Thursday, April 23, 2009

The horse isn't dead, and neither is the Viper…

It was quite a weekend. I headed back up to San Francisco on my way back to Boulder, Colorado, to meet with GoldenGate partners and being there I took some time out to travel Highway 35 “Skyline Boulevard”. This is a highlight for many motorcycle enthusiasts who take pleasure in diving deep into the corners before whipping their bike through impossible-looking turns. And the picture above is of a famous cafĂ© midway along the boulevard where refreshments can be had at Alice’s Restaurant.

And the comparison didn’t escape me – last weekend in the Southern California canyons behind Malibu with a visit to the Rock Store and this weekend along the Northern California ridgeline behind Woodside with a visit to Alice’s Restaurant. And both provided amazing scenery and an opportunity to take advantage of the open road. Of course, while stopping by Alice’s I had to get another T-shirt, as it seemed the right thing to do.

In between the meeting with GoldenGate partners I had the opportunity to flip through newspapers business journals and the many car magazines which I always seem to have nearby. Among the collection was a May 2008 issue of Motor Trend and there was a column titled “The temple of doom.” Author Arthur St. Antoine opened with the observation “the world is running out of oil. Global warming threatens our very existence … in fact, cars as we know them are doomed … life is awful and getting worse.”

And the thoughts expressed by St. Antoine mirrored thoughts I had been having this week as I have followed the much publicized antics of Sun as first IBM, and then Oracle, made offers for the company. IBM initially appeared to have won the deal, but after a very short period of due-diligence, revised the offer price downwards and Sun walked away. Later, when Sun attempted to restart the discussions, IBM expressed little enthusiasm, bluntly telling Sun executives that the deal was now dead. No second thoughts on the part of IBM in this instance – for some reason, they wanted out, and fast!

Imagine the heated exchanges that must have followed between CEO Schwartz and Chairman McNealy. Shareholder lawsuits were being prepared. Key stakeholders were furious as an easy out was taken off the table. But then Scott turns to Larry and probably opens the exchange with something like “want to really get Gates upset – how about you owning Java? MySQL?, and a whole lot of smart people that we have done a poor job managing over the years?” What a deal, Larry!

“And Larry,” Scott probably added at some point. “Remember Grid Computing and how you liked it but how it’s now yesterday’s news? Well, I have some pretty exciting cloud solutions – that’s sure to really get up Bill’s nose and upset his plans for .NET global dominance!” Against this background it is no wonder Larry Ellison immediately called Java “the single most important software asset we have ever acquired.”

Now that Larry has his hands on Sun, he must be waking up each morning not believing his good fortune. But for the industry, will life become awful and get worse? Will, for instance, Larry simply “propriet-ize” java? Already owning BEA, with its wonderful collection of web and application server products, and Java development toolkits, will there be one development program for Oracle, and another for all the partners Sun has attracted over the years?

In a recent Motley Fool column, writer Anders Bylund reported on how he called Tibco’s CEO, Vivek Ranadive, “for a candid rundown of what this merger of not-quite-equals means for his business -- and for Silicon Valley at large.” When asked about the many Sun partners committed to Java. Ranadive was reported as having said "The Java community is going to be concerned.” Ranadive then noted how Ellison "likes to buy commodity has-been companies" and dismantle them by gutting R&D efforts on the newly acquired products. "I'm sure it'll be highly accretive at the start, but it's the end of innovation."

Later in the same article, Byland noted that with the acquisition, “Oracle is also buying into a slew of open-source projects under Sun's wing, including database platform MySQL, the OpenSolaris operating system, and MS Office competitor” This apparently drew a chuckled response from Ranadive "Oracle and open source? That's an oxymoron if I ever heard one!" Oracle certainly doesn’t have history on its side when it comes to supporting initiatives across the open source movement.

IBM may now be having second thoughts, but for many companies that have elected to keep the stack provider separate from the platform vendor, this is beginning to appear as though the infamous Borg of Star Trek were more firmly entrenched in the valley than up Seattle way. Returning to the Motor Trend column referenced earlier, the writer added “why do I bring all this up here? Because the same Armageddon mentality now runs rampant in the auto world. People talk about alternative fuels and smaller cars and the ‘end of the golden ages’ as if it’s all downhill from here.” Again, “resistance is futile, you will be assimilated” comes to mind.

Looking at the key verticals Sun competes in, the telco and financial services markets stand out and here they will come up against stiff competition from HP and IBM as they continue to eat into Sun’s market share. These markets are also important to NonStop’s future as well, as they are two extremely significant and profitable marketplaces. But can NonStop shore up its relationship with partners while attracting new entrants, something extremely critical for any future play in these markets?

It is against this background and with these concerns in mind that I have been impressed with a number of regional activities that are under way and that are aimed at finding new partners and at driving more solutions to NonStop. With solid backing from developers in Java, as well as C/C++, and a commitment to middleware offerings that mask the “NonStop within” from the average developer, there is some evidence of traction developing in the partner community.

