1. DreamSpring is one of the few companies that are still in business since the Psion days.
2. They have developed really outstanding products for ER5, UIQ2, UIQ 3 and now they also produced applications for S60.
3. I interviewed Malcolm over email to ask him some questions that I kept in my mind since a few years ago.
4. I hope you guys will enjoy the interview, as much as I do 
Malcolm Lithgow
Introduction
1. how do you do malcolm? what’s the weather like in australia at the moment?
Well, it’s hot and humid at the moment — getting close to Christmas time.
2. what’s your favourite past time killer? Other than sitting in front of your computer of course
Photography, including editing the photos and creating photo books is probably my favourite, though I also spend a lot of time reading and watching TV (particularly great shows like Lie to me, Lost, Doctor Who, etc.)
3. you use the word “Chief Dreamer”. It must be great to be able to live your dream! No?
The hard part is making the dream reality — or having dreams that have a chance of being realised. It’s good to be able to dream, though.
DreamSpring
1. you are active in the industry for many years now. since the psion days if i reckon correctly. how does it feel being able to survive the changes and challenges especially with regards to application development? many software houses from the psion days did not survive till today. secrets to your success?
Well, we’ve survived simply out of sheer stubbornness, I think. The Psion machines had a good community around them, although it seemed to have happened mostly because of the machines, not because of anything else Psion did.
The switch to Symbian both ushered in a deliberate focus on third party developers, but also an era of constant change and instability for them. We’ve spent most of our time porting from platform to platform, which has limited our ability to grow our portfolio or product feature set the way we would have liked. So far none of Symbian’s decisions have made our job much easier (hopefully Qt will be the first).
2. you only have 2 excellent products that are available to customers, dreamconnect and dreamlife. how do you survive all these years with only these 2 products? if it’s possible, would you mind sharing some numbers?
See the answer above.
3. you offer symbian products only. is there any plan in the future to dwell into other territories like Android, iPhone, Maemo, Windows Mobile, webOS etc? Surely, the market looks attractive enough to bring your products accross other platforms.
We keep our eye on all these platforms. I’ll share my thinking on each of this here:
iPhone:
Pros: easy to develop for; reasonably large market; good buzz
Cons: crowded marketplace; uncertain ability to reach market (depends on Apple’s whims, especially since we replace standard apps); crucial functionality missing from SDK (calendar alarms)
Result: not worth the risk for our current apps
Android:
Pros: growing market (for now); good buzz; various channels to market
Cons: poor development environment (multiple, different targets and UIs); small market; uncertain app delivery framework; uncertain future
Result: not worth the bother (I think Android is hopelessly over-hyped, like much originating from Silicon Valley)
Maemo:
Pros: growing market; easy development (Qt base plus full source); strong app shop (Ovi)
Cons: tiny market; relatively immature OS
Result: too small to bother with
Windows Mobile:
Pros: easy to develop for; various channels to market (and getting better); mature OS
Cons: uncertain future UI (like Symbian); crowded market
Result: best proposition, however still much smaller than Symbian
4. how about the challenges in developing the applications. surely, porting from one platform to another involves planning, structuring, documentation etc. how do you manage that?
Since we’ve done so many ports now, we have our engine pretty well separated from the UI. In fact, we have three types of code: engine (which is identical on all Symbian platforms), common UI (shared between UIQ and S60), and S60 (shared between 3rd Ed and 5th Ed). That’s a lot of flexibility and complexity. If we were writing from scratch we certainly wouldn’t do that.
However, the UI is always such an unknown. Developing UI’s is a much more iterative process than developing engines (the iteration is much finer grained). This makes it harder to plan for. Our S60 port, for example, surprised us by how hard it was to implement our vision using the S60 framework. It involved lots of attempts that had to be discarded eventually. The S60 framework is far cruder than UIQ, particularly UIQ 3. There were lots of controls we had to implement from scratch, and whole frameworks had to be improvised.
