Dear Palm, it's just not working out | Hacker News Reader

Dear Palm, it's just not working out

jwz.livejournal.com

115 pointsblasdel73 comments

It's too bad, I thought the Pre's card metaphor, along with its pervasive notifications and multi-input open-API contact/message/lifestream datastore was the best hope yet for a great multitasking handheld.

The Verizon Droid is the only hope for an iPhone competitor at this point. It looks like it fixes all the hardware issues with the previous devices:

  * shipping the latest virgin Google build — no carrier or manufacturer fuckery
  * CPU + RAM as good or better than everything else (600mhz OMAP3 + 256mb RAM)
  * a capacitive touchscreen with 3x the iPhone's resolution
  * only a bit thicker than a current iPhone
  * a motherfucking headphone jack
  * nice hardware (landscape) and software (portrait) keyboards
  * 16gb of built-in flash AND a micro-SD slot
  * a camera that they made a cursory attempt to make decent (a flash!)

Now the only problem will be the Android software — the spartan builtin apps, the craptacular APIs, and the abysmal/non-existent third-party apps. Hopefully that will all improve quickly once Verizon is shipping millions of the things. Having hardware that's not worse than the iPhone will be a huge help.

WebOS, with its cards metaphor, is extremely impressive. The Pre is not. When Palm gets the hardware fixed/updated and the software optimized, I think they will have a very good Blackberry/iPhone competitor. The interaction and interface design in WebOS is great, and in some cases better than the iPhone. And RIM has shown that it is incapable of developing a good UI for touchscreen devices.

All the things you list above are hardware features. None of those help the fact that Android is kind of crappy.

The only thing that really sucks about the Pre's hardware is the build quality. The battery life sucks because the software is too slow, wasting CPU.

You're right that I list only hardware features, save the point about the unmolested software. Android is kind of mediocre -- it's way too much like the Sidekick platform with a Windows Mobile strategy. I think it can get a lot better, but only if it gains a lot of users on good open hardware, where they normally get continuous updates direct from upstream and can try out experimental software on their own. It's the same with Apps -- Verizon selling millions of these without locking them down in their usual capricious manner could really get some high-quality shit in the app store.

Oh come on. It's easy to piss on stuff that could be better, but Android is actually a fine platform to develop for, and plenty of applications are showing up. Sure, not as many as iPhone, yet, but there are some pretty cool ones out there.

And it's not "the only hope" - there will be more Android phones out, and the whole thing will only continue to improve.

There are some huge holes in the Android APIs that make writing well-produced games extremely challenging. The Audio APIs are particularly shitty, making it basically impossible to have worthwhile soundfx. I just looked and it's now finally possible to read incoming Audio without waiting for the user to finish recording! (though they still don't call it streaming)

That you're stuck in their unjustifiably-oddball bytecode-interpreter sandbox (no JIT!) makes it particularly hard to just implement things yourself in 'userspace'.

People have been saying "there will be a glut of Android phones out in a few months" for well over a year. This is the first Android hardware that isn't obviously a pile of shit -- it's actually better than the iPhone in many ways! Now we get to see if the software can live up to it.

> That you're stuck in their unjustifiably-oddball bytecode-interpreter sandbox (no JIT!)

It's pretty evident if you think about it: they wanted something open source under a liberal license (Apache/BSD rather than GPL), and so did not want to touch Sun's Java implementation. Therefore, they did their own thing, which is a much solution than hassling with Sun about licensing. I suspect we'll see performance improvements with time; it's not as if Google doesn't have the resources to create a JIT if/when they want.

For low-level performance-critical stuff you can always go down to C++, though I believe that's limited to operations on your own internal data and does not extend to system-level programming.

Fake Steve Jobs illustrated the biggest Android problem: When your operating system is designed to be modified for each individual phone, and each phone runs a modified version of the system, then you can't reliably develop for it in the way you can for the iPhone.

If I make an iPhone app, then it'll work for every person with an iPhone and every person with an iPod touch. The same's not true for Android, which is why developers are reporting failures with the applications they're designing for it.

