Eleven months since Vista shipped to volume licensing. The cumulative deployment trajectory is now clearer; specific operational patterns are visible. A retrospective note.
This is a longer post because the deployment trajectory is structurally important.
What the cumulative deployment looks like
The cumulative observation across multiple operators (with appropriate confidentiality):
Enterprise deployment is slower than the consumer trajectory. Specific enterprise estates have been substantially slower to deploy Vista than consumer hardware. The cumulative enterprise deployment percentage by mid-2007 is bounded — most large operators are still in the testing phase rather than the deployment phase.
The waiting-for-SP1 pattern is operationally rational. Specific organisations explicitly waiting for SP1 (expected late 2007 or early 2008) have been validated by the specific issues that have surfaced through the deployment period. The cumulative pattern matches historical Windows-release patterns.
Specific application-compatibility issues persist. Specific older applications continue to require XP-mode workarounds; specific custom applications need updating; specific vendor-supplied applications have varying Vista compatibility. The cumulative compatibility friction is real.
Specific user-experience friction is real but bounded. UAC prompts produce specific user complaints; specific older hardware does not run Vista well; specific user-interface changes require training. The cumulative friction is manageable for organisations that planned the deployment carefully; problematic for organisations that did not.
What has worked structurally
Three positive observations.
The architectural improvements deliver as advertised. Specific exploitation that worked against XP does not work against Vista; specific cumulative defensive trajectory is real. The cumulative effect on the broader threat landscape will be visible across years.
The patching infrastructure is more reliable. Specific operators report measurably better patch-deployment reliability under Vista than under earlier Windows versions. The cumulative operational discipline is supported.
Group Policy infrastructure for Vista is mature. Specific configuration disciplines transfer from earlier Windows; specific new Vista-specific configurations are supported. The cumulative enterprise-management capability is operationally adequate.
What has been more difficult
Three operational challenges.
Application compatibility issues exceeded most organisations' planning estimates. Specific portfolios that were expected to be Vista-compatible have surfaced specific issues; specific vendor responses have been slower than hoped. The cumulative compatibility-testing workload has been substantial.
Hardware-refresh cycles have been longer than expected. Specific older machines that were expected to be retained have been replaced; specific budgets have been larger than planned. The cumulative hardware-refresh cost has been non-trivial.
User-experience friction has produced more support load than planned. UAC, the new interface, IE 7 in protected mode — all have produced user-support requests at higher volumes than typical service-pack-level changes.
The challenges are bounded but real. Specific organisations that planned conservatively are absorbing them; specific organisations that planned aggressively are surfacing more friction.
What this teaches operationally
Three lessons from the cumulative deployment experience.
Specific application-compatibility testing should be more thorough than typical service-pack testing. Vista is closer to a major-version change than to a typical service pack; specific testing infrastructure should reflect that. Future major Windows releases will benefit from this lesson.
Specific user-communication discipline matters more than typical. The user-experience changes are substantial; specific communication ahead of deployment reduces support load substantially.
Specific staged-rollout discipline is operationally rational. Less critical hosts first; more critical hosts later. The cumulative experience informs subsequent stages.
What is in SP1
Microsoft has been describing SP1 contents through multiple beta releases. Specific known properties:
- Specific reliability fixes — the cumulative bug-fix volume is substantial.
- Specific performance improvements — the cumulative effect on user-perceived performance should be meaningful.
- Specific platform updates that improve compatibility with existing applications.
- Specific cumulative architectural refinement — no major new features, substantial polish.
The release timing for SP1 is targeted for early 2008. Specific organisations explicitly waiting for SP1 will be in deployment-readiness state by mid-2008.
What I am paying attention to
Three things over the next 12 months.
SP1 release timing and quality. 80% probability of acceptable release. The cumulative trajectory is consistent; specific delays are possible but bounded.
Specific deployment acceleration following SP1. 85% probability. The cumulative pattern of deferred-deployment organisations moving forward post-SP1 is established.
Specific Vista-targeting malware emergence. 70% probability of meaningful incidents. As the deployment grows, specific targeting will follow.
What I am doing
For Gala Coral: deployment planning continues. Specific application-compatibility testing has identified specific issues; specific subsequent deployment is targeted for after SP1 release. The cumulative discipline is to wait for the platform to mature before broad deployment.
For my own infrastructure: Vista on the test machine continues; production hosts continue running XP SP2 with current patches.
For client work where I have advisory roles: standard advice — wait for SP1, plan deployment carefully, communicate with users. The general patterns are consistent across organisations.
A small reflection on the trajectory
Vista has been the most substantial Windows release in years. The cumulative architectural improvements are real; the cumulative deployment friction is real; the cumulative effect on the broader threat landscape will be visible across years.
For my own continued writing: continued tracking of the trajectory. Specific subsequent posts will inform; the cumulative archive grows.
More in time.