TamsPalm – the Palm OS / web OS Blog

A recent report by Standard&Poor’s names Palm one of the few companies endangered by the current sub-prime crisis. I personally am not surprised – here is why I think that Palm is the worst possible place for money invested into mobile computing.

We best – ignore industry trends
In the last years, Palm has repeatedly ignored industry trends. As an example, let’s take proper web browsing. Insider sources have told me that Opera repeatedly approached Palm with an offer to create a web browser for the platform…only to be ignored.

Palm seems to live by the NIH syndrome – if it’s not invented here, it sucks.

We don’t need money – ignoring the Palm faithful
While we all agree that the PDA is a dying breed of handheld, I wonder why Palm doesn’t offer a step-gap solution for its customers. HP still offers PDA’s, and furthermore offers smartphones that combine big rectangular screens with a keyboard and a phone.

We gamble – betting all on one form factor
Continued from above: if Palm wants to make smartphones only, why do they have to bet it all on one form factor. Yes, the Treo is well-done – but it takes many phones to make a world. Why not offer a touchscreen-only device, a flip phone and a slide-out keyboard one?

This diversification of the platform would make selling Palm devices (and the Palm experience) much easier. Instead of forcing customers down one road, why not let them choose which way they wish to go?

We great – kicking third-party developers in the butt
The final – and very grievous – topic involves Palm’s handling of its developers and enthusiasts. Web sites like the PocketPC FAQ have existed for ages; developers have created PocketPCfoo apps enhancing the operating system.

Palm, on the other hand, feels that this is not needed. In fact, it even is impossible to advertise products for Treo on AdSense nowadays – if that isn’t an effective way to slow down the Palm ecosystem, what is?

In the end, it all comes down to the age-old problem of self reflection and proactive management. Palm’s management seems to lack both capabilities: the company’s recent life can be described as luck and turmoil (IMHO). Palm needs to refocus on its old core values of “being the platform that works best” – unless they do that quickly, we can IMHO short their stock…

What do you think?

The new ‘ACCESS Linux Platform™ mini’ (Alp mini) is a little bit different than the normal Access Linux Platform (ALP). ALP mini is specifically for Mobile TV, car navigation, Personal Media Player, Advanced Mobile Phone and set top box devices.

So ALP mini is for smaller, cheaper devices.

From the ACCESS Newsletter:

—————————————————————————————-

‘ACCESS Linux Platform miniACCESS held its annual “ACCESS Day” on Oct 20th in Tokyo. The event drew about 1,000 attendees, including representatives from hundreds of ACCESS partners and ecosystem providers. At the event, ACCESS unveiled the latest company offerings extending its family of Linux based platforms to include the ACCESS Linux Platform version 3.0 and the new ACCESS Linux Platform mini, a mobile Linux platform targeted at price-sensitive and emerging markets. ACCESS Linux Platform v3.0 provides advanced UI and LiMo compatibility essential to major carriers worldwide. The demonstration showcased highly detailed visuals, smooth transitions and animations, and an intuitive and engaging user experience.ACCESS Linux Platform mini is a compact, commercial-grade Linux platform designed for mobile and next-generation Internet-enabled devices. ACCESS Linux Platform mini features a scalable and customizable platform that includes a comprehensive application suite suitable for devices ranging from feature phones and low-end smartphones to Internet-enabled, beyond-PC devices, such as portable navigation devices, portable media players and set-top boxes. ACCESS Linux Platform mini is well-suited for the more stringent cost and performance requirements of mobile and Internet-enabled devices in emerging markets. The product’s market-proven technology has already been commercially deployed in China and Taiwan on devices from manufacturers such as global leader Haier Co., Ltd. and Shanghai Novarobo Technology Co., Ltd. “ACCESS Linux Platform mini provides carriers with the opportunity to seriously contemplate unlocking data ARPU (average revenue per user) potential in all segments of their subscriber base – not just smartphone users – in a cost effective, proven and scalable way,” said Stuart Carlaw, vice president and research director, Mobile Wireless, ABI Research. “This development will be critical in enabling the expansion of the next phase of mobile user experience for the majority of today’s mobile phone users, including those in emerging markets.”

ACCESS Linux Platform mini includes a complete set of middleware and customizable applications, an optional Linux kernel enhanced by ACCESS and development kit. ACCESS also provides customers and partners with technical and professional services to help enhance their competitiveness in the global market, including product customization, localization, integration and QA services.

