Rants And Ramblings About Mobile Technology
Anders Borg writing about the fun and crazy world of mobile and Internet service technologies.
You can also read the blog via Twitter or your phone via wap.abiro.com. See the left menu for more news.
Comments on blog entries are moderated, but I'm rather liberal as long as it's not blatant advertising.
For general comments, advertising and contribution queries, please use the feedback form.
Sunday, May 31, 2009
Omar Hamoui on entrepreneurship
Omar Hamoui is the founder of AdMob, the world’s largest mobile ad broker. AdMob was founded in early 2006 and are now over 100 employees. A success no doubt, especially as Google (with its Adwords/Adsense) doesn’t seem to be able to knock AdMob from the top position, and to my knowledge AdMob is quite profitable.
Interesting “solve my own problem” starting point: “Omar Hamoui wanted to build traffic for his mobile site. He encountered complexity and fragmentation - it was just too hard to engage users. AdMob was born to remove roadblocks and enable mobile web businesses.”
Distorted by marketing speak (“engage users”?), but nevertheless indicates Omar had a real need of ads himself.
On Competitive Threats and Team
On Communication and a Final Thought
Monday, May 25, 2009
Upcoming Android devices
Just browsed through Engadget. Not just phones.
From Motorola:
From HTC:
- HTC Lancaster is a QWERTY slider with Android for AT&T
- Android build for upcoming HTC Hero has revamped interface, social networking?
From Sony Ericsson:
From Samsung:
From SDG Systems:
- Trimble Nomad handheld gets Android 1.5 upgrade
Navigation something or other
From Archos:
From Fujitsu:
There plenty more:
Sunday, May 24, 2009
Developing applications for Android, part 2
I had the Jitter code and most of the UI ported on Friday, and it all compiled then, but nothing really worked. In part because I ran into a problem with the build path for the project, that took a few hours to track down, and I’m still not quite sure if I made a mistake or there’s a real problem with ADT.
Anyway, this is how the application looks right now:
It’s rather user-hostile at the moment, and the only thing that really works end-to-end is Update. I can retrieve Timelines as well, but there’s a problem with accessing UI widgets from another thread. Probably perfectly logical, but I haven’t been able to understand why this is yet, and how to get around it.
Update and Timelines will probably be provided via separate screens rather than on the main screen. I have to test what works best. I will also try to show user thumbnails, format message text, extract clickable links and support emoticons.
As there’s already good Twitter applications (including Twidroid) for Android, I don’t expect my application to become a hit, if it even will be released publicly. It’s mainly a way for me to learn the application paradigm of Android. There might be an opportunity to make a specialized Twitter client, that adds GPS or cell position, sends messages based on incoming SMSs, attaches photos to the Tweets etc etc.
So far I’ve spent almost a work day (read: 8 hours) reading up on Android development, and close to two work days developing (and drinking lots of coffee, and cursing). I will need approx one more work day to get all Twitter features running, and probably one or two more for optimizing the UI, but lead time wise it will take at least a week.
Thursday, May 21, 2009
Developing applications for Android
Not quite yet in my case, but I’ve read “Hello, Android” (recommended) and gone through most of the examples provided with the book, as well as built and tweaked them to see how stuff work.
I’m using the Eclipse + ADT combo. I tried using NetBeans + nbandroid, but never got that to work with Android 1.5. I might try that later, but it really doesn’t matter what IDE is used.
My first project will be to port Abiro Jitter to Android. It’s enough to provide some challenge. Not that the world is void of Twitter clients for Android, but considering such a client can tweet about so much more than just what the user might want to say, it’s still worth the effort, and an interesting platform for other application experiments. It’s also a relatively simple application overall.
The UI is currently my main concern, as the paradigm is quite different from that in MIDP. All the Twitter protocol stuff I’ve obviously already implemented (from scratch; there were no free libs at the time), and it’s all Java (yet with some other classes than ME and SE) and communication is via HTTP, so hopefully not much else should cause problems.
