PalmOS, Symbian, Windows Mobile, Linux
jfk...@shaw.ca
1) What are the pros and cons of each of these operating systems?
2) Which of these PDA/Smartphone operating systems is the most likely
to be the dominant player in the portable computing market 3 years from
now and why?
-Jonathan
Keef
PalmOS lacks development drive. PalmOS 5 seems to be static and no
company seems to be licencing PalmOS 6. PalmOS 5 lacks solid multitasking.
Technologically the other three are stronger - proper multitasking
kernal's and presumably better hardware abstraction.
2) Windows Mobile (or it's successor) will be the dominant player simply
because of Microsoft's marketing drive and advertising budget. The other
three do not have enough money behind them.
For interest my PDA history:
Psion II -> Psion 3a -> Psion 5 -> Palm V -> Palm M505 -> Palm T3 ->
Palm Lifedrive.
But I am still betting on Microsoft!
Regards,
Keith
Steve Maillet (eMVP)
based systems. Symbian is still the dominant player by far but will likely
suffer the dreaded installed base curse. (E.g First you pray for an install
base, then you curse it.) In order to keep up with the latest cool stuff -
after all that's what is selling phones since they all do, you know, "phone
things" - it's the extras that are growing in sales. The Symbian OS was
never designed for that sort of thing. I think industry is gearing up for a
new battle between MS and PalmSource with it's Linux based offerings,
assuming they can get their act together and create a workable business
model based on Linux.
--
Steve Maillet
EmbeddedFusion
www.EmbeddedFusion.com
smaillet at EmbeddedFusion dot com
Jim Burks
<jfk...@shaw.ca> wrote in message
news:1130615169....@g43g2000cwa.googlegroups.com...
>I have a couple questions for the group.
>
> 1) What are the pros and cons of each of these operating systems?
It depends on what you want to do. You also left out RIM (BlackBerry) which
is its own OS/UI.
PalmOS: dominant PDA in the US. Really good as of several years ago with the
PDA application. Not a lot of uptake on mobile phones.
Symbian: high-end phone US used by Nokia, SonyEricsson and others. I've
never seen it in a non-phone PDA. Not a bad 32-bit C++ environment.
Development environment based on Borland tools and others. Mostly C++. Best
for high-end phone stuff (sync, etc.), especially where you are buying apps,
not writing.
Windows Mobile: offshoot of the lightweight Windows CE portable environment.
Heavy on CPU and RAM requirements. Great development environment, if you
like the Microsoft toolset (some do, some don't, some for technical reasons,
some for religious reasons). Easiest to develop for, especially for new
programmers or light apps. C++, C#, VB. If you're syncing to Windows Server
stuff (SQL, Exchange) a slam dunk.
Linux: darling of the ABM (anything but Microsoft) crowd. Low licensing cost
(free, that is). High ram and CPU cost (for a phone). Easy to program, in
everything from PERL to C++ and ANYTHING in between. Not a lot of phone
apps. Haven't seen it much in a phone or PDA, but in a lot of appliances
(TiVo, EMPEG car radios, MP3 players, firewalls, routers, etc.). Great for
alternative desktops (if you don't need MS Office and OpenOffice meets your
needs), servers or network appliances.
RIM BlackBerry: best UI I've seen for handheld email. Absolutely second to
none. If Symbian (or other Nokia phones) license the RIM email package, turn
out the lights on Symbian UI development as well as RIM hardware. That could
be a winning combination. Haven't seen any custom apps on BlackBerry, but
believe they have a developer's kit.
> 2) Which of these PDA/Smartphone operating systems is the most likely
> to be the dominant player in the portable computing market 3 years from
> now and why?
#1 Microsoft: they have an annoying tendency to win in the end. Because they
have an almost unlimited pot of cash, and, if they believe it's strategic,
they keep plugging away at it until they get something that sells. They
understand usability. But, stretching the Windows metaphor to the phone
really makes it hard to use. They also have the tie in with development
software (Visual Studio), servers and the dominant commercial email software
(Exchange/Outlook).
