Slashdot Mirror
6 Smartphone Keyboards Compared
Barence writes "A debate that crops up time and again is whether it's better to have a dedicated keyboard on your smartphone or whether an on-screen keyboard with text correction is adequate. Some phones with screen-based keyboards have started to provide tactile feedback, either using an ultra-quick spin of their vibration alert or, like the BlackBerry Storm2, using clever piezo-electric technology to simulate the feel of a button press. But which system works best? PC Pro's Paul Ockendon gathered six of the most popular handsets around and put them through a timed typing test to see which proved quickest and most typo-free."
What debate? You're telling me that there are people who seriously prefer to *not* have a physical keyboard on their smartphone?
I downloaded swype for my Nexus One and haven't looked back. It's so much faster than the old virtual keyboard for "hunt and peck". The videos of it don't do it justice. It's much easier and faster than the old ways.
The author acknowledges that this test is barely scientific, but I'm left wondering why he didn't disclose which phone he actually uses day-to-day. The muscle memory he's built up using his primary smartphone should give a huge bias to the results.
I've been using Swype on my Nexus One for a month or two and it's incredible. I wonder how it'd perform on that chart (Or is one of those phones the windows phone that includes it? "Swype" didn't show up on the article page at all).
Depending on what you're used to on existing devices and who is typing, these results will vary wildly. I'm used to a physical keyboard on my phone so I have trouble whenever I try to use something else. I tried tactile feedback screens once and the vibrations felt funny making me go even slower.
I've got a really flat, sensitive keyboard with repeat all the way up and key delay all the way down and a trackball mouse. Most people that try to type on it or use my mouse can't because the keyboard is too sensitive and they don't know what to do with the mouse (some try to move the whole unit, some just look at it and seem to poke at it). I however can type faster on it than any other keyboard and be precise in even difficult 3D shooters.
These 'tests' really require a decent sample size of users and a decent sample size of devices with said screens. Not everybody implements the on-screen keyboard in the same way either.
Custom electronics and digital signage for your business: www.evcircuits.com
---
Mobile Phone Feed @ Feed Distiller
Keyboarding, even within the context of a high quality desktop keyboard, is not a natural act. I requires nontrivial practice to achieve speed and accuracy with low mental overhead.
Smartphone keyboards are ghastly little things, whose virtues lie more or less exclusively in being small enough to fit on smartphones. Each different one requires substantial practice and much of that practice isn't transferable between systems.
Having one person try them all for a few minutes is line noise, it tells us nothing.
Is measuring how one particular person performs over a limited amount of cellphone keyboards. Could be measuring here more things than just those keyboards. Take it as what it really mean.
They didn't even include any Palm products in their review.. and Palm, being both the original smartphone, and the best (at least in my opinion), deserved to be included.
The crapberry bold that "won" has an imitation of the keyboard on a Palm Treo 680 and similar.. the Pre has a similar keyboard as well.. probably further refined than the one RIM ripped-off for the crapberry bold.
I wonder just where RIM's money comes into the equation of paying the person who wrote the review to pan everyone else's product.. while being sure to mention one OTHER crapberry in order not to look biased.
I'm sorry, but one person typing a short message, is not going to tell us anything significant. Furthermore we don't know his background (e.g. what types of phones/PDA's he's used in the past), or how "fat" his fingers are. At best all we know is what phone he's best at right now. The performance of the same person when they first used the phone compared to that same person after owning that type of phone for a year will differ significantly.
If someone plans to type on their phone enough for the difference of a few seconds to matter, then they really need to compare the phones in person themselves. A significantly larger sample of people ideally who have never used a phone with a full keyboard may give some idea of which styles tend to work better on average, but that's about the most information you'll get. Whether it works best (and is comfortable) for you is something you need to try yourself.
Sadly, PS/2 was yet another victim of USB, which doesn't care what you plug into it, the electrical slut.
There's a pictutre of a table with the results. For anyone who is not able to read the information in the picture, here's what it says:
Fastest (1m44s), with 0 errors: BlackBerry Bold 2 with physical keyboard below screen.
Slowest (3m24s), with most (9) errors: BlackBerry Storm 2 that has an on-screen keyboard with "click feedback".