Tuesday, May 19, 2009
Sign of the times: Even Smurfs come from China
Sorry for further lowering the quality of this blog, but I guess I’m too busy to write anything serious and balanced at the moment.
I earlier reported about most consumer devices being made in China, and it would not be far-fetched to suggest that cultural icons like Smurfs also would come from China, even though they are distributed by a German toy company.
Here’s the whole set, with the latest in 1” netbooks. The keys are a bit small, but that doesn’t matter much, as the screen only has 176*184 pixels (the resolution of a VIC-20), and it can only run VIC-20 games and applications with Smurf in the name.
Just kidding, or am I? Actually a computer capable of running VIC-20 applications could be made this small. It’s the UI that’s the problem, but at least my Smurfs approve (Papa Smurf drinks to it).
Actually I bought the Smurfs to use as office decoration, and the PC doesn’t work, I think.
Sunday, May 17, 2009
Amazon Kindle DX and ebook evolution
Updated after reading what this device is all about.
Amazon’s Kindle DX is a larger Kindle deriving from the technology in Kindle 2. It’s got a 10” display instead of the 6” on the Kindle 2, has more Flash memory than the Kindle 2, and can auto-rotate pages, but apart from that it’s pretty much the same as a Kindle 2.
It seems optimized for reading documents with pictures and lots of text per page in PDF format, like computing and educational books, rather than literature. It’s also suited for people that need the extra size for larger letters while reading literature.
$489 is even steeper than the Kindle 2, but I still want one. If they had taken out the mobile radio it probably would have been $100 less.
Amazon: Kindle DX: Amazon's 9.7" Wireless Reading Device (Latest Generation)
PC World - A Guided Tour: Hands On With the Kindle DX
TechCrunch - Rampant Piracy Will Be The Kindle DX’s Savior
Engadget - Amazon Kindle DX announced: $489, ships this summer
Electronista - Amazon launches 9.7-inch Kindle DX
Not about the DX specifically, but the Kindle phenomenon:
MSN Money - Kindle users are old
Anecdote: Black holes and … socks
For what it’s worth, here’s a conversation with a guy who’s slightly more well-versed in astronomy than I am.
The mobile angle: “In space, no one can hear you scream … because there’s no coverage.”
And no, brane is not the same as brain, but short for the multi-dimensional membrane that supposedly our universe is a side effect of. I wonder how they came up with that theory though.
Question
I read the answer to the question about the size of a black hole (at Curious about Astronomy?), and I'm wondering what the event horizon has to do with the size. I guess it's rather the size of a quark, or even nothing (a real hole in 3D (or higher) space, into the brane). Analogy: When measuring the size of a vacuum cleaner (read: black hole) you don't count the distance at which socks (read: light) are sucked in. Bad analogy, but you get my drift.
Cheers,
Anders
Answer
I love the analogy actually, but the problem is what if you had to use socks to learn about the vacuum cleaner. ie information is carried via socks (light) and so we need to be able to grab some socks to learn about things, then the best we can do is say there is a region from which no socks are ever emitted and to which socks are constantly lost once they enter the region. Then the best we can do is quantify the size of the region, by giving the size of the event horizon. In fact if socks are light then we would never know that what was in fact responsible was a vacuum cleaner, which means further that we cant talk about the size of the vacuum cleaner itself but only the size of the region it impacts.
For this reason we don't actually know the state of matter inside an event horizon it is possible that some new degeneracy pressure kicks in and actually maintains a finite sized ball for a while inside the event horizon, and that as the black hole grows in mass it eventually overcomes the pressure and has a supernova like explosion (but all contained inside the event horizon). However it is most often stated as you have that the size of the actual matter that creates the black hole is actually infinitesimal, ie a true singularity of completely collapsed matter or energy.
Put simply, our ignorance of physics at such matter densities (an ignorance which is not likely going to go away since an event horizon is created at these densities which obscures the events inside it) prevents us from talking about the size of anything but the event horizon.
