Reader Feedback on IBM’s Reorg: The Good Me or the Bad Me?
August 20, 2007 Timothy Prickett Morgan
|
As is his wont, Brian Kelly struck some nerves with his article last week, IBM’s Reorg: The Good Me or the Bad Me? Here’s a sampling of some of the feedback that readers sent in. Brian: Many thanks for your great IT Jungle article. I believe that you speak for the vast majority of System i (AS/400) customers and the System i ecosystem. IBM clearly has the leading capability in System i hardware, and the woefully trailing capability in GUI applications development, compiler software, and in marketing. You are correct, again, that the best hardware will not save the System i and the System i ecosystem, without IBM delivered and customer accepted native GUI applications development capability. Do you want me to forward your article to Mark Shearer? I do not have the email addresses of all of the pertinent IBMers, including Ross Mauri, Marc Dupaquier, and Bill Zeitler. All the best, –Paul I have worked with System/38, AS/400, iSeries, i5, and now System i for more years than I will admit. I started out as a VM370 systems programmer. Your article states so perfectly my deep frustration with IBM’s failure to develop the incredible potential of the AS/400 (OS400/DB2, not the hardware). The failure to give AS/400 developers a native Windows/Web-like GUI permanently relegated OS/400 and DB2 to the margins. The AS/400 has contributed greatly to my livelihood, and probably will continue to for a few more years, but it could have been so much more. I had hoped that IBM would spin-off Rochester. You say you’ve got to have hope (the Good you), but in this case you are expecting different results from the same repeated behavior; and you know what that defines. No disrespect; I have gone somewhat nutty hoping IBM will do the right thing for us AS/400 developers. I haven’t replied to any article before, but this was so well said. –Roger Hi Roger, Thank you very much for your kind words. When the spirit moves me, I take the opportunity to put into words the things I believe would make the AS/400 product superior again. I am fortunate to have a great person in Timothy Prickett Morgan who, though 20 years my junior, is tuned in to the frustrations of the AS/400 constituency, and he provides me a forum to reach folks like yourself as well as IBM people who may be able to make the platform better for us all. When IBM tells me that my ideas are bad, or because of such and such the resources are unavailable, or there already is a better solution, I do get disappointed and sometimes I think about giving up on it and going in a different direction. Sometimes I even start that direction and gain some technical capabilities that I did not have. In all of these instances, as I see the degree of work that is required to do simple things in WebSphere and Java and even PHP and MySQL, I know why I love the simple elegance of a development and execution environment that does not rely on me knowing the 15 parameter of parser options 1 to 3. Until I looked at these methods in detail, I let IBM do all of my parsing for me. Now, with the “new advanced tools,” such as PHP and even good old HTML, it is my job to do. Seeing how tedious the work is and how much there is to know with HTML, XHTML, CSS, PHP, MySQL, AJAX, JavaScript, J2EE, JSP, JSF, etc., etc., etc., to do something simple on the Web, I herald those brave souls in Rochester who snookered IBM 30 years ago into believing that the System/3 was really a modern 407 Accounting Machine with card sorting capability. Yet its OCL that looked nothing like JCL and its RPG II compiler (even the card version) was leading edge at the time. Rochester’s computer pioneers accomplished with chicanery much more than they ever could have accomplished by straight talk. And, the System/38–wow! No company in its right mind would have invested in a large system architecture with ease of use features knowing how long it would take to pull off. No company–including IBM–has yet to build a machine with such innate computer science facilities included with the OS. Computer science projects such as EROS are the closest the industry has ever come–and EROS is a research project. The 30-year-old System/38 actually implements the notions that EROS is testing. IBM should be able to sell that. But, I have decided that IBM doesn’t want to sell it because it isn’t the way people are already thinking, or some other non-reason. That’s IBM’s loss. But this notion of thinking it is OK to not develop a simple way for developers to design and render a Web panel is what makes this powerful i5 box appear to be nothing special in the industry and it’s what keeps guys like us from making a killing instead of just a nice living. In 1977, 30 years ago and just