We involved testers in this a lot earlier in the process than previously, to ensure we weren’t simply pandering to our own tastes, and to handle the wide diversity of device types supported by S60. Of course, this didn’t make things go faster — quite the reverse.
Part of the difficulty is that, rather than starting with the SDK and following the “line of least resistance” in coding something (which many of the Symbian vendors clearly do) our approach is to start with the customer and then try to build something that meets their needs. This makes our job a lot harder. As an example, the split-screen edit view in DreamLife was an innovation that the UIQ 3 framework supported quite well, once you had figured out the complexities of that same framework. However, S60′s framework is very simplistic, and there’s just no way to support a split screen like that without basically rewriting it, so after various attempts we came up with the popup day grid in the edit view as a solution that worked well for the end user. However all this takes a lot of time, effort and learning.
5. How many people are there in your team?
We’ve had a peak of five or six people in the past, but we’re down to three now.
DreamConnect
1. are you satisfied with the result? i mean, looking at dreamconnect, in my opinion, the UIQ 3 version is the best. with better list view details, thumbnail support and stability. how can you make the s60 version better? what are the things you couldnt do in the s60 version that you wished you could?
Well, the lack of “mini detail view” in the list is one of the missing features of the S60 DC. We actually have a solution half-coded to that, so that would be nice to add. We’d like to add the missing things like Smart Find and Dynamic Messaging (I have ideas for a better UI for the latter), but these are real power-user features.
I’d like to improve the way the contact editor works, especially for things like long notes and for 5th Ed. There are many other features that have been on our todo list for so long (some since the EPOC days) but have never seen the light of day because of the constant porting to new UI frameworks!
DreamConnect ER5 (Psion)
2. for years, s60 built in Contacts application is too minimalistic and too simple for advanced users. for example, if a contact has a long note, you need to press Options and selet Edit to view the note. Not very user friendly I must say. dreamconnect solve this and many other problems. however, it is not entirely integrated to the S60 system. say for example, one wants to save a new contact from the last received calls, the system is still using S60 contacts. what do you think?
DreamConnect S60 5th Edition (Symbian^1)
This is an issue I raised with Symbian way back before S60 even existed. The idea of being able to redirect application “vectors” to 3rd party apps. Nowdays it could be very easily done with ecom or equivalent and good API’s. Still not happening, though… It’s not really in Nokia’s (short term) interests, to be honest, so why should they bother?
3. will dreamconnect be the default Contacts application for the new Symbian Foundation UI? Is it even possible? Because in my opinion, you guys will prevent a lot of headache by including dreamconnect (or even dreamlife!) inside all futere Symbian devices
We can dream. 
Having looked at the source for the Contacts application (as a foundation member), I think it more likely that DreamSpring could contribute certain UI technologies, but we haven’t started down that path yet — that’s a radical change to our development and business practices and requires a lot of figuring out. How do we make money to survive?
4. historically speaking, your ER5 version of dreamconnect is your only product, then you move into UIQ 2 followed by UIQ 3. but all those platforms can be considered dead now. only now you venture into S60. in your opinion, which is the most flexible and better symbian platform?
The best platform so far was undoubtedly UIQ 3. S60 5th Ed has better underlying technology (if only because it’s based on a newer instance of Symbian), but UIQ 3′s UI was wonderfully flexible, and was developing in a good direction. All that was really lacking in UIQ 3 was a decent GUI UI developer, like Qt’s Creator.
DreamConnect UIQ 3
It was very frustrating to see UIQ 3 dumped in favour of S60 (though understandable). But the rapid switch away from S60 to Qt, while also understandable, rubbed a lot of salt into still-raw wounds.
5. What is your future plan for dreamconnect?
Connecting DreamConnect to services has always been our vision. Unifying contacts (and calendar) in a powerful and business-friendly way is something that no-one (even MS) seems to be taking that seriously. It’s something we’re still keen on, and may result in some radical changes to what we think of as DreamConnect.