In the blog posting of February 19, ’09, “Game changers!”, I wrote of the work being done by Californian company Modius and I reported on the port of “their application onto NonStop. Modius provides a Data Center Infrastructure Manager (DCiM) product that is a ‘complete measurement and monitoring solution for generating a single, real-time view of all business critical facilities across the enterprise, including all major data centers, server rooms, branch offices, wiring closets, etc.’ … NonStop in every data center, wow and we should be leaving the management of the data center to NonStop.

In the blog posting of March 19, ‘0, “The Great Divide!”, I took a look at a company in the UK, Erudine, that “have brought to market a ‘Behavior Engine’ with the ability to duplicate business logic – not by cloning code, but by cloning the logic and (optionally) modifying it’s behavior.” And the thought I had here was with this capability, there would be no limit to what application could be built on NonStop adding “after all, when it comes to us ‘learning new things on your HP NonStop’ perhaps we should be letting HP NonStop learn some new things as well!”

Finally, in Asia Pacific – Japan there’s been work to help Opus with the port of their Electra financial switch software. A relatively small company out of India they see the potential to exploit a measure of uncertainty surrounding the availability of future releases of BASE24 on NonStop and although it’s far too early to argue that ACI faces a competitor, I am encouraged to see solutions providers coming to NonStop no matter the circumstances. Opus, with the help of NonStop expertise in the region is doing a very deep port and have implemented a mix of C and C++ code on top of NonStop Tuxedo (as the TP Monitor) and NS SQL/MX (as the data base). The company has been particularly impressed with the scalability NonStop affords, and I will be watching the progress they make in the financial services marketplace as I will be watching Modius and Erudine.

Talking about new partners with new solutions and infrastructure has been something only lightly touched upon in previous blog postings, and when viewed alongside of what we see happening at Oracle, there may be a lot more solutions providers taking a renewed interest in the product offerings of HP. And NonStop should score points with many of them.

The Motor Trend column concluded with “most of us long ago traded horses for cars, but you can still ride a horse. I’d bet that in 2050 you’ll still be able to drive a vintage Viper – and your tire-smoking fuel-cell Vette. How can I be so optimistic about the future? Hey, I have history on my side.” And across the IT landscape we know all too well that change is constant, and watching Sun be swallowed up within the greater Oracle is only the latest example of change and there are no signs I have lost any of my optimism.

Partners are everything in today’s vendor ecosystems – and should you mess with them, they are hard to re-recruit. There will be many looking at their options following Sun’s assimilation into Oracle and I am very positive about the impact this could have on NonStop. After all, with NonStop there’s history, and over the years, it’s been pretty good! Like the horse, and the Viper, NonStop is here to stay!


Anonymous said...

Hi Richard,
on a closer look some things are looking different. Are you aware of the fact that Tuxedo will not be supported on NonStop anymore in the near future? And do you really believe the marketing ... (censored) from Erudine? There is certainly some substance in the product (basically it is a rules engine). But being so long in our business - do you still believe in magic?
Kind regards

Richard Buckle said...

My own observations on Tuxedo and Erudine?

I am aware of the roadmap for Tuxedo, as I am for Pathway, and of CICS. Commentary on the future of TP Monitors versus their more modern counterparts, the Application Server (often built on top of, or re-using, components of the TP Monitor), continues to be waged with many commentators on the side of TP Monitors being on the far right of the product lifecycle.

Not committing to further ports of Tuxedo is simply because there's no real business case - current deployments have been to few to support additional NonStop investment. Is this right, or fair - I am not sure. Does this mean it's foolish to deploy on an infrastructure without a vendors commitment to future ports?

In looking at this and discussing in other online forums - check out the Connect web site and within Forums, the SIG "NonStop SOA" for a thread called:

OLTP - from TP Monitors to Application Servers....

Where there were a number of exchanges on this thought. But bottom line, hardly any investments are being made in TP Monitors and they are stable. They aren't a bad option for a software house to use when facing a port if they are looking to test the market. Should they be overwhelmingly successful then I am sure NonStop Product Management would reconsider future ports of Tuxedo but the business case would still have to make sense.

As for Erudine, then having worked closely with a Business Process Management solution when at ACI called WorkPoint (Now an independent entity – check out: I am familiar with the technology.

I am also familiar with their pro’s can con’s – but having been walked-through one case study and knowing HP has been testing portions of it, then as I have often been reminded then could it turn out to be less than promised? Perhaps. But is it more than I’ve seen others deliver? Yes … Again, it’s a technology and a company I will be watching closely over the next couple of quarters ….

Richard Buckle said...


Further follow-up with HP Product Management.As for your:

"Are you aware of the fact that Tuxedo will not be supported on NonStop anymore in the near future?"

The word coming from Product Management points out that this isn't the case at all, and they explained to me how they "have never made any such statement to this effect. There is a big difference between not porting new versions (about which your statements are essentially correct), and not supporting the versions we currently have. HP is self-sufficient wrt sustaining of NSTuxedo, and we have no plans to drop support for it."

Furthermore, as for partners basing deployment of applications on Tuxedo HP Product Management advises that "there is no reason why current versions of NSTuxedo should not be considered as a deployment environment for applications on NonStop should it provide the best fit in terms of architecture and functionality required by the application."

Again, I would like to thank HP Product Management for helping us out here as it's always greatly appreciated.