#2 RIM BlackBerry: currently the market leader in 'executive jewelry'. They
have a nagging lawsuit that anything could come out of. Might move to being
a software supplier, if they partner with Nokia. Also, rumored in that past
that Microsoft might acquire. Fairly good integration with Exchange and
Lotus Notes.
#3 Linux: as hardware and RAM get cheaper and draw less power, the
disadvantage of Linux will drop. The licensing cost ($0) is a real advantage
in mass market. Still have to develop a unique phone UI for each
manufacturer, as well as a phone device driver.
#4 Symbian: I'd like to put it higher, as I carry a Series 60 phone (Nokia
6620) and like it a lot. Download a lot of apps (most developed in Europe).
Nokia is the main backer of Symbian, and their support seems to come and go.
Still only shipping Symbian on about 40% of the European phone models, and
25% of the US ones (not sure about market share). Nokia's also rumored to be
talking with RIM about licensing BlackBerry s/w for phones - not sure what
that means.
#5 PalmOS: Latest announcement of Treo running Windows means that PalmOS is
walking dead. Expect it to be gone from marketplace in two years, except
possibly as very low end standalone PDA.
>
> -Jonathan
>
David Friedman
"Jim Burks" <jbb...@hotmail.com> wrote:
> Symbian: high-end phone US used by Nokia, SonyEricsson and others. I've
> never seen it in a non-phone PDA.
Symbion was the OS for the Psion 3, 5 and Revo--the best designed PDA's
I know of, and for a while a serious competitor to Palm.
I've been referring to the Nokia 9300 as "Son of Psion."
--
Remove NOPSAM to email
www.daviddfriedman.com
Tero Lehto
> In order to keep up
> with the latest cool stuff - after all that's what is selling phones
> since they all do, you know, "phone things" - it's the extras that
> are growing in sales. The Symbian OS was never designed for that sort
> of thing.
Just curious, what is this extra stuff you can do with for example a Windows
Mobile 5.0 phone you can't do with a Symbian phone? I have one HTC Qtek 8310
running on WM 5.0 Smartphone here, so I can try for myself. If I compare
this to a Nokia's Symbian phone, both are able to:
- read / send email (IMAP4, POP3, SMTP)
- browse the web (e.g. Opera)
- sync calendar and contacts with MS Outlook
- take pictures and videos
- listen to MP3/WMA music
- chat realtime or with instant messages (ICQ, MSN etc.)
- make video phone calls
> I think industry is gearing up for a new battle between MS
> and PalmSource with it's Linux based offerings, assuming they can get
> their act together and create a workable business model based on
> Linux.
Windows Mobile 5.0 is still too difficult for ordinary consumers and devices
are too expensive too, but I believe WM will be getting customers in
enterprises. Currently Symbian has about 85% and Windows Mobile about 12%
market share in smartphones in Western Europe. I believe Microsoft could get
around 25% if they play their cards right.
--
Tero Lehto
http://lehto.net/tero/
Steve Maillet (eMVP)
are all aided or constrained by that in any system.
Tero Lehto
> I'm talking from a software development perspective. UI and usability
> issues are all aided or constrained by that in any system.
In that aspect I agree 100%. I've been involved in a mobile software
development project where the Symbian (C++) application took about six
months work of three people, whereas exactly the same features were built in
a couple of days using .NET CF and targeted for all sorts of WM 2003 SE
terminals (both Pocket PC and Smartphone).
But as this is a consumer service, low installed base of WM terminals means
it's bad/no business.
Do you have any insight into or opinion about why Microsoft hasn't still
been able to transfer their strength in development tools to gaining
significant market share in mobile devices?
Alle
"David Friedman" <dd...@daviddfriedman.nopsam.com> wrote in message
news:ddfr-
> Symbion was the OS for the Psion 3, 5 and Revo--the best designed PDA's
> I know of, and for a while a serious competitor to Palm.
>
> I've been referring to the Nokia 9300 as "Son of Psion."
Wasn't the OS still EPOC at those days (Psion 3 etc)?
