It looks like the “When will OpenInsight 10 (OI10) be available?” question is getting some more interest again and I have had a few people ask me about the release on my return to the UK. Yes we are all itching to take advantage of the new version, after all it does look like a huge ground breaking release and this was even more evident at the conference, but my personal opinion is that the wait will be more than worthwhile.
Why? Quite simply because I believe that, whilst OI10 development continues at a pace, Revelation are still listening to their client base and that they have the balance about right. I could jump straight into one or two of the distractions undertaken for some of my clients here in the UK, but before we look at those, let’s take a step back by a few days and to Mike Ruane’s opening presentation.
Mike opened the conference with his usual state of the company and future direction, it sets the tone for the whole event and settles everyone into the conference. This year, Mike directly acknowledged that the OI10 project is behind schedule, we all know that, we might not like it, but it is a fact and in my mind for some good reasons. One of his slides included the following bullet points:
- Limited Resources
Revelation is a small company of like-minded people and developers with a passion for delivering tools that deliver on their client’s needs at a reasonable cost. I guess that Mike could sell out to one of the big, but we would quickly lose our competitive edge as everything becomes more ‘standard’, we would lose our ability as customers to talk to the top decision makers or to have Revelation work as our partner to quickly address a business need that helps us to land that big customer. I guess that we could double the development staff and double the license fees, but I think that Revelation have things about right, so I’d not want anything to change on that front.
I could think of many more reasons why I would not want to change this side of things, but you get the picture.
- Scope Creep
Developers like to embrace and work on the next ‘fun thing’ but I don’t believe that this is the case at Revelation. Mike runs a tight ship and I know that Bob Catalano also keeps a wary eye on the project’s progress. Where the project has allowed some scope creep, it has been for good reason and often to address Revelation’s customer needs. I’ll touch on that below using some my situations.
- Changes of Direction
As everyone will know, any lengthy development project will change direction during its life. You can scope and document every last detail, but the IT industry continues to change at an alarming pace. Revelation’s needs change, their customers’ needs change, our end user’s needs change. In addition, the technologies that Revelation and their customers work with have changed and the team have had to wait for some of those changes to come on-line for them to be exploited through the OI10 project.
Throughout this massive project which has seen every part of the toolset and millions of lines of code touched in some way, Revelation have understandably had to constantly review, consider and change direction at times.
- Dead Ends
The above changes in direction have also had to occur when some dead ends have been reached. This becomes hugely frustrating for everyone concerned, especially when it becomes evident that a well-conceived plan has to be dropped or re-thought because a brick wall is reached and a way round needs to be found. As we all know, this is a normal hazard that we all have the deal with in software development but the stronger team rise above this and power on through another direction. I believe that what we saw at this conference, shows that Revelation are doing just that.
- Parts more difficult than we thought
Hey, we all have great ideas and then sit back and think why did I begin this. I have a great idea for delivering an OI training resource for OI10, but the more I look into what I need to do, just at the preparation stage, the more I wonder whether it would have been easier to have kept quiet and leave it for someone else. This is not the ethos at Revelation and I have learned that the team, and Revelation customers, are often ready to make the tough decisions and take on the difficult paths to create better products, better applications and better solutions for our customers, even if it does mean that things take slightly longer to deliver.
Despite the delays, these decisions have and will ensure that Revelation deliver more features that were originally planned and most of those are customer driven needs. Most of the team had to wait for the bulk of OI10 to be built by Mike Ruane and Carl Pates so that they could use the tools to build their parts of the new product. In taking this approach, the whole Revelation development team are now actively developing in OI10 and thereby testing the toolset at a real world development level, even before we are into an ‘official’ beta program. This can only lead to a more robust and successful toolset for us all. In addition, as OI10 comes together and the project progresses towards its conclusion, the team are running into fewer and fewer dead ends and the project is finding a much more clearly defined end point – certainly a lot clearer than 12 months ago.
When I talk to people about the OI10 project and the delays, people often cite the scope creep as the main point of their frustration. “Lock those requests out of the project and press on”, I hear people say, “Keep sight of what is important and don’t get distracted”. Good points, but in Revelation’s world we have clients that need solid solutions that support an ever changing world. I therefore think that it is worthwhile to pick up on some of my personal stories that have impacted the OI10 project and I’m sure that Richard Bright in New Zealand and Bob Catalano in the US will have stories from their own regions.