I've never used Android, unfortunately—can't find people with the phone to irritate/mug—but I'd imagine it's good, and I like that Google's released an OS with some focus on being good, because cell phone companies showed in the last decade that they don't give a damn about good when left to their own devices. Where I don't have faith in it, though, is in its attempt to make a developer haven that the iPhone doesn't offer. It seems that there're integral problems to its developer experience that won't fix themselves easily.

It's a problem, but I don't think it's an impossible one, and nor do I think the answer is that everyone everywhere should have the exact same phone. The Android guys, who are not stupid, certainly have J2ME to look at as an example of what could go wrong. And, to tell the truth, even J2ME is certainly painful once you try and taret a wide number of devices, but not impossible.

On the other hand, it shouldn't be too hard really. People have been developing desktop apps for myriad resolutions for years.

Yes, but we're talking very small screens here, with very little room for design. When you can design any application to run at 800x600, then maybe you don't fill every screen but you have enough space to do what you want. When you're dealing with very small screens, some of which are multitouch and some of which aren't, some of which have physical keyboards and some of which don't, then you have to juggle a lot of variables and the quality of your design suffers.

I shouldn't have to point out the level of build in the iPhone apps gallery. Those applications are gorgeous. Better than most PC apps, I'd argue, in their compactness and elegance. Developers have an edge on the iPhone because there's an established design standard, and there's an established screen size. So they can spend as long as they need fiddling to make their app perfect within those confines. On Android, that's not possible.

To use an unnecessary metaphor, iPhone apps are haiku and Android apps are open verse. On the one hand, it's nice not to have limits, but on the other, you find it's much easier to make something beautiful when you're given set restraints with which to riff on.

The point is not to try and out-Apple Apple - iPhone will probably always be prettier. The point is to be 'good enough' while offering the flexibility, variety and freedom that Apple don't give you.

I guess that's what makes it hard for me to comment on this. I don't really care much for flexibility or variety or freedom. I want something to work as well as it possibly can, and don't worry about tweaking it unless the tweaks make it run better. So Android doesn't interest me at all.

> unless the tweaks make it run better

In a sense all software usage is just a series of tweaks. Otherwise you'd just have a big button to push and it would 'do stuff'. If that 'stuff' just happened to be exactly what you wanted then that would be fantastic. But that isn't reality. Reality is sometimes messy and complex so sometimes you need the ability to tweak things deal with the messy and complex problems from the real world.

Which is why I love my Mac so much. Most of what I need it does without my worrying about. At the same time, I can rip it up and change anything I'd like to if I want, and there are communities to help me get it changed. When I do need a tweak, it rarely takes me long at all.

Oh god. You aren't a real person are you? You're an Apple promo bot.

> I don't really care much for flexibility or variety or freedom

Phones definitely fall into the sphere of things I like to hack on. That's definitely the crux of the difference of opinions. I would agree with you when talking about, say, cars, but a computing device that I physically own is something I want access to.

and that (along with slightly minor platform differences) has hurt JavaME so much it's never grew to its full potential... (if it ever had any)

>I've never used Android, unfortunately—can't find people with the phone to irritate/mug

unalone, will you be at Startup School this Saturday? If so, you're welcome to borrow my Android Dev Phone for the day to play around with. I rotate between that phone, an iPhone, and a Nokia E75 depending on what I'm working on.

(back on topic) Considering the amount of attention that jwz gets, someone on the Android team should send him a free phone to tinker with asap.

I would have loved to attend, but I'm on the East Coast.

(Completely off topic) Dystopia, huh? That's what I named my tank back when I was in the army. See you at startup school.

That's a good name for a tank! "The imperfect future is coming right at you!" :-p

Ah, people just thought I was weird. But they thought that anyway.

Besides, in D companies, all tank names have to start with D. People being who they are, 95% were named "Death .*" I thought "a place of utter misery and wretchedness" would be pretty accurate for just about anybody participating in a battle, regardless of side. I never got any closer than training, and even that sucked quite badly.

Yeah, until the next tech refresh for the iPhone. So then you have iPod Touch 1st and 2nd gen, iPhone 2G, iPhone 3G and 3GS. Bear in mind the 3GS has a more advanced graphics card, and then you need to bear in mind some of these have compass, camera, GPS, etc. That's not a small target by any means.

