Home » 2010 » July

Monthly Archives: July 2010

The Mobile Broadband reality by Akamai

July 29, 2010 9:02 pm / 1 Comment on The Mobile Broadband reality by Akamai

Akamai, the leading CDN, has just released Q1 2010 report on the state of the Internet [July 23] which contains quite interesting actual information on average and average maximum connection speeds from 109 mobile carriers around the world. Only networks where Akamai believes that the entire system is mobile were included.

Note: Akamai defines these speeds as follows:

The “average maximum connection speed” metric represents an average of the maximum measured connection speeds across all of the unique IP addresses seen by Akamai from a particular geography. The average is used in order to mitigate the impact of unrepresentative maximum measured connection speeds. In contrast to the average measured connection speed, the average maximum connection speed metric is more representative of what many end-user Internet connections are capable of. (This includes the application of so-called speed boosting technologies that may be implemented within the network by providers, in order to deliver faster download speeds for some larger files.)

Akamai considers a provider having broadband level service when the average connection speed is more than 2 Mbps. From this point of view it is quite interesting that the countries mentioned as being in the forefront of 3.75G (HSPA+) and 3.9G (LTE) implementation (see: “4G” WiMAX vs. 3.75G HSPA+ [July 24] and WiMAX/WiBro <=> TD-LTE and LTE in general [June 28]) had in Q1 much less than that “broadband level” average connection speeds. Moreover, out of 7 results 6 are somewhat lower than 1 Mbps and only one is somewhat higher than 1 Mbps:

Country Provider Average (Kbps) Maximum Average (Kbps)
Japan JP-1 946 4180
Norway ND-1 867 3121
ND-2 867 3121
ND-2 1186 3875
United States US-1 846 1912
US-2 829 2103
US-3 979 2496

Please note that Akamai is not providing the names of providers. Considering their “mobile only” rule for inclusion it is obvious that Japan’s one entry could very likely be NTT DOCOMO (because it is the largest), as well as that one Norway’s two entries could probably be Telenor. At the same time it is quite obvious that TeliaSonera is definitely missing here since there is no entry for Sweden (and also because TeliaSonera is not a “mobile only” company). On the other hand it is quite obvious that Verizon Wireless and T-Mobile US  could well be one of the three US “mobile only” operators included into the Akamai report. (Note: Akamai has included only 3 operators when in a country were more than that.)

It is also interesting to compare these countries in the forefront of 3.75G-3.9G to less well developed countries. I’ve chosen for that the so called Eastern European countries and added to them Austria being in quite close geographic proximity. In addition marked those which were above the numbers of “forefront countries’” with red-ink. As one could see there are quite a few cases when the numbers are higher.

Country Provider Average (Kbps) Maximum Average (Kbps)
Austria AT-1 2553 10769
AT-2 1886 6292
Croatia HR-1 931 3567
Czech Republic CZ-1 626 2588
CZ-2 415 2024
CZ-3 1320 3561
Estonia EE-1 611 2775
Hungary HU-1 1145 5315
HU-2 1280 5037
Lithuania LT-1 1203 5516
LT-2 760 3205
Moldova MD-1 730 2858
MD-2 1269 4907
Poland PL-1 3444 10298
PL-2 750 2947
PL-3 508 2637
Romania RO-1 375 1899
Russia RU-1 4248 13686
RU-2 586 1933
RU-3 498 1570
Slovakia SK-1 105 418
SK-2 2225 6112
SK-3 7175 20394
Slovenia Sl-1 1074 5514
Ukraine UA-1 175 569

Akamai has noted this as well when doing worldwide evaluation of the results when one Slovak provider (SK-3) came atop (they have excluded the UK-3 provider having the highest result “due to their suspected usage of a mobile gateway architecture, which inflated their calculated per IP address usage”). Akamai has made the following remark related to that:

However, it must be noted that a number of mobile network providers make heavy use of mobile gateways and proxies that will result in higher average and average maximum speeds being calculated by Akamai, as these speeds reflect gateway/proxy-to-Akamai communications rather than mobile device-to-Akamai communications. (These top providers may be making use of such an architecture.) Akamai is investigating methods of mitigating the impact of these gateways/proxies on the source data sets that will be used for future editions of the State of the Internet report.

Let’s see the what kind of networks those Eastern European providers have:

• Austria: 1 with HSPA+ @ 42 Mbps, 1 with HSPA+ @ 21Mbps, 2 with HSPA @ 7.2 Mbps (all with HSUPA)
• Croatia: 1 with HSPA+ @ 21Mbps, 1 with HSPA @ 7.2 Mbps, 1 with HSDPA @ 1.8 Mbps (the first two with HSUPA)
• Czech Republic: 1 with HSPA @ 14.4 Mbps, 2 with HSPA @ 7.2 Mbps (two with HSUPA)
• Estonia: 2 with HSPA+ @ 21Mbps, 2 with 1 with HSPA @ 7.2 Mbps (the first two with HSUPA)
• Hungary: 1 with HSPA @ 14.4 Mbps, 1 with HSPA @ 7.2 Mbps (all three with HSUPA)
• Lithuania: 3 with HSPA @ 7.2 Mbps (tw0 with HSUPA)
• Moldova: 1 with HSPA+ @ 21Mbps, 1 with HSPA @ 14.4 Mbps, 1 with HSPA @ 7.2 Mbps (the first one with HSUPA)
• Poland: 3 with HSPA+ @ 21Mbps, 2 with HSPA @ 7.2 Mbps (four with HSUPA)
• Romania: 2 with HSPA+ @ 21Mbps, 2 with HSPA @ 7.2 Mbps (three with HSUPA)
• Russia: 1 with HDSPA @ 7.2 Mbps,  1 with HDSPA @ 3.6 Mbps, 1 with HDSPA @ 1.8 Mbps (none with HSUPA)
• Slovakia: 2 with HDSPA @ 1.8 Mbps (one with HSUPA)
• Slovenia: 1 with HSPA @ 7.2 Mbps, 1 with HDSPA @ 3.6 Mbps, 1 with HDSPA @ 1.8 Mbps (the first one with HSUPA)
• Ukraine: 1 with HDSPA @ 3.6 Mbps
Source: HSPA Operator Commitments survey by GSA [June 30] (registration required)

Please note that only those marked with red-ink have mobile technologies which might explain their (sometimes much) better results.

Final conclusion? Only one I could draw from this: the current networks of “forefront countries” should indeed be more congested than some of the mobile networks of even Eastern European countries. This means that they should indeed be in the forefront of 3.75G and 3.9G adoption!

SAP’s Business ByDesign SaaS to be relaunched on July 31 with mobility as one of key attractions

July 28, 2010 12:46 am / 2 Comments on SAP’s Business ByDesign SaaS to be relaunched on July 31 with mobility as one of key attractions

Yesterday’s Q2 financial report from SAP specifically mentions the expansion it made in the volume business as well as the progress in the small and midsized enterprise (SME) segment:

… said Jim Hagemann Snabe, Co-CEO of SAP. “Our success in the SME segment creates a strong foundation for the new version of our on-demand platform SAP Business ByDesign. The new version will be available on time on July 31st and is ready for volume deployment in six countries.”

The company provided a quite interesting video as well: SAP Business ByDesign Goes Mobile [July 26]. “Get an exclusive first look at the latest SAP Business ByDesign software, running on mobile devices such as iPad, iPhone, and BlackBerry. The software will be available in 2011.” There is also a related blog-entry from SAP TV: Exclusive: See what’s coming 2011 – Behind the scenes with a SAP developer [July 27]

In a later video interview given to CNBC (see SAP CEO on Earnings [July 27]) Bill McDermott, Co-CEO of SAP, was describing his company’s outlook as follows:

[2:30] Others have consolidated legacy markets to downsize companies and leverage margin. In the short-term that’s a viable shareholder strategy, no question about it.  We believe in organic growth, we believe that customers will run their business with on-premise enterprise applications, they’ll run them on-demand in the cloud, using in-memory technology with real-time analytics, and they’ll run their business on-device.

[3:00] The vision that we have is to connect the virtual boardroom to the shopfloor worker, so on an end-to-end process basis companies are run better, and they are actually managed by exception. So all these knowledge workers that are interned to the desktop can actually make decisions in real-time. So Sybase got us into a new category which is the mobile. I believe to be the new desktop. If you look at markets like China, as an example, three quarters of their knowledge workers have bypassed the desktop and gone right to the mobile.

