> I imagine that
> the tests are running all the time in the background, and that when
> they
> fail, I get a red light somewhere.
We're on the same wavelength.
Some languages have asserts, but not sufficient case specification to
evaluate cases in the editor. Imagine we add "e.g." as a keyword so
that example values can be specified in the source:
int productOfSum (int a eg 5, b eg 7, c eg 9) {
return a+b*c eg 108;
}
... Awesome. -- Ward Cunningham 503-245-5633 v mailto:ward@... 503-246-5587 f http://c2.com...
Ward Cunningham
ward@...
Jan 26, 2004 12:14 am
... We're on the same wavelength. Some languages have asserts, but not sufficient case specification to evaluate cases in the editor. Imagine we add "e.g." as...
Ward Cunningham
ward@...
Jan 26, 2004 12:13 am
... Interesting indeed. I see a red squiggle under this code. It would be fun to work on a toy language with this feature. It would be even more fun to work in...
... This seems to be similar to the sometimes Smalltalk convention of putting an executable sample in the method comment, except the tools know what to do with...
... Hmm. I feel an experiment coming on. Squeak's parser is easy to hack. -- Ward Cunningham 503-245-5633 v mailto:ward@... 503-246-5587 f http://c2.com...
Ward Cunningham
ward@...
Jan 27, 2004 3:50 am
... Here it is in Scheme... (defmacro define-eg (name-result parameter-input . body) (let ((name (car name-result)) (eg-result (cadr name-result)) (parameters...
... Another angle on this idea being presented at PyCon 2004... http://pycon.org/dc2004/talks/index_html#literate "Doctest is a system for writing tests within...
I wanted to reply to an old post that i found interesting. Brian, in your post you distinguish between testers who are focussed on finding bugs and programmers...
... It seems to be common wisdom among "test infected" programmers that tests are a design tool more than a bug-finding tool. Tests help you think through, and...
I agree as well. Even in the early days of implementation we started building a re-usable framework for testing, even at a GUI level which many people claim...
STEURS Stefan
stefan.steurs@...
Jan 16, 2004 1:38 pm
Ron, Are you proposing that the automated tests be changed on a weekly or daily basis? Deborah Haight ... From: Ron Jeffries...
I do think that there must be "critical mass". But I've found usually that my management thinks much more critical mass is needed than I do. If management...
STEURS Stefan
stefan.steurs@...
Jan 16, 2004 2:50 pm
... I can write a defect in my first ten lines of code -- that's why I do TDD, and why I test so much. But if I can do that, and if the testing people know...
... daily basis? ... /snip ... This is, I believe, the single most concrete, demonstrable, supportable, business argument that can be made for IID. Since...
... Sounds great, where could we get such a simulation? Ron Jeffries www.XProgramming.com In times of stress, I like to turn to the wisdom of my Portuguese...
... of the ... You're right about the network effect, although I think complexity might be more a function of the number of connections between components, ...
... There has got to be more going on here than this. For example, my laptop processor can directly address over a billion memories, but I still think of this...
Ward Cunningham
ward@...
Jan 19, 2004 12:46 am
From: Ward Cunningham [mailto:ward@...] ... That depends on your definition of a component, and what constitutes a connection. In your example there is the...
... I suggest that a useful measure of complexity might be the number of minutes of study required to correctly effect an unanticipated change. In my...
Ward Cunningham
ward@...
Jan 20, 2004 3:11 pm
From: Ward Cunningham [mailto:ward@...] ... of ... That's an excellent point. While a system may be characterized by some absolute complexity quantifier (by...
... That raises the question of levels of abstraction. If you are specifying your tests at a higher level than your code and using adapters, then the adapters...
Hubert Matthews
hubert@...
Jan 18, 2004 8:06 pm
... ... which leads, perhaps, to a way tests can help programmers figure out where things belong. Here's a sequence of events I recommended to a client last...
Hi Brian, I like your approach and agree that the Fit tables are a (more) abstract user interface. The GUI and Fit tables each drive the design of the lower...
... Sorry, I don't understand the question. ... Unfortunately, I haven't worked with experienced customers, so I can't answer. My experience with delaying UI...
Hi Brian, My questions arise because I'm working on clarifying the issue of where and when to apply Fit tables. At the moment, my focus is on the areas under...
... That seems reasonable to me. The FIT tests might talk about behaviors not accessible through the GUI, so their API needs should be a superset. (I can...
From: "Rick Mugridge" <r.mugridge@...> ... My experience is that web interfaces, GUIs and external APIs each put very different demands on an object...
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.
Send Email