Hope that helps,
Istvan Laszlo
Question
Thanks for the elaborate response, and continuation of the slightly flawed sock model :).
The reason I asked the question was that I've read articles stating the actual size to be that of a particle (or even nothing) as if it was somehow proven. I understand now it isn't.
Would it be right to say that the radius of the event horizon is proportional (not necessarily linearly) to the mass of the black hole, and that is all that can be measured? What about determining the contents of a black hole (relative quantity of types of particles)? Impossible to know? I figure atoms are torn apart already in neutron stars, so the difference in a black hole _might_ be total collapse, but not necessarily.
Anders
Answer
Yes the event horizon radius is 3 times the mass, if you want radius in kilometers and the mass is given in solar masses. (This is true for a nonrotating blackhole, and is a theoretical result which could be tested but has not yet.
This brings up an interesting point, the size of the singularity is in theory zero ie a true singularity, however this cannot be tested, and anything compressed to a size smaller than the 3M limit of the objects event horizon would be a black hole.
Indeed we cannot really say what kind of particles the matter is in, likely it is not neutrons since neutron degeneracy pressure prevents black hole formation once its over come the main particle constituent may be quarks, but we dont know (in fact it is possible quarks would also prevent black hole formation and need to be destroyed to make a black hole.
So basically as you say atoms are torn apart but to what degree we just don't know.
Hope that helps,
Istvan Laszlo
Well, there you have it.
Friday, May 15, 2009
Really thin TVs
And I mean sub-mm thin.
Sony and others will release large displays based on OLED technology where the display layer is just 0.3 mm and the whole kit with a protecting layer etc is 0.9 mm.
Display technologies evolve very quickly now, moving from decimeters, to centimeters, to millimeters in a few years, not by optimizing a certain technology, but by rapidly moving to completely new technologies.
I expect OLED displays to completely take over from LCD and Plasma displays, provided that OLED displays can be made bright enough, and it seems they already are.
They will be used for TVs, computer displays, mobile phones, media players, ebooks, magazines, posters, “paintings”, you name it. Very likely they will (as listed) also open up for new types of devices.
(the English is even worse than at my site, so beware)
There are also very flat speakers available, vibrating the whole surface. Maybe such could be used for smaller TVs put on the wall, or for small media players etc.
Sunday, May 10, 2009
NFC on posters
Not meaning anyone is promoting NFC on posters. I’m rather meaning the NFC Smart Poster Specification.
Actually it’s remarkably quiet about the technology that is intended to turn mobile phones into wallets and ID cards. Also, how many phones have you seen in USA or Europe with NFC?
As most of you know (no? really?) NFC is based on RFID, as in Radio Frequency Identification, but could be said to be a narrowing in terms of radio frequencies and distance, yet an extension in terms of use cases for primarily mobile phones. While RFID now supports numerous different frequencies and longer distance transfers (like 10 meters or so, best case), NFC is specifically geared towards very short distance transfers, like a few centimeters, to avoid eavesdropping (at least in theory) and make device discovery simple (a major problem with Bluetooth, if used for the same types of applications).
To enable the wallet and ID card functionality the NFC component is generally put in the SIM card, that is essentially a Smart Card. That way providing a closed “block” that can’t be tampered with to falsify identity, provided of course your phone is not stolen.
Smart Poster enables actual posters containing NFC tags to convey messages to services, e.g. for rebates. It’s really very simple: The tag can contain information (typically a keyword for addressing the service and associated service parameters) and a destination number, that will be sent as an SMS. The tag can also contain information for accessing the service via the phone’s browser. In both cases the tag houses all this information in a URL, even for SMS transfer. Developing services for Smart Poster is therefore very simple too, similar to handling inbound SMS’s that users have written themselves, or like handling service-to-service queries via HTTP.
The question is when NFC will reach critical mass. It will take years.
NFC Forum – Specification Download
Will the future go retro?
