Wednesday, August 30, 2006
Tuesday, August 29, 2006
Buy our company now! (Please)
Kiloo, who I must confess to having previously been ignorant of, do have some tasty IP in the form of the Lego license and Hugo the Troll (who I'm reliably informed is a big name in Europe in the same way the Germans are obssessed with Alf and the Hof). However from their site the seem to be more of a developer than publisher with few operator relationships, (traditionally seem as one of the cornerstones of value to mobile investors) with the Lego games set to be distributed by the folks at Handsonmobile.
However good luck to them, although someone should explain to them the concept of playing hard to get.....
Monday, August 28, 2006
Is it possible that you were getting too misty eyed and bendy kneed about seeing China Mobile, Google and Nokia in a headline together to read the very small article?
Google have nothing to do with the QR codes. QR codes are a way of encoding a URL or other small piece of info - up to 400 chars or so - and you can create them for free. As people have been doing to good effect in Japan for some time. The QR code handling will be done by a subsidiary of China Mobile, not Google; Google aren't involved (sorry, I'm labouring this but it seems neccessary). The key interesting bit of the single paragraph you didn't read is that Nokia will preinstall the QR code software, because *if* the integration is handled well and *if* this drives take up by other handset manufacturers it will become ubiquitous, as in Japan, and it will be successful in a way that 3rd party downloadable add-ons attempting the same task never could be. There is no world beating revenue opportunity here for Google because any URL can be in the code, not just one passing through Google allowing them to track and bill for advertising. They are entirely seperate, as the article makes clear. If anyone is making money from this it will be China Mobile increasing ARPU, something they are no doubt well aware of.
If you are going to comment on news please read the news first, and make it very clear when you are indulging in wild speculation which has nothing to do with it. Some people appear to actually take your blog seriously and you owe it to these people to do some due diligence.
PS Google aren't involved in the QR codes. Get over it.
Friday, August 25, 2006
However Samsung have come up with a good use for it - varying the D900's home screen:
- Country-specific foreground (Kremlin in Russia, Palace of Westminster in UK)
- Time-specific sky colouring
- Signal-strength dictates the cloudiness of the sky
- Missed calls shown as a plane in the sky
Main take home for me is that 26% of users fail to make a wap push work, even when it is delivering content they paid for and they have correct wap settings. A breakdown by handset type could be very revealing here, how about it?
Wednesday, August 23, 2006
Jim: Where is Sun in the process of open sourcing the code for Sun's Java platform implementations? When can developers expect to see the code released?
Laurie: Sun will release several significant components of Java SE by the end of 2006... In addition, Sun plans to open source implementations of the Java ME platform (both CLDC and CDC). We intend to roll this out by the end of 2006.
So come 2007 we'll be looking at bug free J2ME right? Well, no. Two small problems.
1) The lead time on any bug fix becoming useful is the time it takes to fix the bug + the time it takes to get a handset to market with that bug fix + 3-12 months for that handset to achieve enough market penetration to matter. At that point you have the bug fix and the bug in the wild; the bug cannot be ignored for another 1-2 years when the handsets that feature it drop out of common use. That said, some people still use the Nokia 7650, so the bug may remain in circulation for longer.
2) Most of the serious bugs are in the interface between the low-level bits of the JVM and the proprietary handset OSs. It's possible Sun have broken implementations of Integer or Vector in CLDC, but I doubt it, so the most I think we can hope for is a fix of that PNG palette bug which is still in the WTK, where white is misread as transparent - but we'll still have to keep using the workaround for older phones.
This is hardly an argument not to pursue greater openness - we have to hope that things can get better in the future - but the lead times are so long I can't find myself getting at all excited.
The key advantage that I can see from an open source ME implementation is the ability for manufacturers who use it to more easily integrate the LCDUI APIs with their own phone native look & feel, so we don't have the jarring difference you find on eg. Samsungs when you switch out of a nice colourful phone UI into an ugly mono LCDUI interface with totally different widgets, etc. Though call me a cynic, but I suspect they could do that under their old licencing agreements - and just didn't bother.
Tuesday, August 22, 2006
Infrared has one set of problems: it requires line-of-site, and even users that have it don’t necessarily know they have it, wouldn’t know how to turn it on, or would get bored before they’d manage to turn it on and wait to receive an advertising link to a wap site they have to download.
Bluetooth is newer, sexier, quite widely supported now, and highly unreliable (in my experience) with all the UI issues of IR. No user who values their battery life would sensibly leave Bluetooth on the whole time, unless they didn’t know they had it or they were using their headset. It avoids the line of site issues that make IR unattractive, but this introduces a more serious problem – the Bluetooth box effectively becomes a Bluetooth spam device because you can’t control who gets the advertising being pushing out. The passers-by who pick up this spam will either be those who don’t understand Bluetooth – “Help, how did Ann Summers get on my phone! That’s not my bag baby, honest” – or the people with headsets, who will get instant beeps telling them to check their phone… for spam. This might have legal implications, but it will certainly annoy the hell out of the alleged 85% of passers by who resent being spammed – with branded location-specific spam so they know who to blame. Nice.
