Skype Arrives on the Palm OS

Surur

They said I only argued for the sake of arguing, but after an hour I convinced them they were wrong...

No crashes due to multitasking taking up too much of system resources.

No problems with alarms.

No playing 'russian roulette' with ActiveSync.

If your masters to fix these problems, the I may switch to a WinCE device. Until then, you Propagandists have a 'hard sell' on your hands.

I tried it with mu T|C. It runs and install fine over WiFi, but it still needs:

1) a Windows PC running all time, with Skype also running
2) minutes on SkypeOut

It seems you cannot run it over WiFi "for free". If you don't have enough funds on Skypeout, it won't allow the call.

Last I checked, there was a pretty huge market of Palm OS5 owners. I don't quite understand why developers are ignoring this market. I personally could care less about Skype, but I'm amazed and disappointed that we don't have more and better web browser options for OS5. The Treo 650 is still selling strong and the rumored 700p will still be running OS5 and will be around for another 12 months. Why does Opera ignore this market? Why have the owners of Xiino given up improving upon their product? I just don't get it.

http://Tapland.com
- Tapwave Zodiac News, Reviews, & Discussion -

>> 1) a Windows PC running all time, with Skype also running
>> 2) minutes on SkypeOut

>> It seems you cannot run it over WiFi "for free". If you don't >> have enough funds on Skypeout, it won't allow the call.

Hmmm. Based on what I read on the EQO site - you can't use this just over Wifi - it requires a cellular voice connection.

Here is how i understand it from looking at their site (which dumbs down the details to the point where it is difficult to say exactly how it works):
- when calls are made using the phone java client, the client first signals to the EQO service (via cell data network) to provide the 'presence' info and basically forward the call details.
- the client app then initiates the voice call over the standard cellular voice channel as normal, but the client app actually calls to a local skype-in number (to your skype-in number? a local code EQO in?).
- the plugin recieves the signaling data from the EQO service/network and based on this, it effectively uses skype to re-route the (now purely ip) call (incoming to your PC via Skype-in) to either a skype-out number or a skype pc client.

A bit of a simplification - but that is basically how it works as I understand it.
Does this jive with other peoples understanding of how this works? This is just my half-baked theory based on what I read they don't fully explain everything (this simple diagram just shows the voice call going into the 'skype service' - it doesn't explain how: http://www.eqo.com/download_how.html ).

Also, obviously the EQO Skype IM component is not as much of a big deal: it purely uses the cellular data network (I don't think that it interacts with - or at least there is technically no need to - the PC skype in anyway).

It is difficult for 2 main reasons: on desktop platforms and on Windows Mobile you have nice high-level APIs/objects that do a lot of work for you AND becuase you have multi-threading.
But 'bordering on impossible' is an exageration. More like "making enough money to justify the development effort is bordering on the impossible.":)
There is no technical reason why it cannot be done - but i can understand why it doesn't make sense given the extra effort. There are several browsers for PalmOS and while they are not as good as Opera or others, they do show that building a browser that is capable of rendoring HTML, xHTML, WML, JavaScript, CSS, and SSL is possible. Really, i wish there was a greater focus on improving existing browsers such as Blazer. There is a lot that could be done - including some rather obvious things that would greatly bring these 'up to scratch'.

Do you really think it's going to be that soon? I thought we were going to stuck with WinMob for the next-gen Treos...

This sig is a placeholder till I come up with something good

Maybe. Maybe not. Actually, I have it on pretty good information that this year's Treos are the end of Garnet and that the next round will be Linux--as early as Fall 2007. As I've mentioned before, you should not assume that Palm will wait for ALP to be complete. They have their own "Palm Linux" framework in development, which might be built on parts of ALP (sans MAX) or possibly on top of MontaVista Linux or some other already complete Linux phone platform.

David Beers
Pikesoft Mobile Computing
Software Everywhere blog
www.pikesoft.com/blog

I resent that. We also use them to check our star signs.

This sig is a placeholder till I come up with something good

The whole "huge installed based of tens of thousands of apps" is rapidly becoming something of a Pyrric victory for Palm/POS faithful.

Why?

Well, just look at the evolution (?) of OS 5.x aka FrankenGarnet over the past four years. You first had to take OS 4.x apps and make them OS 5 compliant. Then you had stuff needing recoding for 320*320. Then you had the Graffiti 2 switchover that surprisingly didn't cause AS many headaches as I anticipated. Then there was the T3 and HVGA compliance along with BT support if the app/game called for wireless connectivity/beaming. Next came NVFS and 160*160 OS5 devices (Zire 21 & Treo 600). Then came the LD and its wacky memory architecture. And then the TX changed things again slightly from a DIA standpoint (Fitaly again).

So an app that might run fine on am m515 from '02 may not run at all on a T|C from '03 which might not run on a Treo 650 from '04 which might not run on a LD from '05 etc etc.

