How Did Danger Not Backup Its Servers? How Did Microsoft Allow Such A Failure?

from the one-of-those-times-where-epic-fail-applies dept

I bought the very first Danger smartphone the day it came out (rare for me -- I'm not so much of an early adopter on mobile phones). One of the features I liked the best was the fact that all of the data on the phone was immediately and automatically backed up to Danger's servers. Since then, I've always been amazed that other providers didn't make similar features standard. Danger never fully lived up to its hype, and eventually sold out to Microsoft. It was never entirely clear why Microsoft would want Danger, but at the very least you would think that it would make sure that the servers were pretty safe and redundant. Or so you would think. Apparently Danger had a massive server failure and is warning people that their data may be completely lost. The company is telling people not to turn off their devices, as the only way to keep the data alive is to keep the phone going.

It's difficult to think of a system failure that makes a company look quite this bad. Tons of people have Sidekick phones and rely on server backup to keep their data. Not having a working redundant backup is a stunning sort of failure for Microsoft, and should remind people of the inherent dangers in relying on a cloud based service. While there are lots of cloud-based solutions that are quite useful, people are definitely going to need to be able to have alternative local and remote backups to make sure that, in this kind of situation, they're not totally relying on a company who should do things right, but perhaps did not.

Filed Under: backup, cloud computing, danger, server side, sidekicks
Companies: danger, microsoft

The Law Isn't Quite Ready For Cloud Computing

from the that-could-be-a-problem dept

Michael Scott points us to an interesting discussion among some lawyers trying to grapple with the implications of a rise in "cloud computing." For example, they note that courts usually don't take kindly to excuses such as "the hard drive ate my documents" when certain documents are unable to be found during the discovery process. But, if people are using a cloud computing solution such as Google Docs, this could actually be an issue. Google's terms of service allows it to cancel accounts and delete documents -- and someone who relies on a system like that only to find out later that he or she needs to hand over documents as a part of a court case may be in trouble. Of course, I don't see how this should be any different than any other "out of my hands" issue. What if a fire destroys paper documents? Or should the problem be that the user didn't make backup copies? The problem here isn't necessarily cloud computing itself, but the way the law views the discovery process.

Filed Under: cloud computing, discovery, law

Obama Administration Sides With Technology Over Hollywood In Cablevision Case

from the finally! dept

It's no secret that those of us who have been in favor of pushing back against the worst abuses of intellectual property law have been disappointed by the Obama administration -- which brought in a number of entertainment industry lawyers and seemed to side with IP holders over the public at almost every turn... until now. We've talked about the importance of the lawsuit over Cablevision's remote DVR system, and whether it represented copyright infringement. The appeals court had ruled that just because the DVR was hosted at Cablevision's datacenter instead of in a house it wasn't infringement. This makes perfect sense. Yet the entertainment industry has been claiming that allowing Cablevision to host DVRs on its own premise is infringement and Cablevision has to pay extra for the right to offer the same exact TiVo-type service that anyone can use in their home. Earlier this year, the Supreme Court asked the new Solicitor General to weigh in. At the time, we noted that this could be a good thing, since one of the last things Solicitor General Elena Kagan did before leaving Harvard to join the administration was recruit Larry Lessig to Harvard.

And, indeed, it appears that Kagan has not succumbed to entertainment industry's tortured logic on this issue and has recommended that the Supreme Court not take the case, saying that the appeals court ruling reasonably resolved the issues in the case. Either way, it's nice to see that the administration hasn't been totally taken over by those who believe in twisting copyright law to protect obsolete business models.

Filed Under: cloud computing, copyright, elena kagan, obama administration, remote dvr
Companies: cablevision

Amazon Connects The Cloud To The Postal Service

from the i'd-like-to-send-this-priority-upload,-please dept

Amazon's done a lot with its Web Services offerings to advance the notion of cloud computing, and it's now realized that the state of broadband in the US could do a lot to hold it back. Slow upload speeds could hamper the growth of cloud computing: Amazon's CTO says that to upload a terabyte of data over a 10Mbps connection would take 13 days. So the company is working around that issue by letting people submit data to Amazon Web Services via US Mail. Users send in a storage device with data they want added to the cloud, and when Amazon receives it, the data gets copied over and put online. It's not an ideal solution, but an interesting stopgap until fatter pipes proliferate. Perhaps there are some interesting business ideas here, like setting up local "cloud upload" stations with superfat connections, where companies could drop off their storage devices to have their data uploaded quickly. The issue of slow uploads could pose an obstacle to the wider uptake of cloud computing, and until it's overcome, there will be a necessity for these sorts of solutions.

