Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #16461

Re: Do C++ and Java professionals use UML??

Show all headers | View raw

  To: Robert Klemme
From: "Gene Wirchenko" <gene.wirchenko@1:261/38.remove-nwi-this>

  To: Robert Klemme
From: "Gene Wirchenko" <gene.wirchenko@1:261/38.remove-10ae-this>

  To: Robert Klemme
From: Gene Wirchenko <genew@ocis.net>

On Fri, 27 Jul 2012 13:21:17 +0200, Robert Klemme
<shortcutter@googlemail.com> wrote:

>On 07/26/2012 06:16 PM, Gene Wirchenko wrote:
>> On Thu, 26 Jul 2012 11:35:01 +0200, Robert Klemme
>> <shortcutter@googlemail.com> wrote:
>>
>>> On 23.07.2012 22:53, Gene Wirchenko wrote:
>>>> On Mon, 23 Jul 2012 21:11:29 +0200, Robert Klemme
>>>> <shortcutter@googlemail.com> wrote:
>
>>>>        Why would one use this class?  [If I am a newbie, I may not
>>>> know.]
>
>>
>>       I think it the most important thing.  If one does not know why
>> one would use a class, why even bother?
>
>IMHO the class documentation is reference material which should provide
>the basic facts so I can decide myself whether the class is appropriate
>for the use case at hand or not.  Learning to judge that is part of the
>process of learning to program.  That should not be piggybacked on
>reference documentation.

     That is the reference documentation!  Maybe the class is better
is certain circumstances than the usual implementation.  e.g. "If you want ..., 
then this class may be useful."

>>       I have seen this botched in intro classes where recursion is
>> introduced.  All too often the example used is factorial which is much
>> more quickly and clearly solved using iteration.  The student gets the
>> impression that it is overly complicated and never bothers with it.
>
>Yeah, but the cases where recursion makes the code simpler are typically
>more complex algorithms (backtracking for example).  I suspect all
>methods which can be converted into a loop via tail recursion
>optimization are as simple as factorial.

     My solution to that is that they should not introduce recursion
until they have a situation that would really benefit from it.

>>> Often class documentation in the Java stdlib does contain usage
>>> examples.  But for fundamental classes like String there are so many
>>> potential use cases that you cannot really cover them all in the class doc.
>>
>>       One example of the method in use is a problem?
>
>Many methods do not make much sense alone (for example, Map's
>containsKey()).  Then the question is where should the example be placed
>etc.

     For clarity.  I would not mind a short example that uses several
methods in concert.

>I think we have a quite different expectation towards JavaDoc.  For me
>it's reference material, so I expect to get formal information (allowed
>arguments, semantics) while you seem to be more concerned with providing
>information that helps learn the language and std library.  I think that
>information is better covered in a tutorial or other type of document
>(maybe even a book) and it would get in your way when working with the
>library.  Because then the introductory information easily gets in your
>way and you need to look longer for the important information.

     Yes.  Where is the learning reference for what these classes do?
If one does not already know the class, JavaDoc is not too useful.

>>> I am not sure though whether I agree that the situation is as bad as
>>> your list makes it sound.  For example, java.util.* is pretty well
>>> documented IMHO.
>>
>>       One effect that I have noticed with things like this is that a
>> newbie struggles.  Someone who already knows does not see a problem.
>> He is already past it.  so it rarely gets addressed.
>
>But if it were such a big issue for a large number of people learning
>Java I am pretty sure it would be addressed.  So I conclude the
>situation cannot be as bad as you observe it.

     It is and it isn't.  Why do you think that there is such a big
market for intro texts for a language?  It is interesting to me though that, 
IME, the next step (mid-level) has little, and this is true in many languages; 
Java is not special in this regard.

Sincerely,

Gene Wirchenko

-+- BBBS/Li6 v4.10 Dada-1
 + Origin: Prism bbs (1:261/38)
-+- Synchronet 3.16a-Win32 NewsLink 1.98
Time Warp of the Future BBS - telnet://time.synchro.net:24

-+- BBBS/Li6 v4.10 Dada-1
 + Origin: Prism bbs (1:261/38)
-+- Synchronet 3.16a-Win32 NewsLink 1.98
Time Warp of the Future BBS - telnet://time.synchro.net:24

--- BBBS/Li6 v4.10 Dada-1
 * Origin: Prism bbs (1:261/38)
--- Synchronet 3.16a-Win32 NewsLink 1.98
Time Warp of the Future BBS - telnet://time.synchro.net:24

Back to comp.lang.java.programmer | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Re: Do C++ and Java professionals use UML?? "Robert Klemme" <robert.klemme@1:261/38.remove-z1h-this> - 2012-07-28 18:36 +0000
  Re: Do C++ and Java professionals use UML?? "Gene Wirchenko" <gene.wirchenko@1:261/38.remove-z1h-this> - 2012-07-28 18:36 +0000
    Re: Do C++ and Java professionals use UML?? "Robert Klemme" <robert.klemme@1:261/38.remove-5bm-this> - 2012-07-29 18:16 +0000
      Re: Do C++ and Java professionals use UML?? "Gene Wirchenko" <gene.wirchenko@1:261/38.remove-5bm-this> - 2012-07-29 18:16 +0000
    Re: Do C++ and Java professionals use UML?? "markspace" <markspace@1:261/38.remove-5bm-this> - 2012-07-29 18:16 +0000
  Re: Do C++ and Java professionals use UML?? "David Lamb" <david.lamb@1:261/38.remove-5bm-this> - 2012-07-29 18:16 +0000
    Re: Do C++ and Java professionals use UML?? "Robert Klemme" <robert.klemme@1:261/38.remove-5bm-this> - 2012-07-29 18:16 +0000
      Re: Do C++ and Java professionals use UML?? "Eric Sosman" <eric.sosman@1:261/38.remove-pvc-this> - 2012-07-29 19:19 +0000
      Re: Do C++ and Java professionals use UML?? "Gene Wirchenko" <gene.wirchenko@1:261/38.remove-x1c-this> - 2012-07-31 20:06 +0000

csiph-web