Many of my customers have recently spoken to me about performance. Here in Europe, it is all about performance these days with companies investing heavily in new hardware, bigger and better environments and some pretty exotic set-ups at that. We have seen virtualisation take off big time, the cloud become ever more important and tighter security needs hit the headlines. Revelation could have ignored these things and kept to the original plan, but I’m pleased to say that all of the above challenges have been embraced and they will make for a much ‘MUCH’ better OI10 in the long run. I especially enjoyed Bryan Shumsky’s presentation about ‘OI10 and O4W User Management Changes’ in which he showed the new user management module in OI10 and how things like Stamford standards can be easily and quickly implemented, another thing that some of my customers have been asking for more recently, especially those working with web based solutions.
Anyway, as I mentioned earlier, below are just a few examples of how Revelation have supported me and my clients here in Europe during the OI10 project and which will have inevitably pushed back the development time somewhat:
- The UD 5.x delivers the ability to use VSS for backups, support fail-over for network issues and also the ability to recover the UD from a system failure. The UD 5.x was originally only planned for release with OI10 but I have needed to deliver these benefits to a few of my customers who are running OI 9.4 and Revelation took the decision to take time out of OI10 development to roll the UD5.x back to support the currently shipping 9.4 release. To better support one of my clients, Bob Carten is taking more time out to further enhance the UD5 – something that I am grateful for, but also sorry that it will introduce another week or so of delayed OI10 development.
- During the early stages of the OI10 project, one of my clients had a huge O4W opportunity and I ended up taking a large amount of Bryan Shumsky’s time away from developing the new OI10 features into O4W. This began with helping me to deliver a proof of concept, then to support OpenID and to get the toolset through some industry leading penetration testing. The end result was more than just a much better, stronger and more secure product, it has resulted in securing two (and shortly three) of the biggest names on the internet as users of Revelation’s technologies.
These enhancements were not factored into the original design specs but Revelation took time out to support my client and to deliver more functionality into the O4W tools. This has directly enabled us all to benefit from that and the knowledge that we have some major blue chip internet company users relaying on our technology.
- Several of my clients have hounded me for more performance in recent times. This has resulted in several of those people making the move from 7.x and 8.x to 9.4 and the UD 4.7, but more than that, Revelation has taken time out of the OI10 project to help us to better profile clients systems and to identify causes of performance degradation and to deliver performance benefits. I’d also like to personally thank Andrew McAuley (Sprezzatura) for putting in many hours with regards to profiling some of the systems and to helping to move my clients forward and to work with Bob Carten on this complex topic. Bob Carten’s work now continues as part of the OI10 project, but it has already resulted in table profiling tools, system optimisation tools and huge code reviews to remove bottle necks and to better optimise OI10 code.
Some of these tools have been rolled back for use with OI 9.4 and a show of hands at the conference confirmed the importance of Bob’s tools and how the OI9.4 community is already using those tools to the benefit of everyone.
In addition, I know that every last inch of OI code if being reviewed by the team. For example, .NET naturally introduces some performance issues for us in OI and several of my customers have commented on the same. Like me, they often hark back to the good old days when OpenInsight was totally self-contained. It is not necessarily common knowledge, but I know that Carl is working hard to remove those dependencies as much as possible and this can only be a VERY good think in my book.
I am hugely grateful to my colleagues at Revelation for all the work that they are doing on OI10, but more for taking time out of this huge project to hear the needs of my clients and to deliver enhancements and functionality to help them to move forward today! As we all drifted away from the conference, it was heartening to hear the positive comments about the new features coming in OI10 and how that will benefit developers and end users. However, as I prepared for the long journey home, it was very disheartening to hear people refer back to those missed delivery projections. We all work in the business of delivering software and this comes with the territory.
Mike took it on the chin and posted the words When, when, when in his presentation.
The question that I think we should all be asking is How, how, how.
How can we use the alpha to better prepare ourselves, How can we put OI10 to good use when it is available and How can it be used to benefit our existing clients and help us to attract new business.
With over 3,000,000 lines of code touched, every part of the OI10 toolset being reviewed and enhanced for performance, functionality and more, there are inevitably going to be some unforeseen delays. Personally, I’d rather wait a few extra months and have the toolset that I can take to my clients with pride and to know that they can go on to continue to deliver some market leading solutions that are stable, robust and which more than meet the needs for their clients.
My take on this topic – I hope that Mike, Carl and the team continue to build the toolset we all need regardless of the time needed to deliver this properly and, at the same time, they continue to listen to the needs of our clients at both OI10 and OI9.4 levels. It will be worth the wait in the end and I’m looking forward to growing the community based on more intuitive, more functional and more easy to use tools for the modern application developer.