Smartphones

Two big Android issues Google needs to address

This week Google hosts its annual developer conference, Google I/O. There are two key Android issues Google needs to talk about.

At a time when the company is growing in strength as a platform provider, Google hosts its annual developer conference, Google I/O 2010, this week in San Francisco. We can expect to hear about new features and product plans for the Android OS and the Chrome Web browser, and other products. But, when it comes to Android there are also two important issues Google must address.

The I/O event starts on Wednesday at the Moscone Center and Google is expecting about 5,000 attendees. The company has already scheduled press briefings on four topics:

  • Android
  • Chrome
  • Wave
  • Enterprise

The most anticipated item of the event is Android 2.2 (codenamed "Froyo"), which will add support for Adobe Flash as well as the capability of turning Android devices into Wi-Fi hotspots (a la the Palm Pre Plus). Those are the new features we already know about. Certainly, Google will also unveil other Android announcements at I/O.

Despite the recent news that Android has made major market share gains against RIM and Apple in the first quarter of 2010, there are still two key challenges brewing in Android land that developers and users would like to see Google address at I/O:

1. Android fragmentation

The first is the issue of OS fragmentation. With lots of Android devices being propagated by multiple handset makers and wireless carriers, there are now Android smartphones running lots of different versions of the OS.

This puts a lot of strain on app developers to support different versions of the software and it can impede forward progress on app features. It also frustrates users when they have to wait indefinitely for their wireless carriers to roll out OS updates.

Google needs to talk about the role it can play to help streamline this process.

2. Android vs. Chrome OS for tablets and netbooks

Beyond the momentum for Android on smartphones, there's also a growing stable of vendors that want to use Android to run tablet computers in the same way Apple used the iPhone OS to power the iPad. In fact, enterprise vendors like HP, Cisco, and BlackBerry have all been rumored to be experimenting with Android tablets.

The biggest problem with that is that Google may not be entirely committed to a tablet version of Android. After all, the company has touted the Chrome OS (based on the Chrome Web browser) as its OS for light computing devices such as netbooks.

Google needs to provide better clarification on the differences between Android and Chrome OS and its vision for how they should be implemented.

Follow developments from Google I/O

To keep up with the developments from Google I/O you can follow this blog where I will offer commentary on the most important developments that emerge. You can also watch the keynotes streamed live at the Google I/O channel on YouTube. The Wednesday (5/19) keynote runs from 9:00-10:30AM Pacific and the Thursday (5/20) keynote runs from 8:30-10:00AM Pacific.

Take the poll

About

Jason Hiner is the Global Editor in Chief of TechRepublic and Global Long Form Editor of ZDNet. He is an award-winning journalist who writes about the people, products, and ideas that are revolutionizing the ways we live and work in the 21st century.

50 comments
public_domain
public_domain

HORSEPUCKIE. google has recast LINUX into a small, powerful, versatile, efficient, safe and magnificient OPSYS. and the lackies at windozer cant take it. i can sling more mud at the deffective innefficient and poorly designed opsys at windozer that will stick than you can at google with stickem attached. LINUX BREEDS - WINDOZER BLEEDS.

sjenkins
sjenkins

I think there are two more issues that Google needs to address. First, from many bluetooth headsets you can not voice dail from the headset. You must speak into the phone after pushing the icon on the desktop after you have to unlock it. For the mobile users that are driving and traveling this is a major safety concern. The second is a better integration with M$ Exchange natively; although Touchdown and others have done a good job. Corporate users don't want to pay for additional software for exchange integration. Give us our email, multiple folders, and meeting requests and now we are talking!

d_baron
d_baron

Support X!! Xorg implementations can be made to run on tiny resources such as deli-linux (as little as 16 meg!). This would open up a proper android tablet and enable a tablet or handset to be an operational X-client. I can SSH onto my desktop from my Sony-Ericson. Wouldn't it be lovely to be able to get GUI as well (OK, it can be simply to slow to be of any use).

dcolbert
dcolbert

I'm not sure there is a division that needs to be clarified. Chrome OS for light mobile computing devices like Netbooks... The iPad is not a Netbook. It is less than a Netbook. It is a large mobile phone. There is a market for these devices - so make those devices - Android based tablets. Chrome OS is also over ambitious. The cloud, thin-client future that Google wants to see is probably a pipe-dream. Hopefully Google overcomes their hubris to realize this. Chrome OS will most likely be rejected. Netbooks will continue to need full desktop OS platforms, tablets are best served by mobile OS platforms. Chrome OS doesn't have a place to live.