[3:31] So we are all about innovation. We will let others have their short-term games on the margin. We think customers will appreciate that in the long run, and shareholders too. [3:41]

Mobility has indeed been the major reason why SAP has recently acquired Sybase for approximately $5.8 billion. Other reasons were being very strong in analytics and certainly its traditional database business. See: SAP to acquire Sybase Inc. [May 12]

Strategic Move to Accelerate the Reach of SAP® Solutions across Mobile Platforms, Help Companies Manage and Analyze Business Information and Processes on Any Device

… SAP will accelerate the reach of its solutions across mobile platforms and drive forward the realization of its in-memory computing vision. … For Sybase, SAP in-memory technology will provide the opportunity for dramatic performance improvements to its analytic processing capabilities. Sybase will also be able to bring its complex event processing and analytics expertise, which was built in the financial sector, to customers in other industries, markets and product areas …

“Mobile devices are becoming the preferred interaction point with business applications, whether the user is a factory supervisor, a retail manager or an entrepreneur in a developing nation,” said Jim Hagemann Snabe, co-CEO of SAP and member of the SAP Executive Board. “The combination of SAP and Sybase will give users the option of running their operations from leading mobile devices … In addition, innovation around Sybase’s established database business will pave the way for ‘real’ real-time analytics and finally remove the decade-old barrier between business applications and business intelligence.”

… John Chen, CEO of Sybase, Inc. “… by combining the market leader in enterprise applications with the market leader in enterprise mobility, companies around the world will be able to run their business from many devices. This will drive a new wave of enterprise productivity. … ”

As a new, to be operating as a separate subsidiary of SAP, Sybase has announced a much strengthened mobility offering almost immediately. See: Sybase Introduces Industry’s First Mobility Platform Empowering Enterprises to Better Enable Employees and Engage Customers in New and Innovative Mobile Ways [May 20]

The Sybase Mobility Platform is comprised of the most comprehensive new and industry-proven solutions, including:

Sybase® Mobile Sales for SAP CRM and Sybase® Mobile Workflow for SAP Business Suite have also earned Sybase an SAP award in the “Technology Innovator of the Year” category. See: Sybase Awarded 2010 SAP® Pinnacle Award [May 20]. These applications, by the way, have been jointly announced well before the acquisition. See: SAP and Sybase Deliver First Set of Applications to Millions of Mobile Workers Worldwide [March 2]

… The population of mobile workers and the number of personally owned devices entering the enterprise is growing exponentially. According to industry analyst group IDC, “The worldwide mobile worker population is set to increase from 919.4 million in 2008, accounting for 29% of the worldwide workforce, to 1.19 billion in 2013, accounting for 34.9% of the workforce.” …

  • Sybase® Mobile Sales for SAP CRM automates sales processes, increases productivity and enhances customer service by equipping sales professionals with anywhere, anytime access to SAP CRM 2007 through smartphones such as iPhones and Windows Mobile devices.
  • Sybase® Mobile Workflow for SAP Business Suite enables mobile workers to complete business processes — such as workflow items and alerts, time recording and travel requests that require immediate action — through a familiar and secure email inbox.

CNBC’s earlier Why Sybase? [May 17] interview with Bill McDermott has given more insight on whether the mobilization of  business is to propel [this] M&A:

[1:31] The main reason for acquiring Sybase was one, mobility. They are the #1 mobile infrastructure company in the world. We want to move our applications to mobile devices. #2 – they are very strong in analytics. So we were already #1 in the world in analytics, and they approach industries such as financial services and public services. And they are growing very fast, and in quick markets like China, where China has bypassed the PC generation and gone right to the mobile. The 3d reason is the database where they have a very strong core business. We believe that in-memory computing combined with their core business will give the customer just an at-a-site-solution that currently does not exist on the market from any vendor. So for all three reasons we felt this was a breakthrough to have that asset. [2:18]

[2:32] We believe very heavily that our strategy was reliant on mobility. … We participate now in a 10 billion market. We wanted to be in a 220 billion market, and to do that you had to get in-memory, you had to get mobile, and you had to get on-demand. Sybase itself touches all three priorities. [2:53]

Note: One can put here indeed press releases that prove Sybase strengths in the #1 and #2 areas. For simplicity I am just giving here the ones that are also related to the China market:

Back to the Business ByDesign SaaS! The new customer-centric innovations in the Business ByDesign solution have already been announced and demonstrated at SAPPHIRE® NOW event this year. See: New Release of SAP® Business ByDesign™ Gives Customers In-Memory Analytics, Support for Mobile Devices, and Customizable User Interfaces [May 17]. Of the new attractions mentioned in this headline one needs to mention the incorporation of Microsoft Silverlight technology into the front end. The rest is quite well understandable from the things detailed above.

One innovation mentioned just in the internal text of the press release, however, needs to be mentioned as a key attraction as well:

Single and multi-tenancy support. Customers can now choose either single or multi-tenant topologies as the solution is fully enabled to support both. Other than traditional on-demand offerings, SAP’s multi tenancy concept has been engineered for customer and partner specific business extensions, business configuration and all kind of UI, report and forms flexibility. It leverages SAP’s proven life cycle management system protecting all changes during upgrades. During the planning and engineering phases towards multi-tenancy, SAP collaborated with Intel to create an infrastructure optimized for the massively scalable ‘cloud computing’ environment of SAP Business ByDesign. The Intel® Xeon® Processor is the reference platform for SAP Business ByDesign deployments. The scalability and efficiency of Intel Xeon processors with the Next-Generation Intel® Microarchitecture will enable SAP to achieve even lower TCO for the solution’s infrastructure by delivering equal business capability with a smaller server and carbon footprint, significantly lowering energy consumption compared to previous generation processors.

The reason for full inclusion of that part is quite simple. That single functionality has been missing from this SaaS product and that was one of the major reason why SAP has not been able to enter the SaaS market in a big way so far. Another one has been the clash of its old and new (i.e. on-demand) business models. Here is some earlier information on that:

“What we did discover in the last five or six months is that while we made progress on our TCO model we’re not where we wanted to be (a 10 times TCO reduction),” says Apotheker, in a meeting with a group of Enterprise Irregulars Monday at SAP’s Sapphire conference in Orlando.

… The rumor mill has been most unkind of late: Rumors have been flying that ByD is going to miss its latest release date, that its marketing plans are DOA, and that SAP is on the verge of laying an egg of colossal size and impact.

The rancor surrounding the July 31 release date of the new ByD reflects a complex history inside SAP and the on-demand market that has clouded a lot of otherwise objective analysis about its prospects, and how important ByD’s success is not just for SAP and its customers, but for the on-demand market as a whole. …

… The other point worth addressing about ByD is that its initial entrance to the market in 2007 came in the midst of a not-so-private battle between ByD’s main advocate, Peter Zencke, and fellow board member Shai Agassi, who wanted the concepts of ByD – model-driven, on-demand ERP – to be embodied in the main SAP suite instead of in a completely new, mid-market product. That battle at the top eventually led to Shai’s departure – Zencke left shortly thereafter – but it left a bad taste in many SAPers’ mouths for the product. The fact that it failed to be economically viable as an on-demand product in its initial release further fueled the ire of its detractors. …

… This internal “polemic”, to be polite about it, reminds me of what Microsoft has been going through in the years since Ray Ozzie first discussed Software + Services as the new direction for Microsoft. Even an outsider can imagine the screaming inside Microsoft about such a radical shift and what it would mean to the king of the desktops. The fact that Steve Ballmer felt compelled three years later to give his now-famous “We’re All In” speech is testimony to how hard it’s been internally for Microsoft to grapple with this paradigm shift. …

… SAP is facing a similar dilemma with ByD. Everything about it – on-demand but not true multi-tenant, requires a major channel that doesn’t exist yet, the potential for cannibalization of other products, the sordid release history, the bad blood inside SAP – means that SAP has a similar requirement to make sure that everyone is walking the walk and talking the talk.

With so much at stake, it’s clear that ByD will be one of the great inflection points in the history of enterprise software, and in many ways its success or failure will define the future of on-demand enterprise software for some time. ByD isn’t just a gamble that SAP can create a product like ByD, it’s also a gamble that there is actually a market out there for what ByD represents. This fact is hardly a given: NetSuite’s struggles to meet its market potential has given many proponents of on-demand ERP pause, as has Workday’s slow slow progress towards a critical mass of customers. …

… If SAP can pull off ByD, that success will create a huge opportunity for others, including NetSuite, by legitimizing a market that to date doesn’t really exist as a billion-plus dollar opportunity. If SAP fails, I fear that this much-needed capability will languish on the sidelines another four or five years, and customers will be all the poorer, literally and figuratively, for that failure.

So, baring some unforeseen glitch, ByD re-enters the market in one short week. Its success or failure will hinge on a number of factors, most important of which will be the SDK that comes out at the end of the year. This is going to be one of the most closely watched software rollouts in a long time, with its repercussions felt across the industry for years to come.

So  the relaunch of SAP’s Business ByDesign product should be watched quite carefully on July 31!

