I notice that the University of Oxford’s Software Engineering post-graduate programme includes a specialist course on Software Product Lines that would seem to offer a good general introduction to the topic. Places are still available for the December 2006 start at the time of publishing.

Please let us know of any other similar courses.

Yesterday, I promised to follow-up with more information from the Cambridge Wireless Games SIG event that took place on 28th June.

Jon Hare’s talk had concluded with the slide Mobile Games Development – Matt knows how to do it so it was over to Matt Spall of morpheme to live up to this challenge, although Matt’s talk had actually been billed as “Can creating new brands work on mobile? – one man’s view”.

Matt began by noting that brands can move around platforms (mobile, PC, xBox etc.) pretty easily – so it’s generally easier to make more money by developing a game around a brand.

Examining the top-selling games in the charts also shows that brand-based games dominate, with only a handful of non-brand games entering the top 10 chart at all. An original game, as Jon Hare had also noted, is likely to be a financial loser. However, Matt noted that this situation could ease as direct access channels between user and developer improved.

Matt gave the example of Balloon Headed Boy, a game morpheme have developed and promoted for the best part of several years. Balloon Headed boy is promoted through a variety of channels including his own MySpace page (he has lots of friends), a music single, and various promotional items such as mugs, t-shirts and notebooks. Using alternative channels to promote the game has enabled Balloon Headed Boy to compete successfully in this brand-based market.

In terms of successful genres for original games Matt noted success for pub-style and sport-themed games but warned against RPGs. Only one original RPG game had made it into the top 10 in a year.

Other points that emerged during and after Matt’s session:

• No game has failed for being too easy
• Focus on social aspects of games
• Get your game preinstalled. Only 5% of users are interested in downloading games.
• Nokia’s Content Discoverer should make it significantly easier for people to find games – if the user has a phone with it installed – so should make things better for games developers

The next Games SIG event will probably be in December 2006.

 

The Cambridge Wireless Games SIG met again yesterday. I’d participated in last year’s SIG, enjoyed it, and decided to go along and see what the SIG had been up to since last year.

Jon Hare of Tower Games was first up and gave an entertaining and informative talk on Mobile Games Development & how not to do it. Jon is a veteran of the UK games industry with many #1 games to his credit and it was useful to hear his inside perspective on the state of the mobile games industry in the UK and in general.

UK mobile developers will be familiar with some of the problems Jon noted – difficulty in developing and testing for the huge range of handsets that need to be supported, competition from low labour-cost economies, pitching arrangements that mean that development costs have to be incurred well before it’s know whether a game will actually be bought and a focus on brand-based games rather than original games. There seemed to be general agreement from the other mobile developers in the room on these issues.

So, how to address these issues?

The handset diversity issue is hoped to become less acute over the next 3-5 years with major players trying to get standard platforms in place across a range of devices. Given that the cost of variant development accounts for around 70% of the total costs of a game development (according to Jon) any improvement in this area could be significant.

However, I’m not clear how stable any move to less diversity will be. I’ve argued elsewhere that device manufacturers will still want to distinguish their devices from others in the market and in order to do so the software using those devices will have to take advantage of those differences. Furthermore, even if devices become less diverse, how long will this situation last before the market fragments again?

Offshoring and outsourcing programming work were floated as solutions to the labour cost issue. The UK has a long experience in the games industry and consequently has a wealth of senior people with the skills and knowledge to manage game development. Perhaps the solution is to hook up with cheaper overseas development organisations – and at least one attendee’s organisation was using this approach successfully.

Several representatives from government funding bodies also noted schemes to help game developers, including subsidies for conference attendance, trade missions, partnering, training and joint marketing funding. However, Jon gave an example of a friend overseas who is able to get 75% of his game development costs back from his government. It seems that the UK still has some way to go to match these incentives aside from any argument over whether it should try to do so.

Improvements in the access channels between mobile phone users and game developers are expected to benefit developers too. These could be increased use of the internet to download games to phones and better search capabilities on phones to enable users to find games. Jon gave an example from a few years ago where he and his colleagues took his company’s accountant through the process of using a phone to download their game. Nineteen separate processes were involved! As well as improving download rates, direct consumer access, through simple to use channels, should also help improve the margins for game developers and so ease the funding situation.

One of Jon’s skills is bringing original games to market. However, the mobile market is dominated by brand-based games – think Who Wants to be a Millionaire or Colin McRae Rally. Various factors in the mobile industry conspire to make it difficult to bring original games to the mobile platform – not least of which is the 16 character space on the user’s phone that you have to promote your game. When you only have 16 characters to play with it’s important that the user recognises the name and this is where the brand helps massively. Again, improvements in direct access between users and game developers is hoped to improve this situation.

To enable investment to be returned many times, one suggestion was that developers set up a configurable games platform/production pipeline that could be used to produce large numbers of game variants. This would support game conversion (porting) to multiple handsets, for multiple operators, for multiple locales etc. However, as well as managing the usual Software Product Line issues, in doing this you also need to take care over who owns the IP rights to your games – make sure your first customer doesn’t tie up all your rights and then fail to commission a sufficient number of variants to allow you to make money.

Jon concluded his talk with an slide introducing Matt Spall of morpheme – Mobile Games Development – Matt knows how to do it. (More on Matt’s talk next)

Google have come up with another improvement to their Google AdWords programme. AdWords scheduling allows advertisers to set the days and times (down to 15 minute periods) at which their ad campaigns will or won’t run. An advanced mode also allows percentage multipliers to be applied to campaign bids during specific time periods. 

