Blowing the doors to Palm - Java programming for Tungsten handhelds

[John D.]

WinCE is crap. Why bother supporting it?

Well, the other 2 (Palm and Symbian) are also crap, but WinCE has
the best development tools between these 3.

 

[asj]

sounds like you never tried java apps.... they are very platform specific
(or better put: JVM specific) and buggy!

sounds like you've never actually used java a lot (other than playing
with applets perhaps).

eBay is building its ENTIRE infrastructure on J2EE (it dumped
microsoft's .NET in favor of a more RPBUST solution):

http://groups.google.com/groups?dq=...=UTF-8&oe=UTF-8&group=comp.lang.java.advocacy

* this infrastructure is allegedly architected to be able to server 1
BILLION page views a day - it now runs 400 MILLION transaction a day or
75% of eBay

Java powers a major proportion of all the high end corporate
infrastruture. for example, just looking at one application server,
IBM's Java app server Websphere:

35,000 global customers use WebSphere:

65% of the Fortune 500 companies
80% of the top US healthcare companies
75% of commercial banks worldwide
90% of the top commercial banks in the US
67% of the world's largest banks use IBM messaging servers
15 of the top Wall Street brokerage firms
7 of the 8 largest US telecommunications companies

mix in all the other commercial java application servers (Oracle's,
SUN's, BEA's), as well as the open source application servers (JBOSS,
tomcat, etc) and you get the picture.

 

[username]

sounds like you've never actually used java a lot (other than playing
with applets perhaps).

you are SO wrong about that .....

eBay is building its ENTIRE infrastructure on J2EE (it dumped
microsoft's .NET in favor of a more RPBUST solution):

you say that serverside J2EE is why java will win on a handheld?
my handheld doesnot have to do 400 million transactions a day.....
same applies to ordinary PC users: they run windows and not websphere!

 

[luke]

there are actually JVMs for windows handhelds, eg.:
http://www.blueboard.com/j2me/notes/2002_7_26.htm

however, there is some concern that sun has decided not to move forward
with anything for the latest windows ce, nor does there seem to be
anyone taking the ball in their stead.

Sun has wisely (IMO) decided that the Pocket PC market is not worth
the trouble of getting screwed by the owner of the OS. Obviously,
others have decided the same since no one seems to be anxious to jump
in.

The good thing about this is that it looks like Palm is starting to
squeeze Pocket PC out of most of the enterprise market, and this entry
of IBM into the Palm alliance will only accelerate Palm adoption in
corporations. I believe the latest figures show Palm with more than
50% of the enterprise market and Windows at around 30%+, with Palm
gaining more traction faster after it conveniently forgot about this
market in the earlier years, when it concentrated on the consumer
market.

 

[username]

dude, your comment was about Java in general...you obviously have no
wide experience with java since you're recycling old criticisms that
were leveled against desktop browser applets in the late 1990s. it's the
21st century and java has pertty much become the de facto standard for
writing applications for many small devices such as smart cards,
handsets, set tops (although that is still an ongoing battle), and
others. java is also widely considered to be the best solution for
server side development, because it is robust, open, scalable, and
cross-platform.

if that is true, then why are you flooding the groups with your java
advocacy?

and, no, your handheld does not need to serve 1 billion page views a
day, but isn't it nice to know that the technology and platform that CAN
be architected to do 400 million and more transactions a day runs in
your handheld?

I love java, but no, when considering software for my handheld, I do not
care about mainframe scalability

new java developers from palm and pocket pc developers who will help
architect the same robust solutions in handhelds that we have come to
expect elsewhere are ALWAYS welcome. the job market for java is much
stronger than most others, and face it, would you rather continue to
write to a proprietary solution that shackles you to a particular os, or
develop on an open, standards-based platform that will run on the widest
number of devices?

I am only interested in delivering functionality to my customers.

 

[username]

WinCE is crap. Why bother supporting it?

because your users are using it???

 

[asj]

if that is true, then why are you flooding the groups with your java
advocacy?

so you think that people should not advocate solution that they think
are GOOD?
or perhaps you wish that people should advocate BAD solutions?

I love java, but no, when considering software for my handheld, I do not
care about mainframe scalability

well, that's surprising...i would love it if my handheld apps were as
robust and scalable as mainframe apps...that way, as the capabilities of
the handheld grew, the apps would in turn be able to meet the increased
demands placed on it.

I am only interested in delivering functionality to my customers.

ahhhh!!!! of course! but giving them a long term solution to their
problems would obviously be preferable to screwing them sometime along
the line.

for example, writing a client application using a proprietary language
that "faded" away or was made into a "legacy" technology would certainly
cause some problems down the road if the client wanted some upgrades
done...perhaps there would not be any available programmers to do the
modifications, or perhaps the few available would be very
expensive...then the customer would have to recreate the wheel, as it
were, and create a brand new app that did almost the same thing!

or what happens if the customer upgrades his hardware or system in
future and needs to port the application over to the new system? what if
that proprietary app could NOT be ported over? with a standards based,
cross-platform, widely-used technology like java, the chances are it
would not cause that much headaches moving the app over to the new
system.