Vulpinemac
Vulpinemac

Chrome is perfect for the underpowered, undersized notebook that is the Netbook; the last thing such a weak device needs is a full-powered desktop OS. This has been more than proven by all the complaints raised by Windows users who expected netbooks to have all the same capabilities of notebooks at a lower price. For over a year, netbooks were actually cannibalizing notebook sales for all brands but Apple as people sought to replace dying notebooks with less-expensive alternatives during the past economic downturn. This isn't to say the netbooks couldn't get by, but for anything more than basic computing or web browsing, they were woefully inadequate. In all honesty, the iPad targeted this in-between market and seems to have hit a home run. Probably the only way to 'save' the netbook is to put Chrome OS on it and get away from the resource-hogging desktop OS; and I don't care what desktop you're talking about, here. The netbook can be the perfect Slim Client-type of device where data entry is it's primary purpose and manufacturers may even be able to reduce the cost by avoiding the need to include any version of Windows. If you look at the more capable netbooks today, they're essentially downsized notebooks with nearly the same basic specs as a low-end full-sized machine. That also means that they're priced much closer to the starting point of the iPad where they once dominated at half that price. Get rid of the need to run a full desktop OS and you get rid of the need for larger expensive storage (whether hard drive or SSD), you get rid of the need for heavy processor power and you get rid of the need for (comparitively) massive RAM. Reducing the need for these three things can reduce the cost by as much as 50%. Not reducing these things and keeping the cost and the price where it is will almost guarantee the death of the netbook as anything more than a niche product with a very tiny market. Netbook sales have plateaued and may have already started to slide. If manufacturers don't do something to restore interest, the netbook will fade away just as the PDA has faded from sight. You are right that tablets are best served by a mobile OS, but netbooks cannot be served by a desktop OS; they need something to make them functional in a different paradigm.

matthias_blackshear
matthias_blackshear

I think moving Android to netbooks and tablets is a natural progression of the OS. I don't see myself or other business folks embracing the Chrome OS. I'll stick with my safe, secure, and private Ubuntu for Netbook install. Not everything belongs in the cloud.

eScoop
eScoop

You know, if Microsoft had announced a product that was "supposed" to do a bunch of great things and then did not deliver even a smokey prototype for years - they would be RIGHTLY chastised for that behavior. For years that's what MS did - and got smoked for delivering only Vaporware. When Google does it people on Tech Republic blogs put up polls asking us which we would want to see more? More? Who's seen it in the first place? Google smoke is all I've seen - why would I want to vote for Vapor? Oh wait! It's the cloud! Right - everything's going to the cloud! Kool-aid anyone?

rcummings622
rcummings622

I agree about the fragmentation. On the Android Market, for example, I have no problem paying for apps I want. But I hate paying for updates that are due to some quirk on a different phone.

WDMilner
WDMilner

The versioning problem is not something that will go away. Telco's are notorious for not updating the software/firmware on their various phones in order to controol what the user can do with them. The solution would be a common upgrade site but I don't see the telco's changing any time soon. Your poll needs another option - neither.

Justin James
Justin James

The fragmentation is the #1 challenge for developers and users. As someone who has looked into developing for Android, you can't count on users having the upgrade available. For example, anyone with a MotoBlur device is stuck at 1.6 (unless they force an upgrade and forego MotoBlur), even if the carrier would release 2.X to them. The way I see it, a developer would be losing a lot of sales to make a 2.X-only app, but the 1.5 and 1.6 SDKs are *very* limited (can't even work with Bluetooth!). At the same time, though, the fact that Android *can* be modified is one of the biggest reasons that device makers are putting it everywhere. Things like MotoBlur, HTC Sense, etc. are critical for makers to have some sort of unique value and differentiation in the market. But its the customization which forces the fragmentation, because for each new Android version, the customizations need to be updated, tested for backwards compatibility, etc. And that's why we see brand new devices still being released with Android 1.6. To put it another way, Android's open source nature is a pair of golden handcuffs. J.Ja