“4G” WiMAX vs. 3.75G HSPA+

July 24, 2010 2:09 pm / 4 Comments on “4G” WiMAX vs. 3.75G HSPA+

See also: Mobile Internet (Aug’11) which is a total update on Aug 26, 2011 with a lot of additions to the original July 19, 2010 content on the following subjects:
– LTE and LTE Advanced — HSPA Evolved (parallel to LTE and LTE Advanced) — Heterogeneous networks or HetNets — Femtocells and Picocells — Qualcomm innovations in all that — Ericsson’s LTE Advanced demo — Current roadmaps on evolutions of current 3G+ broadband mobile networks

T-Mobile has just launched its 3.75G HSPA+ service in the US, which they wrongly call 4G (compare to the core information on Mobile Internet where standards are covered properly). See: T-Mobile USA: More HSPA+ [July 21]. Speedwise -however- they are on par or even better than the current “4G” service from Sprint.

PhoneScoop has published the results of their comprehensive testing 6 weeks ago. T-Mobile USA HSPA+ service has been found better on a download, upload and latency basis (especially on the last two).

… Sprint’s WiMAX network and T-Mobile’s HSPA+ network delivered roughly similar download speeds, just shy of 3 Mbps on average. These are real-world, average speeds, not ideal numbers skewed by a marketing department. 3 Mbps is easily twice as fast as your typical real-world speed with 3G, and faster even than many home DSL connections. …

See: 4G Networks Tested: WiMAX vs. HSPA+ [June 4]: Intro • The Numbers • Equipment • Summary

What was a quite interesting finding -however- that T-Mobile’s earlier HSPA 7.2 Mbps network service wasn’t much slower than HSPA+ or 21 Mbps.

One curious data point was comparing T-Mobile’s HSPA+ with their own HSPA 7.2 (slightly older 3G [correctly: 3.5G]) technology. If you just look at the theoretical peak numbers – 7.2 vs. 21 Mbps – you might think HSPA+ is almost three times faster than HSDPA 7.2. You’d be wrong. We compared them extensively, in all six locations. In our tests, the difference was small. At best, our webConnect Rocket USB stick with HSPA+ was only 15% faster than a standard webConnect USB stick with HSPA 7.2.

See also on my blog:

Could China close the gap in mobile Internet? It should!

July 21, 2010 7:07 pm / 5 Comments on Could China close the gap in mobile Internet? It should!

China has a significant gap in mobile Internet with the rest of the world:

– the current [end of H1CY10] share of 3G technologies in China are:
• 1.9% (10.46 million) for TD-SCDMA/China Mobile (w/ 554M subscribers),
• 4.8% (7.56 million) for W-CDMA/China Unicom (w/157M subscribers) and
• 9.6% (7.18 million) for CDMA2000/China Telecom (w/ 74.5M subscribers)
• which is 3.2% (25.2 million) overall 3G share of mobile connections (785.5 million) for China as a whole

– meanwhile W-CDMA (as the dominant 3G technology in the world) has:
• 12% (600 million) of the mobile connections worldwide (5 billion)
• 13.3% (600 million) of the joint GSM+W-CDMA connections worldwide (4.5 billion)

– and even for the more advanced mobile broadband Internet the current estimated share of:
• joint HSPA and EV-DO families is 6.8% (340 million) of the mobile connections worldwide (5 billion)
• HSPA family is 4.2% (~210 million) of the mobile connections worldwide (5 billion)
• HSPA family is 4.7% (~210 million) of the joint GSM+W-CDMA connections worldwide (4.5 billion)
• EV-DO family is 26% (~130 million) of the whole CDMA connections worldwide (500 million)

Closing that gap would be especially difficult because by far the biggest gap –with current 3G share of only 1.9%– is for China Mobile owning 70.5% of mobile connections and providing the China only TD-SCDMA technology as the path to 3G and beyond.

… number of the mainland`s TD-SCDMA subscribers has grown at exponential rate into the second half of this year, especially in the latest two months in which a net one million additional subscribers were signed up each month.

… In light of such rapid growth, providers of wireless solutions are competing to roll out TD-SCDMA solutions. Among them are Leadcore Technology Co., Ltd., a firmware developer held by telecom-equipment supplier Datang Telecom Technology Co. of the mainland in addition to MediaTek, Mstar and Qualcomm Inc.

Leadcore has recently introduced a less expensive TD-SCDMA solution, which is quoted at only US$9. Yulong Computer Telecomm Scientific (Shenzhen) Co., Ltd., a leading supplier of smartphones in China, has reportedly adopted the solution.

MediaTek is developing several solutions including the one called TD Protocol Stack. Industry executives believes the company will make some available by the end of this year.

Spreadtrum Telecommunications Inc., China`s homegrown chip designer, is now the major supplier of TD-SCDMA chips at home.

Pushed by China Mobile, TD-SCDMA handsets and smartphones at retail prices of about 500 yuan (US$75) and 1,000 yuan respectively will be available in the China market in the fourth quarter of 2010.

The TD-SCDMA industry has been developing for 4-5 years, resulting in a supply chain of more than 100 players, according to secretary-general Yang Hua for the TD-SCDMA Industry Alliance. There are 7-8 developers of TD-SCDMA chips currently, with chip prices falling 60-70% over a year to US$8-10 at present, Yang indicated.

Five international and many China-based vendors had launched 461 TD-SCDMA handsets as of the end of August 2010, with Samsung Electronics the largest international vendor, Yang pointed out. However, a single TD-SCDMA model can reach sales of 200,000-300,000 handsets only, a volume too small to be able to push down component costs, according to Taiwan-based handset makers. China Mobile is setting up its fourth-phase TD-SCDMA network of more than 100,000 base stations and expects the number of TD-SCDMA subscribers to increase from 13.42 million currently to 100 million in 2012.

According to iSuppli Ltd analysis, due to wireless carriers providing heavily subsides, the mobile phones consumer prices fell down. It is expected that China’s domestic 3G mobile phone shipments will grow nearly five times, and the volume is expected to reach 42.97 million in 2010.

… TD-SCDMA mobile phone domestic shipments in 2010 will increase to 20.4 million, from 1.3 million in 2009.

… The iSuppli Corporation predicts that by the end of 2014, the total number of wireless subscribers in China will rise to 1.1 billion, 3G subscribers will increase to 230 million. …

It is worth to copy here the latest chart of 3G growth in China, as well as a table and another chart from the above places, to see that the rate of growth shown so far is not enough:

The Q2CY10 growth rate for China Mobile was 36% which is giving a seemingly impressive 242.4% yearly growth if continued that way. With that China Mobile will have just 35.8 million 3G subscribers by the end of next June which would be still between 5% and 6% of the overall number of their subscribers. Meanwhile the number of W-CDMA Family connections worldwide would probably grow to around 16% (to ~960M, from current 12% i.e. 600M), and even mobile broadband connections worldwide for the HSPA family alone would grow to around 5.7% (to ~340M from current 4.2% i.e. ~210M) of all mobile connections globally (by the end of next June).

I’ve come to these results from the following table and the source data behind that another chart (copied here from Mobile Internet [July 19]:

3.9G TD-LTE rollout in 2012 with integrated 2G, 3G and 4G? [July 19] is already a sign of taking additional extraordinary measures in this regard. For the Chinese it is not a question whether they could close the gap in mobile Internet. They are understanding pretty well that they should close that gap as soon as possible! Whatever it takes.

3.9G TD-LTE rollout in 2012 with integrated 2G, 3G and 4G?

July 19, 2010 9:08 am / 8 Comments on 3.9G TD-LTE rollout in 2012 with integrated 2G, 3G and 4G?

See also: Mobile Internet (Aug’11) which is a total update on Aug 26, 2011 with a lot of additions to the original July 19, 2010 content on the following subjects:
– LTE and LTE Advanced — HSPA Evolved (parallel to LTE and LTE Advanced) — Heterogeneous networks or HetNets — Femtocells and Picocells — Qualcomm innovations in all that — Ericsson’s LTE Advanced demo — Current roadmaps on evolutions of current 3G+ broadband mobile networks

A small news item on DigiTimes last week has gone around the technical web quite fast: China Mobile to expand 4G network trial operations [16 July]

China Mobile Communications will expand the trial operation of its 4G networks in the fourth quarter of 2010 and a successful run of the trials may push China Mobile to start rolling out its 4G infrastructure in 2012, 2-3 years ahead of its original schedule, according to industry sources.

China Mobile Communications has reached a consensus with Apple under which the next-generation of iPhones to be sold in China will adopt TD-LTE technology developed by China Mobile, the Chinese-language Commercial Times quoted the China-based carrier’s chairman Wang Jianzhou as saying.

China Mobile has begun voice testing on its TD-LTE experimental networks in Shanghai and commercial operations of the 4G networks in China are expected to begin in 2012, the paper said.

By 2012, China Mobile also plans to set up over 40 experimental networks, 10 commercial networks and 20,000 base stations globally to promote the adoption of the TD-LTE technology, the paper added.