Second best, both in terms of errors (0) and speed (2m0s) was Nokia E 75 with a physical, slide out keyboard.
iPhone 3 GS with on-screen keyboard took 2m43s and got 2 errors.
Other participants: HTC Hero and HTC Touch HD2, that have on-screen keyboards with "vibrate feedback" (3m1s, 2m41s, respectively, with 4 and 1 error(s), respectively).
screen real estate.
/.) where i am going to be reading and typing with regularity it is nice to have the option to slide out a physical keyboard and get the screen back.
i have a moto droid and while the onscreen keyboard is awesome and the predictive text works great, it takes up about 75% of the screen. if i am on a site (like
Since touch typing on virtual keyboards is impossible and hence one must be looking at the screen at all times, why not just replace them with a more efficient method such as Dasher? The only downsides I can think of is need for a language-specific predictive model (though it could learn from the user and still be more effective than a virtual keyboard) and lack of support for chording, which could make text entry even more efficient.
From TFA:
I used each phone in its default mode, as it would present to a brand-new user out of the box. I counted one error for each wrong word in the main text and for each wrong character in the phone number, web address, username and password. In every test I tried not to look at the screen and typed as quickly as I could, allowing the phone to correct any errors. I’m not the world’s fastest typist, so I’m sure some of you could easily beat the absolute times, but as a comparison between devices it’s reasonably valid.
There's a LOT of use of the word "I" in there. Could it be that he went through an exhaustive process to determine which phone met *gasp* his own personal preferences?
Well, Paul, that's fantastic. In fact, I happen to agree with you. But you haven't settled the "debate" for anyone but yourself. I think most (but NOT ALL) people would likely agree that a hard keyboard is really tough to beat when you want to type in a lot of text. I know typing anything into my wife's iPod Touch is, for my massively meaty paws, an exercise in utter frustration. I think entering anything more than a URL in it should be given a "circle of hell" difficulty level. And I've really honestly tried to make it work. For those apps that support rotation, the wide-format keyboard is just barely adequate, but WHY DOESN'T SAFARI SUPPORT THIS!?!?!?
(breathes) But I digress.
I've seen people who can absolutely whiz-bang on soft keyboards. I don't understand it, but they can. I've also seen people who (believe it or not) do not need to enter any major Tolstoy works into their mobile phone browser on a routine basis. For those people, a hard keyboard is an utter waste of what could be useful screen.
Personally, you can have my Blackberry 8310 smartphone when you pry it from my cold, dead thumbs. Or replace it with a newer Blackberry Curve (oooh! shiny! 3G please!) or something else with a hard keyboard in a similar form factor. I don't like the postage stamp of a screen, but I enter text. A lot. And I need a physical keyboard until voice recognition stops getting me visits from HR when "I like your idea" gets transcribed as "I'll lick you my dear". I also want something durable, and slideouts seem rather breakable in my big meaty paws.
So the wide-candybar format with a postage stamp screen is a reasonable compromise. I've been carrying it for about a year and a half now, and while there's always the "wow, if I could get a bigger screen I could have seen that", there's constantly the "oh, dear, I gotta type a whole paragraph, thank FSM for this real keyboard - wonder twin thumbs, ACTIVATE!" I can type about 1/4 to 1/2 as fast as I can on a desktop keyboard, special characters are just one extra keypress unless they are truly bizarro ones, and it just gets done what I need to get done quickly.
J. L. Slimfingers might be able to throw an iPhone in the air and type "Moby Dick" on it while in flight before it lands. For him, a large screen format is an excellent choice.
D. Elly Catflower might keep it in a shockproof case and only bring it out with great ceremony and lay it on a safety pad of fine Corinthian leather before using it. For her, a slideout is ideal. Lots of screen, full-on keyboard, and they'll treat it right.
Me? Big meaty paws, a tendency to bash it against stuff, and a need to enter a lot of text. I got the Blackberry 8310, put it in a big rubbery slipcover, and put that inside an 8800-style leather case. It's at a year and a half, I don't dread typing on it, and it's still going strong with about 2 days of battery life between charges. I haven't even managed to scratch the screen (though the stick-on screen protector helps). I think I chose well.
