Ars Gratia Artis

Ars gratia artis: Art for the sake of art. It goes to bad places… you end up here with this famous piece:

If you don’t know the story, the picture is of a famous art piece named The Fountain by Marcel Duchamp.  If you are an artist or great lover of art, it is an important work that has a bold place in arts history.  However, if you are outside that circle the piece simply reminds you of why you hate artists.  (You can have fun investigating the paradox of it all on your own.)

The point is this: anything done for the sake of itself is only of value to those within the inner sanctum.  If you look at most of the advice pumped out in various books, television shows, and story lines in movies you see this common thread, anything done for its own sake is ruin.

Like all good bloggers, I’m going to wow you by telling you something you already know and have heard a dozen times before, yet in a slightly new way, and then feel all clever about it.

So hear me now: Don’t “do” devops, purely for the sake of devops.  Don’t use configuration management (Puppet, Chef) just for the sake of using a configuration management tool. Don’t adopt the cloud just for the sake of adopting the cloud.

I know that sounds ridiculous now, but think of all the other fads that have become dirty words.  Agile.  Kanban.  TPS/LEAN.  SOA.  ITIL.  ISO/27001.  CISA.  SOX.  CMMI.  ITSM.  Six Sigma.  ISO-9000.  SAS-70.  On and on and on.  This comes to mind because I just got a piece of email about a conference in Washington (which is the first sign its full of itself) and after watching several videos from years past I wanted to chew my head off.  At some point, surely, you’ve been at a conference or read a whitepaper or seen a presentation on YouTube that made you whisper “This guy needs to get out a little more.”

Anyone engaging in this type of feedback loop will ultimate be confronted with the criticism: “I think your missing the point.”  A little cult is formed and you speak in a secret language causing you to attend annual conferences with the other people who know your mystic dialect of acronyms and start using software by CA.

Standards.  Standards are why I’m really passionate about this point.  Ever met someone who worked for a company that adopted ISO-9001?  Did they like it?  No, they probably bitched about it for as long as you’d let them.  Or SAS-70.  People bitch like mad about SAS-70.  But the thing is that SAS-70 isn’t about what most poeple think it is, Statement on Auditing Standards (SAS) No. 70 simply standardizes a way of auditing a companies internal controls and their effectiveness.  SAS-70 does not itself dictate what those controls are…. but the auditors do.  ISO-9001 is the same, the standard is good, but the auditors want things done in a certain way, the way everyone else does them, regardless of whether or not they still fit the requirements of the standard.  And so they get a bad name.  ISO-9001 encourages quality control and establishes trust between organizations, but employees are shoe-horned into obsolete procedures that goes against the original intents of people like Deming.  The people who fight the auditors bitch, the people who embrace them openly become a special cult.

So… while we’re still in that warm fuzzy place with DevOps and Cloud and all the new forms of automation, embrace them only in so far as that you apply what is applicable and leave aside that which is not.  Solve problems, take the mountains of good ideas and make them your own, don’t build a shrine around them.

3 Responses to “Ars Gratia Artis”

  1. Dave Pickens says:

    I hear your message loud and clear and echo it myself to my customers.

    I see this mistake all the time. I call it “Keeping up with the Joneses” syndrome.

    It’s not unusual to talk with the customer and find out they’re going to adopt {ITIL, EA, PM… yada yada} and at the same time no one can articulate why they are doing so. There’s no linkage back to the high-level business requirements or strategic plan. As bad is the fact that many times they can’t summarize what ITIL will even do for them in a short one page document.

    The other part of this equation is trying to swallow ITIL or whatever whole like a python. None of these things are small little critters fit for swallowing whole – rarely in IT is there anything that you can or should swallow whole. Rather this is more like eating an elephant. When faced with the task of eating an elephant, one often asks “How do I eat this elephant?!?”… and the answer invariably comes back as “One bite at a time”.

    To that end, I definitely subscribe to mantra of “Just enough, just in time” when it comes to dealing with things like ITIL, Enterprise Architecture or Six Sigma or {fill in the blank}.

    A failed effort to do ITIL en toto for example simply wastes precious time, money and resources along with leaving a bad taste for ITIL.

    Smaller, continuous adoption (and adaption) of something like ITIL with slow and steady progress is a good thing.

  2. Marcelo Leal says:

    Hello Ben!
    I totally agree with you… i think people needs to realize that all work needs to have a purpose. The job, the work, the “tool” (like the name implies), is to achieve something.
    If you use it because “other do”, or “because it’s nice”, blah, blah, blah. It’s useless.
    Thanks to show me the right expression “Ars gratia Artis”…
    When i’m talking about this, i use the phrase by myself: Music for musicians. ;-)

    Leal