Loading ...
Sorry, an error occurred while loading the content.

Re: [XP] [Slightly OT] OOP Article on DevX

Expand Messages
  • George Paci
    ... Pretty standard stuff, although he persistently misspells XP as OOP . --George Paci Having a precise definition of XP is
    Message 1 of 10 , Feb 1, 2005
    • 0 Attachment
      Dave Rooney wrote:

      >
      > I saw a reference to this article on TheServerSide.com this morning...
      >
      > http://www.devx.com/DevX/Article/26776/0/page/1
      >
      > Thoughts anyone?

      Pretty standard stuff, although he persistently misspells "XP"
      as "OOP".


      --George Paci <george@...>

      Having a precise definition of "XP" is as useful in
      developing software as having a precise definition of
      "rope" is in climbing mountains.
    • Gary Feldman
      ... Whenever I see stuff like this, I am reminded of a letter to a programming journal that I saw many years ago. The author praised the virtues of Forth,
      Message 2 of 10 , Feb 1, 2005
      • 0 Attachment
        Dave Rooney wrote:
        >I saw a reference to this article on TheServerSide.com this morning...
        >
        >http://www.devx.com/DevX/Article/26776/0/page/1
        >
        >Thoughts anyone?

        Whenever I see stuff like this, I am reminded of a letter to a programming journal that I saw many years ago. The author praised the virtues of Forth, particularly it's wonderful ability to use single character symbols -- %, #, etc. -- as words in the language, making it much easier to type than all those wordy programming languages. (I think he meant Basic). He then wanted some suggestions on how to keep these things straight when you were putting together a larger program from pieces in which a symbol meant one thing in one piece and something totally different in another.

        Some people must think that programming is nothing more than stringing together the right sequence of keys to automatically push on a calculator, and if only those programmers wouldn't produce so many typos, programs would all work perfectly.

        Gary
      • Ron Jeffries
        ... You certainly may. That word frock may have a different meaning in my lexicon than it has in yours, however. Still, people should dress as they like,
        Message 3 of 10 , Feb 1, 2005
        • 0 Attachment
          On Tuesday, February 1, 2005, at 3:44:04 PM, Dave Rooney wrote:

          > "And the profession is now guarded by a priest class that benefits
          > from OOP's murk and mystery—the fewer people who can communicate with
          > computers, the more secure their jobs." Eh?! Do I need to run out
          > and buy a frock?

          You certainly may. That word "frock" may have a different meaning in
          my lexicon than it has in yours, however. Still, people should dress
          as they like, that's my belief. :)

          The article and the one it references both remind me of "topmind", a
          famous anti-OO poster on the newsgroups. I don't get it, myself: I
          kind of like OO, because it lets me program in ways that I find to
          be more clear.

          The author does mention the incredible morass of the object
          libraries, and I agree with that. But I guess I'd rather have them
          than not, though they could be simpler, and better abstractions,
          than they usually are.

          Ron Jeffries
          www.XProgramming.com
          First they ignore you, then they laugh at you, then they fight you,
          then you win. -- Gandhi
          I'm on track. When do I start to win? -- Jeffries
        • Dave Rooney
          ... Naw, I have fat ankles. :) ... I m a firm believer that you can have too much of a good thing in any programming paradigm, not just OO. I ve seen crap in
          Message 4 of 10 , Feb 1, 2005
          • 0 Attachment
            > -----Original Message-----
            > From: Ron Jeffries [mailto:ronjeffries@...]
            > Sent: Tuesday, February 01, 2005 6:42 PM
            > To: extremeprogramming@yahoogroups.com
            > Subject: Re: [XP] [Slightly OT] OOP Article on DevX
            >
            > On Tuesday, February 1, 2005, at 3:44:04 PM, Dave Rooney wrote:
            >
            > > "And the profession is now guarded by a priest class that benefits
            > > from OOP's murk and mystery-the fewer people who can
            > communicate with
            > > computers, the more secure their jobs." Eh?! Do I need to run out
            > > and buy a frock?
            >
            > You certainly may. That word "frock" may have a different
            > meaning in my lexicon than it has in yours, however. Still,
            > people should dress as they like, that's my belief. :)

            Naw, I have fat ankles. :)

            > The article and the one it references both remind me of
            > "topmind", a famous anti-OO poster on the newsgroups. I don't
            > get it, myself: I kind of like OO, because it lets me program
            > in ways that I find to be more clear.
            >
            > The author does mention the incredible morass of the object
            > libraries, and I agree with that. But I guess I'd rather have
            > them than not, though they could be simpler, and better
            > abstractions, than they usually are.