For me.
Not for everyone else.
"This post contains words, known to the State of California to cause thought. Wash brain thoroughly after reading."
more screen space
-- if you mod me down, I will become more powerful than you can possibly imagine
I'm very curious how the G1 (HTC Dream) would have compared. The CPU, memory and screen specs are looking a little dated, but from my (albiet limited) observation, the physical keyboard still seem to be a top notch contender. Particularly because of the innate tactile feedback, due to fairly clicky keys, and the perhaps somewhat overlooked factor of the keys being laid out in the standard diagonal orientation, which I've definitely found to be quite advantageous.
I'm not the worlds fastest typist, but I can touch type, and when I began using that phone, the ability to do so seamlessly transferred to that hardware. Which was interesting since completely different digits are involved. (thumbs vs fingers).
Am convinced that the speed bottleneck involved in on-screen keyboards is more due to do placing the users visual system inside the control feedback loop, which is inherently slower and consumes more resources.
I have a Droid. Therefore I don't have to decide if I want to use an on screen keyboard or a conventional keyboard. I have both. I have my cake and I get to eat it too. Although I'm not sure which I prefer, as I use both the on-screen and regular keyboard about equally, I think I'm a little faster with the regular keyboard. But for a short text, sometimes it's more convenient to just use the on-screen. Otherwise, I didn't find this article too worth-while.
I've owned two Nokias with physical keyboards (6800 and 6820) and a BlackBerry (Curve 8330--not the best) and an iPhone, and I prefer the iPhone's virtual keyboard by far. Not so much for speed, though some basic testing by me shows they're all comparable, but for ease. The 6800 is large with plastic between the nicely-rounded keys and it's very easy to hit the right one. The 6820 is a bit smaller but the keys are also nicely rounded and typing on that is pretty easy. Both also have dedicated buttons for numbers and some punctuation--hyphen, comma, period, slash, single quote, and more are all primary buttons. Their layouts also closely mimic a PC keyboard with comma, period, slash, semicolon, quote, and equals in roughly the same spots as on a regular keyboard.
The BlackBerry's keys are smaller and closer together and firmer than either Nokia and I find I've got to press on them with a thumbnail or the bony part of a finger to get them to register and not mash more than one key at a time, and there are no number or punctuation keys AT ALL which makes typing just about anything quite a pain.
The iPhone only shows letters or numbers/punctuation but since it's virtual the secondary and tertiary buttons are big and easy to find, not like the tiny glyphs you get from sticking two images on one physical key. But the thing I like most about virtual keys is that it only takes a very light tough to register a press, and the clickable area is very large, so typing with the biggest, roundest, softest part of your thumb is a cinch. And because of this, it is by far the easiest to use with one hand. (Though the split-keyboard Nokias are pretty much out of the running in this area, but the BB is similar in size and shape.)
But anyway, that's just my experience and preference. All that matters is what works best for you.
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
for typing text
I prefer not to have a keyboard on my smartphone because typing on a tiny keyboard, whether physical or not, is an enormous pain in the ass
Yes-- complete agreement here.
What I really want is a full-size keyboard as a detachable accessory. When I'm using the phone as just a phone, or for most simple browsing, I don't need to carry a keyboard around, but when I do need one, I'd like a full-sized one, not the little toys.
(It doesn't have to have all the useless keys that clutter up most computer keyboards these days-- just the old QWERTY ones.)
http://www.geoffreylandis.com
Well, it was interesting to see the times he posted, but of course everyone here knows that learning a keyboard takes practice. Months of practice.
By the end of it, I could type 100wpm on my Zaurus C760 - almost as fast as with a real keyboard. I could type about 60 on my treo 650. I've got an n900 now, and I've been hanging around 50. So there are substantial variances in keyboard quality and layout (at least on these devices) .. but what was more interesting were the learning curves. My speed doubled on the Treo in the first week. It over doubled on the C760 (slowly, over its 5 year lifespan). But I'm not getting any faster on the n900, sadly.
All of this aside.. the main problem with on-screen keyboards?