Vulpinemac
Vulpinemac

Which is exactly why Android will fail in the long run. What good is having 50,000 apps in the Android Market, when you can't even determine which ones will work on your specific device? What good is supporting Adobe's Flash CS5 [i]when it won't even work on older Android devices?[/i] Do you just abandon those older devices? What does that tell the people who already have them? "Oh, Android doesn't care about you; they already have your money. If you want the latest and greatest, you've got to buy a new device." Come on now, how do you think the customer will feel with that kind of attitude? Yes, I know this exact same argument is used against Apple all the time, but where Apple differs is that they support the older hardware years after it's no longer manufactured. Maybe Snow Leopard won't work on a PPC Mac, but Macs have been on Intel for 5 years now--we're talking about devices less than a year old being abandoned by Android. So really, is this fragmentation necessary? It shouldn't be. And this is Google's biggest problem right now; not that they're fragmenting the OS, but that the manufacturers aren't working with Google to consolidate the OS across devices. It's not that hard to create a vendor-specific start screen within the OS, but please allow updates to be accessible across the board for at least the device's expected lifespan. My point is that, like Microsoft before it, Google is letting the hardware vendors corrupt what should be an excellent OS, and like Microsoft before it, Google will see nothing but problems as a result.

Justin James
Justin James

In theory, Android can match app needs to device capabilities, because the app's metadata specifies what features it requires. And in theory, this is a good way of handling it. In reality, while the basic app compatibility matching *works* (for example, if an app requires a QWERTY keyboard it won't be shown as available to a phone without a QWERTY keyboard), the compatibility is still quite poor, thanks to device-specific customizations. Whenever I look at apps in Android market, it seems like half the comments on any given app are "I use phone XYZ and the app locks/doesn't work right/etc., refund issued." That is simply unacceptable. I can't imagine going into my local store and seeing notes on the shelf like, "oil filter doesn't work on my 2003 Ford F-150" or "software didn't work on my Model 2121 laptop." No way would that fly. As much as I like Android from the user perspective, the deeper I investigate the development angle, the less attractive it is. Apple's development story stinks too, in large part because their agreement is really uncool... the language restriction would be a lot more tolerable if the rest of the agreement was reasonable, but it isn't. Hard to beleive ensuring stability is the goal when C and C++ are allowed languages but a managed memory language like Java or C# isn't... Right now, as a developer, the best options are BlackBerry and Windows, neither one which have a really bright future at the moment. RIM's lost all momentum, and Windows Phone 7 is 2 years too late and the loss of backwards app compatibility makes the current generation a lame duck platform and the new generation effectively a brand new system. And Plam is dead, unless HP manages to ressurect it somehow (good luck). To put it another way... the mobile app field needs to start over from a clean slate, because nothing on the market is a viable option to developers. J.Ja

jmarkovic32
jmarkovic32

Why does everything wrong with Android always have to be something that Apple is doing with the iPhone? It's basically "Two things Google has to fix in order for Android to be like the iPhone". If I wanted an iPhone I'd buy one.

jasonhiner
jasonhiner

The iPhone has its own problems to deal with (terrible battery life, too closely tethered to iTunes, etc). The two Android problems I mentioned are important issues Google needs to preempt before they get worse.

Vulpinemac
Vulpinemac

... [b]IS[/b] something wrong with Android. It's not necessarily that Android is trying to be [i]like[/i] the iPhone OS, but the iPhone has shown what works and Android is trying to match or exceed that standard. As long as Android has problems, it can never be as good as it should be. You may not want an iPhone, but if you want Android, do you want a buggy, inconsistent device that only pretends to be as good as an iPhone?

The 'G-Man.'
The 'G-Man.'

to qualify your answers by please Thanks.

jeff.merkle
jeff.merkle

The poll is confusing - tablets and netbooks are truly two different beasts. Given the impressive touch-awareness of Android, it makes sense to use it for tablets. However, netbooks would be best run on ChromeOS, as Android without touch input is not a compelling OS.

jasonhiner
jasonhiner

And, I figured someone would bring it up here in the discussion. That said, I don't think Android with a mouse would be a bad thing. I think it could translate pretty well, especially on a small screen netbook.

