The i 7.1s Have It; i5/OS V5R4 Extended
November 16, 2009 Timothy Prickett Morgan
Well, it looks that that future OS/400-i operating system release is going to be called i 7.1 after all, and thank heavens because as I said a few weeks ago, I was getting a bit tired of saying the next i/OS release, possibly called i 6.2 or i 7.1, due sometime in 2010. According to various customer and business partner sources who have been briefed by IBM in recent weeks, not only has the name been decided upon, but i5/OS V5R4 has had its life extended.
The naming convention for an operating system used to be a big deal back in the days of the AS/400 because releases implied that new features were added to the operating system at no cost, while a version upgrade meant that IBM had an excuse to raise prices for the new operating system version as well as upgrades to it, or in the case of the OS/400 V3R2 machines back in 1994, radically change from a per-server, software group price for the operating system and key middleware to a mix of user- and system-based pricing. Pricing that, by the way, bears some resemblance to how i 6.1 is currently priced.
I have no idea if IBM is thinking of raising prices for i/OS licenses as it shifts from i 6.1 to i 7.1, but you can bet IBM is sorely tempted to squeeze some extra dough out of i shops who cannot easily move their workloads to other platforms. But considering how expensive the i 6.1 stack already is, this would seem to be a particularly silly move. I want to be on the record now as saying this, and try to encourage IBM to lower its i software prices in an effort to stimulate demand. And I don’t think IBM should wait for i 7.1 to do this, either. In fact, if I were running the Power Systems division, I would be cutting i 6.1 prices now, particularly when the economy is bad, to try to push some more iron.
The word I hear from sources who have been briefed by Big Blue is that i 7.1 will be available in the first half of 2010; everybody knew that IBM wanted to feature i 7.1 at the annual COMMON midrange trade show in Orlando, Florida, in May, but it looks like i 7.1 could actually be shipping around that time as well.
The interesting bit of news I heard early week was that IBM planned to extend support for i5/OS V5R4, and will keep it alive in the market, available for sale, for 12 months beyond the launch data of i 7.1. Equally important for V5R4 shops (both current and future ones) technical support for V5R4 will extend out another 12 months after that. So, in theory, if i 7.1 came out in May 2010, you will still be able to get support for i5/OS V5R4 until April 2012. I guess the move up to i 6.1 is taking a bit longer at some shops than IBM had hoped. But, the program conversion process that is part and parcel of a move to i 6.1 is a hassle, and one that busy V5R4 shops don’t have time or money to mess with.
After I wrote this story–I am working ahead because I am traveling this week–IBM did indeed say in announcement letter 909-285 that it was rescinding its withdrawal of V5R4 in announcement letter 909-003 from this past January, which killed off V5R4 effective January 5, 2010, and will now sell V5R4 through January 7, 2011. Presumably tech support will be available at least for one year beyond that.
I have a bright idea. IBM knows what program conversion costs, after having i 6.1 in the field and hearing the complaints for nearly two years. IBM wants to get customers modern, and customers want to get modern and stay there but don’t have budget for new hardware, software licenses, and services to port their applications to the updated licensed internal code that lets them take full advantage of Power6 and, next year, Power7 iron. So, what IBM needs to do is this: Provide customers with a side-by-side upgrade to Power6 or Power7 iron and free or greatly discounted program conversion services as part of the upgrade. Customers get to see their code running side-by-side on the new and old boxes, and they can make a gradual shift.
These i shops didn’t ask for IBM to change the guts of OS/400 and its programming model as it did in i 6.1. The kinds of changes that IBM made went explicitly against the value proposition of the AS/400, something that IBM did in 1988 when it converted the System/38 to much consternation. IBM may have had good reasons for tweaking this licensed internal code and programming model–reasons it never did articulate very well–but the whole point of the AS/400 was that you never had to recompile or tweak your code to make it run on new iron and exploit new features. This is a covenant that IBM broke, not the customers. And therefore, rich ol’ Big Blue, which can shell out ba-gillions of dollars buying back its own shares to financially engineer quarter after quarter, can do the AS/400 base and its partners a square deal and cover the costs of a conversion process that I think is a barrier to sales.
I also think I will see a big blue pig fly outside my office window, perhaps doing barrel rolls while wearing a white silk scarf, before IBM listens to what I have to say. But I have done my job by bringing it up.