            I'm a firm believer that you can have too much of a good thing in any
            programming paradigm, not just OO. I've seen crap in the OO world, and
            truly beautiful C code that expressed itself clearly.

            What I have experienced myself, and see on a daily basis, is that people
            first learn how to "use" objects but they don't necessarily understand
            them. At some point a light bulb comes on, and you find that you're
            actually thinking in terms of the abstractions, relationships, etc., and
            moving beyond simple inheritance. IME, that light doesn't come on for
            everyone. I suspect that's the case with the author of the article.

            I had a look at the web site the article's author references, and had a
            flashback to my Clipper days in the early 90's! A company I worked for
            had an almost completely data-driven application engine that worked OK
            as long as the rules for the data were straightforward. Of course,
            there were always exceptions to the straightforward rules, and the 80/20
            rule kicked in.

            Besides, the data dictionary shown in the Table Oriented Programming
            example looked to me to be full of repeating groups, especially if an
            item had more than one pre or post-validation rule. When I learned
            relational theory, I was taught that duplication was a very bad thing
            (funny how that followed to the OO world). Of course, the author
            doesn't like relational theory either, which is almost as bad as not
            liking puppies. ;)

            Dave Rooney
            Mayford Technologies
            http://www.mayford.ca
          • Jeff Grigg
            ... His arguments are just so far off in left field, they re not even on the field any more. I m an OO advocate, but I m quite willing to admit that there
            Message 5 of 10 , Feb 1, 2005
            • 0 Attachment
              --- "Dave Rooney" <dave.rooney@m...> wrote:
              > I saw a reference to this article on TheServerSide.com
              > this morning...
              > http://www.devx.com/DevX/Article/26776/0/page/1
              > Thoughts anyone?

              His arguments are just so far off in left field, they're not
              even "on the field" any more. I'm an OO advocate, but I'm quite
              willing to admit that there are cases where other techniques are
              more appropriate. But he just rants against everything and everyone
              he can see. ;->

              Something that seems to explain the behavior and opinions of some
              people in our industry is that some people develop strong
              abstraction skills, while others focus their abilities on
              more "concrete" thinking. "Concrete" thinkers are typically stable,
              reliable, hard-working individuals; they get lots of code written,
              and are good at working through legacy code and tweaking it here or
              there to fix bugs or add requested functionality. But they
              generally don't see abstractions as valuable, as abstraction "gets
              in the way" of seeing all the code and its procedural execution.
              And abstractions complicate the act of patching code in arbitrary
              ways to achieve given desirable procedural results.

              The most successful developers I've met, however, have very strong
              skills at understanding, introducing and managing abstractions in
              the programs they write.


              Beyond that, some people just seem to be filled with hate. 'do'no
              what to do about them. ;->
            • Michael Feathers
              ... When I read it, I had terrible deja-vu. I swear, it was just like reading critiques of OO in the late 1980s. Michael Feathers author, Working Effectively
              Message 6 of 10 , Feb 1, 2005
              • 0 Attachment
                Jeff Grigg wrote:

                >--- "Dave Rooney" <dave.rooney@m...> wrote:
                >
                >
                >>I saw a reference to this article on TheServerSide.com
                >>this morning...
                >>http://www.devx.com/DevX/Article/26776/0/page/1
                >>Thoughts anyone?
                >>
                >>
                >
                >His arguments are just so far off in left field, they're not
                >even "on the field" any more. I'm an OO advocate, but I'm quite
                >willing to admit that there are cases where other techniques are
                >more appropriate. But he just rants against everything and everyone
                >he can see. ;->
                >
                >

                When I read it, I had terrible deja-vu. I swear, it was just like
                reading critiques of OO in the late 1980s.