 

[John D.]

ahhhh!!!! of course! but giving them a long term solution to their
problems would obviously be preferable to screwing them sometime along
the line.

for example, writing a client application using a proprietary language
that "faded" away or was made into a "legacy" technology would certainly
cause some problems down the road if the client wanted some upgrades
done...perhaps there would not be any available programmers to do the
modifications, or perhaps the few available would be very
expensive...then the customer would have to recreate the wheel, as it
were, and create a brand new app that did almost the same thing!

or what happens if the customer upgrades his hardware or system in
future and needs to port the application over to the new system? what if
that proprietary app could NOT be ported over? with a standards based,
cross-platform, widely-used technology like java, the chances are it
would not cause that much headaches moving the app over to the new
system.

From that I conclude that "username" probably makes more money
than "asj". Software is all about functionality. Faster,
full-featured software always sells better that portable but
feature-limited counterpart. Nobody buys applications with
a thought that it will need to be ported sometime in the future.
If you are speaking of custom software then people buy
hardware+software solution. In this combination they want
software to get the most out of the hardware. If the hardware
is specialized then software will likely be non-portable.
Better use of underlying hardware also makes software more
efficient. That means better return on investment. That
means they can shorten the upgrade cycle and keep up
with the technology better.

Besides, what makes you think that current Java API won't fade away
or become obsolete 10 year from now?

 

[Brandon Blackmoor]

From that I conclude that "username" probably makes more
money than "asj".

From that I conclude that you don't work with complex server-side
applications.

 

[username]

so you think that people should not advocate solution that they think
are GOOD?

no, but you seem to be running a large campaign against everything that is
not java. I would say that a product that is already marketleader (which you
are claiming) does not need a campaign like that.

or perhaps you wish that people should advocate BAD solutions?

no, what a strange idea.

ahhhh!!!! of course! but giving them a long term solution to their
problems would obviously be preferable to screwing them sometime along
the line.

ask your user (after you told him how expensive your solution is)

for example, writing a client application using a proprietary language
that "faded" away or was made into a "legacy" technology would certainly
cause some problems down the road if the client wanted some upgrades
done...perhaps there would not be any available programmers to do the
modifications, or perhaps the few available would be very
expensive...then the customer would have to recreate the wheel, as it
were, and create a brand new app that did almost the same thing!

same holds for java.

or what happens if the customer upgrades his hardware or system in
future and needs to port the application over to the new system? what if
that proprietary app could NOT be ported over?

you are assuming all future hardware will have java JVM's ..... think about
it

with a standards based,
cross-platform, widely-used technology like java, the chances are it
would not cause that much headaches moving the app over to the new
system.

think again

 

[Carlos Bazzarella]

: Here is a perspective from a devolted Handheld Java developer for many
: years. If you doubt me, check out our web site and apps.

Where are your web site or apps?

For Handheld devices :

http://www.poliplus.com/handheldproducts.htm

For cell phones :

http://www.formulae1.com

Carlos.

 

[Carlos Bazzarella]

use his email address. pretty small outfit looks like. but again, he
makes some good points.

Are there any 'big outfits' developing Java based software
for handheld/cell phones devices ? That is another sign
that reality doesn't match the hype...

Carlos.

 

[Tim Tyler]

: Here is a perspective from a devolted Handheld Java developer for many
: years. If you doubt me, check out our web site and apps.

Where are your web site or apps?

 

[asj]

now we actually get a post that makes some points!

there have been some BIG issues in MIDP v1 about compatibility with the
implementations on various handsets...this was a valid complaint (and
one i also posted and wrote about). however, this was only the first
version of MIDP, and the fact that there are so many devices (all from
diff manufacturers and with many diff capabilities) only created bigger
problems...

the many vendors and manufacturers have gotten together and have partly
adddressed this issue by releasing an architectural roadmap for future
developments of wireless java that will help guide companies and
developers.
http://www.jcp.org/en/jsr/detail?id=185

with the recent release of MIDP 2, there has been some notable
improvements in its capabilities as well. this will only get better as
time passes.

remember too, midp is targeted at LIMITED devices, the very ubiquitous
cellphones on the medium and high end range...it was never meant to run
on high end handhelds with much greater processing power.

i would also point out that arguments about the portability of midp
among various devices is sometime exagerated, or are made by people who
tend to use APIs which the manufacturer warns are device specific.
again, i will note the notion that "democracy is not perfect, but it's
the best solution out there." in the same way, "java is not perfectly
portable, BUT it's BY FAR the best solution out there"...try running
windows ce apps on a symbian or linux phone and you'll see what i mean.

finally, the success of java on small devices is obviously not
completely certain (in life, what is?), but it is definitely MUCH MUCH
more certain than the success of microsoft technologies, simply because
Java THRIVES in a highly diversified, highly competitve environment,
which describes the handset/smartphone area nicely, and somewhat the
handheld arena.

if you don't think microsoft technologies are in danger of faltering,
note that microsoft smartphones have NOT sold well at all, and even
advocacy sites for these are VERY frustrated about it:

"Smart phones continue to accelerate, but Microsoft smartphone still
fails"
http://www.msmobiles.com/news.php/1104.html

"Cardinal errors of Microsoft in cell phone industry"
http://www.msmobiles.com/article.php/34.html

In fact, handset makers have been forced to add MIDP capabilities to
their microsoft smartphones, simply because carriers ask for it.

