Since our post of last month, where we announced that we could no longer viably develop Dolphin into the future, things have been moving quickly regarding a possible future for Dolphin Smalltalk. The aim of this post is to let you know what's been happening behind the scenes and to answer some of the questions that have cropped up during this period.
Dolphin Shelter
The most important thing that has happened over the last month is that a group of dedicated Dolphin developers have come together under an umbrella that they call Dolphin Shelter to try and piece together a future for Dolphin Smalltalk. Over a period of about two weeks this core group came up with a number of potential routes forward and we (Object Arts) have responded with our thoughts on their suggestions. Please be aware that Object Arts is not running this group (although we are participating in their discussions) and that Dolphin Shelter is not some "secret society" -- anyone is welcome to join. If you're seriously interested in the future of Dolphin then these guys would like to have you on board and you can sign up to see what's happening by e-mailing mailto://davorin.rusevljan@gmail.com, with subject "Dolphin Shelter", and he will set up an account so you can see the groupware discussions.
The main outcome of the Dolphin Shelter proposals have so far been:
- A suggestion for Object Arts to sell source code licenses to the Dolphin virtual machine as an insurance policy against future changes in Windows breaking the environment and applications created with it. As an initial stance we suggested that we would be prepared to sell source code licenses at a cost of, say, $3000-$6000 providing a guaranteed minimum of $60,000 of licences could be sold
- A proposal to stage a Blender-style buyout of Dolphin with a view to Open Sourcing the code. This would also give the community the ability to create Open Source implementations of Dolphin for non-Windows platforms if it is willing to put in the work needed to do this. We (Object Arts) set a value of $230,000 at which we would be prepared to Open Source the Dolphin code base including the VM.
Buyout Survey
Currently, the favoured route seems to be the Open Source buyout and the Dolphin Shelter guys have produced an online survey to try and determine the prospective levels of participation from the community. You can find the survey here.
Valuation
Apparently, some of the early respondents to the survey have, quite rightly queried where the $230,000 buyout price has come from. I'd like to use the rest of this post to explain why we put forward this amount as being one at which we would be willing to proceed with the buyout process.
Before I do this, however, I'd like to touch on a couple of comments that have been made in various blogs/newsgroups since we made the announcement that Dolphin development would cease. It seems that I've riled some of the members of the Open Source community by my comment that: "Both Blair and I dislike the Open Source movement intensely and we would rather see Dolphin gradually disappear into the sands of time than instantly lose all commercial value in one fell swoop". I think it was my use of the word "intensely" that created rather a furore in some circles and begged the question, "if they are no longer selling Dolphin Smalltalk how can it have any further commercial value to lose?".
Hence, I'd like to run through the reasons why we believe the Dolphin Smalltalk IP still has value even though we are unable to commit the time to further develop it as a commercial programming tool. This will then, hopefully, go some way to explaining the value we have placed on the Open Source buyout.
How can we possibly think that the Dolphin IP has residual value and shouldn't be Open Sourced right away, I hear you ask?
- As you know, Dolphin is a complete and robust development system for Windows running on a Win32 base. If a similar product were to be built from scratch as a commercial endeavour today it would cost over $1.5 million to develop (this is conservatively taken from the number of person-years invested in the product to date). Certainly, it would appear not to be worth anywhere near this in the current market where development tools are devalued and Smalltalk itself is under-appreciated. However unlikely, this could change. If we guess that there is a 10% chance that we might pick up a comparable amount in a commercial buyout deal over the next five years then we get an "expected value" of $1.5M x 0.1 = $150,000. Whilst this is not a huge value it's enough not to be ignored. Note, this "expected value" is not the amount we would accept in a commercial buyout; we've actually set this at around $450,000 (readers may now ROFL if they wish)
- So, perhaps a commercial buyout seems unlikely. But consider the Dolphin Shelter proposal to sell VM source code licenses. Whilst we have had around 1500 unique purchasers of Dolphin products over the 10 years we have been in business, it is our estimate that they are currently around 500 "active" users. These are people who have upgraded in the last two release cycles. My estimation would be that we probably have somewhere between 50 and 100 developers who have commercial products that they need to keep running. We know from the history of Visual Smalltalk that is quite possible to keep Smalltalk apps running years after vendor support for the language has dried up even without access to the VM. This is mainly because the Smalltalk image is inherently open source (note no caps) because all class sources are supplied by default. It would seem reasonable (to us anyway) that a commercial developer might readily purchase a source license for the VM for, say, $3000 in order to insure against future problems. Indeed, if I were in the same situation I would probably be prepared to pay $5000-$6000 for such insurance. So if we assume that 75 pro developers buy source licenses at $3000 over the next few years this gives a potential income of $225,000.
- Some of you may be aware that, for some time, we have been developing a range of automated stock trading systems under the Alchemetrics banner. These systems have been built entirely in Dolphin Smalltalk along with a number of relatively sophisticated analysis tools. The systems are currently managing, in a completely automatic fashion, a fund (a "hedge fund" if you like) of around $1 million, made up from a variety of sources. Results from the last three years have been highly promising (25%-45% APR), so it may well be possible for us to interest a financial institution in a packaging of Dolphin along with the Alchemetrics trading tools in future. If this were to be the case, I'd expect that the bundled value would be far in excess of that of Dolphin as a Smalltalk development system alone.
So, whilst we are certainly not interested in Open Sourcing Dolphin for free, we would be willing to do so providing that it doesn't jeopardise to any great extent the commercial value of the Alchemetrics work or the revenue we might gain from an alternative approach such as the VM source licensing.
$230,000?
But there is a further issue with any proposed buyout which also need to be factored in. Dolphin was originally created by a team of 5 developers at UK company Intuitive Systems Ltd. In total, over two years, they probably spent around $700K on development costs. When Dolphin was canned because of the rise of Java, I left and started Object Arts. Part of the deal we had with Intuitive for taking over the code was that they would require 15% of any future disposal value of the IP. Since then, Intuitive Systems have been bought out by a succession of companies but (although I haven't checked) I would imagine that the same deal still applies with the present owners. Although nothing is written into the contract about a possible sale price I would think that they would see it as rather suspicious, and might try to intervene, if we tried to dispose of the IP for an unrealistically low price (whatever that might be).
Since our guestimate of returns from the source licencing is around $200K, that's where our valuation for the OS buyout came from. We've chosen to add 15% to cover the dispoal fees and hence this has brought us up to $230,000.
What Happens Now?
If you are involved with Dolphin, you probably have a will to see it continue into the future. It seems to us that the OS buyout could succeed although it will likely be difficult to reach the target. Please take the Dolphin Shelter survey so these guys can judge the potential level of interest in their proposal.