ACCESS Linux Platform mini is optimized for the following minimum hardware specifications: 200MHz CPU, 32MB RAM, 32MB ROM. Key product features include:

* An innovative UI design
* Open architecture based on components and services
* Development environment: IDE based on Visual C++ (in Windows)
* ACCESS NetFront Browser pre-integrated
* Support for NetFront Browser Widgets (optional)
* ACCESS Media Music player integrated
* Multi-media applications and RSS Sync
* PIM sync
* GPS
* Mobile TV
* Multiple profiles, including mobile, STB, and PND/PMP profiles’

—————————————————————————————-

What do you think?

Read the rest of this entry »

Symbian just finished the schedule for its 2007 roadshow(visit TamsS60 for details), targeted at both journalists and developers…and it takes place in London, Europe.

Being completely honest, Palm’s support for Foleo developers was very good…if you were in the US. However, european developers got little support…
Palm originally had an european ‘devcon’ – please bring it back…

Symbian just finished the schedule for its 2007 roadshow(visit TamsS60 for details), targeted at both journalists and developers…and it takes place in London, Europe.

Being completely honest, Palm’s support for Foleo developers was very good…if you were in the US. However, european developers got little support…

Palm originally had an european ‘devcon’ – please bring it back…

Access has released its Access Linux Platform SDK a few hours ago, thereby officially starting the development process for its developer partners(who weren’t privileged enough to get the SDK before its public release). Anyways, I don’t crank up my compiler yet; and won’t do so in the near future either – for one simple reason:

No one knows how and when this thingy will ship out!

Thousands of man-hours were wasted in PalmSource’s huge Cobalt debacle – applications were rewritten for an OS that simply never materialized(ok, there were 20 Cobalt smartphones – what a market). Back then, developers wanted to be the first to be Cobalt compatible…but why?

Both ALP and Palm’s Linux derivative will run Palm OS Garnet apps

Let the message above sink in carefully, please. You can support both platforms with the same code base and the same executable file…by simply sticking to what you know best(aka developing OS5 apps).

Eventually, you may want to move to one of the two platforms exclusively(and dump StyleTap in the process) – but this is a step that shouldn’t be taken just for marketing reasons IMHO. Unless the technology forces you to move, don’t…why alienate half of your potential customers?

Please tell us how you handle the upcoming Linuxii!…off to learn some PocketPC programming =)…

Access’s Access Linux Platform has been considered the official successor to the Palm OS for a long time – but nobody ever was sure if Palm(the hardware guys) would adopt ALP. Now, Ed Colligan spoke out…and announced that Palm will deploy an own Linux platform.

Essentially, the platform is said to be “compatible with Garnet”; Palm plans to “evolve” its developers from Garnet to Linux. The platform will not be licensed to other companies – so everyone developing for this platform puts all his eggs in the Palm basket…

IMHO, this announcement marks the final departure and breakup between PalmSource/Access and Palm(the hardware company). The two companies have drifted away from one another since the spin off; and Palm has now “cut off” all bonds to its former software department.

As for developer motivation; Palm’s decision to not license the platform is a very bad idea IMHO. Having more than one hardware manufacturer makes development more “secure” as you are less dependent on the date of a single company…

The PalmInfoCenter has the full scoop:
http://www.palminfocenter.com/news/9351/palm-announces-new-linux-based-mobile-platform/

The news that Access decided to rename Palm OS(5) to Garnet OS hit us a few days ago – my first reaction was an outburst of maniacal laughter. Garnet means as much as not at all in German, so this was quite funny. But now, a few days after the laugh, I think that this renaming gives the Palm OS economy yet another problem to chew on.

I as a developer will definitely not benefit from the renaming. Marketing campaigns established the name Palm OS in the minds of PDA and smartphone software customers – if it runs Palm OS, it can run Palm OS apps like FileFind or Binary Clock. Now that we have two (compatible(!!!)) OS’ses with different names, developers need to redo their marketing campaigns and sometimes even product names to reflect the new situation. This will lead to long and clumsy names like Daily Quote for Palm OS and Garnet OS(yikes), even more text on already overcrowded banners and generally effort for nothing.

Licensees are in an even bigger dilemma, loosing the benefit of the Palm OS name in their products. I can already imagine millions of Joe Consumers walking into a store, looking at a GSPDA M70 or Aceeca Meazura and asking the clerk if this is compatible with my Palm Os applications.

