Unconfigured Ad Widget

Collapse

Announcement

Collapse
No announcement yet.

IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #61
    IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

    ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
    Yeah, but then your code is littered with /End-Free and /Free wrapped around every single MOVE opcode. Not pretty. Chris

    Comment


    • #62
      IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

      ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
      Gosh. How about indentation? Code readability goes through the roof. How about coding file I/O without a single KLIST? This makes coding faster and debugging faster. In columnar RPG, if you want to see the KLIST values in debug, you have to find the KLIST definition first, what a pain. In free form, you just hit F11 right there in the I/O opcode or hover the mouse. To me, those are the 2 biggest advantages. Chris

      Comment


      • #63
        IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

        ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
        Well, I question your use of the word "legacy". Code that has been running for 15 years but is being tweaked regularly for changing business conditions is not "legacy" in the sense that you seem to mean it (i.e., old, unchanging code). I'm not worried about shops that have 20-year old systems with no source. I'm talking about the many shops that have large bodies of code that change regularly. First, is training. While I know there are shops that are still happily running RPG II code, I also know that it's easier to train new programmers when you only have one dialect to learn. It's the same argument for moving RPG II to RPG III or RPG III to RPG IV, except that in orer to do so, you have to inspect each MOVE instruction. Second, the bigger issue of broad acceptance is going to be directly affected by how many shops will use /free. At this point, because of the lack of a MOVE operation, a lot of shops choose to simply not move at all. Forced to choose between a single standard and new features, they choose the lower risk route: not changing. Joe

        Comment


        • #64
          IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

          ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
          Please note that I am NOT arguing against /free; I use it all the time. My problem is that a simple %move BIF would have helped tremendously. I personally disagree with Jon and anybody else who says that MOVE caused errors; it caused no more errors than data structures or pointers or any other programming capability (let's see how many people get bit by MOVE-CORR). Any competent programmer can handle MOVE; anybody who can't needs to broaden their career horizons. Me, I'm tired of the industry being dumbed down. I don't know how to read point 7. When you convert to /free, the MOVE simply gets wrapped in /end-free and /free. Maybe Linoma's tool does better, but that's $1000 I shouldn't have to spend. And finally, as to IBM's intention, I well and truly don't care what IBM's intention was: they've consistently gotten that wrong over the years, haven't they? It's IBM's intention to remove support for the RPG II compiler as well, but that's not going to happen is it? It was IBM's intention that we should all use OS/2. It was IBM's intention that Visual Age for Java was the IDE of the future. Heck, it was IBM's intention that we all dump RPG for Java and Linux. IBM's intention and three bucks gets me a Starbuck's coffee. IBM needs to stop telling me as a programmer how to do my job and instead just supply the tools and get out of my way... and the first thing is to provide a conversion path to /free that is inclusive. If Linoma can do it, IBM can do it. Joe

          Comment


          • #65
            IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

            ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
            Is it still 1965? What other languages other than RPG or Assembler still use fixed columns for coding? This may have worked good for punched cards, but who in their right mind who has been exposed to any other language would want to code in fixed-format RPG, trying to figure out the columns & indicators?

            Comment


            • #66
              IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

              ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
              Hassan wrote: "I am sick of this discussions and would not have thrown my two cents if it were not from a guru like you." I agree 1004%. This issue has been debated to death a thousand time before already, both here and on other on-line fora. Everything that could be said about the issue has already been said, and further discussions aren't going to make much difference one way or the other. There can be only one explanation why the host of this forum would allow such an editorial to be published - to increase traffic to this web site and increase the number of ad impressions. As for me (one of the compiler developers involved with the implemenation of /FREE), I'm outa here. I'll be back in a week or so when this debate falls back into the archives. Cheers! Hans

              Comment


              • #67
                IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

                ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
                Hassan, I agree with you about indenting. I made that point down at the end of the article...I thought. :-) Anyway, don't misunderstand my concerns. I prefer writing my code in free form for exactly the reason you mention...its WAY easier to read. I just don't see the "vast language improvements" that I keep hearing everyone talking about. Thanks for your input. Jeff

                Comment


                • #68
                  IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

                  ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
                  I agree 1004%. This issue has been debated to death a thousand time before already, both here and on other on-line fora. Everything that could be said about the issue has already been said, and further discussions aren't going to make much difference one way or the other. And that's what's wrong with the issue. It would take almost no work for IBM to provide a %move BIF. Let us use it if we wish, and not if we don't. Those who agree with the Great Gurus of Programming can eschew the MOVE, while those of us who work for a living can use it and have a single syntax to standardize on. The continued lack of a %move BIF is simply unconscionable. We work around it, but the fact that there's still not one available means that IBM is still ignoring the needs of programmers and listening to Computer Science Academicians. Honest, folks, some of us don't WANT to write %editc(MYNUM:'X') and we shouldn't have to. Joe

                  Comment


                  • #69
                    IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

                    ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
                    Joe, I think you touch on a very important item; resistance to change. There are many developers and I.T. managers that will simply avoid making the switch because they see only the difficulties and little in the way of "return on investment". Thanks, Jeff

                    Comment


                    • #70
                      IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

                      ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
                      Dave, While you are correct that this is a new feature. It is somewhat watered down by the push IBM (and evertyone else) is making to go to SQL. That's not to say its not useful just not as valuable as it once would have been. Thanks for your input. Jeff

                      Comment


                      • #71
                        IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

                        ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
                        Joe, MOVE was fast, but I'm not so sure about simple or elegant. We had to use MOVEL for character fields, and I preferred Z-ADD for numerics. Today, '=' is polymorphic enough for me, and I can use EVALR to right-justify character fields. Loading a field by using MOVE and MOVEL was critically dependent on the lengths of the source and target fields: changing a field length also changed the resulting value, and it was possible for the source fields to overlap each other in the target field. %subst() and '+' give precise control over how a target field is loaded, and the code is more readable. Your point about the number of MOVE statements in legacy code is well taken, but the difficulty in converting to /free is also the difficulty in maintaining the legacy programs: MOVE statements must be inspected to determine whether they're value assignments or conversions, which takes us back to flipping through code. Like you, I have been programming RPG for a living for a few decades, but different experiences apparently lead to different conclusions.

                        Comment


                        • #72
                          IMHO: Is RPG Free-Form a Cat in Tiger's Clothes?

                          ** This thread discusses the article: IMHO: Is RPG Free-Form a Cat in Tiger's Clothes? **
                          I agree with everybody who says that free form is pretty poor when it comes to data transformation. Move(l) etc. is something you have to do with some regularity. What really gives me heartburn though, is the lack of support for MOVEA. That's the one thing that makes me have to get of of free form and back into the 'C' specs.

                          Comment

                          Working...
                          X