IMHO: What Will It Take to Turn the System i Around?

[tleach@sierrapine.com]

I really can't believe that someone would actually compare PC reliability with the iSeries! As for why IBM doesn't get off their duff and market the machine, I have no idea. I did notice that last year when they did make an effort at marketing the "black box", sales went up dramatically. So I know that IBM can do it! They just need to continue to market "OUR" machine, like it really matters to them! That's what I would like to see more of!

[Guest.Visitor]

IBM does need to improve their marketing substantially. But DeCoville's comments about the database don't make sense. It's better than any other database out there: record level access and SQL access are one reason (combined) that I say that. And what's his problem with ASNA and RPG .NET? Their installed base is huge, technical support is unrivaled, and they have the best set of tools to keep current with Windows/Web interfaces to the iSeries.

[Guest.Visitor]

This has been debated like forever. But let's at least frame the topic to why IBM won't change it as opposed to why isn't it happening. Let's dispense with the notion that DB2 is a problem or that it's number 1 in the market. Oracle is number 1 and every other DB is a distant also ran including your beloved SQL server. Second, ASNA for whatever it's worth at 90 FTE's if that's the case is pretty darn good for a iSeries vendor. That said, just how many FTE's you think IBM's got on the "i". I'll bet it ain't much more than that. The reason is that so much of the software is shared these days and remove the shared hardware guys and how many are really FTE's on the "i"? The same reason DB2 is even used is the same reason that IBM machines are still around, they're stable, reliable, bullet-proof and painfully routine. IBM will never cede control of the machine like all other platforms do and that's the difference. First they removed the assembler compiler, then they might have well removed MI from the user base, so what incentive is left for a vendor to port or try and port a product to the "i". The C/C++ compiler is okay but to get anything really done with it you have to know the machine well and to bother to learn the machine well means you want control and access, IBM doesn't allow that. The market thrives on innovation, a vendor isn't going to be able to innovate on the "i" because it's locked down. All other platforms allow total access to the machine, IBM locks it down and they will never cede that control until it's too late. No access, no products, no innovation. The day Oracle(etc.etc.) can run on the 'i' is the day the machine will live on. The OS was never built to run multi-thread apps like all C/S languages thrive on so GUI is not going to be on the 'i' unless there's a significant OS re-write. If you don't believe me just go look at the API's and see how many really run multi-threaded.

[OzzieH]

Hey, I hear ya, but OTOH, I don't know if I can go your way anymore. I have friend of mine who hasn't rebooted a Linux server in 9 months.

[mollykj]

Dear OS: I've seen iSeries run for a year without a reboot.

[Guest.Visitor]

I do agree that IBM's Marketing has, at best, played a reactive role; where as Microsoft is busy proactively getting their hooks in developers by the millions by giving away a "starter" version of the entire Visual Studio suite. And while the traditional Windows developer has always been able to code "in a vaccuum", the appearance of any inferior (i.e. Source-Safe) tool to facilitate development teamwork appears as a "God-send", no matter how rediculously simple the tool may be compared to some of the very mature team development platforms from IBM. Time and time again though, history has shown us that it is not the superiority of a product that wins in the market place, but it is the superiority of the marketing and catering to the masses that help us to understand why an inferior product is better for us. Spending time in my career in virtually every technical position and 6 years of directorship in highly dynamic environments, I have seen technologies come and go. Many deservedly were relegated to the vast waste land of packages that followed the "good idea, bad implementation" model. Taking a peek at history at the other products, you can ask: How was Michael Dell and his dream of competing with IBM able to become the #1 computer company? How was Bill with his MS-DOS able to squash DR DOS? How was Bill with Windows 3.x and 95 able to kill OS/2? What happened to Lotus Notes and Domino? In each of these cases, marketing and the target audience made all of the difference. Microsoft went after the masses with its "Model T" version of everything. While IBM tried to sell superior PS/2 computers to everyone for everything, a typical end-user would not be affected by the additional features of the PS/2, so paying half price for a generic knock off and eventually Dells was a no brainer. MS-DOS.....pure marketing in 1994.....as Microsoft made use of trade magazines and "leaking" inside info on the vaporous DOS 7 product to get people to hold off on bying DR DOS (DR DOS was multitasking when multitasking wasn't cool). MS DOS 7 was never released as a stand-alone product. It was the core of Windows 95 which started selling in August 1995. OS/2 and the Lotus products both followed the PS/2 fate of being "too superior" (i.e. too expensive) for the masses. Sadly, I see the same thing happening with the iSeries and DB/2. Dot Netters will tell you how cool SQL 2005 is with its CLR support. IBM's DB2 for Windows was doing this in 2003. Had I been IBM, my commercials would have had a blast with that. And the iSeries, while it does not support the .Net CLR, all database applications have always had access to every iSeries system resource since the beginning. Now, it's nearly two decades later, and IBM still hasn't seen the light. Young developers and users grow up into the next generation of managers and adults that make the decisions about which computers and software to buy. And I am not sure which IBM chief(s) has directed these strategies of the past, but I am sure that he/she is paid too much. What lessons can be learned here: 1. Marketing is nothing more than cosmetic makeup for a company. It's marketing's role to highlight the strengths and hide the weaknesses. Score Microsoft 1, IBM 0 2. Small companies grow into larger companies. Small SQL servers grow into big SQL servers. Small companies can afford a $2500 Dell server and $4000 worth of software licenses. Score Microsoft 2, IBM 0 3. Young developers (hobbyists and students) don't have much money. They do what they can afford. These developers grow into adults and development managers with money to spend. Then, they do what they can afford and what they know. Microsoft 3, IBM 0 4. You don't drive a Grayhound bus to take one person to the grocery store, and you don't take a caravan of 20 cars to haul 50 people across country at one time. With this analogy, I am purely baffled at how people can even begin to compare Windows with the iSeries.!?!?!? YOU'RE TALKING GRAYHOUNDS AND BMW's!!!! Oh yea, IBM's marketing let us do it. Microsoft 4, IBM 0 At best, IBM should fire anyone in their marketing management structure. If the decision were mine, I would probably sue them as well for pure incompetence.

