Fork me on GitHub

my goodness, asking for peer reviewed academic papers on estimating brings out a lot of people very damaged by bad relations between managers and coders


have you found any yet?


The only research into areas like that I saw at uni was performed on students


so effectively useless


@otfrom: as I've been both (at same time) and so have you, does that make our damage self inflicted?


Isn't this tension between managers and coders a lack of understanding of each others perspectives? Managers not realising that coding is knowledge work that can't be endlessly subdivided without losing all meaning and context and coders not realising that 'I have no idea how much this will cost you, when we will deliver or what you will get' is not an answer anyone can give a stakeholder.


I read Dan North's Blink Estimation blog ( and that's how we used to estimate 25 years ago!


i like to find out what decision requires the estimate, and see if there’s another way to provide the required information for that decision


“What would you do if I said it couldn’t possibly fit?” “We’d still have to do it” “Ok, lets find a way to work with that then” etc


@glenjamin: agreed. You can also ask 'has this value if I only deliver...' but only after you've built trust as stakeholders won't give you a 'get out of jail' card to start with.


However, coming from a consultancy background and having done lots of RFP responses, often these conversations only happen after you made up a number!


Also you deliberately have to make up an optimistic number to win the work in the first place which puts you on the back foot before you even begin.


thankfully i’ve not been involved in pre-contract estimation for some time now


@otfrom I haven’t done any googling yet… but maybe have a look at what Tom Gilb has to say in this context. He is very keen on measuring things.


I've found some things on google scholar that look interesting. Some of them go back to the 70s, some are post 2005


I'm going to write a blog post on "Academic research I've cherry picked in order to post hoc justify our development process"