More information has become available from TMC, last week as well: China Mobile, Huawei roll out China’s first 3G/4G-integrated trial network [July 16]

China Mobile (CHL.NYSE; 00941.HK) and Chinese mobile equipment and network solutions provider Huawei Technologies have finished the first TD-SCDMA/TD-LTE outfield test and completed a series of tests and verifications in Wuxi, Jiangsu province.

The tests show that the 3G/4G-integrated network and smooth evolution are completely feasible.

… the technology will be tested in three Chinese mainland cities in the final quarter of the year, with 100 base stations constructed and 5,000 people invited to use the service.

At the beginning of the trial operation, China Mobile will offer data cards to the users and provide them with smartphones.

Results drew from such tests may facilitate China Mobile to offer 4G services in 2012, earlier than previously expected 2014 to 2015.

An insider at the Ministry of Industry and Information Technology revealed that China Mobile will select three coastal cities to pilot the TD-LTE networks, and Qingdao, Xiamen, Zhuhai, and Wuxi are all candidates.

The insider believes that 2G, 3G and 4G services will all be available to Chinese telecom service subscribers at the same time in future [emphasis mine], rather than high-grade technology substituting the lower-grade options.

A UK report from a day earlier has indicated foreign operator and supplier interest as well: TD-LTE 4G Sweeping the Wireless Broadband Markets in China [July 15]

… a variety of wireless operator companies have actually started shifting to TD-LTE, inside and outside China. Two of the main users recorded are Qualcomm and Yota. … This new technology is expected to hit the markets within two years, after a series of field tests.

Ulf Ewaldsson, Ericsson VP, said, “The Ministry of Industry and Information Technology and China Mobile will start large-scale field tests in the fourth quarter of this year. Ericsson will also take part”.

Ericsson’s strong commitment and technical readiness has already been demonstrated a week ago indeed: First complete TD-LTE solution showcased [July 12]

At a China Mobile event in Shanghai, China today, Ericsson (NASDAQ: ERIC) showcased its complete end-to-end TD-LTE solution for the first time. Together with ST-Ericsson devices, the system successfully showcased super-fast mobile broadband applications such as video on demand (VOD) and video streaming using a live camera.

… ST-Ericsson, a leader in wireless platforms and semiconductors, was the first in the world to demonstrate a handheld LTE device and to achieve LTE and HSPA mobility with a multimode device [emphasis mine]. Available next year, ST-Ericsson’s next generation modem will support both versions of LTE, in addition to TD-SCDMA and HSPA+/EDGE [emphasis mine]. With ST-Ericsson’s products, Ericsson is now the only player with complete end-to-end TD-LTE capabilities in the industry.

NEW DELHI, INDIA: Ericsson in its claim has shortened the time period for availability of TDD (time division duplex)-LTE equipment from 2013 end to first quarter of 2011.

Per Thorzell, head, radio Access Network, Ericsson India added that Ericsson also did a TDD-LTE trial run in India using 20 Mhz spectrum and achieved a download speed of 71 mbps per second. … “There are approximately 95 per cent functional similarity between TDD and FDD-LTE. Therefore chipsets that support FDD-LTE will also support TDD-LTE,” … to support TDD-LTE, Ericsson feels there will be dongles as access devices in the first half of 2011 and later handheld devices.

Of the two companies mentioned in the above UK article Yota is a Russia-based mobile operator expanding to other countries worldwide, while Qualcomm is the most influential wireless telecommunications research and development company, as well as the largest fabless chip supplier in the world.

Qualcomm has a unique role in the possible 2-3 years acceleration of the TD-LTE technology. The Casino Royale: The story of Qualcomm [July 12] article from July 16 issue of Forbes India, republished on-line by moneycontrol.com, “India’s No. 1 financial portal” has been summarized on the cover as:

Qualcomm is one of the great technology companies of our times. Its products drive large parts of telecom networks everywhere. Everywhere, except India. Now after almost a decade of trying to open up the Indian market, Qualcomm’s time might finally be here. And it will all because of what Qualcomm learnt in China.

Going into the article one can see what made Rohin Dharmakumar, the author of the article to arrive to such a conclusion. Some excerpts from the point of view of our subject:

They [Qualcomm] had invented a proprietary technology that became the seed for the wireless standard CDMA … CDMA began to be adopted by the world’s mobile operators from the mid-90s after years of relentless hard-selling … , backed up by equally relentless patenting by … engineers in the R&D department. … Today, the company … sits on a pile of over USD 18 billion in cash while generating another USD 2-3 billion in free cash flows every year!

Qualcomm’s model as it exists is simple. Be inventive, file as many patents as possible in wireless communication, and then build products around those or wait for people to queue up to use those patents and pay a nice royalty to the company.

… Qualcomm’s patents are critical to almost all major wireless standards on the GSM, CDMA and even LTE (Long Term Evolution) sides, helping it squeeze a royalty out of quite literally anyone and everyone. An army of its lawyers and patent experts go after any company that dares touch its wireless IP without paying for it.

… The first star that heralded its new fortune was the successful completion of India’s 3G auctions. Because Qualcomm’s IP is embedded in all 3G technologies, it will make a 4-5% cut off every 3G phone that will now be sold in India.

… The second omen was on March 11, at a press conference in New Delhi’s Oberoi hotel, when Kanwalinder Singh, its India head said, “Today I’m very proud to announce that Qualcomm has invested one billion US dollars into India.”

Singh did not spend that money in building offices, hiring thousands of new employees or buying local businesses. He spent it to buy 20 megahertz (MHz) of spectrum in the cities of Mumbai and New Delhi and the states of Haryana and Kerala. Qualcomm was one of the six companies that won precious spectrum to offer high speed “4G” wireless broadband to Indians. The $1 billion spent in India was its single largest bet anywhere around the world.

The funny thing though is that Qualcomm is betting on deploying a technology that it didn’t think much of to begin with, a commercially unproven technology called TD-LTE originally developed by the Chinese.

Learning from the Chinese

The Chinese government, in the mid-90s, decided to develop a technology to avoid paying royalties to Western companies like Qualcomm. They created TD-SCDMA, a 3G wireless standard. But it never really took off*. The Chinese, in order to avoid paying royalties, had to reinvent the wheel. This made their technology cumbersome and inefficient.

* Personal note: TD-SCDMA in fact took off in 2009, and that was just too late only for international acceptance (being a 3G technology). See the 3G China customer base chart in my earlier OPhone OS (OMS) 2.0 based on Android 2.1 [July 5] article.

The Chinese pumped more resources into it and refined it into TD-LTE. They also roped in two large Western partners, Verizon and Vodafone, to conduct joint trials of LTE combing the Western (FD) flavour with its own (TD) one. Julian Grivolas, an analyst with Ovum in Paris who tracks the LTE space closely, says China’s goal was to make sure that this joint LTE becomes the next equivalent of the mobile world’s ubiquitous GSM standard.

But the company that took advantage of this trend the fastest was the one which was the last to join the LTE party — Qualcomm. For that it had to swallow its pride and abandon Ultra Mobile Broadband (UMB), a 4G technology it had been developing for many years. In late 2008, it abandoned UMB to focus exclusively on LTE when it saw big operators tilting towards it. A few years earlier it had already spent over USD 800 million to acquire Flarion, a company with a significant amount of patents in areas related to LTE.

Qualcomm smartly co-opted its erstwhile foe China into becoming an ally by announcing plans to build new “multi-mode” mobile chipset** that would combine the version of LTE that Qualcomm and a galaxy of mobile companies like Ericsson and Nokia had developed along with what the Chinese had developed [emphasis mine].

** Personal note: see Qualcomm Introduces World’s First Complete Multi-mode 3G/LTE Integrated Solution for Smartphones [Feb 16, 2009], Qualcomm Now Sampling Industry’s First Dual-carrier HSPA+ and Multi-Mode 3G/LTE Chipsets for Global Markets [Nov 12, 2009] and Qualcomm to launch TD-SCDMA and TD-LTE next year [Nov 19, 2009]. Update: Qualcomm Now Demonstrating Products Based on LTE TDD Technology — Company on Track with Commercialization of Products Based on LTE TDD Technology with First Devices Expected in Mid-2011 [Sept 8, 2010]. At the same time we had news that Qualcomm now has 4G licensing deals with the big three [Nov 16, 2009]: Qualcomm told its investor meeting that LG has joined Samsung and Nokia in agreeing to pay to use its technology in future 4G products. “This means they have 65% of the handset market signed up to long term royalty agreements [emphasis mine],” JP Morgan analyst Steven O’Brien told Reuters. “We have more certainty in the royalty streams and cashflow streams from the major handset vendors.”

Given that a significant chunk of TD-LTE patents rest with Chinese companies like China Mobile, Huawei and ZTE, the only way Qualcomm would have integrated them into its own offerings would be by cross-licensing*** — basically sharing patents with them. The Chinese are returning the love because a global acceptance of TD-LTE will mean cheaper equipment prices for its mobile operators and larger markets for its equipment vendors.

