Blog Post

Here’s Why Google Is Holding Honeycomb Back

Google’s Honeycomb version of Android (s goog) may have left the hive for a few tablets, but don’t expect to see any new ones announced in the near future: The company has decided not to release the platform’s source code to the community just yet. Tablet makers such as Motorola (s mmi), Samsung, LG and others have had early access to use Honeycomb, but only through prior agreements with Google as hardware partners. According to BusinessWeek, smaller tablet makers, the open-source community, and even handset builders that were considering the use of Honeycomb for other new products will now have to wait longer to use Google’s software.

My first thought stems from my initial impressions of Motorola’s Xoom tablet, which runs Honeycomb. Last month, I pointed out Honeycomb felt rushed to market, citing a few examples of why: application instability and non-supported hardware features, such as use the of microSD memory expansion slot. Why the rush? iPad 2 (s aapl) was the answer that came to mind:

So if I’m correct, and Honeycomb and the first tablets that run it are getting rushed to market, the obvious reason it’s happening is to compete with Apple’s iPad, which sold 14.75 million units in 2010. And with Apple’s press event next week, which is expected to shed light on the next iPad and possibly new software features for iOS, I can understand the need for speed. However, some of the issues caused by cutting a few corners to launch devices could end up hurting more than helping, at least in the short term.

That was my thought exactly one month ago today, and now BusinessWeek has a quote from Andy Rubin, the head of Google’s Android group, that confirms my suspicions about the rush job:

To make our schedule to ship the tablet, we made some design tradeoffs. We didn’t want to think about what it would take for the same software to run on phones. It would have required a lot of additional resources and extended our schedule beyond what we thought was reasonable. So we took a shortcut.

This news isn’t going to stop Motorola (s mmi) from shipping Xoom tablets, nor does it mean competing products from top-tier vendors will be delayed. Instead, since the platform isn’t getting released to the open-source community, programmers who want to create their own Honeycomb build for existing products will have to wait. As will any second- or third-tier hardware vendor that had hoped to make their own mobile device running on Honeycomb.

While I expect the open-source community to come down hard on Google for this decision, I can see why the company is taking this step. One simply has to look to the earlier days of Android to see what Google is trying to defend against. Practically anyone or any company with enough technical prowess could build an Android device. That’s ultimately a good thing, as it brings choice, but it also allows for a “wild west” scenario in the world of Android.

For example, some hardware manufacturers, such as Archos, built Android tablets in 2009, when Android was geared towards smartphones. By jumping the gun, these devices delivered poor customer experiences and often had no access to Google’s own apps nor the Android Market for other apps. What do customers blame for the poor experience? Google and its Android platform (as well as the hardware vendor), and Google can’t afford such perception if it wants to keep hardware partners interested and attract developers to its new tablet platform.

13 Responses to “Here’s Why Google Is Holding Honeycomb Back”

  1. I bought an Archos 7″ home tablet, and I agree, thanks for clearing this up. Now I blame Archos for the android software, as well as the poorly constructed tablet. From my standpoint, I got what I paid for.

  2. Google rushed it out to try and show people they’re still working on Honeycomb; it’s not perfect but it does exist sort of mentality. If they don’t get the bugs worked out right away, I don’t think that’ll be much of an issue. They can always issue OTA updates unlike iOS devices that need iTunes.

    Not an Apple fanboy / troll, but I think releasing Honeycomb only when it’s absolutely ready is the best approach. People who want an Android tablet aren’t going to buy an iPad and vice versa. The smart thing is keeping your core audience satisfied. If I went out and bought a Honeycomb that didn’t support anything, crashed all the time and frustrated me to no end, I’d rather just have Google delay it a month or two.


    My take is that Google held back because they where pushing all kinds of mobile services that required a working GPS and on the go data connection. And the only way to “force” hardware companies to include those was to dangle Market access on the same hook or else they would have gotten a mass of devices that barely had a data connection.