DreamLife
1. if i remember correctly, you coined the phrase “Life Information Management“, which is still used widely on your website and your products’ info sheet. where does it come from and what is the philosophy behind it? What’s wrong with “Personal Information Management”?
We wanted to find a way to express how this information (contacts and calendar) was connected together and involved the whole of your life (including work), not just personal stuff. Thus we came up with the term “Life Information Management”, which has taken on a life of it’s own, I’m proud to say.
2. And DreamLife is one of the ways to achieve Life Information Management? How? How about other products?
DreamLife takes the first steps of (a) properly linking contacts and calendar so you can navigate from calendar to contacts (and back again at some time in the future); (b) of implementing an information categorisation system that’s powerful enough to handle your whole life — in DL and DC that’s done with heirarchical categories, in future we want to have even more powerful mechanisms.
Most other products on mobile devices don’t seem to understand the way that a mobile device can act as a hub of your whole life — both integrating and differentiating personal and business data. Indeed, mobile devices are so bad at this that most people who need to differentiate between personal and business data simply carry two phones!
We’re certainly not there yet, and the constant race to keep up with Symbian’s course changes are making it very difficult…
DreamLife UIQ 3
3. With limited screen real estate on mobile devices, displaying valuable information will require a lot of compromises. I mean, looking at DreamLife’s calendar pane, I can see you are moving away from the traditional calendar display on mobile phones (weekly view, daily view, monthly view). You are bringing more desktop-like calendar experience to mobile devices with colour support for different categories (which brings reference to modern desktop calendar solution like iCal and Google Calendar). Why DreamLife was designed that way?
I’m a visual planner. To me a list of appointments of tasks conveys very little information. But a grid view shows me at a glance how I’m using my time (especially with coloured categories). It’s that simple. I recognise that there are a lot of people who work better with lists, and at present DreamLife simply isn’t suitable for them. I make no apologies for that. That doesn’t mean that we don’t want to be able to meet those people’s needs, too — we do, but we saw an opportunity to both differentiate DreamLife and meet the needs of visual planners like me, and we took that opportunity.
4. you have a standalone contacts manager, which is DreamConnect. But you dont have a standalone calendar application. You could have 3 products. A contacts manager (DreamConnect) , a calendar (?), and a combination of both (DreamLife). And sell them separately. Right?
We could, and we’ve gone so far as planning such a project, however we haven’t had a chance to follow through on it yet, and I remain doubtful of the value of such an effort.
5. What is the future plan for DreamLife? Will it include it’s own synching engine to sync with the cloud say to Google Calendar or MS Exchange Server or SyncML servers?
We’re well into development of our syncing solution, which I was demoing at SEE 2009. We have our own CalDAV technology which we’ve integrated with some tech from Symbian’s new CalDAV component, so we have a solution for FP1, FP2 and 5th Ed phones. We’re aiming our initial release at supporting public CalDAV servers like Google and Yahoo!, but in future we’ll be implementing our own services.
Others
1. why symbian?
See the discussion above, and add in the fact that Symbian is far larger than any of these markets.
2. what happen to riyala?
We’re still working on that, but we realised that we really don’t want to release a half-baked solution and have the sort of problems that, say, Mobile Me has had. So it’ll be quite a while yet…
3. The future of the new Open Source Symbian is not written on the wall yet. Nothing is concrete. What do you think about Symbian walking the Open Source path? How Open Source will benefit you as an independent software vendor?
Actually, I’ve just written a piece on that in our “Insight” blog. You can see it at http://www.dreamspring.com/news/insight-2009/symbian-moving-toward-openness.html
4. UIQ died and efforts are made to bring some of it’s designs and functions into the new Open Source Symbian. Do you believe there will be some UIQ legacies inside the new Symbian Foundation OS? What UIQ functions that you hope will be made available inside this new OS?
