Search the web
Sign In
New User? Sign Up
scrumdevelopment · Scrum Users
? Already a member? Sign in to Yahoo!

Yahoo! Groups Tips

Did you know...
Message search is now enhanced, find messages faster. Take it for a spin.

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
Customer Acceptance Tests   Message List  
Reply | Forward Message #10695 of 42883 |
Re: {Possible Spam?} [scrumdevelopment] Customer Acceptance Tests

Hi Mike--

As dumb as this sounds, I had an epiphany a few years ago when I
realized that the goal is to "iterate everything all the time."
Iterate the code and iterate the tests.

I like to start a sprint with a user story plus the high level
acceptance tests (or "conditions of satisfaction" for the story). The
programmer starts iterating the code in short code-and-test cycles
(or preferably test-then-code cycles if doing TDD). While she's doing
this, the tester is iterating over the high level tests--that is,
making them specific (so instead of "test it with a big number" he
makes a test to test the feature with 4,000,001). While the tester is
iterating over the tests (adding detail, making tests specific) he is
also automating those tests.

Ideally at the end of the day, the programmer and tester have a bit
of working code done and the tests automated that go with that code.
(For example, the feature works but only with positive numbers.) That
code is checked in for the nightly build. The next morning, they
continue to iterate, perhaps adding support for negative numbers that
day in both the code and its tests.

In this way, there really aren't tests to deprecate.

Regards,
Mike Cohn
Author:
Agile Estimating and Planning
User Stories Applied
www.mountaingoatsoftware.com


On Dec 12, 2005, at 8:38 PM, mpkirby@... wrote:

> So I'm in the midst of a project. We've got a number of
> iterations, and my customer has
> given me some embryonic stories. Very simple ones, with customer
> acceptance tests. For
> reasons related to our domain the acceptance tests are not
> automatically run, but we have a
> test group that will run them on completion of the iteration, as
> well as periodically through
> the development to ensure we don't have regression loss.
>
> The question is that as my embryonic stories turn into more fully
> featured ones, do we go
> back and update the old test cases? Do we create new ones? Do we
> deprecate the old
> ones, and tell the test groups only to run the new ones?
>
> I don't want to create a morass of unusable test cases, but I worry
> that the test cases
> themselves will become a maintenance headache.
>
> Anyone else hit this?
>
> Thanks,
> Mike
>
> ---
> mpkirby@...
>
>
>
> ------------------------ Yahoo! Groups Sponsor --------------------
> ~-->
> Get fast access to your favorite Yahoo! Groups. Make Yahoo! your
> home page
> http://us.click.yahoo.com/dpRU5A/wUILAA/yQLSAA/9EfwlB/TM
> --------------------------------------------------------------------
> ~->
>
> To Post a message, send it to: scrumdevelopment@eGroups.com
> To Unsubscribe, send a blank message to: scrumdevelopment-
> unsubscribe@eGroups.com
> Yahoo! Groups Links
>
>
>
>
>




Tue Dec 13, 2005 3:34 am

mikewcohn
Offline Offline
Send Email Send Email

Forward
Message #10695 of 42883 |
Expand Messages Author Sort by Date

So I'm in the midst of a project. We've got a number of iterations, and my customer has given me some embryonic stories. Very simple ones, with customer...
mpkirby@...
mpk2
Offline Send Email
Dec 13, 2005
3:25 am

Hi Mike-- As dumb as this sounds, I had an epiphany a few years ago when I realized that the goal is to "iterate everything all the time." Iterate the code and...
Mike Cohn
mikewcohn
Offline Send Email
Dec 13, 2005
3:33 am

... Some teams hit it. They solve it by making the tests less sensitive to change ... when they need that. Ron Jeffries www.XProgramming.com The opinions...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 13, 2005
3:38 am

... That doesn't really mean anything. I write a story. The first version of the story allows the user to open a dialogue, provide some input, which queries a...
mpkirby@...
mpk2
Offline Send Email
Dec 13, 2005
11:52 am

... I was quite sure that it did when it left here. ... A test case for the GUI? What kind? Tell me more about this test. It might be that you're talking about...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 13, 2005
10:23 pm

... It was a truism. ... Typically out test cases assume you know how to use the product. It's not intended to be a users guide. But they do describe...
mpkirby@...
mpk2
Offline Send Email
Dec 14, 2005
3:55 am

... Perhaps. It left here a bit more than that. It left here as encouragement that recognizing the issue is a big step toward resolving it, and an offer of...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 14, 2005
1:20 pm