I'll wager that the never ending parade of reworking/rehacking/retinkering & rebreaking has left such a sour taste in most developers' mouths that unless it's an absolutely top tier program like PTunes or DTG they just say the he11 with it and throw in the towel at a particular point. That's why you are seeing SO precious little Palm software updated/released nowadays. And, of course, the Treo's SSS 320*320 is a blessing as it gives developers ONE MORE excuse not to support HVGA.

And even if you do have a legacy app from 2002 that still works you have to admit it's not much looking at an 8-bit color 160*160 program that doesn't even support the 5-way navigator!

Example: I've had Doom in some form or another (the original Doom mind you) installed on every PC I've owned in the past twelve years. By just tweaking the frontend launcher from DOS to Win3.x to Win9x to NT/2K/XP the same basic program code from 1993 continues to live ad nauseum.

Hopefully whatever Linux solution Access or Palm serves up in '07/'08 will at least be able to semi-elegantly run the older POS software without too much of a performance hit (assuming you are not running a POS emulation process on top of too much else). Advanced options such as OS5/OS4 "compatibility" modes or HVGA "stretching" (think a more robust Code Diver built into the OS)

Pilot 1000-->Pilot 5000-->PalmPilot Pro-->IIIe-->Vx-->m505-->T|T-->T|T2-->T|C-->T|T3-->T|T5-->TX

However, there's still a stopgap solution between the 650 of today and ALP/Palm's Linux solution in '07/'08 and it's called the 700P (assuming it really does come out). Hopefully it'll at least address the 650's main software & RAM shortcomings. Depending on carrier availability, of course, it MAY be enough to hold off the competition until something Linux-powered if in the wings.

VersaMail right now is the biggest shortcoming amongst the apps that Palm ships with their devices. Palm either needs to get VM right or just offer a 3rd party solution on their devices as VM creaks mightily under the demands of HTML e-mails/attachments/multiple accounts on one device.

Pilot 1000-->Pilot 5000-->PalmPilot Pro-->IIIe-->Vx-->m505-->T|T-->T|T2-->T|C-->T|T3-->T|T5-->TX

Pilot 1000-->Pilot 5000-->PalmPilot Pro-->IIIe-->Vx-->m505-->T|T-->T|T2-->T|C-->T|T3-->T|T5-->TX

That requires a following wind, a fair sea, and no bad luck.

As I've mentioned before, you should not assume that Palm will wait for ALP to be complete. They have their own "Palm Linux" framework in development, which might be built on parts of ALP (sans MAX) or possibly on top of MontaVista Linux or some other already complete Linux phone platform.

They bought out Tapwave's IP, didn't they?

POSE on MobiLinux would be sad, I think, but not at all impossible.

May You Live in Interesting Times

Marty

May You Live in Interesting Times

(One of the updates is MxCalc 12c - a financial calculator which emulates the old HP12c - must try this out when I have some spare time.)

That requires a following wind, a fair sea, and no bad luck.

It also requires a 10,000 horsepower engine. Last time I checked, Palm doesn't have one of those. A less ambitious Palm-sponsored "PalmLinux" is coming, but PalmOS 5 Treos can be easily made more stable and milked (along with WindowsMobile Treos) easier than it will be to debug the PalmLinux version. Palm can't afford to pi$$ off carriers by shipping a half-baked Treo - they realize carriers will not accept the kind of crap that's been foisted onto PalmOS PDA consumers for the past few years.

As I've mentioned before, you should not assume that Palm will wait for ALP to be complete. They have their own "Palm Linux" framework in development, which might be built on parts of ALP (sans MAX) or possibly on top of MontaVista Linux or some other already complete Linux phone platform.

They bought out Tapwave's IP, didn't they?

POSE on MobiLinux would be sad, I think, but not at all impossible.

No. Motorola (wisely) bought the TapWave IP and also hired several of their star codemonkeys. If Motorola is smart they would quickly buy (outright) the rights to PalmOS 5 and 6 from Access or at least negotiate a license for PalmOS 5 that allows them to hack the OS in any way they want without having to give back any code. For a (relatively) small investment of say, $100 million, Motorola could end up getting the good parts of PalmSource without all the CRAP they would have been stuck with had they actually bought the company.

I just hope Motorola doesn't end up wasting these resource by releasing another stupid device like Gizmondo or Nokia's embarassing N-Gage.

As usual, Palm was asleep at the wheel. Palm should have bought up TapWave's and HandEra's IP and hired as many of their codemonkeys as possible when it was obvious that they were imploding. These two small companies each had a LOT more talent than Palm has ever seen and could have done a lot to help develop a solid, simple, STABLE PalmOS-compatible platform for Palm. Oh well. At least my bunker is stocked with enough Samsung i500 and Sony CLIEs that I don't have to care that we'll never see another good PalmOS device again...

TVoR