IMHO, this move wasn’t exactly smart(yes, I know it was necessary and all, but I still don’t like it one bit) – I wonder if Palm will start to license thew Palm name separately to other “Garnet OS” licensees…

What do you think about it?

The news about Palm’s buyback of OS5 rights has slowly but surely sunk in – we now have two Palm OS ‘owners’. The current version of Garnet has a few quirks, but is a reliable platform…this is what its owners should do to it 2007 to keep it competitive:

Get a GOOD web browser
Face it, all current Palm Os web browsers suck badly. OK, the Treo 680 version of Blazer/Netfront is said to be decently usable at the least – but it still lacks many features in comparison to PDA browsers for other platforms.

This point should be really simple for Access – all they need to do is add a few features to their old Sony Clie Palm OS NetFront browser. Alternatively, Palm could bug Opera or buy up the Universe3 project before it ships.

Give Garnet a competitive UI
Face it – Garnet looks ugly compared to Windows Mobile. Ok, black on white means great contrast, but the average user doesn’t give a f*** about a few nits more. Instead, he wants a cool-looking device…

This is also a no-brainer for a licensee. Buy up either SkinUI or its competitor(probably cheaper) – and party.

Unlock the multitasking API/ARM development API
First of all – I got much of this information from a reliable source inside a Palm OS licencee. However, I cannot disclose the name for obvious reasons.

Palm OS’s core is multitasking capable – people like Dmitry Grinberg have used the Kernel for quite some time. The problem is somewhere else – Palm’s original license for the Kadak kernel forbade them to publicly release the multitasking API(Kadak sold such an SDK, fyi only).

Now, if Palm owns the complete rights to the OS, what speaks against looking at the kernel rights once again and maybe opening up multitasking in one sort or the other? And if you are already at it, why not unlock the ARM native stuff, too?

IMHO, the three steps outlined above should help Access/Palm get their operating system back in-line. Garnet’s days may still be counted and its architecture may eventually need an overhaul(dual core ARM processor, come) – but implementing the 3 commandments should be able to add at least 3 more years of life to Palm OS Garnet.

Now that the dust has begun to settle about Palm’s perpetual license of Garnet, a few seething questions still remain(in developer’s minds). This article is just my personal opinion though – so please don’t buy stock on it, and feel free to discuss!

First of all, this purchase shows that Palm seems to understand the value that its third party developers make for the platform. OK, Windows Mobile now has a load of stunning apps too, but the diverse application landscape of the Palm OS still is mostly unique for a mobile OS. So, keeping developers happy by keeping their API alive pays out.

This insight probably didn’t fall from the sky though – Nokia gave a great example of what happens when you piss off developers by releasing an incompatible S60 revision.

Palm’s purchase of the Garnet environment allows them to do exactly the opposite of what Nokia did – they can now swap out the old Garnet kernel and kick a new one in; or they can create a Garnet emulation environment(PACE, anyone – Garnet is a huge Os4 emulator mostly) on whatever host OS they want to use.

Palm also saves a lot of licensing fees with this purchase on the long run – they must have some kind of longterm usage in mind with the IP that they just acquired, for else they would have just kept paying a per-box licensing fee.

Essentially, this IMHO is a bit of a good sign for Palm OS developers. While now always is a good time to look out for another OS to support(as Seth said – Do or Die), a one-OS shack owner should now be able to sleep a bit better. I will run another article on how Palm’s future lineup will look soon – for now, what about sharing what you think?

PalmSource was once known for their next-to-perfect API documentation. Failures, bugs, hints and other useful stuff were contained right in the Palm OS API documentation. However, as PalmSource staff got less, the quality of the documentation decreased according to developer’s oppinion(I still consider it useful btw – so this is no rant).

Anyways, while coding a little PHP script, I needed help about a call and looked at the official PHP documentation:

http://at.php.net/str_split

I quickly found out that the function I needed wasn’t available in the PHP version that our current host(HostLoco) runs. However, the comments contained like 15 ready-to-use replacement functions and hints on usage!

The PHP API documentation system has two parts – the first part is ‘fixed’; and the second part consists of comments like we have it at TamsPalm’s. So, vandals cant change the core documentation, but users can add hints easily.

IMHO, PalmSource should convert its Developer’s API Reference to this format if they somehow can…developers would greatly benefit from the extra amount of knowledge available!

What do you think?