We think it makes sense to run some of our ad campaigns only during specific time periods so we’re going to take advantage of this feature. Let us know if you think we’re right.

With summer here and our next talk not scheduled until September 13th we thought that it might be nice to have an informal get together to keep up the momentum of the SPA Cambridge group, to review progress so far, to discuss future plans, and to chat about what each other is doing and the general state of the software industry.

So, we’re planning to meet at the Fort St. George pub (http://www.fortstgeorge.com/) on Midsummer Common on Tuesday 18th July from 19:00 onwards. If the weather’s nice we’ll sit outside, otherwise we’ll be indoors. Do come along and say hello, tell us what you think and meet the other SPA Cambridge attendees.

The pub can be walked or cycled to but you will have to park elsewhere if you come by car. See map on the website above.

Hope to see some of you there.

I saw this aphorism on a manager’s whiteboard once.

I guess it was an attempt to guide himself or his engineers away from gold-plating the system they were working on. But I didn’t ask for and he didn’t offer an explanation.

However, if you were determined to develop the best systems possible, then I guess it could have been pretty demotivating to see that.

We had a stand at this year’s British Computer Society Configuration Management Group’s annual Tools Fair on June 15th. The Fair was organised by Robert Cowham, a genial chap who knows a lot about Configuration Management. Robert was remarkably calm when we turned up at 7:15 and remained unflustered throughout the day as far as I could see.

The Fair had a small exhibition and talks on two themes:

• Service Management – Topics covering the ITIL framework from the CMDB to the Release management process.
• Application Development Lifecycle – Topics covering vendor products and services that enable users to actively manage application code throughout its lifecycle.

About 100 people turned up on the day and my impression was that the bulk of them were interested in the Application Development Life Cycle theme.

There were a couple of hours of breaks throughout the day and this meant there was a steady stream of visitors to our stand. The rest of the day was spent talking with the other exhibitors, introducing ourselves to those who hadn’t met us, and showing them pure::variants if they hadn’t already seen it.

One interesting point that arose from these discussions was a general agreement that there are very few good quality software tool exhibitions in the UK at present – the Tools Fair being one of them. This has been our impression for some time and probably reflects an increasing expectation (certainly in our industry) that all essential product information, including demonstrations and downloads should be available from the web. Do let us know if there is anything else you’d like to see on our website.

Some of you may know that I co-organise SPA Cambridge where we have a regular software talk led by an industry or academic expert. Our latest session, Modelling for Maintainability, was presented by Andrew Watson of OMG on 14th June. Andrew’s slides are available here.

After a short sermon on standards and a review of the CHAOS reports, Andrew provided some background on software maintenance costs as a proportion of total costs of software projects quoting from various studies:

• > 90% Moad 1990
• 60-70% Huff 1990
• 75% Eastwood 1993
• > 90%, Erlikch 2000

Maintenance is hard because of lack of documentation, changing platforms, changing program structure and the sheer difficulty developers have of understanding each others code and old code. (> 50% of maintenance time is spent understanding code). Furthermore, the volume of code under maintenance is doubling every seven years – Andrew’s claim, based on work by Paul A. Strassmann, is that we’ve now reached the point where we can no longer afford to throw away our existing software assets when we change our underlying programming technologies.

OMG’s solution (and that of others) is to raise the level of abstraction from code to higher-level models and Andrew provided the results of various studies to help support the claim that doing this would lead to significant benefits.

Models (of programs), it is claimed, will be easier to maintain than programs themselves. (There was some good discussion on whether this would be the case and the risk of lock-in to particular modelling paradigms.)

It’s also worth mentioning one anti-pattern that is occurring now and is likely to occur every time the level of abstraction is raised in this way. The antipattern is the rearguard action fought by those who claim that we will never be able to provide a sufficiently rich modelling tool and sufficiently rich optimisation tools to map models to lower levels.

This argument was used by assembly language programmers to resist the introduction of high-level languages and while say 5% of assembly language programmers could write code that is better than could be generated by a compiler in say 5% of cases the fact is that for most programmers, most of the time the compiler will do better.

Now I’m not saying that the current crop of Code Generators are at this state of development, although some are sufficiently good that they should be considered, at least for work in well-defined domains.

Thanks to Andrew for an interesting and informative session. SPA Cambridge’s next meeting is on September 13th 2006.

The people at Proven Software Solutions have come up with an interesting concept – setting up a brokerage for proven embedded software. Until they came along I had only been aware of similar approaches for desktop and server components.

What they mean by proven is covered in their FAQ and each product they broker comes with its own datasheet that specifies this in more detail. Given that the company is a start-up they seem to have done quite well to offer as many products as they do. Some people have also started asking for specific products to be developed.

I guess one question I have is just how much reusable embedded software is out there that isn’t already being sold by other organisations. Also, will people who are currently developing non-reusable embedded software be willing to make the extra effort to make it reusable now that there is an emerging marketplace for reusable products?

Only time will tell and we wish them luck.

Steven Kelly of Metacase has waded into the on-going debate between domain-specific modeling fans and fans of the OMG’s MDA approach.

Steven has written a new article for the Code Generation Network arguing that Domain-Specific Modeling languages are superior tools for Code Generation when compared to UML. Steven also claims that direct model-to-code tools offer much better solutions than model-to-model (to code) solutions such as MDA.