*** Personal note: There is no information regarding these cross licensing agreements, although they definitely exist. We only know that the Chinese were quite successful in licensing negotiations even before. See CDMA success brings Chinese royalty gripes [Feb 12, 2009]: “… in 2000 Irwin Jacobs [the father of current CEO] negotiated government permission for CDMA to be used in China, Qualcomm granted Chinese manufacturers the best rates in the world — around 2.65% as opposed to 5% for other makers. (Qualcomm neither discloses nor confirms its royalty rates).”

Such an advantage even prompted the following: “In fact, outside of China, TD-SCDMA is generally dismissed as a negotiating ploy by China to obtain acceptable patent cross licensing agreements with Qualcomm, Ericsson and the like.” – see An update on TD-SCDMA, China’s 3G technology [Oct 18, 2007]

With TD-LTE under his belt, [Paul] Jacobs [their CEO] knew he had WiMAX cornered.

What happened to WiMAX in the last month I’ve already described in:
• WiMAX/WiBro <=> TD-LTE and LTE in general [June 28]
• Intel dismisses WiMAX Program Office [July 1]

In addition to that it is also worth to note that LTE is Changing the Landscape of IPR Wealth [May 18, 2010]:

… the 2G, 3G, and 3.5G IPR [Intellectual Property Rights] landscape was dominated by Qualcomm, Nokia, and Ericsson … “While Interdigital and Qualcomm are clear leaders in the global LTE patents portfolio with 21% and 19% market shares respectively of the total number of patents, Huawei comes in third position with 9%, Samsung in fourth with 8%, and Nokia, LG, and Ericsson in joint fifth place, each with 7% market share,” [said Malik] Kamal Saadi [, principal analyst at Informa Telecoms & Media].

… Until now only a third of current global LTE patents could be described as essential but about 60% of them are recognised as having the potential to become essential in the future. Whether or not a patent is viewed as being essential could also vary from one market to another. For example, from the 182 LTE patents contributed by Huawei, 178 are registered in China and only a handful of these could currently be described as essential. …

… Informa Telecoms and Media believes that more than 60% of LTE patents from likes of Qualcomm and Nokia, 50% of LG’s portfolio, 40% of Samsung’s patents, and less than 33% of Ericsson’s portfolio could be described as essential LTE patents so far.

The second omen was on March 11, at a press conference in New Delhi’s Oberoi hotel, when Kanwalinder Singh, its India head said, “Today I’m very proud to announce that Qualcomm has invested one billion US dollars into India.”

Singh did not spend that money in building offices, hiring thousands of new employees or buying local businesses. He spent it to buy 20 megahertz (MHz) of spectrum in the cities of Mumbai and New Delhi and the states of Haryana and Kerala. Qualcomm was one of the six companies that won precious spectrum to offer high speed “4G” wireless broadband to Indians. The $1 billion spent in India was its single largest bet anywhere around the world.

The funny thing though is that Qualcomm is betting on deploying a technology that it didn’t think much of to begin with, a commercially unproven technology called TD-LTE originally developed by the Chinese.

Windows slates in the coming months? Not much seen yet

July 13, 2010 2:06 am / 8 Comments on Windows slates in the coming months? Not much seen yet

On the first day of the Microsoft Worldwide Partner Conference this week there was not much about the Windows slates. It is not surprising given that Intel has just begun real and honest marketing of its sufficiently low power SoCs (see: Intel SoC for Cloud Clients [June 25]).

… what you’ll see over the course of the next year is us doing more and more work with our hardware partners creating hardware-software optimisations with Windows 7 and with Windows 7 Media Center …

  • Even more, Mr. Ballmer is emphasizing that whatever they will be doing on Windows 7 and Windows 7 Media Center basis will still not be the real thing, and everybody should wait for Windows 8 with real “Big Button touch” — i.e. (I would add) waiting till 2012:

… Media Center is big and, when people say ‘hey, we could optimise [that] more for clients’ I think what they generally mean is ‘Big Buttons’.  Big Buttons that’s, I think, a codeword for Big Buttons and Media Center is Big Buttons not Little Buttons. I’m not trying to trivialise that – it’s a real issue.

We’re not going to do a revamp of Windows 7 over the course of the next year for that purpose.  Whether we should, or we shouldn’t, we’ve put all our energy around doing a great job on that and other issues in the next version of Windows …

  • Related post on my other blog: Windows 7 UI overlays from Microsoft and elsewhere [Aug 30]. This is clarifying the UI situation for Apple iPad competitive Windows slates.
  • Related: Compal president pessimistic about non-Apple tablet PC shipments in 2011 [Digitimes, Sept 2]. His prediction is only 15 million units at max while Apple tablet PCs ~10-12 million. Also: “Wintel netbook sales have recently been devoured seriously by tablet PCs and if the two firms do not consider dropping prices or improve performance, sales will continue to drop.”
    Notebook supply chain still has no clear visibility for December orders [Sept 6]: “First-tier notebook makers Quanta Computer and Compal Electronics have recently adjusted their third-quarter notebook shipment forecast from positive growth to an up to 10% drop; however, they still forecast to see growth in the fourth quarter and are working aggressively to bring up their shipments for the quarter.”
  • Update: Even the most Windows slate concious manufacturers are publicly admitting now their dependence on Intel’s lagging chip capabilities (finally): MSI waiting on Intel Oak Trail for Win 7 tablet, Android version will hit before end of the year [Aug 23]. See also Mary-Jo Foley’s Another Windows 7 slate dropped from this year’s Christmas list [Aug 24].
  • Update: Toshiba’s dual-screen Libretto W100 laptop on sale in America for $1,100 [Engadget, Aug 16]. Excellent price for this unique, Intel U5400 based device which has only 0.8 kg weight and 3 cm thickness (when closed). See the complete offering on the Amazon site: Toshiba libretto W105-L251 7-Inch Dual Touchscreen Laptop (Silver/Black). Look at the picture on the right (copied here for convenience). Toshiba’s characterization is nothing less than: “ultramobile companion goes beyond slates and tablets to deliver something more: a full Windows 7 experience to be enjoyed across two touch screens. So now you can enjoy many different things – games, ebooks, movies, music, TV and so on – from a single handheld device.”
  • Toshiba is, however, mum on the battery life, just stating 36 Wh with the large capacity 8-cell Li-Ion battery. Considering Intel U5400‘s 18 W max TDP this is clearly not enough for even 2-3 hours of continuous use. This is why vendors should wait for Oak Trail’s Q1 2011 arrival for their Windows slates.
  • Update: and surely we will have the notebook convertibles with full touch screen capabilities like this: New Smaller LIFEBOOK T580 Is The Gateway to Fujitsu’s Tablet Line Up — Slate-beating LIFEBOOK tablet PC combines touch screen mobility with the convenience of a keyboard [Sept 2], see also the detailed specification. Notable differentiator: “Even in direct sunlight, the screen display is bright and clear thanks to high-definition LED backlighting with an Ambient Light Sensor that automatically adjusts display brightness, while a smear-proof coating means no distractions from fingerprints.” Price, availability: “The LIFEBOOK T580 is the lowest-priced tablet PC in Fujitsu’s line-up. Exact pricing varies by region.  Fujitsu’s LIFEBOOK T580 will be available for all regions in late November 2010”
  • Update: Acer to launch three tablet PC models featuring 5- to 7-inch panels [and Android 3.0 in the first quarter of 2011] [Sept 14]: “Before the Android models are released though, Acer will launch an x86 model using an Intel processor and Windows 7 to test the water in the market”.
  • Crucial issues for the future: What if a Microsoft Surface like functionality will go into this kind of handheld design? I mean a case when one of the screens is configured to work (via software added to Windows 7) as a Microsoft Surface which is even more enhanced with a tailored very general input functionality. I came to this idea when watching the concept video on YouTube (also available on the Amazon site). Even more: What if the Microsoft Courier prototype will also be adopted to this design and brought to the market? Those two things will change the whole market by themselves! And these are “just” software additions to the existing Windows 7. The screens are already multitouch!
  • This little device has been the best what Microsoft could demonstrate at this partner event. See all the detailed information below on that Toshiba Libretto W100 as mentioned at the partner event as well as elsewhere around the web.

This is what Microsoft could tell and demonstrate on this important event:

Microsoft’s Ballmer: Windows 7 slates are coming this year [Mary-Jo Foley’s “All about Microsoft” blog on ZDNet, July 12]

I think it’s way overdue for Microsoft to tell customers what’s coming on the slate front. We’ve heard that there will be Windows Embedded Compact tablets and slates coming, but it’s doubtful (I’d think) that those will be able to run Windows apps.

Microsoft execs can’t — with any credibility — keep pooh-poohing slates, claiming they’re nothing more than PCs, given how quickly the iPad has been gaining traction. Microsoft is known to be focusing on the slate form factor with Windows 8, but that operating system is still likely close to two years away from release.

