Jun 262016
 
 26th June 2016  Posted by at 11:57 am 2 Responses »

Slightly belated Summer Solstice Greetings! It’s time to provide an update on the latest events and progress from Archi world. These are interesting times we live in, and it’s an interesting time of the year. Hence the title of this update – referring of course to Shakespeare’s play dealing with the themes of confusion, trickery, and dreams. And hopefully, as in the play, after a night of foolery and ambiguity, it all works out in the end.

Archi 3.3.2

I released a maintenance release of Archi, version 3.3.2, at the beginning of June. This contained a few bug fixes and some new features for the Visualiser and HTML reports. Both contributions were provided by Jean-Baptiste Sarrodie, now working for Arismore in Paris, who has over the years proved to be an absolutely amazing supporter for me personally, and tireless evangelist for Archi. As we move forward, J-B’s vision and involvement in the project will prove to be more vital. It was a pleasure to finally meet J-B in London at The Open Group’s event earlier this year.

There was quite a gap between the previous release, version 3.3.1 in October 2015, and 3.3.2 released this month, and the reason for this was to allow the dust to settle on last year’s release of The Open Group’s ArchiMate Exchange Format, and to wait for the official release of ArchiMate 3.0.

ArchiMate 3.0

The latest version of ArchiMate was officially released on June 14th. This is a significant development, as we shall see. I won’t go into all the details of what’s new and what’s changed but it’s as if a new broom has come in and swept up all the bad stuff, and re-arranged all the furniture at the same time, hopefully for the better. There are new concepts, such as Capability, Resource, Course of Action, a new Physical layer, relationships to relationships, and a whole lot more. This is a major update to the language. There’s a summary of what’s new here.

Archi, ArchiMate 3.0 and the future

Archi does not currently support ArchiMate 3.0, only version 2.1. I’m now seeing the same question being asked repeatedly on Twitter, in forums, in webinars, in emails, and in person – “When will Archi support ArchiMate 3.0?”.

As I type this, I’ve stopped to consider my response to this burning issue. There are a number of possible responses and I’m not sure which one to choose. Should I mention the significant amount of work involved in this? Should I ask where I might find the time for such a non-trivial task? Should I question whether I am prepared to work for several weeks, if not months, to single-handedly code, document, test, deploy and support a major new update to a piece of software used by hundreds (although it’s probably thousands) of professional enterprise architects. For free?

Let’s be clear about one thing – since it’s initial release in 2010, Archi has become an extremely popular tool in this domain, and many organisations (and some very major ones, too) use it and depend on it. Archi is downloaded on average about 3,000 times a month. Every month. Archi is a game changer and it is a major contributor, if not the major contributor, to the uptake of ArchiMate globally. This is a significant responsibility for one person. For free?

And let’s be clear about another thing – updating Archi to support ArchiMate 3.0 is not a trivial undertaking. It is not simply a matter of adding a few new concepts, compiling the code and then heading down the pub to celebrate. There are many factors to consider – a whole new refactoring of the code, backward compatibility, documentation, testing, asset management, build processes, and of course providing unpaid support. Typically, this process is managed by a team of employed developers, not one (unpaid) person. Some people have suggested that the work can be done in a weekend, or that their nephew has just learnt Java and will do it for a packet of chewing gum.

This has to change, and I have personally come to a crunch point. To be blunt, either I get funding to continue with Archi or I’ll do something else.

Having said that, there is hope. As Robert Fripp says:

In strange and uncertain times such as those we are living in, sometimes a reasonable person might despair. But hope is unreasonable and love is greater even than this. May we trust the inexpressible benevolence of the creative impulse.

Indeed, there is hope. But there are dark clouds, too.

Consider the overall strategy of promoting and supporting the ArchiMate language with an open source implementation and data format. A central plank in this strategy is The Open Group’s ArchiMate Exchange Format. This has proved to be a significant thing, with more and more tools supporting the format, and users discovering further uses for it and realising many benefits of an open data format.

However, the introduction of ArchiMate 3.0 means that a new version of the exchange format needs to be developed, and, indeed, this initiative is ongoing.

But this means that I will need to develop a new implementation of the exchange format in Archi. Archi currently supports the exchange format for ArchiMate 2.1. But, here’s the thing – Archi cannot implement and support a new version of the exchange format if it does not implement ArchiMate 3.0. Obviously, implementation of ArchiMate 3.0 is a pre-requisite for implementation of a new ArchiMate exchange format in Archi.