[mollykj]

but 'fess up - who are you ?

[R.Daugherty]

At best, IBM should fire anyone in their marketing management structure. If the decision were mine, I would probably sue them as well for pure incompetence. Here's the problem. They're incompetent from our perspective, but they are doing their job from IBM's perspective. The sooner people understand that IBM's only concern is selling Websphere, and tying people into it so that they have to keep paying for it, the sooner we start talking about reality. Any sales of an AS/400~iseries is purely coincidental, as in these are type i folks to nail with Websphere. In addition, these worthless (to us) IBM'ers have designated the iseries as a small business computer. This is ludicrous and along with Websphere being their product to sell is the other nail in the iseries coffin. This is not a minor philosophical point. They essentially have categorized the AS/400 as a partitioned disk server and are trying to convince small businesses (small in the realm of corporate America, the category is SMB) that this computer is easy to operate and cheap because they don't need admins. The basis for this is "one backup command". However, anybody but a clueless IBM'er knows that putting OS/400, Linux, AIX, and Windows on one box doesn't eliminate the need to admin those operating systems. So the alleged, and when I say alleged remember that IBM has decided this is what the iseries purpose in the world is, the alleged savings are from common virtual partition savings available everywhere, to solve problems Windows and Linux people have with underutilized disk and CPU in their one server for each app world. So it doesn't even apply to us to start with. You will always see IBM and their shills talk about selling the iseries (only as a container for Websphere, by the way) to small business. I could go on, but you get the point. Somehow the AS/400 has been categorized as a System/36 by IBM while real business is slated for IBM's favorites, AIX and Z/OS. This despite the fact that I've worked for several multi-billion dollar companies over the last 12 years running on the AS/400. But you'd never hear that from IBM. It's not in their Websphere marching orders. rd

[bharder@nlrha.ab.ca]

I have said for some time that a key IBM strength on the iSeries has become a boat anchor. The problem is the backwards compatibility. It's so good that no one actually has to change. Programmers, vendors, software architects, everyone can keep doing things the same old way. Current versions of applications can still use 30 year old interface technology. They often do (not every one of course, but far too many). On a somewhat different note, I remember reading a book about IBM years ago. It talked about the Thomas J. Watsons, both Senior and Junior. It quoted someone whom I believe was TJW Jr. This was back in the mainframe days of Sperry, Burroughs, Honeywell, Control Data and the like. The essence of the quote was that IBM's competitors would repeatedly release computers that were better than IBM's. Nevertheless, IBM would overpower those competitors over and over again. The source of IBM's strength was marketing, service and support. Now Microsoft has taken those historical lessons and triumphed over IBM.

[bharder@nlrha.ab.ca]

Oh, and something else I forgot. Much of computing is now a commodity, and we have to adjust to that. Sure the iSeries is reliable. But now, so is Windows, Linux, Mac, Solaris, and many others. You think the iSeries value proposition is unchallenged? Look around! I don't think Windows is as reliable or scalable as the iSeries. However a business can get started with a low-end competitive solution for much less than an iSeries. By the time they are hooked and paying much more, it's too late for them to make an easy change. Many businesses consider this "water under the bridge" time, if they consider it at all. The reality is that by the time a business is relying upon a specific computing solution, careers in that business depend upon going forward with that solution. Even database technology is becoming a commodity. Why do you think that overseas outsourcing is so viable? It's a commodity too. The commoditization of our industry is here and it's changing everything. Don't agree with me? Take a look at the glory that was Cray, and Silicon Graphics, and Tandem, and Thinking Machines, and Apollo, and many other high-end computing companies. Cheap, commodity machines have eaten out the bulk of the markets they used to sell to. At the low end, IBM sold off their PC Division to Lenovo to get rid of a commodity business arm.