All that said, there’s more to a slate than just the physical form factor. If there isn’t longer battery life, instant on/off and some kind of app store with not just the usual business apps, but also consumer-focused apps and games, I’m not so sure users are going to bite…

So let’s see in the speech transcripts what has exactly been said today about the slates:

Steve Ballmer: Worldwide Partner Conference 2010 [July 12]

… over the course of the next several months, you will see a range of Windows 7-based slates that I think you’ll find quite impressive. Tami is going to show you some of those today. They’ll come from the people you would expect, from Asus, from Dell, from Samsung, from Toshiba, from Sony. Windows 7-based slates, they’ll come with keyboards, they’ll come without keyboards. They’ll be dockable. There will be many form factors, many price points, many sizes. But they will run Windows 7. They will run Windows 7 applications. They will run Office. They will accept ink as well as touch-based input. And they will be very good for the kinds of scenarios that all of us are going to see for knowledge workers in the business that we serve that want to have something that works super well at work, but also supports their kind of personal interests as they travel.

Tami Reller: Worldwide Partner Conference 2010 [July 12]
(Video record: select from the presspass video gallery)

[Ryan Asdourian demoing for Tami Reller, [8:05 – 11:10]: none of the devices shown in this period of time are releated to the slates or tablets, they are rather variations on notebooks and netbooks]

… [11:10] but I’ve also got some surprises. I’ve got this right here, this Sony P. It’s a beautiful machine. Open this up, this is running the full version of Windows 7 right here. And it’s just really light, it’s beautiful, easy to carry around.
[11:35] Let me hand this one out as well. Just play with this for a second. While you do that, I’m going to show you this one last machine I’ve got. This is called the Toshiba Libretto. And it’s got two capacitive touch screens. You see here, I can actually interact with it right here, and it’s got a core i5 chip so it’s actually running an HD video and I’ve got a document open at the same time. It’s a beautiful machine. [11:50]

… [32:30] seeing apps is really what brings it to life. And seeing some of the great apps that have been built by you. So, let me ask Ryan to come and show us some of the great innovation out there. Welcome back. [[32:40] Ryan Asdourian demoing for Tami Reller] … the first demo I’m actually going to show is actually on this [Hanvon] slate I’ve got here. This is the Copia Reader. And what I can actually do here — I go ahead, I click into this, and what I see is a lot of articles here. I can actually go ahead and drag these right here to my reading queue. I can go ahead, save some of these in my favorites, and this app is actually tied into a bunch of my social networks, so I can see what my friends are doing as well, and they can comment, I can comment on their stuff. It’s a really interactive, rich way to interact and this is all built with Windows, WPF, and has Azure as a back end, so it’s a great example of rich client app and putting the cloud with that. [33:20]

The Sony P-series Lifestyle PC: Just don’t call it a Netbook [Jan 7, 2009]
VAIO P Series Lifestyle PC | Sony:
Estimated Battery Life with Standard Battery6 (included)
– Default Settings: Up to 4.5 hours
– Max. Brightness: Up to 4 hours
Estimated Battery Life with Extended Battery6 (sold separately)
– Default Settings: Up to 9 hours
– Max. Brightness: Up to 8 hours
Power Requirements: 68W + 10%

Toshiba Libretto W100 resurrects the classic UMPC brand with dual 7-inch displays [Engadget, June 21]
Toshiba libretto® Dual-Screen PC:
Toshiba libretto® W100 Dual-Screen PC (coming in August): “Enjoy up to 5 hours battery life” (see here) Warning[Aug 24]: This reference is for Toshiba’s UK site. Elsewhere Toshiba is mum on the battery life, just stating 36 Wh with the large capacity 8-cell Li-Ion battery. Considering Intel U5400‘s (used in current W100/105) 18 W max TDP this is clearly not enough for even 2-3 hours of continuous use. Vendors should wait for Oak Trail’s Q1 2011 arrival for their Windows slates! No question about that.
Toshiba Libretto W100 preview [Engadget, July 1]

So, what’s our overall takeaway after spending an afternoon with the W100? It’s definitely working better than the model we saw a few months back, but even when it did work there’s not much you can do with it. It’s neat as a web surfing device, but very few things take advantage of the two screens — for instance, we’d like to see a compelling e-reading app (eh hem Toshiba Book Place). In the end — even if Toshiba gets all the hardware and software kinks worked out — we’re far from convinced that there’s a place for the W100 in our lives for $1,100.

The latest showing of the lonely Hanvon slate device was just last weekend:

Hanvon Touchpad B10 tablet computer at CES China [July 12]

… new Touchpad B10 [linking to Hanvon’s product page] tablet computer. The 9.96 x 6.61 x 0.70 inch, 1.98 pound device is powered by a 1.3GHz ULV 743 Celeron processor and GMA X4500 display chip from Intel on a GS45 chipset with 2GB DDR2 memory and a 250GB hard disk drive.

Here is an earlier report from Asia already declaring local availability and even superiority: Chinese tablet maker Hanvon throws down the gauntlet to the iPad [May 20]

To mark the availability of its first slate tablet in China, the TouchPad, Hanvon chairman Liu Yingjian smashed an Apple-shaped ice sculpture with a hammer, insinuating that his products are superior to the iPad. He claimed that the Apple slate is like a mere toy which cannot satisfy the real needs of a consumer or business user, unlike the TouchPad B10 and B20 tablets which run on Windows 7.

Quite an ambitious company! E-Reading the Market – Hanwang sells nearly all the digital readers in China. Founder Liu Yingjian knows that won’t last for long. [Forbes, May 28]
(You could see a collection of related videos as well: Hanvon’s E-Reading Market [June 1])

China-based Hanwang Technology has denied rumors that its Hanvon-branded e-book reader sales have been seriously dampened by competition from the Apple iPad and its current inventory has piled up to as high as 500,000 e-book readers, according to a Chinese-language report on sina.com.cn.

Hanwang chairman Liu Yingjian is cited saying that although the second quarter was the weak season for e-book readers, Hanvon’s sales was significant, and the company expects at least 300% growth in net profit in 2010, the report noted.

The 3d party application shown on that slate is not less ambitous: Welcome to Copia

– We’re not just changing the format.
– We’re rewriting the whole reading experience.
– Introducing Copia, the world’s first social reading experience — reading, learning and sharing all in one.

And the latest report regarding the availability: Copia Coming in July? [June 20]
(There is an embedded video demonstration [June 7] of the software as well.)

Google App Inventor Beta for Android

July 12, 2010 6:11 pm / Leave a comment

Google’s Do-It-Yourself App Creation Software [The New York Times, July 11]

… has been under development for a year. User testing has been done mainly in schools with groups that included sixth graders, high school girls, nursing students and university undergraduates who are not computer science majors.

… The Google application tool for Android enables people to drag and drop blocks of code — shown as graphic images and representing different smartphone capabilities— and put them together, similar to snapping together Lego blocks. The result is an application on that person’s smartphone.

App Inventor for Android [Google Blog, July 12]
About App Inventor
App Inventor research project launch [Google Research Blog, July 31, 2009]

Android App Inventor lets you be the developer (video) [Engadget, July 12]

Google is following in Nokia’s footsteps today by offering its users a simple-to-use DIY app maker. Employing a design scheme that relies on visual blocks rather than oodles of arcane code, the App Inventor — still in Beta, of course — has functions for “just about anything” you can do with an Android handset, including access to GPS and phone functionality.

Is Google App Inventor A Gateway Drug Or A Doomsday Device For Android? [TechCrunch, July 11]

Because this new tool makes it easy for anyone to make their own apps, it makes the idea of trying to create your own app a much less daunting one. And that’s the powerful thing here. If this tool can get some kid to start messing around with app creation, maybe they’ll get more interested and start learning actual Java. And then maybe one day they’ll create the next killer app.

… they [Google] have to hope it doesn’t backfire and simply flood the Android Market with more junk apps than already exist. Google already has a problem with surfacing good apps in their market—interesting, given that they are the ones that surface good webpages as mentioned earlier—the problem could get worse if this tool is a success.

Still, I’m going to be cautiously optimistic that this tool is a good thing. Potentially a very good thing. And it’s something Apple should be taking very seriously.

