Search the web
Sign In
New User? Sign Up
extremeprogramming · Extreme Programming
? Already a member? Sign in to Yahoo!

Yahoo! Groups Tips

Did you know...
Real people. Real stories. See how Yahoo! Groups impacts members worldwide.

Best of Y! Groups

   Check them out and nominate your group.
Having problems with message search? Fill out this form to ensure your group is one of the first to be migrated to the new message search system.

Messages

  Messages Help
Advanced
[agileprogramming-ottawa-canada] New Blog Entry - Agile Circling the   Message List  
Reply | Forward Message #146421 of 152316 |
[XP] Re: Local adaptations (was: New Blog Entry - Agile Circling the Drain?)

Manuel,

--- In extremeprogramming@yahoogroups.com, "Manuel Klimek" <klimek@...>
wrote:
> > It's hard to sell the notion that "sometimes" my acceptance tests
will
> > tell me that I forgot something. Wouldn't it be cheaper to find a
way
> > to remember in the first place?
>
> Sometimes my unit tests will catch bugs. Wouldn't it be cheaper to not
> insert them in the first place? I don't see what you're aiming at...

I don't use TDD to catch bugs, I use TDD to prevent bugs.

So yes, it is cheaper to not put them in in the first place... and I use
TDD to help with that. It's the cheapest way I have found to do this.
If someone tells me a cheaper way I will probably do that.

<snip/>
> So how do you know for every method that Presenter and BusinessObject
> agree on the semantics of the method?
>
> Cheers,
> /Manuel

I don't know what "agree on the semantics of the method" means.

Do you mean the signature? Our mocking framework lets us test arguments
on mocked calls.

Matt





Fri Nov 21, 2008 5:06 pm

maswaffer
Offline Offline
Send Email Send Email

Forward
Message #146421 of 152316 |
Expand Messages Author Sort by Date

... Sometimes my unit tests will catch bugs. Wouldn't it be cheaper to not insert them in the first place? I don't see what you're aiming at... ... So how do...
Manuel Klimek
manuel.klimek
Offline Send Email
Nov 21, 2008
4:42 pm

Manuel, ... will ... way ... I don't use TDD to catch bugs, I use TDD to prevent bugs. So yes, it is cheaper to not put them in in the first place... and I use...
Matt
maswaffer
Offline Send Email
Nov 21, 2008
5:06 pm

... Sounds like nit-picking to me: if I write a line, and a test breaks unexpectedly, then I call that "catching a bug". If I had not caught this bug, it would...
Manuel Klimek
manuel.klimek
Offline Send Email
Nov 21, 2008
5:47 pm

Manuel, ... If you read the thread, I haven't said we have zero bugs... but our bug count is incredibly low considering we don't have a large suite of ...
Matt
maswaffer
Offline Send Email
Nov 21, 2008
6:02 pm

Hello, Matt. On Friday, November 21, 2008, at 1:02:12 PM, you ... Please say what "incredibly low considering we don't have a large suite of acceptance tests"...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Nov 21, 2008
6:30 pm

Ron, ... a ... In the project in question we do have more integration tests because the mocking capabilities aren't quite as good as they are in our other ...
Matt
maswaffer
Offline Send Email
Nov 21, 2008
7:17 pm

Hello, Matt. On Friday, November 21, 2008, at 2:17:26 PM, you ... Yes. When I TDD, I write whatever test interests me, and many of these tests use more than...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Nov 21, 2008
10:51 pm

On Fri, Nov 21, 2008 at 1:30 PM, Ron Jeffries ... LOL! Just had a 'milk-out-my-nose' experience reading this. You owe me a milk. Chris. [Non-text portions of...
Chris Wheeler
chris_h_wheeler
Offline Send Email
Nov 21, 2008
8:32 pm

... The old game: how do you know acceptance tests would not further reduce your bug count? Or is the point that you simply think that would not give you any...
Manuel Klimek
manuel.klimek
Offline Send Email
Nov 21, 2008
7:23 pm

Manuel, ... bug ... Because they didn't. We tried acceptance tests and they weren't that useful for us. We stopped doing them and our bug count didn't go up....
Matt
maswaffer
Offline Send Email
Nov 21, 2008
8:14 pm

... I have found acceptance tests that exercise the software through the GUI tend to be brittle. Is this a possible explanation for your observation. I have...
Steven Gordon
sfman2k
Offline Send Email
Nov 21, 2008
8:41 pm