Turned into Symbian when the bigger players (Ericsson, Symbian) took over.
I think Psion has had very little (or nothing) to do with Symbian OS 7 or
later (so including 9300 also).
-A
sarah...@gmail.com
no MS certification specific to mobile apps developers on this
platform. These things really hinder capturing a market. They need
the developers behind them and this market is growing rapidly. The
other thing that really doesn't help is the lack of support from the
device manufacturers such as HTC and lack of access to some of the
underlying device API's. But hopefully this will be resolved in the
future.
The other problem with device development is the number of hoops that
developers have to jump through to actually get software out into the
public domain for WM devices. First you develop the app. Then if you
are a good developer you get it tested and certified to MS standards.
Then before you can release it if it uses any connectivity such as GPRS
etc then you have to get it certified for use by each mobile provider
eg Orange, O2, Vodafone. It's not easy to satisfy all groups all the
time.
Keef
software (like PalmOne/PalmSource). My assumption is that the Symbian OS
is an evolution of EPOC32 which was the OS for the Psion 5. I thought
the Psion 3 used EPOC16 (which may or may not be a different beast).
Anyone confirm or deny my assumptions?
David Friedman
"Alle" <nyyss...@R-E-M-O-V-E.koti.luukku.com> wrote:
> "David Friedman" <dd...@daviddfriedman.nopsam.com> wrote in message
> news:ddfr-
> > Symbion was the OS for the Psion 3, 5 and Revo--the best designed PDA's
> > I know of, and for a while a serious competitor to Palm.
> >
> > I've been referring to the Nokia 9300 as "Son of Psion."
>
> Wasn't the OS still EPOC at those days (Psion 3 etc)?
> Turned into Symbian when the bigger players (Ericsson, Symbian) took over.
Judging by a little googling, you seem to be correct. EPOC release 5 was
waht the 5 and Revo ran on. Symbion v. 6 was the next release, sometimes
referred to as ER6.
The numbering suggests, however, that it was a name change, not a shift
to a new OS.
> I think Psion has had very little (or nothing) to do with Symbian OS 7 or
> later (so including 9300 also).
I don't know about development involvement, but Psion didn't sell its
stake in Symbian until after v 7 was released.
JF Mezei
actually 2 units in the same cabinet. This is a must for palm and
microsoft based interfaces sicne their OS is not able to do hard real time.
EPOC-32 was originally marketed as being able to do this, but I am not
sure if any phone was ever built with EPOC/Symbian actually running the
phone itself.
Linux, from what I have been told, doesn't have hard real time
capabilities, but can be configured to do a pretty good approximation.
On the other hand, what is not clear to me is whether the USA vs intl
mobile standards (CDMA vs GSM vs UMTS (3G GSM)) require different
hardware or whether a software change alone is sufficient to change a
handset from one protocol to another. If they require different
hardware, then combining the phone into the UI may not be desirable.
But if the same hardware can be used to drive any telephone standard,
handset manufacturer may wish to have an OS that can handle both the UI
and the telephone which woudl allow them to simplify the handset
hardware and thus lower costs.
--------
On the other hand, the marketing department may dictate that having the
microsoft logo on the handset is a must and thus throw out any economic
or technical considerations.
JF Mezei
> Psion split into Psion who did the hardware and Symbian who did the
> software (like PalmOne/PalmSource). My assumption is that the Symbian OS
> is an evolution of EPOC32 which was the OS for the Psion 5. I thought
> the Psion 3 used EPOC16 (which may or may not be a different beast).
> Anyone confirm or deny my assumptions?
the original EPOC, which ran on SIBO hardware platform was 16 bits. It
ran on the Series 3 as well as a bunch of industrial devices.
Then PSION set out to build a totally new 32 bit OS designed to be able
to run on phones. It created a new subsidiary called Psion Software and
released what was called EPOC-32. Then, PSION convinced mobile makers to
buy into it and Psion Software became Symbian with PSION retaining about
a 20% share in it. This was done at a time when EPOC32 was still
unfinished (versions 1.0). It took many years for the prpmosed features
to appear n the form of the ER5 version which, contrary to early
promises, was not available to the Series 5 owners.