 

[Carlos Bazzarella]

is SHARP big enough? what about SAP? what about IBM? what about SEGA?
what about RIM?

I am talking about big Software developers and not hardware makers.
Hardware makers jumped on Java like an Elephant on a mouse just to
protect themselves from the evil empire (Microsoft) and they also
figured more software would be written for their devices since they
weren't going anywhere with C/C++ and their proprietary APIs (talking
about RIM here).

Now let's see the big hardware players.

Sharp, recently dumped Jeode (amazing implementation of Java) for
Sun's J2ME Personal Profile (on latest C760 device). In the process
not one single previous Java application can run on this device.
Whatever happened to Java's amazing capabilities.

SAP, no comment since the discussion here is for Handhelds and cell phones.
Same for Sega.

Now IBM has J9 (they recently changed the name though) and their first
deal was with Palm for future Tungsten. So all of this is future things
so there is also nothing to comment right now. We'll see what we, developers,
can do with their JVM on Palm's future Tungsten.

For RIM there isn't still that many Java based apps out there, not
to mention that RIM is not very happy about MIDP 2.0 and so far I
don't think they have committed to it. Another blow to Java don't
you think ?


Carlos.

 

[asj]

no, but you seem to be running a large campaign against everything that is
not java. I would say that a product that is already marketleader (which you
are claiming) does not need a campaign like that.

(1) sorry, i do not have to claim anything. in the realm of handsets and
many small devices like smartcards, the numbers speak for themselves.

(2) you must not have any marketing experience - just because you are
doing well does NOT mean you just sit back and watch the parade flow
by...it would be like coke stopping all advertising and marketing
dollars "just because they're on top".

ask your user (after you told him how expensive your solution is)

ah! but this presupposes that java is a MORE EXPENSIVE solution. in
fact, java is firmly aligned on the open source side, and there are TONS
of FREE tools that will always give you an ROI that is less than any
completely proprietary solution!

in this way, the COST to you and your client can be MUCH LESS than
using, say, microsoft technologies, where microsoft tries to gouge every
penny it can from developers and users in its quest to continue growing
fast enough to justify its high stock valuations.

for example, we could create a robust J2EE solution by using the open
source web app server JBOSS, the servlet/jsp container Apache Tomcat
(also open source), running on open source Linux server (the fastest
growing server OS in the world) with apache (the dominant web server in
the world by far). Total cost of software components: $0...ZERO!

and you'll create a solution that most likely is more stable, secure,
and robust than a much more expensive microsoft solution using
security-challenged software like IIS and windows 2000.

same holds for java.

since java's chances of success are much greater than a proprietary
technology with only a limited number of programmers, then how could
this be the same for java?

in fact, look at how this microsoft developer is complaining about how
the frequent churning of microsoft technologies has left him tired and
dispirited:
http://www.angrycoder.com/article.aspx?cid=1&y=2003&m=7&d=17

the fact is, 10 years along the road, it'll be much easier to get a java
programmer, than say, a microsoft COM developer, or microsoft J++
developer, or even a microsoft VB programmer, simply because these have
now become legacy...again, java does not CHURN, proprietary technolgies
whose owners just want more dinero tend to churn a lot, to the detriment
of users and developers alike.

you are assuming all future hardware will have java JVM's ..... think about
it

well, it sure is in a hell lot more hardware than anything out there.
like is said, it won't be in everything, but nothing elese will come
close - certainly not microsoft windows ce (unless you're really into
stripped down embedded devices).

Think again

uh....it look like you have minor real world experience with java so
short comment like this actually make no sense. try adding some
elaboration and examples.

 

[asj]

Where are your web site or apps?

use his email address. pretty small outfit looks like. but again, he
makes some good points.

 

[username]

(1) sorry, i do not have to claim anything. in the realm of handsets and
many small devices like smartcards, the numbers speak for themselves.

(2) you must not have any marketing experience - just because you are
doing well does NOT mean you just sit back and watch the parade flow
by...it would be like coke stopping all advertising and marketing
dollars "just because they're on top".

I will ignore these 2 statements, as they seem totally unrelated, and they
are not responses to my points.

 

[asj]

I will ignore these 2 statements, as they seem totally unrelated, and they
are not responses to my points.

yes, please do run away when you suddenly find yourself forced into an
indefensible position.

and what happened to the rest of the post? you know, where i show that
java can actually give you an ROI (return on investment) less than
microsoft technology? or that even microsoft developers are complaining
about being on an endless treadmill as it continues to churn away its
technology?

 

[asj]

Are there any 'big outfits' developing Java based software
for handheld/cell phones devices ? That is another sign
that reality doesn't match the hype...

is SHARP big enough? what about SAP? what about IBM? what about SEGA?
what about RIM?

as the market matures, i'm sure consolidation of the various and many
small players will occur and result in a few very large software houses.
that's just how life works.