of assets. Not managing configuration (or versioning) is always a cheaper and easier option, with no management overheads, and far more agility – until something goes horribly wrong, costing the business hundreds of thousands of pounds managing the business consequences.

So, what can Perforce do to promote success with versioning everywhere, including within the business? I’m sure it has thought of this, but it would be useful to see a body of auditors, PAs, managers and business analysts at Perforce Roadtrips, instead of just developers. Although if Perforce concentrates on winning new business evangelists – as it should – at the same time, it has to make sure its existing developer supporters don’t feel unloved.

One of the other things Perforce is doing is to expand and invigorate its partner program. In the past, I’ve heard people complain about Perforce going it alone and not supporting its partners. If true, this has to be overcome quickly (and I get the impression that Perforce is really working hard on doing this). Better a small slice of a large pie than a large slice of a small one and perhaps Perforce can learn from the embedded database market – it wins if “everyone” builds products on top of OEM’d Perforce versioning technology without ever mentioning Perforce’s name! This implies really strong “win-win” management of a rich partnering community – and I think that’s where Perforce is now investing.

Perforce’s real competition isn’t Subversion and GIT (there’s an interesting Perforce “federated server” technology story there I don’t have time to write today). It’s, for example, the legion of business workers using Word to print off different versions of stuff to store in filing cabinets, keep in the top drawer, stuff in the back of their wallet “just in case” (and, yes, I do know that Word does a bit of rather clunky version control these days). The driver for “versioning everywhere” is the amount of money businesses would save if these people could be persuaded to trust a business-friendly automated versioning system on top of rock-solid configuration management technology the business doesn’t see. To its credit, Perforce is starting to try to exploit this largely untapped market.