Furthermore, we need now to recognise that Archi has become an Enterprise in itself. Jean-Baptiste Sarrodie made the compelling case for this with reference to Tom Graves. To use Tom’s definition of an “Enterprise”, Archi has become

a bold endeavour; an emotive or motivational structure, bounded by shared-vision (purpose), shared-values and mutual commitments.

And, to me at least, it’s clear that Archi incorporates a powerful shared vision, and one held by many stakeholders. It is too important to fail.

As I said, there is hope. Hope lies in the fact that funding may have been secured from a benevolent source so that we can proceed with our bold endeavour. On the other hand, the dark clouds that I mentioned are manifesting as opposition to this funding from some not-so-benevolent quarters.

So, please, if you feel that you are part of this “bold endeavour”, speak up and assert your support for Archi. The future is open.

Phil Beauvoir, June 2016

Jun 052016
 
 5th June 2016  Posted by at 11:56 am Comments Off on The Archi Philosophy

Jean-Baptiste Sarrodie and I have been emailing back and forth for some time with some positive ideas and we now agree on what amounts to an “Archi Philosophy”, a set of principles that guide our future development of Archi and associated services. But these principles have already been written about in detail in the book, Rework, so all I can really do is talk about key points that relate to Archi and quote relevant passages from the book that resonate with us. I want to expand upon these principles and where we see Archi going (a.k.a the “roadmap”) in future blog posts, but, for now, regard Rework as our manual of truth and guiding light.

In this post I want to talk about open source and some of these guiding principles.

Archi’s code has a liberal open source licence, the MIT licence. It means that anyone can take the code and build a commercial product based on it. You can build a commercial product from the code, Microsoft can build a commercial product from the code, and your great aunt Edna can build a commercial product from the code. Heck, even I can build a commercial product from the code. So, what’s to stop somebody else from hijacking Archi and making something commercial from it? Actually, nothing. And, in fact, we want this liberal licence to stay in place because many organisations and developers have already built some interesting things based on Archi’s code, and they use it in a commercial setting. Also, the MIT licence is compatible with other licences.

Here’s a relevant passage from Rework that summarises our view:

Decommoditize your product

If you’re successful, people will try to copy what you do. It’s just a fact of life. But there’s a great way to protect yourself from copycats: Make you part of your product or service. Inject what’s unique about the way you think into what you sell. Decommoditize your product. Make it something no one else can offer.

Look at Zappos.com, a billion-dollar online shoe retailer. A pair of sneakers from Zappos is the same as a pair from Foot Locker or any other retailer. But Zappos sets itself apart by injecting CEO Tony Hsieh’s obsession with customer service into everything it does.

It’s unlikely that somebody would sell a product that has the exact same functionality as Archi using the code but, even if they did, here’s the thing:

Make you part of your product or service.

You see, Archi is a way of doing things and this is what sets it apart. So let’s list some of the guiding principles that make my and Jean-Baptiste’s philosophy part of the product:

  • There will always be a free and open source version of Archi
  • Archi is agile, intelligent and lightweight
  • We believe in elegant and simple design
  • We do not ask you for your contact details (but please get in touch!)
  • We believe in “open”, in open standards and in open source, and are therefore open and transparent in what we create
  • We want to build services based on trust
  • We believe in sharing
  • We want to create new ways of doing things
  • We want to make Archi and its services fun

So, if somebody does take the code and tries to sell another product based on Archi, then good luck to them because they ain’t us!

But let’s be perfectly clear about one thing. When we say that “there will always be a free and open source version of Archi” this does not mean that we will necessarily work for free or that we might not develop some paid-for services that would support Archi in the future. ArchiMate 3.0 will be released on June 14th and it’s important that Archi implements it, both for itself, for its thousands of users, and for the sake of The Open Group’s ArchiMate Exchange Format. This work is not trivial and will take a lot of effort, but unfortunately I cannot do this unpaid as I like to eat and pay bills. So we are now seeking sources of funding that would support this development. We are also thinking about how we can generate some form of income that would help sustain Archi for future support and features. Jean-Baptiste and I have some great ideas for Archi, but first we need to implement ArchiMate 3.0. Let’s work together in bringing you some great tools with the Archi Philosophy.

Image courtesy of Jean-Baptiste Sarrodie

Image courtesy of Jean-Baptiste Sarrodie