A public forum for discussing the design of software, from the user interface to the code architecture. Now closed.
I don't really think the tools work very well. You are better off using your own actual times for similar work, do horizontal dive on the problem, and come up with your best estimates.
son of parnas
Thursday, June 09, 2005
I've used Cost Xpert with some success. It has a number of different estimating methods, and you can choose the one that you have the understanding of (read faith in).
Our current company standard tool is SEER. It's okay, but the version we use only allows SLOC or Function Points as the basis of estimate.
Another popular tool is COCOMO due to it's basis on the research of Barry Boehm.
Steve McConnell has one available on his company's web site at: http://www.construx.com/resources/estimate/
Uses COCOMO and Putnam forecasting models for its projections and allows you to develop pretty sophisticated project models based on cost, schedule, and resource constraints.
I personally like to use @Risk for project, from palisade software. Doesn't use standardized estimators like the one from Construx mentioned above, but provides a clearer picture to me about how it makes its predictions.
Friday, June 10, 2005
Some of Joel's earlier articles were on spec writing and task scheduling. You might want to look at those. Read them all starting from:
I don't know how you could come up with a good estimate without writing up a pretty complete set of functional specs.
Also, don't know if it's quite what you're looking for, but several people have written simple apps based on ideas in Joel's software scheduling article. E.g., http://www.uil.net
Tell us how you currently estimate manually, and we can try to find a tool to match that.
If you don't really estimate manually, and you want a tool to make the problem go away ... well ... too bad ... the place to start is with manual estimates. Functional Decomposition is a good start. IE, Break the project into very small chunks, add them up, add a buffer from 10-40% at the end depending on risk, etc.
Joel has an article on this, someone linked to it above ...)
Friday, June 24, 2005
This topic is archived. No further replies will be accepted.Other recent topics
Powered by FogBugz