|
More Reader Feedback on EGL, State of System i, Pricing Disparities
Published: October 15, 2007
by Timothy Prickett Morgan
Even though they ran a few weeks ago, readers of The Four Hundred are still sending in feedback on two stories that ran recently, State of the System i: What's Going On in the Market? and EGL: The Future of Programming for the System i? Both stories asked questions, and tried to answer them, and the state of the System i story is but one in a continuing series that I am working on now.
One reader also reacted to the lead story in last week's issue, which covered a System i upgrade deal and discounts on BladeCenter gear. (See IBM Offers System i Blade Deal, Nixes i5 550 in Upgrade Deal for details.) IBM is requiring customers in Canada to spend a lot more in this deal than customers in the United States, even though the respective dollars in these two countries are now worth the same. (It is kind of shocking to me that the Greenback and the Looney are worth the same amount, but there are good reasons for that.) In any event, I suggested that Canadians insist that they only have to spend as much as U.S.-based companies, since the currencies are equivalent. Here is what one reader had to say.
TPM,
First, I want you to know I thoroughly enjoy your technical articles and have used them to influence my company's i5 buying. Second, the IBM pricing "unevenness" does not just apply to Canada, you should check into prices in Puerto Rico and the Caribbean Islands. I believe that they are all considered in the same IBM pricing group. It goes beyond currency exchange, it appears to be a matter of being able to charge what the market will bear.
--Mike
TPM,
Moore's Law seems to work against the AS/400, at least from a marketing point of view. Few companies see a doubling of ERP transactions every two years requiring the extra horsepower. And it is always easier to throw another dedicated PC server at a new, non-critical, graphical browser server application than to upgrade the main box. Granularity is attractive, even if more costly in the long run.
I contract at a company whose AS/400-based business activity is doing almost $200 million a year. Their eight processors and 600 GB of disk serves 300 users nicely, and has for at least four years.
You will like this: The former mainframe application development manager once said in frustration, "I don't understand how you can store seven years of data online and still get sub-second response time!" I started to explain single-level storage and 64-bit binary radix indices, but gave up. The company's mainframe (to be fair, probably a small model) uses a wicked combination of core IMS database surrounded by a forest of VSAM files, each with a 32-bit 4GB file size limit. The company has to copy off completed transactions to a SAS server for BI inquiries every five months.
To add insult to injury, we are de-commissioning the AS/400 legacy ERP to allow the mainframe legacy ERP to continue. I need a mainframe-to-AS/400 project to re-balance the karma.
--Doug
Brian,
Please accept my thanks and praise for taking a stand about EGL and RPG. As one who has long been both a customer and vendor of IBM midrange systems (I started on the System/3), I applaud your anger. It's time for us to take a stand. Hopefully IBM is reading these reactions.
As for IBM being angry with you, take pride you were able to awaken at least some faction of the sleeping giant. Perhaps IBM's anger will force it to take a good look at what it is doing to the System i platform. At least IBMers are expressing some emotion instead of continuing to ignore it.
I've been involved in software development for over 30 years now and have worked in several languages and platforms, including COBOL on IBM mainframes. That makes a good business platform for companies with enough money to support the infrastructure. As we all know, however, there is no comparison to the System i, regardless of programming language.
However, RPG made this platform and all that preceded it. If IBM continues to ignore the obvious, RPG faithful will also break this platform. I have stood strong for the System i, but my faith continues to be eroded and my defense of the platform is questioned more than ever. In the past couple of months alone we have seen IBM's commitment to the platform erode further with their decisions about Domino and EGL. I was considering moving a large Notes + SQL Server application to the i + DB2. Now, I'm just not sure that makes sense.
IBM should wake up to the true indication of the importance of their non-Intel and non-Unix platforms. Instead of number of servers installed, why don't they count number of end users serviced? That equates to the Windows desktop world and would give IBM a different (better?) perspective on their installed base.
Perhaps IBM should look to the ISVs for some good examples of how it is possible to offer a good RPG-based product for creating browser-based applications. Our choice (Profound Logic's RPGsp) hinged largely on its ability to code RPG in-line with HTML. Wow! Big learning curve for HTML, JavaScript, and AJAX, but we're doing it. Is IBM so weak now that it can't do that one better? Or did IBM just wait so long it is now afraid to produce a product that would impact the ISVs business?
To quote Ira, who stated our frustrations so eloquently: "Damnit!" Sorry for the diatribe, but my patience is wearing thin.
--Gary
Hi Gary,
I went back to the archives accidentally after writing that article and I noticed that it was not 10 years of Java. Java was announced May 3, 1994, and it has been miserable for the System i contingent ever since. Around this time or slightly before, Lou Gerstner forced IBM's mainframe and AS/400 divisions to build TCP/IP for real for the platforms since the Pascal programs were so poorly written and so low in function that neither platform was Internet worthy. You may recall that IBM thought it had TCP/IP and client/server covered with the PC and the RS/6000. IBM believed that its other systems (AS/400 and mainframe) that were SNA-oriented would not need a productive TCP/IP stack. I knew the Internet architect for AS/400 back then and he told me that the new Gerstner directed TCP/IP actually ran better than SNA, but he was afraid to tell the mainframe division guys that. I think Gerstner meant that they not just put the TCP/IP plumbing in, but that the mainframe and the AS/400 divisions do something with it. He also directed that client/server be a priority. The logical next part never got done as IBM became more concerned with platform code reuse than doing exceptional things with an exceptional platform.