I’m afraid I’ve given up on that one. However, one of the greatest UI innovations in UIQ 3 (apart from the complete touch/keyboard integration) was the list view with it’s “mini-detail view” expansion. That was very clever, and I’d like to see a list view widget like that again.
5. But it’s still a long way to go. Symbian^4 which is said to unite S60, UIQ and MOAP under one roof, wont see the light untill somewhere in the end of 2010. That’s like a gazillion years in the mobile industry. Technology moves forward so fast where something would become obsolete and old-fashioned very fast. Your opinion?
Symbian’s current attitude that S60 is simply abandoned come Symbian^4 is very foolhardy to my mind. After all, Symbian^4 is both some time away, but also relatively close. Thus you’re telling developers don’t bother developing for S60 because we’re chucking it in a couple of years, but don’t bother developing for S^4 either because we haven’t created it yet. Does that make sense to you? And the only reason I’ve heard for discarding S60 from future phones, from a Nokia engineer, was ludicrous: the S60 libraries would take up too much RAM! Come on, how much has RAM grown in the last couple of years — Nokia needs to get a grip on reality here. I attribute MS’s successful dominance of the PC industry to two things: anti-competitive practices (which they’ve been convicted of) and unstinting attention to backward compatibility. Without the latter the former would not have been sufficient — ordinary people buy Windows PC’s because they know that virtually anything that want to run on them will more-or-less work. When that breaks down, MS suffers and quickly moves to fix it. Nokia and Symbian should be paying attention to that valuable lesson.
6. Qt will replace Avkon. Does that mean you will have to re-write all your codes from scratch? What is your opinion about Qt on Symbian’s future?
No, just the UI component, which is a huge part of our code, anyway. I have very mixed feelings about this, see above. However, Qt really impressed me with the demo I saw at the show, and if Nokia can get the Symbian widget set right (and preferably shared by other platforms) then it could make things a lot easier for us in the long run.
7. Does DreamSpring contribute to Symbian Foundation? In what way?
Apart from our membership fee we are currently too busy porting to S60 to contribute much. However we’re always looking for ways that we can contribute. There’s much less benefit for a small products company than for a large company (like Sun) or a services company (like Ixonos), so we need to bear that in mind.
8. The trend nowadays is to sell applications through application stores. You know, Nokia Ovi Store, Sony Ericsson PlayNow, Samsung Mobile Club and so on. Do you put your products inside these stores? How’s the market acceptance? Your opinion about application stores? Is it just a trend?
No, I think it’s here to stay. However, I think subject-specialist app stores are a likely development, and I’m glad to note that Horizon supports this sort of development.
9. Tell us 5 favourite 3rd party Symbian applications that you use daily and cant live without? Is it true that the value of smartphones and advanced mobile devices are measured according to what the users want? Not what the devices can do?
Obviously DreamConnect and DreamLife get a heavy workout, and not just for testing but because I really use them. Also Opera Mobile 10 is great, Mobireader (which really needs a 5th Ed version), Worldmate (or now PsiLoc World Traveller) is very handy, and Olive Tree bible software.
10. Will you see Symbian Operating System evolving into something bigger, not limited to smartphones? Say, in the future, people create their own Symbian based distributions for say, netbooks, Mobile Internet Devices (MID), satnav devices etc?. Look at Linux, there are hundreds of distros available. Will we see Symbian going to that route? It’s Open Source anyway.
It’s possible, but I think Linux has really filled that niche, certainly in terms of mindshare if not competence. Symbian’s core strengths are its great real-time kernel and its strong suite of comms and PIM support. Maybe an eMagazine reader, or the old communications hub idea (a Pocket Server instead of a Home Server), or something like that?
Thanks a ton Malcolm!
—
Visit DreamSpring’s official website http://www.dreamspring.com to loearn more about DreamLife and DreamConnect.
Read my review of these products at allaboutsymbian.com :
- Asri al-Baker