Category Archives: Code

Software Development

Estimates and NoEstimates

We had a debate&discussion at XP-Man on NoEstimates for which I did some notes. Reading the NoEstimates stuff, I was most attracted to the sense of “Let’s not be satisfied with second rate” and of a thirst for continuous improvement.
I was left with the sense (possibly because I already believed it) that there are contexts in which NoEstimates works, and contexts in which it doesn’t. But I was very glad to be provoked to ask in each case, “What value if any is our estimate/planning effort adding?” and “Isn’t there a better way to deliver that value?”

What is an Estimate?

An estimate for a Project is (1) a list of things to work on; (2) a cost-range for those things; and (3) a list of risks, that is (3a) of Dependencies that 1&2 rely on, and (3b) of things that might cause significant change.

An estimate or plan for a sprint is (1) a list of things to work on, (2) a “cost” (eg story points) for those things and (3) a list of things we are uncertain about, or (4) need to get help with.

The value of a project estimate is to feed-in to (1) A go/no-go decision and (2) seeing things we want to see sooner rather than later (e.g. should we hire more people, do we need help from specific 3rd parties, is releasing in time for Christmas possible)

The value of a sprint estimate is, to see things we need to ask for in advance (ie external help or resources); to give everyone a sense of confidence about what we’re doing; to fail-faster, that is to see sooner what we can’t achieve.

Money doesn’t make the world go round

“Who are we kidding? Apps are built to earn money. They are a product created to generate profit, and there are a number of ways we can get them to do just that. The simplest, and therefore most common method is …”

So says a typical email in my inbox. But. It ain’t true. Apps are also built because people love to build things. Many, many apps – and websites; and Real Stuff; — are built because people love to built things, and if they didn’t have to earn a living they’d still be building things. For some people, earning money is the pesky thing getting in the way of building the apps they’d love to build.

Should a Software Architect Code? is the Wrong Question

It’s a long-running argument in software architecture and I’ve seen some quite emotional comments on it. But it’s the wrong argument. The right question is surely one about ratios:

“What ratio of non-coding architects to coders will work best for us?”

A ballpark answer to this question would follow logically from calculating how many full-time-developers worth of work results in an amount of non-coding-but-still-requires-deep-or-broad-technical-understanding-and-vision work that adds up to one full time job.

I suggest, based mostly on e-commerce & similar SMEs, something like:
(1) 25 developers’ worth of coding results in 1 full-time non-coding-architect’s worth of work.
(2) Every team needs at least one coding-architect-or-architecturally-competent-developer per 5 developers

Which gets me to a ratio of

1:5:25

for

non-coding-architect : coding-architect-or-lead-dev : developer

Architecture features plenty of non-coding work. Dealing with people, plans, business change, technical design, design decisions, frameworks, principles, patterns, catalogues, quality attributes… There’s more than enough of it if you have 25 coders’ worth of development going on.

So in a workplace with 100 developers, you may want a chief architect, 3 or 4 non-coding architects and 20 coding-architects-or-architecturally-competent-developers. But if you are chief architect or even CTO of a company with 15 developers, you probably still code.

There’s no right answer to the question, how much coding does a coding-architects-or-architecturally-competent-developers do. It can vary from nearly-nothing to 95% as projects progress. Perhaps 50% is a good average?

Discussions

There having been not one but two still-live threads on this on LinkedIn software groups since 2012:
LinkedIn – 97 Things Every Software Architect Should Know – Should architects continue coding … ?
LinkedIn – IASA – Should software architects code?
Anthony Langsworth’s 2012 post

Clojure (and some Lisp) in 10 mins for OO developers

Notes from a short talk at xp-manchester:

No doubt the most useful slide is 3, which is a 1-slide “how to read lisp” tutorial.

We did a series of 10 minute talks on the evening, including a set of 3 on features of functional languages: Haskell’s strong typing, mostly inferred; & Erlang’s agents. Erlang’s agents are almost like an OO language in which every ‘class’ (agent) gets its own process; which is what OO seemed to suggest to me when I read descriptions of the grand idea in the 80s/90s. My headlines for Clojure are largely the points made in http://clojure.org/rationale

Max de Vries Interview – What every project manager should know

I was recently fortunate enough to catch Max de Vries on a trip to the UK. Max has overseen complex IT projects in Europe, Asia and the US, in finance, public sector and now games for over 20 years, most recently being in charge of Popcap’s Plants vs Zombies franchise, overseeing the launch of PvZ 2 and PvZ-Garden Warfare. I managed to fire off two quick questions before he had to go:

Q: What are the top 3 things every project manager should know?
Who to trust; how to communicate; and what would success look like.
Q: What are the common mistake that inexperienced project managers make?
  • Not realizing they are inexperienced
  • Making estimation a political process
  • Miscalculating the cost of saying no vs the cost of failure
  • Not thinking about the critical path hard enough
  • Thinking that people know what they want
  • Underestimating the value of a high functioning team
  • With agile processes: Modifying a process without understanding the reasons why the process works and doesn’t work

So now you know.