So here we are, 13.5 years after Java, and Gerstner's job is not done yet. And, yes, if Infinite Software can emulate the whole platform and if Profound Logic, ProData Computer Services, BCD, LANSA, looksoftware, ASNA, System Objects, and others can come up with a solution, why not something integrated into the platform so it separates the platform from all others--like it once was with interactive terminals?
Your response helps make me think that it is worth the fight, but the part of me that deals with common sense in life says to me that what we seek is just common sense.
--Brian
Brian,
I read your EGL rant and fully agree. I do very little programming anymore, still old school using PDM on the green screen. But a number of years ago I got into Web development on the iSeries using Net.Data. This is a great tool for connecting RPG to the Web. You can call an RPG (or any other language *PGM) and pass parms with ease. It is easy to learn and it's FREE with the OS. I have never understood why IBM didn't promote Net.Data.
Just my two cents.
--Doug
Thanks for your two cents, Doug I think with inflation everybody's two cents has gone up substantially in value. The more I read what I wrote, the more it seems like a rant, but I feel better that you agree.
--Brian
Concerning the comments in the EGL article about "cross-platform" software. Windows is not cross-platform software; it runs on X86 and X64 machines (what used to be called IBM compatible) hardware. Well, now X64 runs Apple software, too. No one talks about Windows being obsolete and legacy-based.
For years (I mean 40-plus), IBM made fun of other systems and software as being outdated and obsolete. I guess the shoe is on the other foot and IBM is not willing to change and accommodate its user base.
These developments are so sad to see for an individual who thinks the RPG/AS/400 combination is the best development environment ever. I might add that this is an independent evaluation after many years of experience on almost all platforms and software combinations. I would expect most AS/400 fans have also come to their independent opinions also.
Another issue is Microsoft. For years, Bill Gates and Microsoft have had a reputation of being leading edge and innovative. They are neither. All of Microsoft's "innovations" have either been stolen or bought. Every company that does business with Microsoft, including Apple and IBM, ends up being robbed or raped. On top of the disheartening news on the AS400 concerning the Web, it is so discouraging to see everything going to the Microsoft-based systems when they are far from being the best solution.
Thanks for your hard work and effort to help the RPG/AS400 community. Keep up the good work and try to keep IBM's feet to the fire.
--Chuck
Hi Chuck,
You are certainly right about Microsoft not being cross platform. You are also so right about IBM deferring to Windows and not competing against it like you and I would if the AS/400 were our product.
Only IBM is to blame for the plight of the AS/400 and a plight it is when it should be held up as the best example of a best system.
IBM is so big and has so many businesses that are doing so much better than the AS/400 that as much as I would like to see it, I have a hard time believing that at the very top of IBM there can be any time for the AS/400 platform--as good as it still is. IBM is a services and software company, and a lot of its software runs on Windows. And since Windows is not as easy to work with, IBM sells a lot of services for Windows. Not much services or software revenue comes from the AS/400.
What other great technology or innovative notion does IBM need to have exclusive marketing rights to that gets its chance for success? Look at IBM's technological success record over time and you see the company that invented the IBM PC is out of the market and the company that invented the disk drive and held the most patents in disk technology got out of the disk business. What about the AS/400 as the most advanced computer science machine in the industry--64-bits for years before Intel and no hooplah? And there are lots more.
The only CEO that paid the price was John Akers as Big Blue frittered away opportunity after opportunity and lost over $13 billion in two years. Akers got fired over the money, not the lost opportunities. I sometimes wonder if Sam Palmisano, an IBMer promoted from within to replace Gerstner, ever heard of the AS/400?
I wish IBM the best and I urge the company do its best and I tell them what I think is the best. Maybe one day IBM will listen and focus enough energy on our platform so that it can sustain itself over the long haul. Wouldn't that be nice? Thanks for joining the choir.
--Brian
TPM and Brian,
I don't know why IBM keeps tripping over itself to announce and then bury Web and GUI products for the AS/400. Remember Net.Data, VisualAge, Java, JSP, WebSphere, WebFacing, HATS, iSeries Web Access, PHP, and CGIDEV2? Frankly I have tried almost every one and the only product that stuck with me was CGIDEV2.
All the other major industry payers, Microsoft, Oracle, and Sun Microsystems for instance, have worked this problem out years ago and have been releasing generations of coherent evolutionary solutions. IBM has absolutely fumbled here and is paying the price.
--MC
RELATED STORIES
State of the System i: What's Going On in the Market?
Reader Feedback on EGL: At Least It's Not Java, But It Ain't RPG, Either
EGL: At Least It's Not Java, But It Ain't RPG, Either
EGL: The Future of Programming for the System i?
 Post this story to del.icio.us
 Post this story to Digg
 Post this story to Slashdot
|