In the new Star Trek movie (watched it yesterday; two big thumbs up for acting and effects, thumbs down for major plot holes and physical impossibilities) young James T. Kirk is driving in a pretty old car with a dashboard phone from Nokia. Here’s the clip up at Engadget.
Nokia, cheesiness featured in new Star Trek movie
Apart from the fact that it’s a blatant product placement, hasn’t Hollywood yet heard about mobile phones, as in carried-along phones, and where’s the heads-up display on the windshield, the speech control, the video, the caller’s personal info, etc etc, things we can do today.
Maybe it was meant to be shown as retro technology also in the phone department, or Hollywood can’t figure out how far person-to-person communication will come in 200 years from now. Mobile phones have been around for 1/10th of that, so expect communication to be pretty different then, but very likely not go retro.
ESS coming to Lund?
Lund in Sweden is considered for European Spallation Source, a very large (understatement) lab for neutron research. Maybe it doesn’t matter that we have a telecom crisis, and Sony Ericsson and ST-Ericsson (the merger between Ericsson Mobile Platforms and ST-NXP Wireless) are laying off people in the region, if we all with some level of engineering and management skills can get paid by EU to develop this lab.
I’m a physics layman at best, and I might not be able to split atoms and I don‘t know much about quarks, but I’m good at hairsplitting and I’m quirky. Does that count?
Wikipedia is as always to the rescue: European Spallation Source
Scandinavia – the perfect host for ESS (promotional document)
Android, for more than mobile phones
Here are indications that both netbooks and media players will come with Android soon.
I have lots of music, yet 500 GB should be enough for my whole collection, even if all of it was ripped to FLAC.
Tuesday, May 05, 2009
BlackBerry Curve now sells more than iPhone 3G in USA
Supposedly due to the campaign where you can get two BlackBerry phones for the price of one. Sounds a bit desperate, but I guess Verizon recoups it somehow. Interesting all the same, but it’s important to note that it says nothing about the after market opportunities (read: additional revenue) in terms of music, videos and applications compared to iPhone. Most likely not at all on the same level.
Surprising note: “Sales figures do not include corporate/enterprise mobile phone sales.”
Meaning what? Doesn’t Verizon sell subs to companies?
Interesting is also the fact how relatively more mature the US market is in terms of information-centric phones (23% this year, 17% the last). I don’t believe Europe is close to this.
Note that the first ever Android phone is at position 5, and #3 and #4 are BlackBerry phones too. Not a single Nokia, Sony Ericsson, Samsung or LG in the top positions.
RIM Unseats Apple in The NPD Group's Latest Smartphone Ranking
Sunday, May 03, 2009
The state and future of mobile applications
Update 20090504: I guess I should have checked how Apple App Store handles payments. You register as a member and give out your credit card information once. After that, purchase is done simply by choosing an application. No more to fill in. Excellent, provided users get over the initial “bump” of at all giving out the credit card information. Seemingly they do.
Enrique Ortiz at About Mobility has written a sobering, yet also carefully optimistic, piece on mobile applications.
He goes through the situation at Android Markets, including that it contains a lot of crap applications that generate no revenue for anyone (which by the way very much also applies to Apple AppStore).
I don’t agree with assessment 5 though (paid applications came later than free ones, and it supposedly set the wrong expectations): The answer is rather: People simply usually don’t want to pay for mobile applications. Period! See my comment to MobHappy – Striking the balance in ad-supported Content for more on this.
Paying with a credit card (or even paypal) is neither a good solution for mobile applications, due to the hassle and the security concerns for such a small amount of money. Mobile applications should be charged to the phone bill, or be ad-supported free. Period!
Enrique provides good descriptions of the mobile application platforms, focusing on the essentials. I have one comment on J2ME (or Java ME): The security model is completely misdirected. Why on earth should I get a question whether I will allow the application to e.g. take a photo? In an average non-certified, non-game application you get several of those messages, that are of no use: Even if the application would be fraudulent (remember Red Browser?) the user would anyway answer yes to all the questions (a la “get on with it”). If certification should at all be needed I as a developer should be able to do it quickly, easily and for a low cost via the Web, and the application should then not generate any warnings what-so-ever.