Following this, no more work was done by Symbian on behalf of PSION and
Symbian was able to start doing real work on EPOC32 , now called Symbian
OS and started a pretty huge transformation process of the OS, adding
many features which PSION had refused to support (such as file converters).
And in recent year/years, PSION sold its shares in SYMBIAN to Nokia and
bought into some company Teklogix which makes industrial devices based
on Windows/DOS. The SYMBIAn OS is now totally independant of PSION and
majority owned by NOKIA. Some of the early adopters of EPOC32 such as
Motorola have produced non-Symbian based phones since this. So it isn't
a given that shareholeders of Symbian produce all their phones based on
Symbian OS.
JF Mezei
> I don't know about development involvement, but Psion didn't sell its
> stake in Symbian until after v 7 was released.
Psion ceased its interference with Symbian once ER5 was released. It
sold its shares much later, once it was possible to sell them at a
profit. Between the initial released of EPOC32 and ER5, there was a long
gap (2-3 years) which has never been explained. During this time,
Symbian did nothing significant.
PSION just changed the CPU and found slightly better LCds wben it
upgraded the Series 5 to 5mx and got ER5 installed to support the
featues that had been promised when the Sries 5 was released. (just as Java).
Note that EPOC32/Symbian versiosn produced for hansets were compiled
with UTF character set support, whereas the PSION devices just has
microsoft's proprietary deviation from ISO-LATIN1 character set support.
So applications on one won't work on the other because of the different
character sets even if the OS is the same.
Steve Maillet (eMVP)
>been able to transfer their strength in development tools to gaining
>significant market share in mobile devices?
Unfortunately these are orthogonal items. Microsoft bet on Lot's of
functionality over and above the basic PDA capabilities of the Palm Devices
at the time. The systems were bulkier and more expensive at first and people
laughed saying "who needs all that extra stuff" but lo and behold - it's
just what people wanted and developers were able to take advantage of it
all. Same transition is happening in the phone markets. It's getting to the
point where calling someone is almost a secondary feature of a more capable
"Portable Device" Windows CE was designed to support that kind of system
from the start. Other OSs were not, so it's a longer reach for them to make
that happen. So why is WM behind? Because it's still more expensive to make
the more feature capable devices, the technology is changing literally on a
weekly basis, and device vendors are refusing to deal with software and OS
updates. All of that makes people skittish about buying the new feature rich
devices. Then there is the usability. So long as these things are sold as
phones with the basic number keypad as the primary input mechanism they will
continue to be overpowered phones. A new UI model is needed, so far the
Pocket PC phone edition type devices look the best for that, but they've got
limits as well.
Steve Maillet (eMVP)
This is FALSE, it astounds me how anyone can continue to make this claim
after it's been shown false on many occasions over many versions of the OS.
Windows CE (The underlying OS for Windows Mobile) IS able to support hard
real-time systems, See the analysis performed by dedicated systems for
details of independent tests. I've built many a real time system based on
Windows CE.
Linux does not have real time natively, it was never designed for that.
There are some additions and add-ons that deal with the issue in a variety
of different ways some commercial others GPL with varying levels of success.
(Usually targeted for a specific industry or solution type)
The biggest reason why the phones run with 2 processors is that the radio
vendors want to tightly control the radio interface code and to standardize
throughout the industry on the "AT command Set'. The radio vendors want to
control the "stack" to protect their IP AND to protect the carriers. It's
possible to wreak havoc on the network with the low level radio stacks. By
having radio modules with hardened, tested and certified stacks built-in
that a device OEM can't mess with in any way, the carriers and device
vendors can get devices onto the network faster while maintaining network
integrity. The OS and CPUs are perfectly capable of doing it all. But the
security concerns keep it separate. Not a bad design really.
RobR
to CE, but yes, we built realtime high speed memory testers for Teradyne
based on CE because of its RTOS capabilities (which NT was not
capable of).
