Post Agilism: Moving beyond the outdated and misdirected aspects of “Agile”

Cell Phone 1992, Credit: SF Chronicle

The Software World is moving beyond “Agile” into the post-Agile world — and for good reason.

The main problem with Agile, is that it was invented in the 90s, to solve the problems of the 80s.  Problems that are different and unrelated to what we have today.

Consider the world that XP and Scrum, were born in. (We’ll get to Kanban later).

Cell phones were the size of lunchboxes. Most computer ran DOS. Fax machines were state of the art inventions.

Although the internet existed, the modern browsers did not, nor did skype, Webex, Goto Meeting, Twitter, cell phone email, or anything else we take for granted.

Windows 95 was years away at that point.

C#, Java, and PHP did not exist… Most software was painfully developed in C.

Am I getting somewhere yet?

Of course there were big documents and CASE tools back then — there had to be — we didn’t have rapid application development environments like we have now.

But the world has changed — while the Agile practices and mindset have not kept pace.

Take colocated teams — as I mentioned above, communication was primitive back then. Even sending an email between companies was difficult — and faxes were never that fun.

Today, we have numerous, powerful tools that bring people from around the country together in ways they never dreamed about when inventing Scrum and XP. But still, they harp about colocated teams and how verbal communication is best.

Windows 3.1 ca 1992

200 Page documents? Case diagrams? Sure maybe those happened back then, but that was long before the average attention span declined to what it is now, before rapid fire web startups became the new rage, and before it became productive to roll out software incrementally on the web.

So called “waterfall” — eg traditional business — has adapted to those changes by working in smaller increments with more modest documents. The Agile field still lives in the 90s.

We don’t need less planning — we need more! Planning as well as attention spans, have declined over the past 20 years. Need proof? Watch a movie from the 70s — there will be 10 or 20 or even 30 seconds between camera shots. Watch a movie from today and see if the camera says parked for more than 6 seconds before an angle change.

Todays companies don’t write anything down, don’t plan enough, don’t do enough architecture, then they go agile and exacerbate their problems, because agile treats them as if they are an 80s era overplanner.

We need to treat the real symptoms, not fight the last battles.

One way to do that is listed here. I look forward to hearing more from people and their post agile solutions.

As far as Kanban goes, Kanban goes back the 50’s and before. It’s designed to address how factories worked in the 70s.

Not only have factories evolved since then, and made many Kanban concepts redundant, but software development is not an assembly line exercize. Kanban can probably work fine for repetitive tasks, but today’s software is not repetitive in nature.

Coders could be working on a web site one day, database backend another, facebook integration the next. It’s not screwing in bolts one after another.

So we can see that Agile focuses on the problems of the past, and misdirects it’s energies on things that are no longer germane to the issue, and possibly never were.

It’s time to move forward, not dwell in the past.


About postagilist

Architect, Manager, Developer, Musician, Post-Agile thought leader
This entry was posted in Uncategorized and tagged , , , , , , , , . Bookmark the permalink.

9 Responses to Post Agilism: Moving beyond the outdated and misdirected aspects of “Agile”

  1. Actually the Kanban model works really well for the problem you described – website one day, database backend another, yada, yada.
    I was saying in the most recent podcast that regardless of Agile the software development world would have shifted anyway – we had no choice with the advancement of technology.
    p.s. My dad had one of those phones!

    • PostAgilist says:

      I think the “task board” makes sense for what you are talking about, but “Lean principles” in terms of reducing waste do not apply to the case we’re talking about, since their isn’t enough repetition to analyze and reduce waste.

      The “task board” can and has been used for anything and everything for a long time; to make “Kanban” into anything useful besides being a task board involves the nebulous “Lean” and as I mention Lean and waste reduction are best suited to repetitive tasks.

      How do you see Lean helping out concretely in this case?

    • PostAgilist says:

      Also the programming world DID shift — it’s just that Agile pretends it didn’t. I was doing customer collaboration, short releases, frequent releases, unit testing, as early as 1989. Before agile. I developed my own process and it’s still as good or better than any agile process I’ve used (incremental traditional).

      That is why agile has never been important or necessary to me, and I’m sure that’s true of many thousands of developers/shops.

      Forget the past — look to the future

  2. Limiting work in progress to reduce context switching is an easy implementation. Root cause analysis would be another (5 whys). Nothing amazingly surprising but still effective.

    • PostAgilist says:

      Broadly speaking, at the end of the day, Lean is is about improving operational efficiency.

      Although improving operational efficiency is useful, and can be done with or without “Lean”, it has nothing to do with with innovating new products; seeking operational efficiency at the expense of innovation is what turns mature companies into irrelevant companies, at least in the software world…

      Post agilism isn’t about improving operational efficiency as a core function, it’s about developing new products.

      When you create prototypes at Toyota, you don’t worry about building them efficiently as the highest priority. Software development is more like that — you are then optimizing for solving problems and creating new ideas. When you get into serial production, which in the software world is deployment and maintenance, then you would worry about things like that.

      They are different functions not interchangable.

      I’m sure Microsoft, with all its offshoring and internal scrumming, has improved it’s operational efficiency over the past 10 years. But have they really innovated anything in the past 10 years?

      WP7 was DOA and Windows 8 Metro looks to come out of the gate with a thud.

      Limiting WIP is something that’s easy and it’s common sense — but it won’t in and of itself aid in the software development process, anymore than giving people adequate lighting well in and of itself, create great products. They are good ideas but they only go so far.


  3. fijiaaron says:


    Glad you posted a comment on my blog so I could read yours. But I almost thought your comment was spam since it didn’t mention the subject of Agile process.

    My favorite statement here is “We don’t need less planning — we need more!”

    Although I think your timeline is off a bit. The Agile Manifest was written in 2001, Extreme Programming Explained by Kent Beck was written in 1999. The Chrysler project that was the earliest germ of Agile was in 1997. Email was readily available by then. XP/Agile really didn’t take off (and indeed was primarily a response to) widespread outsourcing which took off at the same time.

    I would still agree with their tenet that co-located teams are better ideally, but it’s a cost-benefit analysis, not a binary choice And 200 page documents are definitely not a thing of the past, though they’re definitely less useful than they were 10 years ago.

    • PostAgilist says:

      Hi Aaron

      Thanks for commenting 🙂 I really loved your blog post.

      I don’t think my timeline is off that far; early Scrum was being done in 1992, at Easel, and both Scrum and XP, as well as the cross-functional, colocated team concept, is derived from the paper “The New New Product Development Game” which came out in 1987.

      Maybe Kent’s book came out in 1999, but it’s based on concepts from the mid 80s, as well as Scrum.

      Yes we need more planning and we need to move the pendulum back towards the center, not just switch it from long term planning to short term planning and call that an innovation like agile has done…

      As far as me commenting on your website I loved your post there it hit many of the salient points I cover in my “New New Agile Manifesto” post.

      As far as my post being understated, that’s my style. Compared to all the luminaries who spam boards constantly, I don’t even use my last name most of the time.

      They can read it or not based on whether what I say makes sense or not…

      Thus in my own little way I differentiate myself from the vast herd of shameless self promoters out there…


  4. Pingback: My most Relevant posts on Agile organized for your reading pleasure | PostAgilist's Software Architecture Blog

  5. Pingback: DevOps Is The New Agile | Secure Software Development

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s