Personal note: One thing is clear, there is a significantly increased momentum for Google Android as has been well indicated by my previous 4 infonuggets:

Android 2.2 (Froyo) excitement is just the tip of the iceberg for the current Android momentum

July 9, 2010 10:25 pm / 9 Comments on Android 2.2 (Froyo) excitement is just the tip of the iceberg for the current Android momentum

Ars technica had two articles in the last 3 days which have created quite an excitement on the web (66K+ hits):

  • This is, however, just “the tip of the iceberg”.
  • To understand that you could read first my two previous infonuggets on the subject if you want to grasp the extent of the current Android momentum:
    – Beyond Android 2.1 [July 4]
    – OPhone OS (OMS) 2.0 based on Android 2.1 [July 5]
  • The mobile Internet technology momentum behind the Android in general could be well understood from my another infonugget:
    – 3.9G TD-LTE rollout in 2012 with integrated 2G, 3G and 4G? [July 19]
  • In addition, Android has just joined the ranks of the 1st tier cloud client software platforms, as has been indicated recently by my another infonugget, titled:
    – E-reading SaaS wars next to e-reader wars [June 30]
  • Finally, below is the whole discussion of Android 2.2 (Froyo) significance which will make the whole “iceberg” completely visible.

Ars reviews Android 2.2 on the Nexus One [July 6], giving the following conclusion about this latest release:

Android 2.2 is an incremental update, but the performance enhancements alone make it an important upgrade for Android enthusiasts. The improved responsiveness and smoother interface transitions really boost the user experience.

Several of the new features—particularly batch updating—offer immediate and obvious benefits. Other features, such as SD storage and cloud backup, aren’t going to start delivering real value until they get broader uptake among Android application developers. I’m particularly enthusiastic about the cloud messaging service—it seems like a capability that will deeply enrich Android and open the door for some innovation in the application development community.

Google’s aggressive development efforts are moving Android forward at a rapid pace. The platform has matured considerably since its initial launch two years ago. The introduction of a JIT [just-in-time compilation] in version 2.2 has largely resolved Android’s performance problems, making it more competitive than ever. As Google works towards version 3.0 and a rumored user interface overhaul, it’s likely that we will see even more innovation.

Android 2.2 demolishes iOS4 in JavaScript benchmarks [July 8] because:

In our recent review [i.e. the previous article] of Android 2.2, we conducted some tests on the Nexus One to measure the extent of the JavaScript performance improvements. SunSpider and V8 benchmarks show that JavaScript execution in Froyo’s Web browser is almost three times faster than in the previous version of the platform.

We compared these findings with that of our tests of Apple’s mobile Safari browser on the iPhone 4. The results show that the Android device delivers significantly faster JavaScript execution than the iPhone, scoring over three times better on V8 and almost twice as fast on SunSpider.

It is worth to combine the results of those two tests in a single chart (recalculated in order to show the performance improvement vs. the performance of the benchmark on the Nexus One with Android 2.1 as the baseline):

The performance improvement is especially remarkable for v8 benchmark which is quite understandable since the V8 Benchmark Suite by Google is:

… a suite of pure JavaScript benchmarks that we have used to tune V8 [JavaScript engine brought to the Android browser as part of 2.2]. …

With this the browser in Android 2.2 (Froyo) is having the same JavaScript performance as Google’s mainstream browser since:

V8 is Google’s open source, high performance JavaScript engine. It is written in C++ and is used in Google Chrome, Google’s open source browser. [see: V8 JavaScript Engine]

SunSpider is just delivering about half of the performance improvement of the V8 benchmark suite which could be because:

This [SunSpider] test mostly avoids microbenchmarks, and tries to focus on the kinds of actual problems developers solve with JavaScript today, and the problems they may want to tackle in the future as the language gets faster [see: SunSpider JavaScript Benchmark].

Note: The main characteristics of a microbenchmark are the following:
• Small program
– Datasets may be large
• All time spent in a few lines of code
• Performance depends on how those few lines are compiled
• Goal: Discover some particular fact
• Remove all other variables
[See: slide #8 of the How NOT To Write A Microbenchmark presentation from JavaOne 2002]
The V8 benchmark suite used for tuning the engine might be more “microbenchmarkish” than the SunSpider benchmark which is explicitly trying to avoid microbenchmarking!

Developers of Android are themselves declaring that:

This has resulted in a 2-3X improvement in JavaScript performance vs. 2.1. See: Android 2.2 and developers goodies [May 10, 2010].

which seems to well correspond to the performance improvement result by the SunSpider benchmark suite.

Ars technica, however, didn’t carry out benchmarking of the Java engine used in Android 2.2 (so called Dalvik, which is not a regular JVM) albeit vast majority of Android applications are coded in Java, and:

The new Dalvik JIT [just-in-time] compiler in Android 2.2 delivers between a 2-5X performance improvement in CPU-bound code vs. Android 2.1 according to various benchmarks.

This is from the same, Android 2.2 and developers goodies [May 10, 2010] blog post which contains other significant improvements from developers’ point of view as well.

Endnote: My statement in the name of the post, that the type of Android 2.2 (Froyo) excitement shown by Ars technica is “just the tip of the iceberg” is even more true from software developers’ perspective. A lot of professional developers in technologically well established and long mature camps, like Microsoft .NET and JVM based enterprise Java / community developed application frameworks (eg. Spring), are considering the Android framework quite underperforming and immature. I’ve met even misunderstanding among them, that Android is programmed in JavaSript or even C++.

With the introduction of Dalvik JIT there are no more reasons to underrate Android (2.2 and above)! To dismiss the wrong belief of “JavaScript and C++ only” programming myth for Android I am usually providing the following links and quotes as well:

http://developer.android.com/guide/basics/what-is-android.html

Android includes a set of core libraries that provides most of the functionality available in the core libraries of the Java programming language.

… Android includes a set of C/C++ libraries used by various components of the Android system. These capabilities are exposed to developers through the Android application framework.

Since June 2009 there is also a Native Development Kit (NDK) for which it is good to know that:

Android applications run in the Dalvik virtual machine. The NDK allows you to implement parts of your applications using native-code languages such as C and C++. This can provide benefits to certain classes of applications, in the form of reuse of existing code and in some cases increased speed.
… Please note that the NDK does not enable you to develop native-only applications. Android’s primary runtime remains the Dalvik virtual machine.
http://developer.android.com/sdk/ndk/index.html

As you know, Android applications run in the Dalvik virtual machine. The NDK allows developers to implement parts of these applications using native-code languages such as C and C++.
http://android-developers.blogspot.com/2009/06/introducing-android-15-ndk-release-1.html

… the NDK is a companion to the SDK that provides tools to generate and embed native ARM machine code within your application packages. This native code has the same restrictions as the VM code, but can execute certain operations much more rapidly. This is useful if you’re doing heavy computations, digital processing, or even porting existing code bases written in C or C++.
http://android-developers.blogspot.com/2009/09/now-available-android-16-ndk.html

Finally, the use of JavaScript requires absolutely different considerations, namely: Building Android Apps with HTML, CSS, and JavaScript, extensively elaborated in a so named and actually excellent book under development (and hence available online as just linked) by Jonathan Shark, which has a Chapter 1 with the following observations (to get a feeling of difference):

What is a Web App?

To me, a web app is basically a web site that is specifically optimized for use on a smartphone. The site content could be anything from a standard small business brochure site to a mortgage calculator to a daily calorie tracker–the content is irrelevant. The defining characteristics of a web app are that the user interface is built with web standard technologies, it is available at a URL (public, private, or perhaps behind a login), and it is optimized for the specifics of a mobile device. A web app is not installed on the phone, it is not available in the Android Market, and it is not written with Java.

What is a Native App?

In contrast, native apps are installed on the Android phone, they have access to the hardware (speakers, accelerometer, camera, etc.), and they are written with Java. The defining characteristic of a native app, however, is that it’s available in the Android Market–a feature that has captured the imagination of a horde of software entrepreneurs worldwide, me included.

…

Which Approach is Right for You?

Here’s where it gets exciting. The always-online nature of the Android phone creates an environment where the lines between a web app and a native app get blurry. There are even some little known features of the Android web browser that allow you to take a web app offline if you want.What’s more, several third party projects–of which PhoneGap is the most notable–are actively developing solutions that allow web developers to take a web app and package it as a native app for Android and other mobile platforms.

For me, this is the perfect blend. I can write in my native language, release a product as a pure web app (for Android and any other devices that have a modern browser), and use the same code-base to create an enhanced native version that can access the device hardware and potentially be sold in the Android Market. This is a great way to create a “fremium” model for your app – allow free access to the web app, and charge for the more feature-rich native version.

OPhone OS (OMS) 2.0 based on Android 2.1

July 5, 2010 10:17 am / 9 Comments on OPhone OS (OMS) 2.0 based on Android 2.1

… the carrier announced on Monday it was preparing to establish its Ophone Innovation Alliance, said Bill Huang, general manager of the China Mobile Research Institute.

“Our Ophone strategy is constantly moving forward,” Huang said during in a speech at a conference in Beijing. Huang did not give any specific details about the alliance, but encouraged potential partners to participate once more information is publicized. The company has spent over a year planning the alliance.

The announcement follows China Mobile’s effort to promote its OMS in April, when Huang urged developers in Taiwan to build more mobile applications for the operating system. At the time, Huang said there were around 600 apps developed specifically for Ophones. In contrast, Apple’s App Store features more than 300,000 third-party applications.