"Steve Maillet (eMVP)" <nos...@EntelechyConsulting.com> wrote in message
news:%23GtfqJc...@TK2MSFTNGP15.phx.gbl...
> >This is a must for palm and
>>microsoft based interfaces since their OS is not able to do hard real
>>time.
> This is FALSE, it astounds me how anyone can continue to make this claim
> after it's been shown false on many occasions over many versions of the
> OS. Windows CE (The underlying OS for Windows Mobile) IS able to support
> hard real-time systems, See the analysis performed by dedicated systems
> for details of independent tests. I've built many a real time system based
> on Windows CE.
>
<stuff deleted>
Sander van der Wal
>
> 1) What are the pros and cons of each of these operating systems?
Who cares.
> 2) Which of these PDA/Smartphone operating systems is the most likely
> to be the dominant player in the portable computing market 3 years from
> now and why?
Symbian. Phone manufactures like Nokia, SonyEricsson et al don't intend to
be commoditized by the OS vendor in the same way Microsoft commoditized the
PC market.
Whether that is relevant to ISV's is a different matter. For them, it is
which phones sell software.
--
Sander van der Wal
www.mBrainSoftware.com
Christopher Browne
> the fact that an increasing number of devices that use it, such as the
> Treo smartphones, are just plain crap when it comes to quality.
>
> I just converted to a Blackberry after having 2 Treo 600 handhelds
> go dead on me within a year. I love the functionality of the Palm
> OS, and the Blackberry's not as slick as I'd like it to me, but at
> the end of the day, if a device is not reliable, what good is its
> functionality?
Hmm. Yeah, that makes a lot of sense.
I bought my last phone based purely on considerations of battery life;
it came with a spare cell, and had a pretty long life battery to begin
with. It's not that I kill off battery quick; it's that the batteries
just tend to tail off and die, even under a pretty *low* usage
pattern.
(My mother has a cell phone that she almost *never* uses; if often
hasn't worked when she wanted it to because she doesn't regularly
charge it.)
I would certainly get pretty exorcised if I paid the huge bucks for a
Treo phone and then watched it go dead in a year.
We use Blackberries for oncall stuff at work; one thing that *really*
pleases me about the model we have is that the batteries are generic
and we can buy replacements for about $30, eminently reasonable for a
cell battery.
Nothing Palm offers today compares to that, simply evaluating based on
battery handling. Forget about whether or not the Treo's are
more/less robust; that's a separate issue.
Actually, I get the sense that Apple has a *most* fascinating take on
things with the iPod that may be a potent step forward.
- You can take iCal calendar files (in a "standard" vCal format) and
shove them in an iPod directory, and it'll turn that into an
on-the-iPod calendar.
- You shove pictures into a directory, and it'll display 'em.
- You shove vCard address information into a directory and there's a
local phonebook.
- Shove text documents in the "Notes" directory, and you can read
documents
- And of course, it plays MP3s ;-).
As far as "conversion/SYNCing" user interface is concerned, that's as
simple a scheme as I can imagine. It's not two-way, but updating
information on PDAs always sucked pretty bad. (And I say that as
someone who has used Graffiti with reasonable success since the Palm
1000.)
The iPod approach to these "PDA" features would have the possibility
of getting us out of the horrid interface lock-in that PDAs and cell
phones alike have so often suffered from.
A phone that lets you plug in SD cards with iPod style directories of
data would be most sweet. A sort of shocking form of simplicity...
--
(reverse (concatenate 'string "moc.liamg" "@" "enworbbc"))
http://cbbrowne.com/info/
If we were meant to fly, we wouldn't keep losing our luggage.
Tero Lehto
> Symbian. Phone manufactures like Nokia, SonyEricsson et al don't
> intend to be commoditized by the OS vendor in the same way Microsoft
> commoditized the PC market.
Do hardware manufacturers actually have the power to choose this? If large
operators like O2, T-Mobile and Vodafone want Linux or Windows Mobile
devices, I bet there are manufacturers who are willing to offer them.
