Following the trumpet calls last Monday as Compaq announced a $9 billion-plus takeover of Digital Equipment, a three-month vow of silence has descended around the companies as they seek shareholder approval.
Customers, however, must wait anxiously for at least that long to discover which of the many competing products, processors and operating systems will, in golfing terminology, 'make the cut'.
When Compaq completes the takeover of Digital, it will be the proud purveyor of three different microprocessors, three different brands and up to 10 different operating systems.
Quite a heady mix, requiring some difficult decisions on what to axe and what to integrate. What is clear is that there are not sufficient research and development dollars to go round.
Per Andersen, senior analyst at IDC, believes Compaq will have to back one Unix operating system long term, possibly retaining SCO Unix as well, as its package offering to the low end.
"It is clear that Compaq cannot sustain three different versions of Unix. Chances are it will put first priority to Digital Unix as its high end offering on Alpha and Intel," he said.
What is clear is that, whatever Compaq's strategy for Unix, it will always be the minor partner to NT.
"NT is Compaq's strategic product but the market for Unix today is still growing. Compaq has the ambition to become an enterprise player so it will not want to be premature in cutting off Unix. This would be a barrier to entering the enterprise," explained Andersen.
Unsurprisingly, Compaq's chief executive Eckhard Pfeiffer was upbeat about his plans for Digital, although Compaq refuses to discuss specifics until a 90-day quiet period is over, allowing Digital shareholders to vote on the offer.
"We are committed to ... investing in Digital's strategic assets, particularly its worldwide service organisation, as well as its 64-bit leadership with Alpha microprocessors, OpenVMS, Digital Unix and Windows NT enterprise systems, open storage and software products," he said.
Also complicating the issue is Sequent's agreement with Digital, signed at the beginning of the year, to jointly develop a 64-bit Unix that can run on both Intel's forthcoming 64-bit Merced processor and Digital's Alpha.
Based on Digital Unix, with the addition of Sequent technology, the companies claimed it would have full compatibility for source and binary code across both processors, meaning applications can be developed to run on either architecture.
Compaq will now have to decide whether the market really wants another Unix operating system, although one Compaq insider said there was no reason why not if it conformed to all industry standards.
Compaq has sold Unix systems for some time, currently shipping SCO Unix on its Intel servers. Hugh Jenkins, Compaq's UK enterprise server marketing manager, said SCO Unix currently runs on around 10 per cent of all server shipments, mostly into the application server space. This addresses a far lower end market than Digital Unix, so they should not conflict.
Although Jenkins was unable to comment on any future strategies regarding Digital he said Compaq had always supported a number of different operating systems. For example, he noted that Compaq is one of very few vendors to actively support and certify Banyan Vines on its servers.
"We have a long history of taking care of our installed base. What operating systems a company supports is something that the customer base defines for you. Whatever happens it will be very customer-led because that is Compaq's culture," said Jenkins.
No one is suggesting Compaq will completely sever any operating system. But agreeing to support and maintain a product is a very different commitment from planning significant investment in developing new versions and functionality. OpenVMS, for example, has a very large customer base, but its future is even gloomier after Compaq's takeover.
IDC's Andersen is unequivocal. "OpenVMS - it is a maintenance-only candidate in the very near future," he warned.
The culling has already begun. Last week Compaq confirmed it was to drop some of Tandem's NT server range. Also likely for a swift visit to the chopping block is Tandem Unix, which appeals to a market that is too small and proprietary for Compaq's liking.
Tandem's third OS offering, the high end Tandem Nonstop Kernel, has a very solid user base among banks and financial institutions requiring truly fault tolerant computing. However, the architecture does not fit in with Compaq's proclaimed quest to be 'open' and the company has begun hinting that the Mips processor, adopted by Tandem, could be replaced with something more mainstream. Possibly even Alpha.
On the processor front many issues remain to be resolved. Merced looms on the horizon and US regulators may yet block the sale of the Alpha business to Intel, particularly with Compaq taking over Digital in the mean time.
Analysts generally have been cynical about Alpha's prospects of lasting much beyond Merced's arrival.
"The Alpha line is dead. Not immediately but in between three and five years' time it will be downgraded and put on life support," predicted Andersen.
Alpha also creates issues for what the two companies claimed was one of the most logical fits of the deal - the respective Windows NT businesses. The version of NT that runs on Alpha is not the same as the one on Intel.
Applications require recompiling to run on the different versions, effectively meaning Compaq would be selling two NT lines - negating some benefits of the overlap in product ranges and creating another problem with keeping Alpha long term that Compaq will seek to address.
Commentators have focused on the benefits to Compaq of taking on Digital's higly rated services division and how it propels them into the top three computer companies. All of which is true, but as so many mega-mergers in IT have proven, the proposed benefits of the deal often take significant pain and costs to achieve.
As one analyst noted, Compaq may have to put so many products on life support that its portfolio will resemble a hospital ward. Customers can only hope the mercy killings will be painless.
A nuclear strike has been considered, but Bruce Willis is nowhere in sight
Spray-on antenna could enable seamless integration of antennas with everyday objects
Parker Solar Probe, TESS and GOLD missions will deliver exciting data, claims NASA
But deep learning pulls ahead for complex tasks