Sunday, November 26, 2006

Ouch

Siemens pay BenQ €50m to take away their mobile unit, then have to pay for the redundancies of the workers when BenQ reneged and canned them. Just think of the poor executives who had to put aside that 30% pay rise, barely enough to keep up with inflation (~2%) let alone reward them for the amazing performance and sale price of the mobile unit. Cynically you might assume they just rolled this year's pay rise into next for PR reasons, but who knows.

If only they'd thought of making good phones about 2 or 3 years ago, it could all have been averted.

One Step Forward, One Step Back

One of the many QR code reader companies, Kaywa, have now struck a deal with Kerrang magazine to link articles to mobile content. Whoever knew Technokitten was a Kerrang fan? But I digress...

She states exactly what I have maintained for a long time: QR codes are great if they are transparently integrated into the phone, as they are in japan where they are immensely popular and rightly successful. If the reader is separate from the handset's camera function requiring the user to learn a new set of behaviour to use it, it will fail - particularly when that path is longer than the path to the camera.

I see three possible ways to present a QR reading application:
  1. The best - you take a photo as normal, the phone tells you there's a link embedded in it, done.
  2. Second best - press whatever button(s) you need to get to the camera function, one of the options (next to 'take a pic' and 'record a video') is 'read a QR code', click it, done.
  3. Won't get used - any function that does not sit on the camera menu. This includes, sadly, Java apps.
Japanese phones use 1 and/or 2, depending on the make.

If the app is awkward to use, or it is not stored where the user expects it, the user won't use it. If the app is easy to use and is stored where the user would logically expect it, the user might use it, if they remember it's there and it offers sufficient value.

Sadly today, that means 3rd parties can only offer this functionality on some smartphones, ie. well under 10% of the market. For all other phones, the manufacturer must put it in the firmware - at the operator's behest (pioneered by DoCoMo in Japan) or their own.

I'd love to see a world where Java could be used for this type of phone extension effectively, but it isn't an area even being addressed currently as far as I can see and certainly it isn't practical with today's phones. VCs should perhaps bear this in mind before pouring cash into the latest bandwagon.

Thursday, November 23, 2006

888 Go Mobile

888.com, an Internet casino, appear to believe they are about to sign up with one or more UK operators to take their mobile product, which can already be downloaded. Interesting that they should tell Reuters about this upcoming stuff before the deal is signed without naming a partner, I am certain my mother said something about counting chickens before they are hatched... perhaps they don't know about the notoriously flirtatious nature of operators who will keep you almost 'in there' for a long time without ever offering that final release. Who knows, will be interesting to se the follow up...

Just for fun I looked round their site. The single thing I find most impressive is the handset compatibility list: it does not feature any Series 60 devices but does cover the oldest of Series 40, a reversal of the usual lazy coding practices. That hasn't stopped them using a Series 60 device (6680) to display screenshots etc on almost every page of their site though, including detailed instructions on how to text a shortcode (specific to Series 60) with a broken Next button that drag them out for ages, and a broken link at the end to the Login tutorial (broken on Firefox, anyway). Couldn't test the S40 MIDP1 version sadly, as it wouldn't deploy to our 7210 - their wap site is XHTML only, which means a lot of the handsets they claim to support can't download the content. Doh. The jad/jar are there though - you can easily spoof your way through to them in Firefox.

More impressive is the Sony-Ericsson compatibility list, which includes a lot of handsets - including the entry-level J200 series, which have no Java. Love to know how they pulled that one off. The incredibly tedious DHTML handset selection interface put me off looking any further - what is wrong with standard browser widgets? Why must sites hide information and force the users to use non-standard incredibly slow scrolling controls?

Security is an interesting area of their code - they use some, apparently, supported on 120 leading handsets. Great, but their latest release claims support for 340. Do the clients for the other 220 handsets just not feature security? They aren't telling, which is never a great sign.

