Work In Progress #78
Replies: 7 comments
-
|
Have you ever looked at McConnell's "Software Estimation"? |
Beta Was this translation helpful? Give feedback.
-
|
Hey Bret, Short answer, no... and I'm leery of shelling out again for one of his books on Amazon as it's a) 13 years old and b) I didn't get the value out of the last one I bought of his. Is there something specific that you want me to know? From this post, https://stevemcconnell.com/blog/17-theses-software-estimation/, this is all very predictable software estimation fare: measuring past projects, measuring velocity etc. Having said that, he seems to have been having an interesting discussion with Ron Jeffries about estimation lately. Although from the look of it, they're not making much headway! In this, they're basically discussing whether estimates are or aren't useful, and in what context. I'm really hoping I can dig a bit deeper than this. That's why I'm trying to look at the axes along which projects lie and figure out the domain in which classical extrapolation works, and how it fails. Personally, I now have a feel for certain types of task, where I can tell how long they're going to take. And I also have some kind of alarm-bell for the types of task that are going to go off the rails. Based mainly on bitter experience. I'm trying to explore this... but if McConnell already has something around this then let me know. |
Beta Was this translation helpful? Give feedback.
-
|
I've got it in hard copy ..... somewhere? I might have it on a kindle.
…On Mon, Nov 4, 2019 at 2:51 PM Rob Moffat ***@***.***> wrote:
Hey Bret,
Short answer, no... and I'm leery of shelling out again for one of his
books on Amazon as it's a) 13 years old and b) I didn't get the value out
of the last one I bought of his.
Is there something specific that you want me to know? From this post,
https://stevemcconnell.com/blog/17-theses-software-estimation/, this is
all very predictable software estimation fare: measuring past projects,
measuring velocity etc.
Having said that, he seems to have been having an interesting discussion
with Ron Jeffries about estimation lately
<https://stevemcconnell.com/blog/noestimates-response-ron-jeffries/>.
Although from the look of it, they're not making much headway! In this,
they're basically discussing whether estimates are or aren't useful, and in
what context.
I'm really hoping I can dig a bit deeper than this. That's why I'm trying
to look at the axes along which projects lie and figure out the domain in
which classical extrapolation works, and how it fails.
Personally, I now have a feel for certain *types* of task, where I can
tell how long they're going to take. And I also have some kind of
alarm-bell for the types of task that are going to go off the rails. Based
mainly on bitter experience.
I'm trying to explore this... but if McConnell already has something
around this then let me know.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<https://github.com/orgs/risk-first/teams/risk-first-team/discussions/21/comments/2?email_source=notifications&email_token=AABUIS5T6PZJK6YGTXCOLELQSASGLA5CNFSM4JIRN5P2YY3PNVWWK3TUL52HS4DFWNCGS43DOVZXG2LPNZIG643UKJSXA3DZVJRW63LNMVXHIX3JMTHAAAS45E>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AABUISYO3ITRSLJYZP3WWA3QSASGLANCNFSM4JIRN5PQ>
.
|
Beta Was this translation helpful? Give feedback.
-
|
What about the risk that whoever asks you to pull an estimate considers it
a quote :)
…On Mon, Nov 4, 2019 at 2:56 PM Bret Weinraub ***@***.***> wrote:
I've got it in hard copy ..... somewhere? I might have it on a kindle.
-
On Mon, Nov 4, 2019 at 2:51 PM Rob Moffat ***@***.***>
wrote:
> Hey Bret,
>
> Short answer, no... and I'm leery of shelling out again for one of his
> books on Amazon as it's a) 13 years old and b) I didn't get the value out
> of the last one I bought of his.
>
> Is there something specific that you want me to know? From this post,
> https://stevemcconnell.com/blog/17-theses-software-estimation/, this is
> all very predictable software estimation fare: measuring past projects,
> measuring velocity etc.
>
> Having said that, he seems to have been having an interesting discussion
> with Ron Jeffries about estimation lately
> <https://stevemcconnell.com/blog/noestimates-response-ron-jeffries/>.
> Although from the look of it, they're not making much headway! In this,
> they're basically discussing whether estimates are or aren't useful, and in
> what context.
>
> I'm really hoping I can dig a bit deeper than this. That's why I'm trying
> to look at the axes along which projects lie and figure out the domain in
> which classical extrapolation works, and how it fails.
>
> Personally, I now have a feel for certain *types* of task, where I can
> tell how long they're going to take. And I also have some kind of
> alarm-bell for the types of task that are going to go off the rails. Based
> mainly on bitter experience.
>
> I'm trying to explore this... but if McConnell already has something
> around this then let me know.
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <https://github.com/orgs/risk-first/teams/risk-first-team/discussions/21/comments/2?email_source=notifications&email_token=AABUIS5T6PZJK6YGTXCOLELQSASGLA5CNFSM4JIRN5P2YY3PNVWWK3TUL52HS4DFWNCGS43DOVZXG2LPNZIG643UKJSXA3DZVJRW63LNMVXHIX3JMTHAAAS45E>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/AABUISYO3ITRSLJYZP3WWA3QSASGLANCNFSM4JIRN5PQ>
> .
>
|
Beta Was this translation helpful? Give feedback.
-
|
You mean where you give an estimate and suddenly all the risk of the project is on you? Hold up, I think I started a diagram for that! Seriously though, that's where things start getting interesting. I haven't written about that yet, I'm still in the foothills. But that's seriously fun territory. Here's a big question: does that gotcha exist because we're bad at estimating, or do we get bad at estimating because that exists? |
Beta Was this translation helpful? Give feedback.
-
Some kind of game theory at work here...... |
Beta Was this translation helpful? Give feedback.
-
|
Definitely. Let's say a non-technical manager Alice hires Bob to add a feature to her application. She has very little idea of why the feature takes the "man-months" that it does. She could attribute it to her codebase's quality, Bob's competence or his working conditions. But there are a few sources of information she does have: she can ask him (take his word for it) or hire another developer to look at his code (get a second opinion). Or she can see what difference reducing the budget (# of hours) makes on quality while requesting/demanding the same scope and quality of work. If this tactic works and neither scope nor quality slip, she's discovered producer surplus and has now managed to reduce (or eliminate it). If it doesn't, she's sabotaged her own application. |
Beta Was this translation helpful? Give feedback.
-
So a couple of months back I started working on a few new articles about estimating. I wanted to try and write up something on the different approaches to estimating software development, and the limits to our ability to estimate.
At the same time, I had some ideas about complexity, and writing up the main tools we use for managing complexity in a code-base.
But the whole thing has turned into a bit of a rabbit-hole, and both of these concepts ended up being linked together.
I still have a long way to go on this. Nevertheless, I thought I'd share the first five articles on here so that I can get some feedback, and see what you guys think.
Many apologies - this is not as clear or polished as I'd usually like, but I think over the next month or so it should shape up a bit.
https://riskfirst.org/Estimating
cheers,
Rob
Beta Was this translation helpful? Give feedback.
All reactions