Building less expensive smartphones without sacrificing quality is critical to China Mobile’s business strategy, Huang added. “If we can effectively make an advanced smartphone less expensive, then more users will come to use China Mobile’s services in a shorter amount of time,” he said.

China Mobile is also working to introduce more middle and higher-end smartphones for its 3G network, said company chairman Wang Jianzhou during a speech at the conference.

While China Mobile may have a relatively small number of applications for OMS, the company’s app store continues to gain in popularity. Since it opened in 2009, the carrier’s Mobile Market has recorded 70 million downloads, Wang said. The store features apps and multimedia downloads for different mobile operating systems.

  • Update: Despite of recent reports on weakening OPhone support within China Mobile (see the below update) there are newer reports pointing just to the opposite direction:
    – China Mobile to procure 6 million TD-SCMA handsets, says Chinese media [Oct 8]: “Of the total, 3.6 million will be of entry-level models and 2.4 million mid-range products. … the latest procurement effort is largely due to the fourth phase of the China Mobile’s TD-SCMA network construction. When completed, demand for TD-SCMA end-use products is expected to increase substantially.”
    – China market: Inexpensive TD-SCDMA handsets to be available in 4Q10 [Oct 13]: “Pushed by China Mobile, TD-SCDMA handsets and smartphones at retail prices of about 500 yuan (US$75) and 1,000 yuan respectively will be available in the China market in the fourth quarter of 2010. … China Mobile is setting up its fourth-phase TD-SCDMA network of more than 100,000 base stations and expects the number of TD-SCDMA subscribers to increase from 13.42 million currently to 100 million in 2012.”
    – China Mobile: 4G network coming soon [Sept 15] is stating that “4G data card is close to debut and the carrier and partners are working on the research of 4G handset chip … China Mobile is expected to launch 4G mobile communication services as early as 2011 to boost its high-margin data services, according to the GSM Association.”
    – China Mobile to set up device sourcing company [Sept 17] is telling that “The planned device-sourcing company will begin to purchase TD-SCDMA-enabled feature phones with prices below 1,000 yuan (US$148) at the end of the year and then shift to smartphones priced below 2,000 yuan in the first half of 2011“.
  • Update: OPhone fails to connect [Aug 24] which reports that with the change of leadership in June all the technology initiatives started under the previous GM and managed by Bill Huang (Huang Xiaoqing), head of  China Mobile’s Research Institute, will get less support — this is the case of OPhone — or even have been put on hold — the case of mobile payment project. Reasons are the insufficient progress. The estimations are, for example, that OPhone sales are less than half of the number of iPhones bought in China. Quite important article on what is going inside China Mobile.

• OPhone 2.0 Platform Debuts in Beijing [June 28]
• Samsung first to unveil handset for oPhone 2.0 [June 29] – important to note: “Nokia is also planning a smartphone for the carrier using the oPhone overlay and hardware specs, but running on Symbian.”
• Samsung I7680 (Oscar) parameters – Google translated from Chinese
• The world’s first mobile phone Samsung I7680 evaluation OMS 2.0 [July 1] Google translated from Chinese
• Page 3: Fusion Android 2.1, OMS 2.0 magnificent transfiguration [From Android2.1 transfiguration OPhone2.0 evaluation Samsung I7680] – [5 July] all Google translated from Chinese

The developer of the platform is the Beijing based Borqs: http://borqs.com/index.jsp. They describe the OPhone OS (short for Open Mobile Phone Operating System) platform here: About OPhone OS. See also the recent interview with their CEO: BORQS Ophone OS & more & the China mobile revolution [May 27], important to note: “Borqs is the only company that has done significant Android customization.”

But the supporter is the world largest mobile phone company: China Mobile wants Android apps for its Ophones [April 7], launching the first version of the platform last year: China Mobile Ophone hopes [Sept 1, 2009], China Mobile Launches OPhone [Sept 1, 2009], China’s first OPhone-based mobile phone hits market [16 Dec, 2009]. The last, OMS 1.5 was Android 1.5 based: OPhone SDK 1.5 New features and API changes ENG.1 [Nov 19, 2009].

Note [updated with June data on July 20]: The 3G customer base of the China Mobile is still embryonic with just 1.9% (10.46 million) of the overall number of customers (554 million) at the end of June 2010. Also this number has not been increased significantly for the last 18 month, as shown by the chart below. Meanwhile the rival China Unicom has 4.8% (7.56 million) of the 2G+3G customer base (157 million) on 3G (by end of June 2010):

Another rival, China Telecom, is not publishing 3G numbers. The MIIT* Q1’CY10 3G numbers for China Telecom (CDMA2000) were 5.57 million, i.e. 8.5% of their total CDMA subscriber base (65.45 million).
*Ministry of Industry and Information Technology

Overall 3G market share in Q1’CY10 according to the above MIIT report:  42.5% China Mobile, 30.8% China Telecom, 26.7% China Unicom. For end of May a China Telecom “official said it now has between 6 million and 7 million 3G subscribers”, i.e. about the same number as China Unicom (6.5 million).

The follow-up MIIT report for end of June 2010 is indicating 7.18 million 3G subscribers for China Telecom. When combined with the company’s July 20 stock market closing report this would be 9.6% of their total CDMA subscriber base (74.52 million). Overall 3G market share at the end of H1’CY10 corresponding to the follow-up MIIT report would be: 41.5% China Mobile (-1% vs. Q1’CY10), 28.5% China Telecom (-2.3% vs. Q1’CY10), 30% China Unicom (+3% vs. Q1’CY10). A significant gain for the W-CDMA!

The 3G numbers in China are well below of other geographies. See: Could China close the gap in mobile Internet? It should! [July 21] The Chinese government and all the three mobile operators are therefore making extraordinary efforts to close the gap between China and the rest of the world.

There is also an OPhone SDN (OPhone Software Developer Network) community site: http://www.ophonesdn.com/. Here you can find the OPhone SDK Documentation in English, currently for v1.5: http://www.ophonesdn.com/documentation/index. The rest is in Chinese.

The platform is also going out of the mainland China:
first, to Taiwan: Taiwan-based III to cooperate with China Mobile to develop OPhone [brief, June 10], China, Taiwan agree on Android, Ophone, WiMax, TD-LTE, more [detailed, June 10]
then to US and RoW: China’s OPhone to find its way to US as Android+ [May 28, 2010]

China, Taiwan agree on Android, Ophone, WiMax, TD-LTE, more

Beyond Android 2.1

July 4, 2010 7:04 pm / 3 Comments on Beyond Android 2.1

UnwiredView: Android 3.0 Gingerbread details: 1280×760 resolution, 1Ghz minimum specs, mid-Oct. release [Jun 30]
then: Revisiting Android 3/Gingerbread details post. Some corrections, clarifications [July 2]

Between these two original English posts has been an avalanche like reporting and discussion (almost half million hits!) of the Android future, all based on this source alone (directly or much more indirectly).
Note: The funny thing is that UnwiredView’s Staska (definitely a Russian speaker) has used as a trusted source the [30:36 – 38:45] time duration part of the Russian podcast Айфономания, серия четвертая, Выпуск 19 (or this one). One thing I would include here from that podcast: “As of today the number of daily Android activations has reached 160 thousands. With this Android will be #2 behind Symbian before the end of 2010”.

This is a typical reporting based on the UnwiredView source:
• Android 3.0 may ship in mid-October, carry minimum specs [Electronista, June 30] originally named: Google may fork Android 3 to guarantee experience
which – however – worth to mention here because it has a whole chain of links to previous/other posts on the pressing issues with current Android versions, and these also validate very much the original rumor (indepedent of its origin): 
• Android 3.0 to hasten the end of custom UIs? [Electronista, June 16], originally named: Android’s Gingerbread may hurt HTC Motorola
• Android Team “Laser Focused” On The User Experience For Next Release [TechCrunch, June 16]
• Android to end app fragmentation by 3.0? [Electronista, March 29], originally named: Android to detach core apps from OS updates
• Exclusive: Android Froyo to take a serious shot at stemming platform fragmentation [Engadget, March 29]
• Sprint Android 2.2 plans trigger outrage, highlight OS woes [Electronista, June 25] originally named: Evo 4G and other Sprint phones reach 2.2 soon
• Sprint expects to launch Android 2.2 in near future [Sprint News, June 25]
• Adobe posts Flash 10 for Android, leaks Android 2.2 upgrades [Electronista, June 22] originally named: Adobe outs mobile Flash 10 only for android 2.2
• HTC promises Android 2.2 updates in 2010 [Electronista, May 20] originally named: Froyo coming to company’s latest Android phones
• Samsung won’t upgrade Behold II, shows Android fragmentation [Electronista, May 28, 2010] originally named: Samsung says Behold II can’t go past Android 1.6

Post navigation

← Older posts