Filed Under: cloud computing, snail mail
Companies: amazon

Is Cablevision Caving On Remote DVR?

from the hopefully-not... dept

One of the more important copyright lawsuit decisions was last years appeals court ruling in the Cablevision remote DVR case, which we discussed at length at the time. Cablevision built a remote DVR system. It was almost exactly like a TiVo, except that the device sat in a Cablevision datacenter, rather than next to your television. Functionally, it was no different. It really was just the length of the wire and where the box sat. Now, using a DVR in your home to record TV is perfectly legal. But the TV networks hate that, even though DVRs may actually have helped the TV industry by making it easier for people to watch their favorite shows (no one ever said they were good at figuring out the big picture). So they sued Cablevision, claiming that because the box sat on Cablevision's property, it was no longer legal and now it was copyright infringement.

The networks love to set up absolutely ridiculous explanations like saying that Cablevision is like the person who sets up a gun to go off when a door opens, so it's not the person who opens the door who commits the murder, but the person who set it up. Except... that makes no sense. Murder is illegal. Recording a TV show for personal use is not. A more accurate analogy would be like setting up a pillow to fall on someone when you open the door. That's not illegal for either the person who opened the door, or the person who set it up... because the action (falling pillow/recording a show for personal use) is perfectly legal. But the networks want to ignore this, and tried to twist copyright law by saying that because Cablevision's remote DVR creates a buffer version for a fraction of a second, it's making a copy, and thus violating copyright law. Seriously.

While a district court bought the argument, the appeals court (thankfully) pointed out how ridiculous this interpretation of the law was, and said the device is legal. The networks are now appealing to the Supreme Court, and the court has asked the Obama administration for input. I know there's been massive lobbying from a lot of different parties trying to get the administration up to speed on the detailed issues, and hopefully the important points get across. While this may seem like a trivial issue, it could impact nearly every online service, that suddenly becomes liable for making a "buffer" copy on its own servers based on something you do on your computer. Lots of "cloud" computing services could suddenly face massive copyright liabilities.

Still... while we wait for the Supreme Court to go one way or the other with this, it appears that Cablevision has been negotiating a compromise on the device, which (as Broadband Reports notes) probably means making things a lot worse for consumers (funny how that works). Once again, we'd have a scenario where content companies are killing off innovation because they're unable to adapt themselves -- and that's a really sad outcome. However, it might also lead to an end of the lawsuit, which could leave the appeals court ruling standing (which would be a good thing, rather than risk a Supreme Court overturn).

Filed Under: cloud computing, copyright, dvr, remote dvr
Companies: cablevision

Privacy Group Wants FTC To Shut Down Gmail... Again

from the give-it-up dept

I think privacy is a very important issue that often is given short-shrift... but I've never been able to understand some of the positions staked out by the Electronic Privacy Information Center (EPIC), who seems to have decided long ago that, even if people are making a conscious choice, anything that puts their privacy at risk is downright evil and must be stopped. When Google first launched Gmail back in 2004, EPIC went ballistic saying that it needed to be shut down as a privacy violation. Most people responded by getting Gmail accounts as quickly as they could.

Apparently, EPIC isn't giving up this fight, even though five years have gone by and Gmail has become a popular email service for many, many people online. EPIC has now asked the FTC to shut down all Google online applications, from Google Docs to Gmail, claiming that they're unable to "adequately safeguard the confidential info" of users -- and comparing those apps to a faulty car seat for kids (hyperbole, much?).

This all seems designed to get EPIC attention rather than to actually help consumers. The likelihood of the FTC agreeing with EPIC seems slim (which even EPIC seems to admit). People are pretty aware of what risks they're taking on by putting stuff on Google's servers, and Google has a pretty clear track record of doing its best to keep that info private. But most people feel that the risk is slight and the trade-off and value from the services is obviously worth it. Thus, it's not actually a privacy issue at all -- because most people are comfortable with the situation. So, why is EPIC trying to take away such useful services from millions of people who have come to rely on them?

Filed Under: cloud computing, epic, gmail, google docs, privacy
Companies: epic, google

Cloud Computing Has To Be About Openness And Ease; Not Locking Developers In

from the let's-see-something-real dept

There was a ton of press coverage today over Microsoft's not-particularly-surprising announcement of Microsoft Azure, its attempt to get into the cloud computing business, competing with the likes of Amazon, Google's AppEngine and (now) Rackspace, among others. Microsoft entering this space isn't a surprise at all, so it's a bit disappointing to see the sheer lack of details surrounding the announcement. Amazon has succeeded in the space because of two main things: incredibly cheap prices and ridiculous ease-of-use.