They take up half (or more) of your display!
This is one point I'm amazed people are able to get beyond. Sure - it's a rare use case... but terminals are unusable with an on-screen keyboard. More commonly, web forms are confusing. I can't even imagine trying to work on a document or spreadsheet. The screen constantly changes as you need to enter data!
I believe that if you can get by under those conditions, the speed difference between hardware and virtual keyboards is mostly irrelevant. How much data entry are you really doing?
A government is a body of people notably ungoverned - AC
"it’s a debate that hotted up recently"
Normally I criticize the summaries for having typos, but this time the article itself has a pretty awesome one!
This does not work on my Nexus One (android), but I'm not sure about the other platforms.
Another alternative, which is probably faster than any of the tested keyboards is swype, where you just draw a line which represents the word. There is a public beta at http://beta.swype.com/ (Android only)
I think this is a completely legitimate question, but I agree with the author...it's barely scientific. Let's do this again with about 50 people, all who have little to no PDA experience and therefore no bias to one type or another. Then another 50 who are used to physical, and yet another 50 used to on-screen. Run the test and see what the results are. I'm sure more scientific /.ers out there have even better methods but this would at least offer more accurate results than what the author did.
Loading...
Translucent Keyboard
I don't like it as much as my old Droid and Palm physical keyboards but it does let me see more of the screen when using an on-screen keyboard by letting me see THROUGH the keyboard.
Where I could by a decent phone with wifi connection and qwerty keyboard? I could buy some fancy phones (iphone, xperia, blackberry,nokia e72) but no dirt cheap models. I tried ipod touch but it feels cold and too heavy on the hand.
Samsung GT-B3410 sounds most reasonable option. Lacks wifi.
http://www.pda-247.com/wordpress/2009/12/a-smartphone-for-under-80-meet-the-samsung-gt-b3410/
All virtual keyboards are NOT created equal.
One of the primary differences is the backing material of the touch screen. The cheaper phones utilize a plastic backing on the touchscreen, this plastic will bend, warp and cause 'typos' even if your finge is precisely where it's supposed to be. Glass does not flex, or warp - but is more expensive. This is why the iPhone gives such a superior performance on the virtual keyboard, as they have a glass backing.
I think many of the problems with virtual keyboards is due to the cheaper touchscreens utilize the flexible plastic backing behind the flexible membrane - thus adding distortion to the pressure point matrix - resulting in typo's that are indeed the "phone's fault".
It would be interesting to see this sort of study conducted with external keyboards, virtual (glass) and virtual (plastic) keyboards.
I'm switching to the Droid for the option of not only abandoning my cheap plastic backing on my touchscreen LG Dare; but also because I'll have the option of the slide out keyboard.
He also didn't mention what order he tested the phones in. If he wanted to be fair (not even going to stretch it to "scientific") he would type different passages, numbers and URLs on each phone. I'm sure by the last test he had the text almost memorized, so the latter phones probably ended up with shorter times and less mistakes. Just guessing, but the Nokia E75 and the Bold2 were probably last to be tested.
I have followed this company for sometime, with a novel keypad idea: http://www.digitwireless.com/ I have yet to see it in the marketplace though. I once spoke to the inventor who holds the patent.
So, the author decides to post about experiences with those six devices, leaving out MANY other devices. The old Palm Treo is very dated at this point, but still good for comparison purposes. How about the Palm Pre, or any of the other lesser known phones that have a keyboard? You may as well just go into your local hardware store and compare surge protectors and then claim it is news, without mentioning that a hardware store isn't going to give enough of a selection to make any sort of comparison meaningful.
I have a Nokia E51 so it has the good old basic 3-letters and a number per key (except 7 has 4 letters). Today I ordered a 18€ bluetooth keyboard for my cellphone so I can type like mad on the cellphone. Symbian-version of Putty + server at home + small, pocket-sized bluetooth keyboard + 3.5G connection = enjoyment
As someone who has never owned an iphone, I never understood why iphone users raved so much about their on-screen keyboards, until I got my current phone. My previous phones were the Blackbery Pearl and then the t-mobile wing. The Pearl of course has no touch screen and the Wing's screen was too small and the hardware too underpowered to make a decent on-screen keyboard feasible. My current phone, the touch pro 2, has the finger friendly on-screen keyboard and on-the-fly error correction that the iphone has had since it's inception and I've come to really enjoy it for typing emails and notes. I still value the slide out keyboard for tasks that require numbers and symbols though. The T-Mobile version of the Touch Pro 2 has a dedicated number row at the top, which is incredibly useful and, IMO, makes it superior to other versions of the Touch Pro 2.