I seriously wonder what will happen when Apple upgrades the screen resolution to 800x400 or attempts to fit in new functionality such as multitasking without altering the current applications' VMs.

I can run any application on my 1st gen touch. Some things run a bit more slowly for me, but they work. Most things work perfectly. Games sometimes exceed me, but they still look perfect on the screen.

Apple's not stupid. If they up the screen resolution, and I'm not so sure they will any time soon, they'll do it in a way that maintains aspect ratio and offer a way for older apps to adjust. As things stand right now, the resolution is brilliant. Not much reason yet to up it any further.

the point here is that the iPhone is a lot less 'fragmentable' than Android since it's controlled by one single manufacturer. Sure, it will eventually change features so much that apps made for the latest hardware will no longer work on previous ones - but it will probably be a lot easier to manage than having a thousand different versions of the hardware, all being released at once...

Plus, AAPL's "throw away and buy the shiny new one" ethos has kept the majority of users constantly updating their already new ipod/iphone/mac for quite some time, so this fragmentation might as well be yet another strong point on their evil marketing plot. "1st gen iphone? Sheesh, that's too old, go get a new one"

Is that really the ethos you see Apple emanating? Because I've always seen the opposite. One of the things that I thought coolest about Apple back when I was on Windows were the people who talked about installing Leopard on a Powerbook and having it still work perfectly. Ditto the iPhone now. In fact, the line I hear directly from Apple is "Because we have no physical controls, we're allowed to keep our 1st gen iPhones in line with our freshest models, so the iPhone will outlast many of the other phones its age." Certainly my iPod is showing no strain in its second year. I haven't even upgraded to 3.0 and it still feels great.

Absolutely agreed on the lag. My fiancee has the Pre and the experience has been going downhill from day 1, worse and worse each day.

The lag is absolutely unbearable and one thing jwz forgot to mention is the shit battery life (< 1/2 day).

She just uses her work old blackberry now with like a 5 day battery life. It gets the job done.

I can't agree on jwz's Android opinion based on some random baseless rumors. I have an Android phone and it's a lot better than Pre (almost no comparison) and to me it's better than the iPhone. With Droid's muscle, it will be the best phone (unfortunately it won't be available for Sprint).

It's a quality control issue for sure. I got a pre on day 1 that just had its touchscreen stop working. Sprint replaced it with no questions asked.

The performance on this new one is better in every way:

2 Day battery life (up from 1/2 day)

Many cards run at once with much less lag

The slider actually feels solid

ironically, lag was one of the reasons i ditched my iphone. bringing up the camera application would take 5 seconds to be able to take a picture, apps would take a long time to start and then end up crashing shortly after.

now i'm on an android phone (the t-mobile mytouch) and lag is there too. a phone call comes in and it's slow to respond. the web browser is painfully slow to navigate and clearly chokes on javascript-heavy sites.

i think android and webos will get better over time, but more focus needs to be put on responsiveness. even if the phones don't have all the whiz-bang capabilities, if they are quick to respond to the user, they will at least not frustrate them.

JWZ isn't being dramatic. The slogan should be "Plam Pre: your life, just slower". I had a Pre for a month, returned it on day 30. It was like living in molasses.

Plus the hardware was way way too cheap (fragile plastic screen, sharp edges on the keytray, poor fit between the sliding halves). I was afraid to use the phone for fear of breaking it. You could hear the plastic crunching everytime you typed on the keyboard.

I was afraid to use the phone for fear of breaking it. You could hear the plastic crunching everytime you typed on the keyboard.

O/T: I'd like a concise term for this "author switches from 'I' to 'you' and ends up suggesting the reader's involvement where there couldn't have been any. (This reader couldn't hear the author's phone crunching).

It's like the reverse of the "royal we". Maybe the "common you" (meaning: I).

Hopefully after this drawing of attention to it, you will see and hear it everywhere and you, er, I can promptly forget about it.

http://en.wikipedia.org/wiki/Generic_you

I love my Palm Pre, but I agree, the calendar load time is really absurd. One of the things I loved about my old Palm Treo was being able to be out and about and quickly add an event to it when I came across an interesting flyer or article about something that was happening. It's such a royal pain on the Pre.

The phone features on the Pre are horrible and utterly unacceptable on the Pre -- you often can't conference two calls together, and if there are two calls on the line (conferenced or not), you can't hang up only one of them.