They’ll apparently be receiving anything from a link (ooh, compelling) through ringtones and wallpapers to a game. Presumably the tag can do handset sniffing to deliver appropriate content (if it can’t it’s really a waste of space), so for brands which can really benefit from this type of content its got some potential value. When considering the value of links to content on the network it’s also worth remembering how great the user experience is when struggling with network settings.
However, all the VC capital is not lost – Hypertag have a solution to keep them out of court. They will make users change their Bluetooth IDs to include a special unique Hypertag code, which acts as an opt-in: a wireless signal saying “Hi marketers, do your worst” whenever you walk down your high street. Who will do this? Let’s consider the potential market segments:
1) Bluetooth users – experienced users will have to change their Bluetooth IDs and probably re-pair all their devices so that they can receive advertising messages. That’s a wonderful added incentive to make sure they don’t.
2) Inexperienced users – don’t know they have Bluetooth. Wouldn’t know a Bluetooth ID if it jumped out of the Hypertag box and slapped them on the face, but might have a quick go at changing the config. Given the plethora of phones out there this is destined to failure even with the best instructions, potentially causing annoyance.
3) The rest – users who know what Bluetooth is, but don’t use it much. They’ll have it turned off, but if they see the advert they might have a go at changing their ID because this won’t impact the rest of their lives, and because they know a bit more they might even succeed.
So we probably have a target market of people who are kind of curious. Will that curiosity dwindle after the first few messages? The first dozen? Hundred? What is the cost of acquisition to the people running the campaign? I seem to remember hearing these devices were really not cheap to rent, though that might have been a competitor's product.
This leaves Hypertag with a system which will be used by at most 15% of the population who might like indiscriminate alerts popping up on their handsets; a large percentage of these strange people will fail to get the system to work, and only the really persevering masochists will actually be able to download very interesting content. Hmm, maybe Ann Summers (NSFW) should look into it after all, could credibly target a nice niche market for them...
Thursday, August 17, 2006
Is A Locked-In Feature Better Than No Feature?
The reasons are clear - lock customers in and increase data traffic.
However, cost-concious users or customers who are roaming for an extended period (at £8/Mb) will probably be sufficiently turned off by all this unneccessary additional hassle that they won't use it at all. Without this lock-in they might have enjoyed the convenience of GPRS syncing whilst on the move, for example when they pick up an important new contact or schedule a big meeting, whilst doing more regular syncing at home to avoid the tariffs.
Am I wrong in thinking that this sort of strategy is self defeating for the operators? If the service were truly open, some customers would certainly use the service exclusively through a PC and thus be an unprofitable resource drain for Orange, but many more would surely be tempted to start using the service in a cheap way (possibly via a PC), then as they started to perceive the value they would become more regular and profitable customers?
To my mind if an operator wanted to truly drive data usage they should offer this kind of service unrestricted for free, permanently in such a way that a user will latch onto anything they find useful, naturally increasing their usage as they become comfortable, and driving increased data revenue alongside sales potential for service upgrades. Accept that there will be freeloaders, and realise real growth with the majority of users who don't mind a small cost here and there when they can feel the value, but would never have signed up for a premium service on day 1.
100% of very little is less than 10% of a lot.
Just back from a rather hectic week in Japan sorting out some iMode DoJa content - sadly my site-seeing rather curtailed by the realisation that whilst DoJa is pretty standardised and reliable compared to the morass of MIDP phones in the West, they do have bugs. In particular certain types of HTTP connections will work in vanilla HTTP, but fail subtly under SSL with byte injection - a nice recipe for 16 hour days until the bug was tracked down. Killed the blogging a bit too.
Since my return I've been investigating mobile gambling for a series we plan to do on the blog, because connected mobile games have yet to take off in a big way for all sorts of reasons and we view the gambling sector as a key enabler here (in Europe at least, where gambling online is broadly legal) - because they have the financial incentive and the leverage to find solutions to Java connectivity, both with the operators and despite them.
This brings me on to the point in the post, a rather masturbatury press release I just read from Slotland. Razrs are big right now and Slotland clearly want to leverage that for all they can, so they have written a load of cobblers saying it's a great development phone, etc - much more contemporary and enticing than a puff piece stating they have a product, some people have used it, they have made an awards shortlist, yawn.
"If they (J2ME developers) could choose one phone as the standard, it would be the Motorola Razr."
Oh really? I have yet to meet a developer who wouldn't standardise on an SE K750i because it's fast, slick and reliable, with more features that are better implemented. Whilst I've made it clear I hate the UI on the Moto, I would rank it as a so-so development phone - quite quick, relatively bug free (though the random networking exceptions thrown when working with a bad signal can be a bitch), not especially attractive (no alpha transparency, ugly font, ugly header bar) but equally nowhere near the bottom of the pile.
You can't really complain though, the entire piece is crammed with as many self-congratulatory claims as their PR bunny could squeeze in. Accuracy doesn't matter when you're acting the expert to an industry and public that don't understand the technology. Peer recognition doesn't matter for a product-based company - credibility like that is only really interesting if you're doing consulting.