                Michael Feathers
                author, Working Effectively with Legacy Code (Prentice Hall 2005)
                www.objectmentor.com
              • Jim Standley
                As so often is the case, Ron said it better than I would have. OO seems to work for me. There are bits that aren t perfect, but they re still better than most
                Message 7 of 10 , Feb 2, 2005
                • 0 Attachment
                  As so often is the case, Ron said it better than I would have. OO seems
                  to work for me. There are bits that aren't perfect, but they're still
                  better than most other things I've tried.

                  That article ends with a glowing link to the "OopBad" page. I thought
                  parts of the article sounded too familiar. Somebody was quoting this
                  stuff on another discussion board last year. It's mostly unreadable but
                  if you suffer from too-low blood pressure ...

                  http://www.geocities.com/tablizer/oopbad.htm

                  Ron Jeffries wrote:
                  > On Tuesday, February 1, 2005, at 3:44:04 PM, Dave Rooney wrote:
                  >
                  >
                  >>"And the profession is now guarded by a priest class that benefits
                  >>from OOP's murk and mystery—the fewer people who can communicate with
                  >>computers, the more secure their jobs." Eh?! Do I need to run out
                  >>and buy a frock?
                  >
                  >
                  > You certainly may. That word "frock" may have a different meaning in
                  > my lexicon than it has in yours, however. Still, people should dress
                  > as they like, that's my belief. :)
                  >
                  > The article and the one it references both remind me of "topmind", a
                  > famous anti-OO poster on the newsgroups. I don't get it, myself: I
                  > kind of like OO, because it lets me program in ways that I find to
                  > be more clear.
                  >
                  > The author does mention the incredible morass of the object
                  > libraries, and I agree with that. But I guess I'd rather have them
                  > than not, though they could be simpler, and better abstractions,
                  > than they usually are.
                  >
                  > Ron Jeffries
                  > www.XProgramming.com
                  > First they ignore you, then they laugh at you, then they fight you,
                  > then you win. -- Gandhi
                  > I'm on track. When do I start to win? -- Jeffries
                  >
                  >
                  > To Post a message, send it to: extremeprogramming@...
                  >
                  > To Unsubscribe, send a blank message to: extremeprogramming-unsubscribe@...
                  >
                  > ad-free courtesy of objectmentor.com
                  > Yahoo! Groups Links
                  >
                  >
                  >
                  >
                • Gary Feldman
                  ... That s my take, too. Still, I think it s a legitimate question to ask whether practices appropriate to team development are also appropriate when dealing
                  Message 8 of 10 , Feb 3, 2005
                  • 0 Attachment
                    Jeff Grigg wrote:

                    > His arguments are just so far off in left field, they're not
                    > even "on the field" any more. I'm an OO advocate, but I'm quite
                    > willing to admit that there are cases where other techniques are
                    > more appropriate. But he just rants against everything and everyone
                    > he can see. ;->

                    That's my take, too. Still, I think it's a legitimate question to ask whether practices appropriate to team development are also appropriate when dealing with a programming project that is being done by exactly one person, will always be done by that same person, and will never be given to a different person. Certainly that's what I expect from the type of business he seems to be in; it's unlikely anyone would ever take over his code.

                    > Something that seems to explain the behavior and opinions of some
                    > people in our industry is that some people develop strong
                    > abstraction skills, while others focus their abilities on
                    > more "concrete" thinking. "Concrete" thinkers are typically stable,
                    > reliable, hard-working individuals; they get lots of code written,
                    > and are good at working through legacy code and tweaking it here or
                    > there to fix bugs or add requested functionality. But they
                    > generally don't see abstractions as valuable, as abstraction "gets
                    > in the way" of seeing all the code and its procedural execution.

                    I'd have to think about this a bit. I tend to be a very abstract thinker, and not the sort who gets lots of code written. But it's because of the latter point that I tend to do much better with legacy code. When writing from scratch, without being very strict about TDD, I'm always finding new abstractions or other ways to improve my never-ending design. In other words, I do better with legacy code because it suppresses my abstract distractions.

                    Your description of concrete thinkers makes me think of the sort of programmer who's quite happy working in a shop where the designs get thrown over the wall by the architect/design/analyst team, and they're just coding to spec.

                    Gary
                  Your message has been successfully submitted and would be delivered to recipients shortly.