    I would rather applaud Archos for attempting to make the best of a bad situation, using proper internals to get acceptable performance (unlike the million and one rockchip devices out of shenzen). This while cutting back on extras that was relatively useless without a persistent data connection in the first place.

    In the end i suspect we see the end result if Rubin rebooting Danger without a firm independent goal of the end product beyond making a better X (x being first blackberry then iphone).

  4. While I can sympathize with Google’s problem, the solution to hardware compatibility isn’t holding back the code. Rather, they should implement a hardware testing program that ensures hardware works with Android. With a “Works with Google Android xxx” sticker, they could offer a strong incentive to manufacturers to build devices that work in a consistent fashion.

  5. Developer

    Android isn’t a platform. You can’t develop for it and expect your code to run on all android devices like you can with iOS or wp7.

    This is what reporters and analysts are missing. Fanboys can talk about it being open but developers have to ship code.

    Who wants to customize your app 250 times?

    iOS requires 2 versions- iPad and iPhone optimized, not 250.

    • Developer2

      Oh really… so you are saying that all apps built today run well on all IPhone 1, IPhone 3G, IPhone 3GS, IPhone 4 ?.. because I have a 3G one and have come across quite a few apps that wont run on it. Yes segmentation does exist in iOS too. and the latest system updates completely screwed up my phone, crashes and quite slower then before.

      Truth be told what Google is doing is the same that Apple does and others, simply that the others you are used to them not sharing the goods, and just because Google wants to optimize before doing so, people are complaining. Its a free product guys. FREE!

  6. Archos tablets were better than ipod touch and ipad1, so really, this is only Google just making sure Ice Cream source code also is ready, and that Laptop-versions of Honeycomb web browser are ready, and Google TV for ARM also be ready for open sourcing, so that when Honeycomb source code is released, that all the hardware manufacturers know what to do, take the parts they need for the types of product they want to make.

    Eventually, Google will be able to be totally open even with the daily development process and allow external developers from submitting code change to the core Android source code, but for now, as Smartphone industry is completely insanely exploding, and the tablet market is exploding even faster, Google has to make sure they point Android in the right direction in the beginning.

    Anyways, Google has to open up Google Marketplace to more devices such as Archos, that is what I expect. And they could announce the CDD earlier than they release the source code.

    • Archos tablets were better than ipod touch and ipad1

      Charbax, I know you’re a shameless Archos fanboy, but this is too much even for you. Neither sales figures nor user reviews for Archos products offer any support whatsoever for this assertion. (To be fair, *nobody* shipped an Android tablet that met even the low-bar expectations of the diehard anyone-but-Apple crowd until the Galaxy Tab.) Thanks for playing anyway!

      • Archos has 25% of sales in certain European countries, such as France, and that is for a company about 3000 times smaller in than Apple in terms of valuation, revenues, profits and number of employees. Apples large money reserves is the ONLY reason Apple sells more tablets than Archos, if Archos had more money, they would be able to manufacture more tablets and WOULD be selling many more tablets than Apple even today.

        If you want numbers, use a reputable polling institution of any kind, and make them ask consumers what they think in the store, while the store has to exhibit both Archos choices and the Ipad2 next to each other. At given prices, Archos being $250 and Apple being $500, you will simply see that 90% of consumers will choose Archos in those situations. Even to top it off, people who really care about features regardless of price, will choose Archos, as people who care about features care about the much lighter weight, built-in kick-stand (absolutely compulsory for use to watch video or play games for more than 5 minutes), built-in real HDMI connector, built-in real USB host connector, webcam (vs ipad1), real full video and audio codecs support, anyone who cares about features, cares about downloading all their tv shows, movies and music for free on BitTorrent, those files all work on Archos, none work on Apple.

  7. Hardware benchmarks (e.g., between the iPad2 and Xoom) point to the same reason. Honeycomb, today, just isn’t optimized. The early releases probably took some of the steam out of the iPad2’s unveil, but now all the kinks need to be worked out ASAP.