The three clients (Roulette, Blackjack, Slots) are pretty basic - can't be bothered to review them, but pros:
  • They don't pack the mages or sounds in the Jar, and use stacks of classes, so if you're inclined to hack it apart it's very easy. Schoolboy.
  • They do use full screen graphcis where available, and nav is generally consistent relying only on two soft keys and arrows on any given screen.
  • Roulette is a very hard UI to do well on mobile, and their implementation is relatively intuitive.
  • Roulette practice mode allows you to place bets without reducing your balance. Nice!
    * actually it does take the money once you spin, but only after you've placed the bets - until then you have to do the subtraction in your head... ah well.
And cons:
  • Uggers. Really, Roulette is not nice to look at, though I've seen far worse of course.
  • Blackjack UI is not very intuitive at all. The whole two soft key thing breaks down when there are more than two options (eg. Hit/Stand/Double), and they haven't dealt with this well.
  • No sign-up in game - you have to go to the web. Surely a crippling feature for a product hoping to be included on an operator deck?
  • "Options" on the main menu takes you to instructions, but no options to change. Uh?
  • They produce versions for older phones, then make it unreachable from the wap site. New to mobile, boys? Doesn't inspire confidence about their QA.
  • The rules for Blackjack, on the D600 at least, are apparently "ArrayIndexOutOfBounds=128" (in a branded window, though, so give them points for style).
To be fair, it's their first crack, and they obviously haveb't done much mobile work before, so presumably they'll get better. Operators should maybe be a bit wary however, especially without in-game signup (assuming the operators will be acting as affiliates following the conventions for casinos, and have some interest in the future gambling of the players they introduced). I am hoping they do get better, because as I've said before I view mobile casinos as a force capable of pushing the J2ME networked app/game world into the mainstream, forcing operators to fix the wap settings issues etc which plague all developers with the age old incentive of serious cashflow. 888 need to fix some basics before they can be a part of the solution not the problem.

Misleading Headlines

It's always easier to nitpick someone else's work than create your own, so I'll just point out two unhelpful headlines in the last 24 hours from Slashphone:

1) SEVEN Delivers Push Email to the Popular Sony-Ericsson Phones (emphasis mine)

"The Popular Sony-Ericsson Phones" does sound so much better than "the two latest niche phones", but the P990 and the M600 really are two niche UIQ3 phones which I suspect are in no way as popular as the Walkman range, the K750 etc. Coming off the back of impressive sales and profitability for SE this appears to be a SEVEN PR bunny trying to create an exagerated impression of sales/market penetration/whatever among those who skim read the news, and Slashphone just printed verbatim.

Interestingly, SEVEN's blurb says they support "all major mobile phone platforms, including BREW, J2ME, ..." but their product actually only supports Symbian and Windows Mobile, ie. native code on smartphone platforms making up a small percentage of worldwide handsets. I don't have anything against them, they should maybe just be a little bit more honest in what they can and can't do...


2) Nokia Emerges As Market Leader
OK call me a pedant, but doesn't the use of the word "emerges" suggest this was some kind of new development, rather than a continuation of the status quo?


The world will not end, but hitting me twice in quick succession with silly headlines was enough to warrant a post methinks. One day soon I may move on and do one of the many posts I have planned containing actual original thought and opinion...

Wednesday, November 22, 2006

Just Doing Our Bit

(off topic)

Normally it's obvious when you're dealing with cranks on the net and they can easily be filtered out, but obviously sometimes we can help that filtering process. This kind of stuff gets me annoyed even more than people abusing stats...


Martin Luther King

Martin Luther King

Martin Luther King

Martin Luther King

Martin Luther King

Martin Luther King

Martin Luther King

Martin Luther King

Martin Luther King

Useful stats?

QuicklyBored have directly reprinted MobileRated's press release of their top 10 phones for downloads (I like QB, but I wish they'd make the division between PR reprinting and editorial clearer); of all the non-news pretexts for a press release MobileRated could have come up with to get some free publicity this would appear to be one of the most useful. However does it really tell us much? Two of the absolute latest Blackberries in the top 10 phones for content downloads? Maybe, but I'd be surprised...