No doubt browser or SMS based services have way less problems with device compatibility, and SMS is excellent for pushing content to a user, with the content completely in the message or accessed via a link. For interactive mobile marketing, SMS is still the best choice.
I continue to be a Java ME fan from a development perspective, partly of course because all phones (except iPhone and Android-based phones I guess) support it, and that you can do a hell of a lot more via a MIDlet than via a browser. Yet generating revenue through existing application stores has generally failed. GetJar still doesn’t have a scheme for commercial applications (even though they have started an in-app-ad trial), and purchase frequency at ClickApps is dismal at best. Kalador/Mobilerated might be more right on target, but is way less known than GetJar.
Apple showed the direction, but success for other application stores is not given, as Apple runs a tight network of interacting/complementing services and offerings. They are totally in control of their business (I’m not really sure why though), where not even operators dare to question what they do. Ask any other phone manufacturer, and they will tell you the operator is totally in control of them. Apple is also extremely business-oriented. They do creative things no doubt, but only if it long term generates more revenue. You can’t call Apple a media player or mobile phone manufacturer, because that’s missing the point of what they do. Apple is a business mesh in itself. I’m not saying this to praise Apple, just to indicate the difference between Apple and most other businesses.
Whatever the setup, and as Enrique also concludes, the key thing is of course to make money now. That’s why we are in this business.
Michael Yuan made a very important comment:
“What developers want is to address the “maximum number of people who are willing to pay”. Today, it means the US iPhone market.”
Whether it’s USA or e.g. China (where people generally don’t have PCs) is though open to discussion.
About Mobility - On Mobile Applications, Platforms and Monetization — “Show me the Money”
Corporate requirements on IT
A tidbit from the previously mentioned seminar on BlackBerry, here’s a list of requirements from companies regarding IT. I don’t know if this is RIM data or somebody else's. As always I have to comment on them too. The last one might be especially interesting.
Complete end-to-end security
Feeling safe that no operator or even IT department is eavesdropping on your communication is crucial, even if what you communicate is nothing to be ashamed about (I’m not talking piracy and … other things here, but confidential information). You should feel as safe when out and about as when at the office accessing services and information locally. Of course that also means if somebody steals your laptop or phone, they shouldn’t be able to get to your data or be able to pretend they are you towards the services.
Scale and reliability
This is a given: What you invest in now you should be able to stay with for years, even if the amount of users grows considerably. Also, you don’t want down-time nor cause a lot of labor cost and wasted time for your staff.
Low TCO
That’s a golden oldie, as everyone has said it, but I wonder how well it can be quantified. What about all the PDAs (with or without mobile radio) that were handed out a couple of years ago. Were they ever recouped? It’s anyway a given and needs to be considered whenever buying those tens of thousands of consulting hours for customizing a system that really should fit your needs from the get-go, at least through configuration.
Manageability and integration with critical systems
Being able to manage user devices and services centrally and ideally automatically is key to low overhead for IT. Also, e.g. mobile access to a corporate service needs to be well integrated, so that data is directly exchanged with the same database used internally. Otherwise you run the risk of data irrelevance, latency and/or even corruption.
Architecture for mobilizing all key systems
I guess this is a very interesting one for this blog’s audience (if there’s any). Mobilizing is not just about mobile phones of course, but generally being able to access the corporate services untethered from the office, be it via a laptop PC, or an information-centric or voice-centric mobile phone. Phones have advantages, as I’ve described a couple of times, like always-on, always-with-you, always-on-line, with integrated functionality like camera, GPS etc. Especially information-centric ones of course, but also standard voice-centric phones will do for many cases, and in the latter case you don’t have to make a truck upgrade of all users’ phones. Field staff often still collect data off-line and sync back at the office. Being on-line all the time means data is transferred when it’s needed, rather than too late.