I guess one point would be acceptance tests are closer to the business value than unit tests -- acceptance tests verify that user stories actually work. And as...
Olof Bjarnason
olof.bjarnason@...
Send Email
Nov 21, 2008
9:29 pm

Olof, ... value ... work. ... really ... quality ..) I agree with this. In our case our paying customers don't know or care that we have *any* tests and our...
Matt
maswaffer
Offline Send Email
Nov 21, 2008
9:46 pm

Steven, ... Thanks for the observation. The only GUI testing we tried to do was in the web world with Watin. It works fairly well and we actually still have...
Matt
maswaffer
Offline Send Email
Nov 21, 2008
9:42 pm

... This looks interesting. Tests that you have to change all the time are clearly a red flag. The interesting thing for me is that I see that problem more...
Manuel Klimek
manuel.klimek
Offline Send Email
Nov 21, 2008
10:33 pm

... Brittle tests are a red flag to us. We saw this a lot when we were trying to do a lot of FitNesse tests on our C# project. I will be the first to admit...
mrlukelanphear
Offline Send Email
Nov 21, 2008
11:43 pm

... Acceptance tests may reduce our bug count, but at what cost? It is VERY rare for us to ship a critical bug OR to have a bug fix take more than an hour or...
mrlukelanphear
Offline Send Email
Nov 21, 2008
10:47 pm

Hello, Matt. On Friday, November 21, 2008, at 11:32:20 AM, you ... Um ... what would be cheaper and actually work? Ron Jeffries www.XProgramming.com ...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Nov 21, 2008
6:24 pm

Hello, Matt. On Friday, November 21, 2008, at 1:06:16 AM, you ... There are tons of root causes. I think acceptance tests address the possibility of having...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Nov 21, 2008
11:06 pm

Ron, ... I think we have addressed these concerns with other process improvements. ... "Mistake on the story" is an interesting one. Mistakes in code are my ...
Matt
maswaffer
Offline Send Email
Nov 21, 2008
11:40 pm

Hello, Matt. On Friday, November 21, 2008, at 6:40:41 PM, you ... Could you please give us a quick list of things that you do that address these? Should be...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Nov 22, 2008
1:17 am

Ron, ... All three of these stem from communication issues. During our release and iteration planning meetings the team used to flesh these details out and...
Matt
maswaffer
Offline Send Email
Nov 22, 2008
9:35 pm

Hello, Matt. On Saturday, November 22, 2008, at 4:35:05 PM, you ... Yes. That's why in my writing I call them Customer Tests ... ... Well if you're happy I...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Nov 22, 2008
10:24 pm

Matt, Loud responses don't equal concensus. What I hear is that you tried developing with acceptance tests and without, and without works better overall....
kentb
kentlbeck
Offline Send Email
Nov 21, 2008
11:59 pm

Kent, I appreciate the sentiment. My intent with exploring this was not to prove or disprove the value of any given practice. I am sincerely looking for...
Matt
maswaffer
Offline Send Email
Nov 22, 2008
12:29 am

... Well, you're observing a phenomenon that I cannot observe from this distance, hence my question. ... How much does the functionality change over time?...
George Dinwiddie
gdinwiddie
Offline Send Email
Nov 21, 2008
6:51 pm

George, ... has ... The business rules change whenever the government decides to change regulations... major changes every 12 to 18 months. Just this morning...
Matt
maswaffer
Offline Send Email
Nov 21, 2008
7:14 pm

... And what is the norm? What is "quite often?" ... What made the cost so high? Did the tests need a lot of maintenance of old tests? Is there a lot of...
George Dinwiddie
gdinwiddie
Offline Send Email
Nov 21, 2008
10:03 pm

... else? ... Yes that is right. Writing and maintaining Fitnesse fixtures took up the extra time. "Designing" the fitnesse pages, then writing the fitnesse...
mrlukelanphear
Offline Send Email
Nov 22, 2008
12:17 am

Hi, Luke. Thanks for chiming in. I hope that Matt will also respond to these questions, as he may have some different observations. ... It sounds like there...
George Dinwiddie
gdinwiddie
Offline Send Email
Nov 22, 2008
1:20 am
 First  |  |  Next > Last 
Advanced

Copyright © 2009 Yahoo! Inc. All rights reserved.
Privacy Policy - Terms of Service - Guidelines - Help