The fact that handset number 5, the Sony-Ericsson K810, does not exist (outside the confines of student imagination) also makes me question the figures a lot, though presumably they meant W810 and just didn't bother to proof-read or don't know much about phones.

The presence of the 6620 suggests we're looking mainly at the US market here, the rest are all very new phones except the 6600 (lovely phone for development) and the ubiquitous chav phone.

All this leads me to one conclusion. If this really represents MobileRated's top 10 phones downloading their games, they are only being checked out by advertising and marketing execs owning the newest / most fashionable top-end phones. This is the only group I have seen who really do all have handsets like this, so maybe this is telling us something - just not what they wanted to reveal ;) Maybe we could see some more quantitive download figures?


UPDATE: finally found that very useful Bango popular handsets page - a nice public counterweight to the suggestion that people who download mobile games only have the absolute latest and greatest handsets. Real people downloading content appear to use a mixture of phones from across the spectrum, so MobileRated appear to be missing out on a lot of their target audience right now. It remains to be seen whether people reading their press release will realise what they're really being told.

Positive Vibes

I was just idly sitting contemplating writing a post on kittens or puppies, because there has been an overwhelmingly negative vibe to the blog as a whole and the mobile space rarely gives you an opportunity to shake that off unless you are the eternal optimist hoping mistakes won't be repeated when the newest latest and greatest ideas are implemented.

Kittens are great, but i think instead it would be more positive to round up some of the recent events in mobile which are good:
  • 3's x-series. Finally they've realised that they're getting nowhere with their walled garden approach, and they've done the right thing. They're in the ideal position to do it because they have little to lose. Lets just hope they remember to configure the phones with the right wap settings else it'll all be for nothing...
    (yes I was going to post when this news was fresh, but everyone else under the sun did so it seemed a little redundant... as so often, MobHappy summed it up best)
  • Sony-Ericsson still make great phone UIs, and the K800 is very nice.
  • HSDPA rollout has started in Japan.
  • I recently discovered that the original Symbian Monty JVM (observed on Nokia 6600) is able to hold the value -255 in a byte, which any Java programmers among you will have to admit is an impressive achievement, though ultimately not very useful.
    (If you're interested, a static byte that usually holds a negative number has a thread write 1 to it, that thread can read it back as 1 but other threads read it as -255 for some indefinite period afterwards. Consider the 32 bit nature of the Java heap, the int binary representation of 1 vs -255, thread variable caching and the known sync/thread instability of early Monty JVMS to get an idea of what's going on).
Hmm, now I am a little stuck and already veering towards the negative... ah well, good to at least try to get that positive vibe out every so often. One day maybe we'll be included on Carnival of the Mobilistas ;)

Tuesday, November 21, 2006

Hotxt - When do we get to meet Doug?

Seems that Mr Doug Richards of Dragon's Den fame has been meeting with the great and the good of the mobile technology blogging world (see Mr Hume, The Mobhappy folk and SMSTextNews - BTW read the comments on this one, they are brilliant) but no invite for myself and Raddedas :-( Our great sadness aside (we actually prefer the wannabe Simon Cowell that is Peter Jones) Hotxt is something that we've been looking at for a while.

The business model is nothing new, there was actually a German company doing this back in 2001 or such when GPRS was in its infancy and application development was even younger. There are also a crop current competitors including the likes of Simtext and MobiSMS (and even Reporo seem to have added messaging to their shopping application – if the biz model isn’t floating throw in the kitchen sink eh lads!) although what they all lack is user-base, or indeed the ability to generate one.

This is where Hotxt would seem to have a distinct advantage, £4million quid in the bank and the nice Mr Richards to front things. Securing such funding is impressive, and was no doubt tied in no small part to the publicity draw of the kindly Doug and lots of mentions of "the mobile Skype" (as a complete aside has anyone else noticed the number of mobile application businesses currently describing themselves as either the mobile Skype, the mobile Flickr or the mobile Google? Nothing seems to have VCs reaching for their chequebooks faster - but I digress). I’d imagine that the aim for the business plan is probably to build up a sizable user base, and then flog the outfit.

