IMHO: The Day the Software Died

[J.Pluta]

** This thread discusses the article: IMHO: The Day the Software Died **
** This thread discusses the Content article: IMHO: The Day the Software Died **
0

[frankgw@adelphia.net]

** This thread discusses the article: IMHO: The Day the Software Died **
If you haven't read this excellent article by Joe Pluta, I strongly recommend that you immediately do so. It does an excellent job of documenting many of the reasons that my current frustration level is at an all time high when trying to determine what the best moves are today for a smaller mostly custom coded ILE RPG AS/400 shop. "Some of IBM's tactics have shaken me to the core." Agreed! However, as you go on to document, many of these tactics and trends are industry wide - not just limited to IBM. All the more reason to be "shaken to the core". Joe - Thanks for writing this excellent article. To the rest of you - Are Joe & I being unduly pessimistic?

[J.Pluta]

** This thread discusses the article: IMHO: The Day the Software Died **
While I hate to come off sounding so pessimistic about a platform I have supported for so long, it's becoming obvious to me that there are some trends being set, and I don't like them. This is just one of them. As another example, I recently posted a long soapbox rant on one of the mailing lists about the decreasing attention being paid to backward compatibility from release to release. What makes me so angry is the fact that the IBM midrange, as it has evolved, is without a doubt the most powerful, resilient, flexible platform ever created for developing business applications. But much of that power comes from the legacy code that has been developed over the past couple of decades, and perhaps more importantly, the legacy programmers that developed it. I don't care how many neat bells and whistles you have, if you don't have the knowledge base of 20 years of ERP experience, then you aren't gonig to be able to write a decent ERP application. I've been told that SQL syntax is "easily understood by any competent database administrator". All fine and good, but how many database administrators know what an MRP gen is? While I'm all for continued integration of the midrange with all the newest technologies, if it happens at the expense of that incredible body of legacy knowledge out there, knowledge that is housed in the minds of the legions of programmers who have created those wonderful applications, then it's too costly. I may suffer from a vastly overinflated sense of worth of both mself and my programming peers, but I believe that the midrange community has created the great majority of all real business software in place today. And turning the most productive business application development platform on the planet into an SQL-serving also-ran seems to me to be just about the most ill-advised decision made since SSA decided that Unix was the future of BPCS. But hey, that's my opinion, and I'm sticking to it. Joe

[R.Daugherty]

** This thread discusses the article: IMHO: The Day the Software Died **
I just rated it excellent, Frank. Joe was right on. IBM has been fixated on changing from selling computers to selling access to Websphere at $450 per user *per server* per year for some time now, and I have gone over the top railing about how they are steadily migrating a required connection to it for any IBM software to work, but they had bamboozled everybody with a free version. I saw the exact same scheme with OS/2 development. IBM will never get it, they're just not a software company, and it used to not matter, but they have now bet their future on it and in the process sacrificed the AS/400 at the alter of Websphere. For more of this trend, PC Week has an article this week on Lotus developers being bent out out of shape over a rewrite to Websphere. I noted here a few months ago how IBM usurped the Lotus knowledge based project Raven to require Websphere, followed by major defections from Lotus. Lotus Domino was never cheap, and Websphere will soon reflect that, believe me. I understand there is app server competition out there, and that Websphere is a J2EE compliant app server with hooks to require IBM's Websphere instead of any other J2EE app server. It has always been this way with OS's, and to some degree hard to picture a viable business model without proprietary extensions or at least hooks, but this is clearly a battle of the Apache Foundation type worldwide developers against IBM hooking everything they sell into Websphere and migrating from selling mainframes to selling connections to Websphere. I'll place my bets on open source software and against Websphere, but the AS/400 is a casualty either way it goes. As I say, I'd love to be wrong, but the architecture we ran the business world with is being steamrollered by the IBM juggernaut of Websphere and SQL generic database access by Somers marketing people executing a plan, and the AS/400. Ralph