But I still think it's a fantastic device overall.

Yeah, I'd agree. I don't find it slow for a smartphone other than the Calendar app, which is abysmal. I wonder if he did something to the unit when mucking around with it (I know I've applied a couple hacks that really messed it up) and just needs an OS Doctor reinstall.

I haven't tried the Pre myself, but if jwz's experiences aren't unique (most likely), it's really a shame. The specs don't look so bad- is it webOS dragging it down?

It looks like Motorola's Droid is the best hope at this point to drive some innovatin' by competing at the level of the iPhone.

Or is there another contender I've missed? Is the HTC Hero/G2 worth looking at?

If all webOS applications are coded in JavaScript (so I've heard; didn't look into it too much), the VM overhead could well be dragging things down. Even V8 is ten times as slow as Java or C#, and v8 isn't even available for ARM.

Yeah, the problem is that every single app is written in JavaScript. It works better if you open the apps you use frequently and just leave them open. Then the page remains rendered and is much faster.

Interestingly, it would be possible to code native apps on it via browser plugins. Some parts of WebOS do use this it seems, but only to interact directly with the hardware.

I'm really disappointed to learn this. I've been really curious as to how Palm succeeded to make their apps based on javascript so fast as to be even comparable with iPhone apps.

Turns out they haven't, they've just got really low standards as to what's good enough to ship. Really sad, I was hoping they'd be able to offer some strong competition for Apple.

Why does everyone think Droid is the only hope? Nokia's Maemo-based N900 looks promising as well.

Their latest Symbian/series60 phones (e.g. E71, E75) are also pretty decent nowadays, although I would compare those to Blackberry rather than the iPhone.

The N900 is full of cool hardware but it's 2cm thick, will cost $700+, and is unlikely to ever be available subsidized in the US.

Edit: http://gigaom.com/2009/10/19/nokia-n900/ - An early review

Nokia is doing a full rewrite of Maemo for the upteenth time after buying Qt — the software platform is constantly going in different directions. They're so proud of shipping a Mozilla browser, but how the fuck are they going to integrate that with Qt? Don't they also maintain multiple Webkit ports for their other platforms? The mind boggles...

They're so proud of shipping a Mozilla browser, but how the fuck are they going to integrate that with Qt?

I don't see the contradiction there. Mozilla and Qt are at different levels in the GUI stack. (KDE is a Qt-based desktop, yet it runs Firefox just fine.)

You have 256mb of RAM to work with, most of which you would like to use for rendering webpages. Do you really want to blow a huge chunk of it on something that just gets you an inconsistent look-and-feel? A handheld device is the last place you want two completely different file-picker dialogs, and you can't cover that shit up with theming.

Do you really want to be shipping second-party Gecko in the when Webkit is better and you're already paying people to work on it?

A handheld device is the last place you want two completely different file-picker dialogs, and you can't cover that shit up with theming.

What does Mozilla have to do with file picker dialogs? Firefox uses native dialogs on the platforms it runs on, and so does Nokia's Gecko-based browser on Maemo.

Do you really want to be shipping second-party Gecko in the when Webkit is better and you're already paying people to work on it?

What's preventing them from switching to a WebKit-based browser when it's ready for production? Why should they have tried to squeeze that into the Maemo 5 dev cycle, when they already have a fine browser?

I don't understand your argument at all, because OS and device vendors switch browsers all the time as new alternatives become available. Mac OS X 10.2 still shipped with Internet Explorer 5 as the default browser...

Firefox uses native dialogs on the platforms it runs on

All three of them -- Win32, Carbon, and GTK+

What's preventing them from switching to a WebKit-based browser when it's ready for production?

Inertia is a bitch

They're so proud of shipping a Mozilla browser, but how the fuck are they going to integrate that with Qt?

http://arstechnica.com/open-source/news/2008/08/nokia-helps-...

Says that they did the easy thing, mostly, and wrote a Qt backend for Cairo.

It also states that Firefox/Qt was usable over a year ago, but needed some more work on stability before it would be ready.

I don't see it as being a major obstacle to them at this point...