However they need a better app than the one which I downloaded the other day. On my Samsung D600, it downloaded ok, but on loading looked hideously amateur (no full screen, badly sized and laid out icons) and then asked permission to send Socket information and then when that didn't work it tried to connect via Http (a far more sensible suggestion) and I got it working. However the load screen simply hadn't prepared me for the actual UI. It was shocking. Using only basic on phone components, it looked like a first year computer science student's pet project. I showed it to some programmers and they actually laughed out loud. Did Doug write it himself?

And this is where, for me, the whole thing sort of falls down. Whilst they are doing a commedable-ish job of promoting the app and will no doubt end up getting some ok user numbers. Technically the client app looks so amateurish that for a company with their level of funding its embarrassing and certainly isn't going to get your average consumer rushing to use it more than once.

And that’s why I won’t be investing today.

Tuesday, November 14, 2006

JSR238 - why? Why???

The new JavaME internationalization API, JSR238, appears to be a lost opportunity - the originators rightly spotted that the basic MIDP/CLDC spec has no real support for internationalisation of apps, so they've chosen to add it with an optional JSR adding three classes. At first glance that might seem useful, but it isn't.

This JSR will presumably be supported by a small collection of handsets (including many of those which are remote updatable with whatever that JSR was). Developers will have three choices: continue handling things with custom code, develop nice international apps purely for this JSR and lose all other handsets, or handle both. All three choices have disadvantages, but the sensible developer has to go for the first unless they are an enterprise handling the rollout of all new devices to a controlled user group. This is certainly a valid scenario, but what about legacy handsets? How many enterprises can afford rollouts to their whole workforce like this? Why wouldn't they just spend a day writing their own solution, scheduling 15 minutes to moan about it not being in the standard APIs before forgetting the problem ever existed?

Looking at the details, it uses XML content descriptors and hierarchical directory structures to hold all resources, so clearly it is only aimed at the enterprise. This is jar eating stuff capable of adding to the download size dramatically, with added parsing memory and clock-cycle overhead, so it really won't be hitting the mainstream handsets with users who pay per Kb and the gaming or mainstream app world any time soon.

I fully understand that if we tried this argument with every new API, we'd never have any progress. That's a valid viewpoint in the general sense but I think this is an area where it is not valid. Optional JSRs make a lot of sense when they rely on underlying hardware (Bluetooth) or device capabilities (3D), and can't easily be mandated in the Profile. But for something so trivial, why not just add this to MIDP3? MIDP3 will definitely be rolled out faster and wider than a single niche JSR like this, and so over time it would become a tool the whole community might choose to make use of. Like this, I really don't see it adding value any year soon.

Friday, November 10, 2006

You Paid $4.5m For What?

PicDeck from Ontela appears to have some potential to be a useful app, allowing users to easily sync pictures with other services (in theory Flickr, etc, who they would obviously like to be purchased by). The Windows and Symbian versions might even integrate with the phone with enough depth to be useful, though I'd say nothing would be likely to beat a feature intergrated into the OS by a manufacturer who understood usability.

They even make the right noises about compatibility: "We’ve focused this on the mainstream users, and that means Java and Brew" a PR droid said on behalf of Dan Shapiro, CEO - and the droid is right, in the US those are the two platforms to focus on to hit the mainstream (most of the rest of the world has luckily escaped the lobbying power of Qualcomm and is relatively BREW free).

