What are the real motivations of each carrier?
Docomo and E-mobile are both in favor of unlocking, though E-mobile would like to see a more fair allocation of frequencies, since they only have the 1700 MHz band, which no one else really uses. Docomo comes off looking like the good guy to customers, but it doesn't take a genius to see what they really want is the iPhone and iPad and could just as easily use their muscle against SIM unlocking if they already had what they want.
As they said in the press conference prior to the meeting, AU is against it because they are left out of the party with their incompatible CDMA-2000 network, which will most likely be used for voice even after the introduction of LTE. If this is the case, AU should focus on selling high-speed data plans for the slew of LTE data-only tablets sure to follow the iPad.
Softbank is naturally against it, offering up a number of arguments based on questionable logic, but then again it is easy to understand their real thinking, especially since Son already admitted that Softbank's network is inferior to Docomo's and an unlocked iPad would have left them at a severe disadvantage.
Will SIM unlocking happen?
No, it didn't.
I haven't had a chance to read the new guidelines released two days ago, but the government will only "urge" carriers to unlock phones sold from 2011. Docomo may do it, Softbank most certainly will not. So no iPhone or iPad on Docomo, at least for the time being.
Given the circumstances this is probably the correct decision, though not because I think the government was convinced by Softbank's arguments, which I would classify as FUD.
The government shares in the blame for softbank's comparatively poor network by denying softbank's request for an allocation of the 800 MHz band, which gives Docomo's network better penetration inside buildings, underground, and in mountainous areas. Softbank is trying to compensate by supplying free femtocells to customers. Government mandated unlocking would almost certainly result in a one-way flux of customers from Softbank to Docomo. That would be very unfair.
What about the incompatible platforms?
The argument that incompatible platforms are grounds for not mandating unlocking is erroneous.
The walled garden internet and carrier mail platforms (i-mode for Docomo, EZweb for AU, and Yahoo Keitai for Softbank) will simply not work if a "feature phone" is taken from one carrier to another. This will effectively prevent people not using smartphones from switching carriers.
But, as far as I know, the reason the MIC began reconsidering SIM locks was due largely to increased usage of smartphones, which are incompatible with the above platforms in the first place (the iPhone being somewhat of an exception). And, as shown by E-mobile, it is projected that smartphones will soon account for around 40% of all handsets globally.
What will be the effect of LTE?
Probably none at all, in the beggining at least.
Docomo is planning to start LTE service at the end of the year (See AU's chart). For the next couple of years, Docomo will be the only option for LTE (3.9G) handsets, so there is no reason for the MIC to reconsider unlocking. Of course LTE handsets will also be compatible with 3G, so it is conceivable that a "4G" phone could be released by AU or Softbank.
So if the "iPhone 5" is a "4G" phone, the only place for it will be Docomo? Yay!
According to AU's LTE chart, Softbank plans to begin operation of a DC-HSDPA network in January 2011 and begin offering service in July 2011. Who knows what real world speeds this network will get, but assuming the average iPhone user gets 2 - 3 Mbps, it is likely that users could see a 10x increase in speed. That is a significant increase and would certainly seem like "4G".
And of course, Apple releases the new iterations of the iPhone, pretty much like clockwork, right around July of each year, 6/29/2007, 7/11/2008, 6/19/2009, and 6/24/2010, respectively for the four iPhone generations. If this pattern continues, the "iPhone 5" should be released right about the time softbank can begin boasting of a ten fold bump in network speed. Hmm....
Could this result in IMEI Registration with Docomo?
Perhaps, but not necessarily for overseas phones.
It is well known among the smartphone-toting foreign community that Docomo refuses to sell ala carte SIM cards. While an unlocked phone is fully capable of working on their network, it is prohibitively expensive to do so because the unlimited data APN is filtered by IMEI number, even for people who have properly subscribed to the "biz-houdai" plan. This means that if you didn't buy a phone from Docomo, you cannot get unlimited data. Rather you will pay the full packet rate, which would result in monthly bills easily in excess of ¥100,000. Check out this post for an epic screenshot showing a guy on softbank who used ¥13,805,678 (currently $156,217.01US) worth of data packets in one month prior to the flat rate being applied (Login my be required.) That amount is so huge, I initially misconverted it in my head to only $13,000!
Softbank's presentation included the threat of very heavy network usage from handsets that can do things like tethering and heavy streaming, such as skype over 3G. Now that I think about it, this is probably the reason Docomo doesn't register IMEIs.
Now that the MIC is only recommending that cellphones be unlocked, after being so strongly for unlocking, Docomo would appear very hypocritical by not accepting unlocked phones. On the other hand, if no other compatible carriers (i.e., softbank) unlock phones, then there will be no unlocked phones to accept. Fortunately, according to Docomo's comparison slide, Docomo also uses the 1700 MHz band from Osaka to Tokyo, covering a large chunk of the population. So technically, E-mobile is also somewhat compatible with Docomo.
But overseas phones could still be excluded.
The preliminary unlocking guidelines made no mention of overseas handsets, and I assume neither do the final guidelines (will have to confirm later). In fact, the guidelines require carriers who participate in unlocking to begin sharing IMEI numbers to prevent the use of stolen phones, but this database would contain no information pertaining to overseas phones. The easiest and logical way of setting up a database would be to share blacklisted IMEIs and deny service to those devices. But what if the carriers decided to share all IMEIs and only approve service to those on the list? This would not be an efficient way of doing things but it would have two benefits (from the carrier's point of view):
- Remove the risk of/responsibility for providing service to someone in possession property stolen from overseas.
- Soften the anticipated decline in handset sales that would accompany unlocking.
Anyone living in Japan long enough has encountered the bazooka/fly swatter scenario. Though a fly swatter is perfectly capable of fixing the problem of an unwanted, winged insect, a bazooka is brandished, effectively killing the fly while taking out the entire wall and anyone unfortunate enough to be standing behind it. A good example of this is, as a means for combating spam, the default bouncing of any email not sent from a Japanese cellphone, which assumes all mail from PCs to be spam. Similarly, any phone purchased overseas could be stolen.
If the carriers perceive there to be even the slightest risk of one phone stolen from overseas being used on their network, they could just outright prevent all overseas phones from being used. A good "tatemae" for the "honne" of selling handsets, indeed.
Is SIM unlocking really going to increase phone theft?
Of course not, but that doesn't matter. What matters is whether risk is perceived.
For people involved in the fencing of stolen handsets, with the technical means of spoofing the IMEI number to circumvent blacklisting, a simple SIM lock isn't going to be anything more than a mild annoyance at most.
In this respect, locking down of hardware is similar to DRM in digital content, which serves only to annoy and alienate loyal law abiding customers while doing absolutely nothing to prevent the spread of pirated video.
(Side note: For those of you who's first experience with linux was Android, watching a DVD on a linux computer is actually in violation of the terms under which one bought - I mean "licensed" the right to posses - said DVD because the CSS decoder must be licensed for a fee by the maker or the hard/software that decodes and plays the video, which is something that obviously no one involved in linux development has done. Many popular Linux distributions do not contain the required library pre-installed, but it is typically in the corresponding repository and easily downloaded. This is also the case when watching a DVD with the very popular VLC.)