Vulpinemac
Vulpinemac

... how long will it remain valid? Apple aimed its tablet at the netbook market, and from what I've seen so far, Apple appears to have hit a bulls-eye. The netbook was an attempt to improve the mobility of desktop computing and honestly the only reason it succeeded to the point it did was the very low price of the hardware. Quite literally, netbooks were sapping the sales of laptops for almost every brand and laptops were replacing desktops. If you look now, laptop sales are significantly higher than desktop, yet visibly lower than they were before the netbook was created. Only Apple saw continued growth there by comparison while netbooks were seeing 50% growth period over period for more than a year. However, netbooks are a poor execution of the mobility concept. Yes, they are more capable in some ways than tablets, but this is due to their running a desktop OS at less-than-ideal speeds. The problem is, they're not truly mobile. They're almost impossible to use while walking anywhere and even with their reduced size are still inconvenient in crowded commuter situations. As a presentation tool, they're almost unusable unless you connect it to a projector of some sort and as a sales tool the display is almost too small for comfortable viewing. In other words, the netbook is an underpowered, undersized notebook. This is where the tablet format has its advantage. Android really shouldn't want a mouse. In all honesty, you shouldn't even want to see Android on a netbook-type device. Chrome OS is effectively dead out of the gate because the only device that could take advantage of it is faltering and likely to collapse within a very few years. Android as a 'Widget layer" on top of a desktop OS could work to permit integration of data, but Chrome as an OS doesn't look any more to be a good idea.

devolution23
devolution23

where's your impressive list, then? you're obviously ready to throw down the gauntlet here, so wow us with your big... credentials -- not that it makes much difference when we're talking about OPINIONS. you could be pulling it out of your a$$ just like anyone else on here, so what's the point?

The 'G-Man.'
The 'G-Man.'

First you have indicated.... You should have quite the list...so we are still waiting....unless that was just total bull and you have no facts to back this claim up. You're just a general Joe Other then.

Vulpinemac
Vulpinemac

Sorry, not buying. I expressed my opinion and if you don't like it, then ignore it. With over 30 years in computing, you really shouldn't worry about my credentials.

Contradiction
Contradiction

These people keep making the same mistakes. Can?t they learn from the mistakes of the past? It has to be ONE manufacturer, creating the device AND the operating system! Android will end up just like Windows Mobile. One million versions of it, each one customized for a certain device. No consistency, no success. HINT: Look at Apple.

Hazydave
Hazydave

One device, once manufacturer, one OS.. that's the way we did it in the 70s and 80s. A few companies still do, but it's the flawed way of the past. The future is OS and device independence from one another. The reason integration was useful back at the dawn of the personal computing revolution was that OSs were very "thin", there was no easy way sometimes to separate OS from hardware support layer, some OSs did even have proper abstract device models, etc. That's ancient history. We are, after all, talking about handhelds running some version of UNIX or Windows... OSs that have had fully functioning device abstractions for over a generation (a human generation, not just a device generation). Google and the Android folks really should have taken a page from Windows NT and some of the other late 80s/early 90s computer science. Or even the PC. You don't need HP or Microsoft to push you a customized version of Windows for your new PC; you can go to a store and buy it. NT extendeds this well into non-PC-Clone x86 architectures, by means of the HAL: the Hardware Abstraction Layer. Every Android device should have had a small HAL that defines the hardware on the device, and presents simple drivers for low level stuff like GPS, touchscreen, tilt sensor, etc. Then a fairly generic version of Android can "plug" directly into this software socket, perhaps loading up full blown Linux or Java drivers for things that need performance. There are only so many different CPU and GPU models out there... not that many actual differences. Done correctly, Google could offer 2.0 or 2.2 updates exactly as downloads in the store. That's just the OS... they'd have to worry about who's supposed to get "premium" Google apps or the HTC or Motorola homescreen replacement, but overall, this kind of thing was worked out very nicely twenty years ago. This is precisely tuning the hardware and software together, it just leaves the "tuned" bits as a permanent, low-level part of the hardware, and lets the OS talk to that layer. Most of the time, the performance is actually better, simply because the people who understand the small bits of hardware being addressed do the bulk of the driver for them, rather than counting on generic Linux drivers or whatever. We're well past the days of Apple ][ and Commodore 64, folks (and for the record, they actually had their own separate hardware interface layers....).