However, this all reveals two flaws:
  1. Camera integration will be very poor for all existing handsets (you can't force the user to take pics with MMAPI inside a Midlet, it is just awkward , bug-prone and won't happen) which means the user must take a picture, exit the camera app, click through to the Java folder, and launch the camera app then find the picture and choose to upload it. That's a lot of clicks for a service which is a minor value-add that should work as a time-saver, and more expensive to use than a quick PC sync.
  2. Most mainstream Java phones don't support the File APIs, so won't be able to read the photos. So all the talk of the mainstream only counts if you define it carefully... which is maybe why they don't list compatible handsets, only reiterating the argument that J2ME is mainstream and they support J2ME so they support the mainstream.
So actually, it innately can't be a useful app not through its own faults but through the faults of the phone OSs, and it won't actually be able to run on a lot of phones - I wish I had figures, but you have to assume it's under half of the J2ME phones out there (the US is very backwards, with a huge number of Series 30s and T720s still in use as well as a load of Sprint phones I'm not too familiar with). You have to imagine they pitched to those VCs using something nice and modern, or more probably some form of RAZR as they are American, but basically a higher-end feature phone; and because all the developers and all the VCs had nice phones, they all immediately extrapolated the potential market accordingly. Obviously the kind of user who has a high-end phone is the kind who is in theory more likely to want the service, but I doubt they raced to clarify the details with the smell of $4.5m in the air.

So what else did the investors get access to for their millions? Well, not many features, apparently. “We wanted this to be ridiculously easy to use for everyone, and that can mean removing features,” said the PR bunny in the CEO's voice. “We gave away all these features for the sake of simplicity and reliability”. I am all in favour of simplicity, but we basically have a glorified photo upload form here and I cannot believe it was impossible to create a super slick easy to use product which also had the ability to tag a photo - with an appropriate amount of effort.

However, the investors do get one amazing new thing the market has never seen before: "Patent-pending technology automatically recognizes the phone handset and supplies the proper installation package". Hmm, bet that patent stands up in court (it's the US, so of course it will be passed). Maybe something like this really didn't exist over there in 2005 when Ontela started, but you have to assume there's going to be some prior art - I wrote my first server to do this in early 2002, and I've worked with quite a wide range of others since then and there have to be dozens more I haven't worked with, and it all boils down to matching user agents to handset databases in the end. Not rocket science and not something to get excited about unless you're a VC gagging for some mobile action.

They are very proud of their installer though, and their uploading, and the fact that both are rock solid and utterly reliable - though it's difficult to see how they achieve this when mobile networks are so flaky, especially in the US. Maybe they're glossing over that fact here, trying to counteract a general perception of mobile networking as unreliable by emphasising their code is reliable - fine, but that really should be taken for granted...

It's maybe a bit rude of me to pick apart this company like that, plenty more could get the same treatment and one day they might roll out something interesting - though I persist in my opinion that handsets are too locked down for the level of integration required to make this type of service usable. But hell, it's Friday and I just don't care...

Wednesday, November 08, 2006

Operators Please Read

Vision Mobile's recent suggestion of a future strategy for operators makes an immense amount of sense and seems to offer exactly what consumers want whilst also making the operators money, so much so that it stands no chance of ever being adopted by the current management at the operators.

After all, these are the same people who despite having zillions of employees in charge of everything under the sun are incapable of specifying a small settings file on their handsets to make GPRS work from within J2ME, thus denying themselves revenues and a potentially booming content market for the cost of a few hours of work, which they could probably force onto the handet manufacturers anyway if they wanted.

Tuesday, November 07, 2006

Pedantry Rules

OK, I may be wrong, but I think there's an attempt to add authoritative context to this short Reg article on Symbian buying UIQ which manages to get a number of key facts wrong:

"Most famously, Nokia made just such a move, and developed the precursors to its current Series 60 and Series 80 - aka S60 and S80 - UIs, both of which..."
Maybe just poor phrasing, but the latest OSs are S60 and S80 and their precursors were Series 60 and Series 80. Ambiguous but we'll give them the benefit of the doubt.

"UIQ remained as the UI for Symbian licensees unwilling to develop a user interface of their own. Siemens was originally a UIQ fan, but its support went the way of all flesh when it sold its mobile phone division to BenQ, never a vociferous Symbian supporter."
BenQ developed the P30 and P31 using UIQ, though it shifted to Windows Mobile for its successors like the P50. Motorola also created some hefty UIQ 2.0 bricks which could be useful to have on your side in a fight. Siemens on the other hand released the SX1, a Series 60 phone, but nothing on UIQ that I ever spotted (could be wrong though). Of the remaining Symbian licensees, DoCoMo have created their own UI layer used by six manufacturers and Samsung have toyed with a handful of Series 60 handsets (one of which may even have been launched), but only SE and Nokia have really pushed the OS and so purchasing UIQ makes sense for SE.