N900 is a great platform, free as everyone is asking. The point is that is not for US because of the subsiding US people is looking to have. 700$ is the same price of an iPhone, the two are on the same league (more or less depending on the feature) but it seems that the marketing of 199$ + n hundreds $ in 2 years contract wins. The only reason I can imagine (not being in US) for this being reasonable is that there is no reasonable dataplan without phone "embedded in". S60 is OK from the user point of view (barring the UI from 1990), phones are cheap given the features, but programming for it is a nightmare...

Yes, I agree totally that S60 as a programming platform is horrible. I hope Nokia has learned the lesson with Maemo. Because it's based on a Linux distribution which is not that far from your standard desktop Linux, maybe they just aren't able screw that up too badly. To Nokia's credit they are not the open software development freedom-haters that Apple and maybe Palm now too have become.

OTOH, I don't think that Nokia really get open source in the same way that Google does. Doubtless, they have people that do, and their circumstances are quite different: Android can't be doing anything but losing money for Google, and is not really their core business anyway, whereas for Nokia, opening up their crown jewels cannot be an easy process. And Nokia has always been a bit ambivalent about open source. Google seems much clearer about what is and what is not going to be open source, which makes it a little bit easier to predict what they'll do and how they'll do it.

I agree about the poor performance. I really hope they can fix that somehow, but it's starting to look like the heavy reliance on javascript was a design mistake.

However I don't have any problems with Mac sync. I've always used Spanning Sync to keep my Mac synced up with my Google account. Once there, the Pre just finds everything automagically.

It has been a bit of a shift in thinking to go from the old model of "I need to sync my device with my Mac" to the new "My device will just find what it needs in the cloud". But after using it for a few months, I'm really happy with the change. Palm synergy just works, and I don't at all miss the old way of doing things (cross your fingers and hit the hotsync button).

"but it's starting to look like the heavy reliance on javascript was a design mistake."

They can always compile it to binary code. Or drop in a faster interpreter, a persistent JIT perhaps. Or even something that spits bytecode to files as needed (like Python)

I got my Pre last week. I really like the UI and the idea of multitasking. However, I already sent it back because of the spacebar and the power button not working properly (aparrently not only an issue with my Pre). From my experience from using it for a few days all I can say is that the battery life really sucks. You can't get through a day with moderate use of email, twitter, phone...

I am also a little disappointed by the build quality. It just does not feel smooth when you slide out the keyboard. I, however, did not experience any issues with the calendar. It takes some seconds for every app to load but I did not see a big difference in the loading time of the calendar and the other apps.

Now I have to wait two weeks before I will get it back since they would not just give me a new one. Reading this article really makes me wonder if I made a bad decision buying the Pre.

In my experience with mobile development, disk I/O is especially slow. So the calendar load time might depend on how much data you have saved in your calendar.

The Pre needs a serious software upgrade. The processor itself should be fast enough, the application layer is a bit too webby, so probably wastes a few bogomips with all those layers -- but still, with some clever load optimizing and caching, at least launch time should be faster (and/or appear to be faster).

I just got one yesterday, and both speed and build quality are slightly disappointing. One of the main reasons I got one was due to the fact that it's much cheaper on a monthly basis (well, here in Germany), and I thought about having another mobile platform to develop for (my iPod touch should suffice for most iPhone dev work I'm going to do).

But the app catalog is somewhat disappointing. That should've been settled before the device was released, especially if you want to be compared to the iPhone.

I had some of the same issues. Happily (sadly?) I had them so badly and soon that I was able to return the Pre for a refund. Details here:

http://kylecordes.com/2009/06/06/palm-pre-first-impressions/

Not sure why suddenly jwz is considered to be a consumer electronics guru? There's always churn and any particular device isn't for everyone. He's just very vocal about it.

The Pre suffers from some slowdown issues which as some commenters point out is almost certainly a software issue, which means it's fixable :) Would like to see Palm take up the community created patches, but it's the wealth and quality of these open modifications (tweaks/patches) as well as the rate of new additions to the app catalog that leave me hopeful. It isn't gonna be a market buster, but I think webos is fairly solidly staying around.

Not sure why suddenly jwz is considered to be a consumer electronics guru?