Vulpinemac
Vulpinemac

Maybe you should look at that history again; while the abstraction layer (i.e. Windows vs PCs) is what made Windows so bad--what gave Microsoft such a terrible reputation. Yes, Microsoft succeeded in the short run, because honestly they drove almost all the competition out of the market. But at the same time, but permitting such loose control over the hardware used, every version of Windows has had its horror stories as well as its successes. Quite honestly, it's why Compaq (as an independent brand), Gateway and even eMachines have essentially died--absorbed by larger companies in each case. The problem is, those brands still use the cheapest possible hardware and tend to give the worst possible service--which Microsoft tends to get the blame for. Apple, on the other hand, has almost never let that happen. Even the clones they permitted in the early '90s had to be built under pretty tight specifications which is why the clones were barely $100 less than Apple's own machines. The mistake there wasn't that they required that control, but that they entered the 'clone' market far too late to have any effect on their expansion. It was a case of [i]follow the leader and hope some of his luck rubs off.[/i] Problem was, they followed the wrong path; it wasn't the cloning per se that put Microsoft where it was, but getting itself grandfathered into the enterprise that really launched them. Yes, it was tying itself to IBM's coat tails that got them in the door, at which point the less-expensive PC Clones and Compatibles drove IBM right back out. If you look now, IBM holds a very tiny share of the desktop market--less even than Linux boxes even though I'm sure Linux is running on many of them. No, Apple is proving now that consolidation, not fragmentation, is the way to excel in the computer market. OS X's growth over the last few years has been remarkable, often anywhere from 2 to 5 times higher than any single competitor and is encroaching into the enterprise by demonstrating a superior ROI despite the perceived higher initial cost. Yes, we're past the days of the Apple II and Commodore 64, but we are also rapidly moving beyond the days of a splintered OS/hardware market. HP has made a very visible step in that direction themselves with their acquisition of Palm and its WebOS. Their greatest advantage at the moment is the fact that they don't have to get their foot in the door, they're already one of the largest enterprise computing vendors on the market. If they manage to incorporate WebOS into their enterprise solutions, they could well start driving Microsoft out by offering more client-centric packages that don't require thousands of licenses over and above the cost of the hardware.

Neon Samurai
Neon Samurai

Even an eighteen button wondermouse will provide basic industry standard interfaces and resulting mouse controls. It may not provide teh extended wizzbang features of the mouse but it also won't be a brick if not plugged into a single vendor approved general purpose OS. If only more hardware vendors could stop screwing teh consumers and agree on industry standard interfaces for the generic functions they all do. (These days, basic 3D is as generic as Vesa 2D was when it was adopted as a standard) Keyboards, Mice, touch pads and keyboard dots, hard drives, chipsets.. all respect a generic level of industry standardization.. what makes sounds and video so special?

jasondlnd
jasondlnd

Who needs an Apple branded mouse? You can pick up any USB mouse, plug it into a Mac and have it work.

chadbo
chadbo

Wow, it is amazing how fast we forget. If Apple had won that war initially we would not be nearly as far as we are in technology. One vendor = One version of the truth, do you believe Apple would be as technologically advanced if it didn't have Microsoft to compete with? One O/S per hardware vendor cannot work, you will end up with two choices on two platforms that will build towards their niche and neither will advance nearly as quickly as they do today. Besides for all of the great things Apple has done recently, I still have a serious problem with having to pay twice as much because they control it, ever seen an Apple mouse for $10.00?

Daniel Breslauer
Daniel Breslauer

My wife has a Samsung i7500 Galaxy with Android 1.5 (Cupcake), via the Israeli carrier Cellcom. It's a horror story though things are getting better now, after 6 months. * Google Maps keeps on crashing whenever we try to use Latitude, especially when there is no GPS signal (like indoors). * The battery was so horrible it barely lasted 12 hours without use. After 6 months of complaining, numerous trips to service stores, and 4 replacement batteries, we now finally, since last week, have a battery that seems to WORK. The worst thing is that all of the carrier's service reps tried to tell us that it's perfectly acceptable and normal for a touch screen device to have a horrible battery life. They didn't seem to understand it's WORSE than horrible. There WAS a serious problem with this device - it has now been solved with the 4th battery we got. Combined with the lack of an Android update, lack of tethering (preferably USB because of the battery), limited usability of Bluetooth, we're thoroughly regretting purchasing this device and would not make the mistake of buying Samsung ever again. I do have to say that the screen is awesome. Now that the battery is getting better, we still need a fix for the Google Maps issue (at the carrier, they didn't even know what Latitude was) and an Android update.

