Managing Software Development
A handbook for the people who have to decide: what to build, how big it is, in what order, and whether it is going well. It accompanies the MSD course at Innopolis University and is written to be read on its own — one page per method, self-contained, with its sources on it.
The six areas
Each area answers one question, and each supplies the next — which is why so many “planning failures” were settled long before anyone got to planning.
People — who does the work. What a technical person gives you and how to stop measuring the wrong thing; the limits of personality models; how teams form, why effort per person falls as they grow, and how groups decide worse than their members. After it, you can tell which team problems are fixable by a manager.
Process — how work flows. Lifecycle stages and process frameworks, and how to choose between them by context rather than by fashion. After it, you can justify a process choice instead of inheriting one.
Needs — what to build. Customer expectations and how they are set; requirements that can be tested; risks named early enough to act on. After it, you can write a requirement someone can disagree with.
Scope — how big it is. Work breakdown structures, and estimation from analogy, counting, judgement and parametric models. After it, you can produce an estimate with its assumptions attached.
Planning — in what order. Agile against plan-driven, milestones and activities, the critical path, and release planning under a fixed date. After it, you can say which slip matters and which does not.
Tracking — is it on track. Progress monitoring, earned value, burndown, and reporting that survives the person being reported to. After it, you can read a green dashboard sceptically.
Start here
- Revising for a session or a quiz → Materials — a study note per lecture and a set of revision questions per area.
- Looking up a method → open its area above, or use the search box.
- Wondering whether to believe a page → its How solid is this? section, at the bottom.
How this handbook is built
It began as the skeleton of Eduardo Miranda’s and David Root’s Managing Software Development at Carnegie Mellon, together with the classics that course rests on — Brooks, Herzberg, Tuckman, Steiner, Goleman.
That skeleton is re-grounded one topic at a time. Each gets a literature search and a read of what it turns up — the corpus stands at 24 books and 153 papers and grows as the review moves through the areas. What survives becomes a short brief: one method per page, under 800 words, with a worked software example and at least one limitation.
Two consequences show on every page. Sources are cited where the claim is made, not gathered in a bibliography nobody opens. And each page ends with How solid is this? — what the claim rests on and where it is thin. Some of the best-known ideas here are the least evidenced, and the handbook says so.
Contributing
Corrections and additions are welcome, from students and practitioners alike.
- Fix a content issue — typos, wrong facts, a citation that does not check out → guidelines
- Add a method page — one technique, following the method template
- Write an analysis — a comparison across methods, following the analysis template
Every page is held to the page standard.
Acknowledgments
Much of this handbook is heavily inspired by and adapted from lectures by Eduardo Miranda and David Root on software project management — the structure, examples and pedagogical approach throughout reflect their teaching materials. Pages built on that material name them in an Acknowledgments section, with full citations in References.
Disclaimer: AI is used for text polishing and explaining. Authors have verified all facts and claims. In case of an error, feel free to file an issue.