They seem keen on their ME award shortlisting, but when I read that list I can't think of any other competent mobile gambling offerings out there. ME are lucky they didn't try to pick ten companies for the shortlist, they'd have run out.
Blatant idiocy or abuse of the facts always annoys me, but I think I've calmed down now. Coming up soon, reviews of the full user experience with those five shortlisted companies and any others we can find, to see how they are leading us into the brave new interactive world! Alongside a comparison of the Japanese and UK mobile markets, of course.
Friday, August 04, 2006
It's easy to be cynical in the wireless world, both conventional mobile and the WiFi bubble, because everything really is twenty times as complicated as it should be and good ideas are rarely implemented well. But it doesn't change the fact that he's right; I feel like I should add something more but WiFi's not really my main area, and as Dean explains it probably won't be for quite a few years...
Thursday, August 03, 2006
My Chemistry teacher would not be impressed. I remember being told in no uncertain terms that drawing a graph with a few points in the bottom left and then extrapolating those across the rest of the paper with your ruler in a long straight line was Very Bad Science. Obviously no-one told AlwaysOn this.
(Guesstimation: following the extrapolation the size of m-commerce in Japan is multiplying by 10 every 5 years. The world economy is currently $34,000bn rising to $140,000bn by 2056. So Japanese mCommerce overtakes the world sometime in the late twenties)As an aside, it's very dangerous to try to draw parallels between the Japanese mobile market and the rest of the world - this is something I want to go into properly in a future post as I've been spending a long time working on the Japanese market recently, but in brief DoCoMo launched iMode on a market with very low PC/internet penetration and from that point on owned the connected experience for a lot of consumers. Combine that with sensible revenue share models (unmatched in the short-term greedy West) and a rigid control of the handset from spec to UI, and you have a very mobile-friendly market lacking the hideous usability issues plaguing everyone else. It will be many years before gaijin will be in a position to benefit from this ubiquitous access and the operators show every sign of not caring.
I've been a fan of the Vodafone Simply concept for a while, and the post points out a few interesting facts - mainly that Sagem may be the nominal manufacturer but they have very little to do with what makes the handset tick, ie. the UI customisation and software which presents the user with a stripped down subset of core functions (voice, text, clock, not much else). The article suggests that, were it not for fashion being the ultimate driver for handset desirability, most people would be happy with handsets that removed 80% of the functions that only 20% (or even 2%) of users actually want. As a developer that scares me but as a UI designer I can't fault it. My takeaway feeling from that is that handset manufacturers have to start doing some serious consolidation work on their UIs - new features are nice, but make what you've got work.
Simply phones address the (large) market segment of users like my parents who want a phone to talk on, and might brave SMS once a year; smartphones running Symbian or Windows cater to subtly different forms of power user; Blackberries cater to those who want no personal space, ever, and so on. These niches still leave a big fat chunk of the market - probably over 60% - who want different recombinations of phone + camera + calendar + browser + MP3 player, give or take, all wrapped up in every permutation of case design that manufacturers can come up with.
If we stick with this 60% we move into the realm of the other article, on brand vectors. This assumes two things which I feel are only partially true:
- Users wantt o fully customise every inch of their phones
- One day any person will be indexable by the list of brands they follow
The brand vector concept is sadly likely to come ot pass, but it may not give as much fine control to marketers as expected. I think a lot of fields will need to be nullable - I for example actively search out clothes with no visible brand affiliation, whilst my preference for airline can neatly be expressed as 'cheapest' (including hidden costs - which is why I avoid RyanAir like the plague - does the vector allow for negative associations?). Will brand vectors result in a more granular targeted world, or will we just find most people lumped into categories like they are today, because so many follow basically the same tribal affiliations?
I suspect most people will still end up in coarse groupings, with distinctions being made on sub-categories of the vector - music, travel, fashion - which may often be closely correlated. Some will stand out more, but maybe not enough to justify much content or product customised to their needs that doesn't already exist.
The relevance for phones in all this? I think manufacturers need to consolidate their interfaces, cut the fat out, simplify and generally improve them. This may result in interfaces which can be customised and blended for different market niches - camera users, music users etc - and certainly it will also result in the ability to reskin and fine tune UIs if the customer is interested enough to do so. But to the needs of most users will be best met with simplicty out of the box combined with off-the-shelf fashion skins.
Where does this leave content developers? As a mobile developer, I hope we'll see hooks for properly integrated programmable content (J2ME or whatever replaces it) so users can download the niche killer apps that will work for them, and have quick access to them - when this happens we'll have a trully powerful platform. But currently the trends are diverging and neither looks good - on the one hand simple phones where Java is jettisoned as a preium/complex feature, on the other hand feature phones and smartphones where Java is there but integrated under a growing maze of menus so content is too many clicks away and never used.
Please, someone, draw a line under current mobile UIs, start with a blank page and build a simple, modularised, extensible interface that can be tailored for niches without overwhelming a user. Without it the mobile experience will never fulfil its potential and that will be very sad.