You're right, it's shameful that we listen to a guy who might not even have filled in all the application forms to become an officially recognised certified consumer electronics guru :-)

Eh, I meant exactly what I said. An opinion from a person is an opinion from a person. There are ways to become an expert in something. I'd be stupid not to listen to at least some of the things he might say about browser design.

This isn't about empty credentials but about track record. Dude has demonstrated to do X,Y, Z well. I will value his opinion on such matters highly. But just because of XYZ, valuing his opinion on Q doesn't make sense to me. Because most people don't have XYZ, perhaps I'll weight his opinion on Q a tad higher than the average, but not much more than that.

Does this not make sense?

Edit: I should add that any and all complaints of his are valid. Some of them are objectively accurate (loading the calendar app is slow) some of them are subjective. The issue is that both kinds seem to be weighted by people to be highly representative.

This isn't about empty credentials but about track record.

His track record is that he has used the device, that he has written software for it, and that he helped investigate the issue. He says the device is slow, and he says what he considers slow in terms of seconds.

I don't think reporting a simple observation like that or paying attention to someone reporting it lays claim to any kind of usability expert status. He credibly said Q. I don't know what that has to do with X, Y or Z at all.

If someone tells me the traffic light is red, I don't reply by asking whether he's a traffic lights expert.

But if you're just expressing general astonishment about waves of popularity and public attention, I'm with you. It's futile to resist that though. I think the only way to deal with it is to judge each statement on its own merit.

I have had similar experiences with my Pre, but FWIW not consistently. As expected, the more apps left running load times for new apps definitely increase. I tend to keep most apps closed when not actively using the phone and rarely experience more than a one or two second delay. For example, I just took a picture and capture to view was less then 5 seconds. I closed the camera app as a phone call came in and saw the keypad immediately. On the few occasions where the phone has started creeping, I popped out the battery. Once the OS reloaded all was smooth again.

I just had a play with this in a local shop. The bottom of the pre keypad is knife like in its sharpness and the interface is cricket ball like in its sharpness.

For a guy complaining about iphone and palm, he sure fails at a good argument against the G1--it's probably what he actually wants.

Hm, I would've taken jwz for a G1 kinda guy...

As someone who'd previously been on the business end of subpoenas for data and legal discovery during the Netscape lawsuits, he's paranoid about giving any of his data to Google.

That's why he was trying to get his Pre to sync solely to his local computer, despite that really not being what its makers intended.

I am curious. As plenty of business people use gmail/docs, what are the subpoena implications of that?

I am sure the first case with a tech-clueless judge and an overly broad subpoena will be hilarious. The subpoena will probably bring back every document they have on the service since google will not likely be able to spend the time / money or have the obligation / right to edit the request.

Me too. But look at where he lives: SOMA SF is the heart of iPhone / Apple country.

That's disappointing to hear, but sort of inline with what I expected.

Too bad, looked nice :)

my pre has been pretty snippy and fast most of the time, but was getting progressively slower and doing weird email syncs. after a wipe and reinstall the device its back to being fairly fast.

I love my Pre. It's not perfect, and I've given up writing an app for it because of Palm's crippling of the O/S, but as a phone it's cute and works.

Sure load times suck. I just keep frequently-used apps open all of the time, then I don't have to worry about it.

They still think it is possible to write apps in javascript? Good luck!

Actually you're probably right on the mark..

Acceleroto's blog post about their experience with developing the Pre version of Air Hockey tells the full story:

"Memory management (aka Garbage Collection). WebOS is allows users to run multiple apps at once. Users beg for multitasking. You think it impacts game performance? Absolutely. Even if every app behaves nicely, they all consume memory. Memory management is handled by the OS without any direct say of the individual applications. Whenever the OS decides, it does a Garbage Collection. Currently, this takes about 300ms (about a third of a second). Air Hockey contains specific code to lessen the impact of these pauses, but developers can’t turn them off. A lot of time was spent making sure objects weren’t needlessly allocated to minimize the frequency of the Garbage Collections, but you’ll see the pauses. That may change in the future – we don’t know."

http://blog.acceleroto.com/2009/10/06/the-first-paid-app-is-...

Stop-the-world? Really??

Shameful in this day and age especially for the slick multimedia experience they are selling the Pre as.

Reply on news.ycombinator.com