I don't always use unix-like operating systems; but when I do, I prefer FreeBSD.
Disclaimer: I own a G1, nonetheless I'm fairly certain my view here is correct:
The hero's with it's well spaced out keys would of crushed the others, and is the reason I didn't choose a clque, or mytouch/hero and why I haven't upgraded to the N900, droid or nexus one. Moreover, I previously had a tilt (htc kaiser), and had tried the N900 for a couple weeks, the keyboards for both are very similar to the nokia 75. Without the spacing between the keys, there's no way one can be as accurate as the hero (g1).
All this article shows is that the author can type better on a physical keyboard. Nothing else. What a waste of 30 seconds of my time.
Ah yes, the obligitary Iphone mention.
Why pick the Iphone? By that logic, there are billions of phones without a real keyboard. But it's ludicrous to claim that therefore, people prefer not having a keyboard, all things being equal.
You could only test this if there were two phones, equal in all ways (including price), except that one had a real keyboard, and the other did not. The Iphone adds nothing to this discussion, no more than my 5800, or indeed my old Motorola V980, or the dumb phone I had before that.
I have an Android phone, and while the on-screen keyboard is okay, it does have severe disadvantages (as explained by others above).
The thing that I miss the most is definitely handwriting recognition -- it allows a small device to have a relatively big UI during input, it is not affected by regional layout differences, and it's fairly easy to extend and personalise by the individual user. TealScript is a prime example, really, doing full-screen transparent handwriting capture for any application.
Too bad most (all?) modern phone OS'es are too badly broken to even allow something like this to be implemented by a 3rd party (or even without serious core OS design changes).
"Good news, everyone!"
A radio station in California (KOME) used to claim it is physically impossible to over cum.
And remember - don't touch that dial, there's kome on it.
The sun is the same in a relative way, but you are shorter of breath and one day closer to death
Regrettably, the best keyboard I've used on a Smartphone was on a Danger Sidekick, by far.
Test didn't include the Nokia E90, far and away the best keyboard of any smartphone (6 rows).
There is nothing wrong with yr Internet. Do not attempt to adjust the picture. We are controlling the transmission - NSA
never mind the learning effect of typing the same phrase multiple times, it normally favours the later trials so maybe a baseline offset is required and the last phone might come in a bit slower than thought. Of course to be truly scientific he should have worn two blind folds? FWIW anyone debunking evidence based research would do well to participate in a double blind crossover trial of efficacy of parachutes since they have never been proven to be effective through the use of a control group ;)
Brisbane Aikido Republic
Everyone who has ever tried both the E75 and the N900 knows that the N900 beats the E75 hands down. I don't know how they could not include that in the test. Pity, really.
How in the world did this make it? Are there no hardware stories at all?
If you are good at typing you can do it in the dark at high speed. Impossible on a virtual keyboard with any accuracy. I have abandoned all my devices that use virtual keyboards. Netbooks for me ... never a tablet!
Ian Soutar
Victoria BC
http://www.microsec.net/
After 18 months with an iPhone type at around 30 words per minute. The auto correct fixes typos in every tenth word or so, giving near perfect accuracy. It's nothing compared to the 100 wpm I get on a PC keyboard, but definitely good enough (I'm writing this on it).
Lately I've started "touch" typing, where I look at the words rather than the keys.
When I pick up my gf's nokia with a hardware, I'm screwed. I can barely type at all... I can't seem to get out of the habit of touching the keys without physical force.
For me, it not even a matter of feel, I don't want to have to push a key, a light tap feels much faster. I even wonder if I could beat my 100 wpm on conventional keyboards with a touch surface big enough to use 10 fingers.