At the moment it seems Symbian has been the most powerful in the consumer
smartphone segment when it comes to tempting applications and services.
Tero Lehto
> I avoided chiming in because I wasn't sure how closely related WM
Is there something unclear in this? Windows Mobile is just a brand name
which means a device based on Windows CE operating system and Pocket PC or
Smartphone user interface (UI) that has to match different kinds of hardware
and software requirements.
Pocket PC and Smartphone 2003 ("Windows Mobile 2003") are based on Windows
CE 4.2.x and Pocket PC / Smartphone 5.0 ("Windows Mobile 5.0") are based on
Windows CE 5.0.
The difference between Windows CE and Windows Mobile is that there are also
other kinds of embedded Windows CE devices than Windows Mobile specified
devices.
A nice tour to different kinds of embedded Windows CE devices is available
at MSDN Channel 9 archives:
http://channel9.msdn.com/ShowPost.aspx?PostID=10924#10924
Or even this is a Windows CE device: http://www.14mz.com/
Steve Maillet (eMVP)
>> still is to CE ....
>Is there something unclear in this?
Apparently so, a quick scan of the newsgroups indicates that's close to the
#1 most frequently asked question. (E.g what's the difference between
Windows Mobile/Pocket PC/Smartphone and Windows CE. Sometimes it's phrased
differently but there's a lot of confusion as the branding for Windows
Mobile makes little or no mention of Windows CE as the core.)
--
Jim Burks
"Sander van der Wal" <svd...@mBrainSoftware.com> wrote in message
news:43660ded$0$11051$e4fe...@dreader11.news.xs4all.nl...
>
> <jfk...@shaw.ca> wrote in message
> news:1130615169....@g43g2000cwa.googlegroups.com...
>>I have a couple questions for the group.
>> 2) Which of these PDA/Smartphone operating systems is the most likely
>> to be the dominant player in the portable computing market 3 years from
>> now and why?
>
> Symbian. Phone manufactures like Nokia, SonyEricsson et al don't intend to
> be commoditized by the OS vendor in the same way Microsoft commoditized
> the
> PC market.
I wouldn't bet my company on it. Microsoft has an annoying tendency to keep
plugging away at a market, coupled with an almost unlimited basket of cash.
Also, the US handset market is a captive of the carriers. A
Microsoft<->Cingular or Microsoft <->VerizonWireless alliance could catapult
a 2nd or 3rd tier handset maker into a large position.
Also, as more people start running apps on handhelds, and developing small
line-of-business (custom) apps, the development platform and tools will have
an influence.
Don't get me wrong - I carry a Symbian handset, and it's better than the
other smartphone offerings. But, I wouldn't place a large bet against
Microsoft.
Jim Burks
Mirko Grewing
> 1) What are the pros and cons of each of these operating systems?
IMHO:
Palm OS was the smartest OS. Easy to use, light, with long battery life.
Actually it's going to die. I mean that is going to die AS Palm OS. It's
looking for other ways of development. Palm OS 5 is obsolete: I imagine
that realize a "HTC Universal" with Palm OS would be frustrating... So
PalmSource/Access is pointing to Linux (better: to Linux kernel).
Windows Mobile is going to conquer the crown of the market. WM5 semms to
have solved all the problems that his ancestors had. I personally
dislike his "philosophy": I think this OS is too much similar to Windows...
It should be simpler...
Symbian OS: It conquered the smartphone market. Now it's going to face a
crossroad: it must follow the user wishes without losing its qualities.
Nokia preferred Linux for its 770. Palm OS is going to migrate to a
Linux kernel. Symbian OS risks to be relegated to the small phones...
Linux: It was never able to conquer users as Linux. Maybe it will be
able as engine for other OS...
> 2) Which of these PDA/Smartphone operating systems is the most likely
> to be the dominant player in the portable computing market 3 years
> from now and why?
Impossible to say.
I think that WM is going to be the leader in 2006, with Symbian OS in
second place... Palm OS will bite the dust!
But in third quarter of 2006 we would see the first Palm_on_Linux PDA...
maybe something will change! ;)