"That's left Sony Ericsson as the major UIQ licensee, adding to its P series of smart phones, its M600 smart phone and a number of its higher-end Walkman-branded handsets."
Well, one is unambiguously a number but I think most people would read this sentence and assume more than one Walkman handsets run UIQ. Actually, ten Walkman phones (not including those pesky i/a/c variants) run the SE proprietary OS, leaving the W950 as the only UIQ Walkman.

Not a bad mistake rate for four paragraphs. Now some of this is nit picking, for sure, but I think if a respectable news outlet is going to go to the effort of providing some background detail, they should go to the extra effort of actually getting that detail right, especially when they are more than happy to lay into the minutiae of other people's statements.

That said, they'll continue to be my main general technology news source, so I can't be taking this slip up that seriously ;)

UPDATE: of course I criminally forgot to mention that Nokia have also licensed UIQ, though I would tend to view this as a Chinese market anomaly and quite why they bothered escapes me when the phone is so old...

Monday, November 06, 2006

Vodafone ditching Blackberries/UIQ?

Well, probably not, but they have just stated Windows joins Symbian S60 and Linux as the other approved OS for Voda handsets over the next 5 years - though they'll happily carry other OSs onlow-end handsets. Maybe I'm missing something here, but doesn't that just mean that they'll continue business as usual, with maybe a few more Windows devices and a few less UIQs and Blackberries?

Seems like an excuse for a press release for me, but a minor win for MS given that Vodafone haven't really jumped on the rebadged Windows bandwagon to date.

Friday, November 03, 2006

Statistics rant Abuse rant

ABI Research appear to have carried out an exercise in futility recently, asking 1725 US teenagers if they would replace their existing iPods with a product not yet launched, that no-one has really seen and few have used. 58% said they would, therefore Apple is doomed. Apparently.

I hate this kind of research almost as much as I hate publicising it (though I think our readership is small enough that that doesn't matter too much ;) ). Firstly, what were the exact questions, and who paid for the research (closely related I'm sure)? Secondly, once quite a few people have got their hands on a Zune, and some word-of-mouth has spread, will the results bear any relation to this?

It is unfair to say that the Microsoft Zune will not be a quantum leap in music playing - maybe it will be easier to use than the iPod, look sexier, have fewer bugs and have less intrusive DRM and a better syncing solution. Its Microsoft heritage suggests otherwise but we won't know til the reviews are out, and asking this kind of question now is meaningless and only goes to show how you can buy opinions - I can't 100% guarantee Microsoft paid for this research but I bet they're not upset by it, and I also bet if Apple had carried out the research it would have reflected roughly the opposite opinion...


Another example of this recently, in a press release picked up by MobHappy, Nokia claimed "two out of three consumers whose phones can play digital tracks already use it for that purpose". Two thirds of users? 67% use their phones as music players??? Intuitively to me this feels wrong, as I pointed out in the comments - but it was nice to find m:metrics agreed with me (UK nearer 3%, US trailing on 0.7%), and I respect their figures more than the unverified soundbite of a Nokia PR bunny trying to justify a $60m purchase of a music portal. Nothing against Russell or MobHappy, they have their stand and they have every chance of being proven right, but not by corporate misinformation like that "2 in 3" quote...


UPDATE: and again, I got carried away without fully researching, so I should stick to mobile phones where I have reasonably up-to-date knowledge... A few people have been allowed to touch Zunes and write about it, though whether the 1725 US teenagers had read the summaries is not mentioned. So I've corrected the earlier false assumptions but the sentiment remains the same... and suitable chastised for not thoroughly researching etc. Certainly for me, the problem with blogging is you fire off a missive when still enthusiastic or annoyed by something, spur of the moment, and don't have time to fully research - if you try, you never end up actually posting the thing