Neon Samurai
Neon Samurai

The Razr looked so nice but was decidedly not the best hardware. When I asked the carrier about firmware updates they looked at me like I was crazy. I did find out that Motorola offered a firmware updates though not until the generation after my phone. I would say that the carriers need to put more effort into pushing updates. Having it available from the original manufacturer and the service provider is better than just one source.

merumaru
merumaru

how fast is booting up an Android tablet VS Chrome OS tablet? if the tablet runs Chrome, does it mean that it has to stay connected to the Cloud all the time, to get updates, download apps, and store data in the cloud?

Neon Samurai
Neon Samurai

There have been rumours of some kind of temporary storage which would be good for network flicker but I don't know if it would survive a reboot. The ChromeOS is meant to be in a networked environment though. At home you have your ISP and google's services along with whatever other browser apps you use. At work, hopefully you've got an in-house managed browser app or citrix server making it a handy thinclient terminal. So far though, what you see as Google Chrome (the browser) is what you will see as ChromeOS (the OS) though I'm sure Google will make some customizations. By contrast, Android is a full blown netbook/smartphone OS with non-volitile local storage and all the rest of it. Both are worth a look on your handy VM setup. Actually, that reminds me; must go find a current Android VM image and take a browse around the market.

Timespike
Timespike

One of the nice things about the ipad is that not all iphone or ipad apps are web browser plug-ins or even need the web at all. Android has a similar advantage. Chrome is JUST a browser - that doesn't offer nearly the functionality.

sky.dyver
sky.dyver

Chrome Browser, which is what we can see in the wild now, IS a browser. Chrome OS is an OS ... Based On ... the Chrome Browser.

bboyd
bboyd

Inseparable, windows folder views are just browser windows. Chromes plus is that is designed from the get go for isolated reduced privilege execution spaces. Should be more secure by a long shot. Not sure that android was designed with the strong security mindset that Chrome has now.

scubaj
scubaj

The first point doesn't make a lot of sense - why would anyone put that onus on Google? The company I used to work for had many different OS's out on RIMM, and if a user wanted an upgrade to the latest, we could push manually, otherwise it's the carrier. I would assume with Android it's the same. If you want the upgrade, download it from the Market place, if you want it hand delivered, wait on your carrier - plain and simple.

Vulpinemac
Vulpinemac

... that on some Android devices, it's almost impossible to upgrade the OS, which is one reason why Android is currently so fragmented. Add to this that some users have no concept of even the need to upgrade the OS. A line from the original MIB said it quite well: "A person may be smart, but People are stupid." Unless Google does something definitive about the Android fragmentation, they're going to lose much of the progress they've made in the mobile market. Consumers won't put up with devices that don't work right for long. Their concept is, [i]Fix it or Forget it.[/i]

chadbo
chadbo

It would seem to be fairly straight forward to use push technologies in this environment. You are always connected and when there are updates to the O/S and then to the apps they should be noted and applied without user intervention. This takes the user out of the equation and forces the OS and Hardware vendors to be on the same page. The post before was correct, a universal API to the hardware is essential.

jerrycline
jerrycline

The inability to voice dial through a Bluetooth is a major safety and convenience issue. How can a more powerful platform fail to at least match functionality that has been available for years, particularly with the personal safety issue of having to perform multiple manual steps on the phone to place a call.

greg.harry
greg.harry

I would have hoped readers of TechRepublic would know the difference between a technology and a product category. Perhaps you meant a "Bluetooth headset".

mark
mark

All Android apps have to be installed in internal memory, not SD Card for them to run. Well, ain't that a worthy issue to address ASAP?

XnavyDK
XnavyDK

But it has some buggy issues which are merely annoyances rather than fix it or forget it. I think the OS has promise. I wish it could be updated from a self download from the market rather than waiting for big red to let me have it when they feel like pushing it. I prefer to have some control over my destiny and my tools. I do enjoy the phone though. Also as a side note, I liked VZ navigator better than the current Navigation on the Droid and the gps takes forever to connect if it does. The navi was just updated, I have not used it yet so the jury is still out on the current update. I can see this going to a slate/tablet platform. I would buy one if it can print without laying it on a copier machine, attach an external drive, or sd card, and of course decent battery life and lets not forget one that I can swap out myself if it kicks it.

Editor's Picks