two years after the System/32, IBM built a far better mousetrap. Instead of a built-in screen, the monitor and the keyboard could be detached and there could be up to 16 of them locally on a system–and even more remotely attached. IBM proved with the System/34 that green-screen terminals could be natural devices to the system and its compilers. Before then, small systems such as the System/3 needed CCP and mainframes needed CICS. Mainframes still need CICS because IBM chose not to give them the better way–the mainframe—to pay for what you need. Surely IBM could have buried CICS in the MVS OS if it chose to, but it chose not to since it was OK for mainframes to have an associated major degree of difficulty. IBM in Rochester got beyond needing a special monitor like CCP or CICS. The same IBM could deliver the Web without a PHP engine or CGI engine or JVM or a Web Application Server (WebSphere). Why all the complexity? If there were a renegade left in Rochester, the browser would already be a device to RPG and COBOL, and a GUI designer with the same ease and facility of SDA would be available to create the “Web File.” And maybe that Web file would consist of a lot of DDS statements that today already exist and another bunch that right now do not exist. I’d feel better if IBM said they know that this facility is needed and they are working on it with great dispatch than to pretend like the emperor that the look and feel of his clothes are just fine. So, regardless of whether IBM likes what I say, because I hear from folks such as you, Roger, who know how much this capability would mean to AS/400 developers, I will probably be inspired again. –Brian AAAAAmen!!! The AS/400 did not die; its daddy (IBM) slit its throat with all of the practices you mentioned in this article. I have been watching IBM advertising to the general public for 15-plus years and I have yet to see an ad that mentioned the AS/400 (these are only to be found in AS/400 trade journals). And not that many in the trade journals, either. Of course, the Good Brian doesn’t really expect IBM to follow his suggestions. Why, oh, why would a company with a product that has already separated the display definitions into a separate (compiled) module not develop a GUI and Web version of this function? About 90 percent of the work could be segregated to the DDS code or the display (GUI or Web) file module. Thanks for your support, –Chuck Hi Chuck, Thank you for your kind response. The Bad Brian had a lot more material to work with than the Good Brian. Will this mean anything to the Bad IBM or the Good IBM? I am with you. What is wrong with DDS? It is well conceived, easy to read, clean, straight-forward and it’s even keyword oriented–ahead of its time. It is a perfect companion to the keyword-oriented RPG IV language. Who says that DDS is legacy and HTML and XHTML are modern? Thank you for your comments and your encouragement. –Brian Hi Brian: Just read your article today. Got a little long but had a lot of good points. I was curious to know if the community has ever filed a formal complaint about the lack of native GUI on the System i5? Similar to the battle we won (are winning) with WDSC packaging. Thoughts? –Aaron Hi Aaron, I just got back from the shore. Thank you for your response and thank you for enduring all the set up work in the article. I do not know the meaning of your term, “formal complaint” in the IBM context. I am very aware of the WDHT battle and in fact, I added a new wrinkle to it in my article about the Web Runtime Tax a few weeks ago–a tax that as I see it is just as onerous as the Advanced Edition being needed for designing green screens. I send note after note to the IBM System i hierarchy and occasionally I go higher than Mark Shearer, and to me they are formal requests if not complaints. I normally offer a solution, but my message is clearly both a complaint and a recommendation. IT Jungle has given space to the GUI issue in a number of pieces I wrote over the last several years. Other than TPM and myself, however, I am unaware of any other person or group really hammering IBM formally or informally about no GUI on the system and no GUI in RPG. In fact, because I roughed up a few IBM people in one of my articles there were a few AS/400 industry gurus who attacked me personally for suggesting that GUI was more important than a few more BIFS in RPG IV. As you know the BIFS are very nice but, compared to GUI, they are a pittance to provide. If you know of a place where we can all lodge a formal complaint, please make me aware. Thanks again for your comments. Below are a few links you may enjoy. –Brian RELATED STORIES Wanted: Native RPG and COBOL Support for Browsers Readers, Vendors Weigh In on Native Browser Support IBM’s Reorg: The Good Me or the Bad Me?
|