PalmSource recently sent out an email to its developer partners covering GTK, the new Palm OS application framework. The tutorial that they sent out is available here:

http://www.palmsource.com/developers/newsletter/20060425_GTK.html

However, I still didnt start porting BinaryClock and AutoSync to ALP, and dont plan to do so in the next few months for a simple reason: it isn’t clear yet what ALP will really be. Will it be a smartphone OS? Will it be a PDA OS? Will it even make it to the US? Questions over questions, and if you add in the constant rumors about how Palm is planning to create its own new “Palm OS”, nobody can really say anything about ALP yet.

Many developers ported their apps to Palm OS Cobalt and paid a dear price for it-the 10 Palm OS 6 powered phones sold at a Palm DevCon for sure don’t create enough revenues to finance the huge costs for the port.

OK, but won’t I loose cash by not porting? The answer is simple, nope. Both ALP and Palms rumoured OS are said to be fully compatible with “wellbehaved” Palm OS Garnet apps. So, users of phones/PDA’s with the new operating systems can continue to use their existing Palm OS 5 applications while developers slowly but surely create their ports.

Are you preparing for a port?

The Haier n60 was in the news very much recently, as PalmSource decided to use it as its “test bed” for ALP. PalmInfoCenter recently got a nice pack of screenshots-find them online here:

http://www.palminfocenter.com/news/8514/access-linux-platform-shown-at-linuxworld/

But this is not the problem of this editorial. Today, we will focus on one of the biggest stupidities ever made by PalmSource IMHO, sorry PalmSource, but it as just dumb not to get into end user sales.

When Acer bailed a few months ago, people began to notice one thing: PalmSource is dependant on licencees for its survival. And this is a problem-licencees can come and go. Now compare this to PalmSource beeing an independant OS vendor too. If a company like Microsoft looses a licencee, they dont care. Users can purchase the operating system in a store and install it…

So, what PalmSource should do IMHO is offer a ‘palmization service’ for existing Windows CE handhelds. The rise of Palm OS emulators shows that many people dislike Palm hardware, but want the software. So, why use a emulator, if you could install your OS directly.

And last but not least, don’t forget that you usually earn much more cash from a direct sale than from a OEM one…

Now, a few days have passed since the appearance of the first few MAX screenshots. At first, they really looked confusing. No category-style launcher like we know it from the Palm OS, no Palm-Ossy look, no nothing. WTF was this showing??

To be honest, I believe that this was a very phone-centric demo of PalmOS on Linux running an existing NetFront product. Any guesses on what it was? Yeah, it damn looks like Acess Netfront Mobile Client. In case you don’t beleive us, here are a few screenshots:

http://www.access.co.jp/english/products/nf_mcs.html

But what does that tell us? In fact, it tells us very little except that they now have Palm OS Garnet(maybe with some cobaltish elements) running on Linux, and this is good news. We are still about a year away from the publication of the SDK’s, so nobody knows now if they won’t create a “glue library” that will allow PalmOS developers to still use the API’s they know while benefitting from GTK.

Do you believe that the Palm OS as we know it(the API) has a future? Or do you believe that there will be a huge lot of learning and porting ahead of us? Commenting is free and anonymous!

Recently, we saw the first few screenshots of the ALP, the “future” PalmOS platform. The screenshots showed more than just the classic OS; it also showed a few “included apps”. And those appeared to be meager. Well it still is a beta, but lets just put this clear:

A device without a good software library in the ROM is a dead device.

The reason for this is easy. Many stores have handheld displays where user can test out handhelds. These handhelds are hardreset every one to two days, and they usually are not connected to the internet or to a computer. So, the users dont get to see the great apps that are included on the CD-they just see whats in the ROM, and ion devices like the TT, this is more-less empty.

What licencees need to do is increase the “outta-the-box-oomph” factor of their devices. Include music, images, videos, games-just do whatever you can to make the device be more attractive after the initial power up. This may appear to be very expensive at first, but it actually isnt(if you go with the underdogs). For example, there are many DocsToGo replacements on the market that go unnoticed mostly but offer excellent quality neverhteless-those are cheaper to licence and still improve your device’s look.

So, to cut a long story short, a great OS does noit make for a great user experience. The market already rejected many operating systems that lacked enough third party products, and carriers also will reject phones that can’t do anything out of the box! Licencees, use your “volume purchase discounts”, and include as much great software onto your devices as you possibly can!