... Hmm. I think there is an insight here. Let me play it back. If we define our tests well enough, would it be possible to gather a signature from running ...
mpkirby@...
mpk2
Offline Send Email
Dec 15, 2005
1:04 am

... It's a start ... I can't wait to see what's next. Ron Jeffries www.XProgramming.com I could be wrong, of course. It's just not the way to bet....
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 15, 2005
2:20 am

... I was struck by this comment. "Clicking around in a GUI" is clearly the hardest kind of testing, whether you automate it with some sort of scripting, or...
murdocj2000
Offline Send Email
Dec 14, 2005
2:34 pm

... The test is likely to become an invalid test, yes. It's not necessarily the kind of test mostly likely to break due to a software problem. ... I would...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 14, 2005
7:58 pm

... Well, there is usability testing. Any time the user experience has changed, it needs to be validated. There isn't a set of sample data behavior that you...
Jim McCusker
jpmccusker
Offline Send Email
Dec 14, 2005
8:23 pm

Even a good part of (certainly not all) usability testing could be automated. There is a very large body of usability knowledge out there in literature. Even...
Pascal Roy
pascal_roy_1967
Offline Send Email
Dec 15, 2005
3:59 pm

... Do your projects repeat usability testing every time they change the GUI? Ron Jeffries www.XProgramming.com Knowledge must come through action; you can...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 16, 2005
8:25 am

... Of the parts that change, yes. Actually, we try to test or do heuristic evaluations before every release. They aren't always full-bore formal usability...
Jim McCusker
jpmccusker
Offline Send Email
Dec 16, 2005
3:13 pm

... 2 aspects: 1) The GUI usually consists of components that we don't build that process user events in ways that we often can't control. I've spent many...
murdocj2000
Offline Send Email
Dec 15, 2005
1:55 am

... You've not heard of real world software where different sequences of clicks don't break it? ... Yes ... but every week? Ron Jeffries www.XProgramming.com ...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 15, 2005
2:18 am

... Pardon me. I exaggerated. Have you *not* run into software that performed differently based on different command sequences that should have done the same...
murdocj2000
Offline Send Email
Dec 15, 2005
1:38 pm

On web development projects, we usually set aside a preview site where stakeholders (customers, qa, product manager) can experiment with ongoing development....
Tiran Dagan
tirandagan
Offline Send Email
Dec 15, 2005
1:53 pm

Before we count a story done, we demo it for a customer (usually, immediately after the daily standup). It is not unusual for the customer to say that it does...
Steven Gordon
sfman2k
Offline Send Email
Dec 15, 2005
2:18 pm

... Of course I have. But I don't have to /write/ software that works that way. It's rather straightforward to do otherwise. If the lottery ticket thing was...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 15, 2005
3:06 pm

Ron Jeffries writes: It seems clear to me now that you are imagining manual tests. I recommend against those. The main reason is that they will kill you, but...
Dave Bly
davidably
Offline Send Email
Dec 15, 2005
3:53 pm

... I'd be happy to ... in fact probably nothing could hold me back. But before I start, could you say a bit more about your question, your thoughts, and such,...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 15, 2005
6:38 pm

I'll provide a little input from my perspective: My team writes acceptance tests from user stories that, as with most stories, are emergent. Therefore, as the...
Schiel, James (MED US)
JASchiel
Offline Send Email
Dec 14, 2005
2:17 pm

An important aspect of iterative, agile software development is collaborating with customers and users to discover how the software will really be used, as...
Steven Gordon
sfman2k
Offline Send Email
Dec 15, 2005
5:08 pm

Steven, I absolutely agree with your statement. Just to put my experience in context, I’ve been doing agile as best as I can for a couple of years now (even...
Pascal Roy
pascal_roy_1967
Offline Send Email
Dec 15, 2005
6:34 pm

I agree that the tradeoffs do change a bit once the software goes into production. Persumably, there is also less to discover about what is already in...
Steven Gordon
sfman2k
Offline Send Email
Dec 15, 2005
7:05 pm

Dear Friends, I wanted to run this by group members. If you or someone you know is qualified for this position, please email scrumjobs@... ...
Tiran Dagan
tirandagan
Offline Send Email
Dec 15, 2005
8:04 pm

... Don't want much, do they? Even my record isn't THAT good, and I won't show up at dinner for $150 an hour. ;-> Ron Jeffries www.XProgramming.com Wisdom...
Ron Jeffries
RonaldEJeffries
Offline Send Email
Dec 15, 2005
8:15 pm

... --Hubert...
Hubert Smits
hubert_g_smits
Offline Send Email
Dec 16, 2005
1:35 pm
Advanced

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