Amazon recognized early on that its "cloud computing" efforts were a true utility offering. It needed to be pure plug and play with very low, easy to understand pricing, and absolutely no hassle to get started. It was so easy and such a good deal, many developers couldn't come up with any reason not to use Amazon's web services. While I initially thought Google's AppEngine might provide serious competition for Amazon, today I'm less sure. Google is using AppEngine more as a way to get startups to build their technology to work on Google's tech platform. That's the same thing that Microsoft will clearly be doing with Azure. Yet, for Amazon, it was never about locking developers in to Amazon's platform: it was just about making use of spare computing cycles. The fact that it wasn't so tied to Amazon's core business may actually be a benefit here, in that it lets Amazon be a lot more open and free about how people can use it, keeping it super cheap and easy. Google and Microsoft, on the other hand, get too focused on using their cloud offerings to tie developers to their own tech stack. I agree that someone else could come in and create a cloud computing solution that beats Amazon by being even more open and easier -- I just don't see either Google or Microsoft being that company.

Filed Under: azure, cloud computing, ease of use, lock in, openness
Companies: amazon, google, microsoft

Rackspace Wants To Take On Amazon's Cloud Computing Efforts

from the interesting-battles dept

Over the past couple of years, Amazon has successfully built up quite a business in commoditizing and renting out its server and process power through its Amazon Web Services efforts, such as S3 and EC2. These days, a lot of startups don't even bother getting servers in a hosting facility, knowing they can just scale up on Amazon's machines. That, of course, could represent a threat to the big hosting facilities, such as Rackspace. And while many thought the eventual competitors against Amazon in this space would come from the likes of Google or Microsoft, it looks like Rackspace is trying to be a bit proactive here. It just bought two companies and announced a competing web services platform. The two companies are Slicehost, which does virtualization, and JungleDisk, an online backup service that is built on Amazon's S3 storage system.

The JungleDisk deal is particularly interesting, as Rackspace knows that JungleDisk users (myself included) are effectively all Amazon S3 customers. If it can offer an easy and convenient way to switch over, then it can take a bunch of customers away from Amazon and move them right over to its own platform. If you're unfamiliar with JungleDisk, it creates a virtual mounted drive on your computer that connects to Amazon's S3 platform. So, as long as you have internet access, you have an unlimited size hard drive that you can reach, where you pay based only on what you use. On top of that, it includes backup software (or you can use other backup software) so that you can regularly back up anything on your computer onto this network drive regularly. JungleDisk customers pay for the software, but the ongoing costs are all paid to Amazon. It's actually quite useful (and crazy cheap compared to some other backup services). It would be interesting to see if Rackspace also allows a service to let you back up to both Rackspace and Amazon, so if one goes down, you still have access to the other. Either way, it looks like the competition in the so-called "cloud computing" space is about to heat up.

Filed Under: cloud computing, online storage, web services
Companies: jungledisk, rackspace, slicehost

What Sun Should Be Doing With MySQL

from the good-advice dept

Earlier this year, Sun scooped up MySQL for a cool $1 billion. However, while Sun has been somewhat friendly towards open source software, there were plenty of concerns about what the company planned to do with MySQL. Some of those fears are now being realized. After first disappearing some of MySQL's public statements on the evilness of software patents, Sun caused quite a bit of controversy with a plan to close source certain new features in an attempt to push people to upgrade to a premium, paid version of the software. While that's certainly one strategy, it could be a dangerous one, ticking off many MySQL users who will go searching for alternatives.

So, what could Sun do with MySQL to help build a bigger and better business that doesn't involve locking up any software? The blog Milking The Gnu has a very interesting suggestion that makes a lot of sense (and certainly fits in directly with the economics we discuss around here). The idea is not to worry about locking up the software, but to turn MySQL into a cloud computing web-platform. The reasoning makes a tremendous amount of sense (much more than Sun's current strategy). Basically, on the low end, you have folks who will never pay for a premium version of MySQL anyway. At the high end, most of those companies (if pushed) will probably lean towards Oracle or IBM. But in the middle-tier there's a real opportunity -- not to be a database software company, but to build that all important web platform we've been discussing.

Already, Amazon and Google are trying to build that platform, with Amazon seeing a fair bit of success (and Google just starting). Sun has promoted the concept of cloud computing for years, so why not flip things around and make MySQL the database part of a cloud computing offering. With so many folks already comfortable with MySQL, it will be much easier for many of them to embrace this offering, rather than having to figure out the details of Amazon's SimpleDB or Google's AppEngine/BigTable setup. Then, the more people you get to adopt the free open source version of MySQL, the more likely they are to make use of Sun's cloud computing offering over the alternatives. And, then, Sun can charge for the use of cloud computing resources (scarce resources) while knowing that the infinite nature of MySQL promotes that scarce good. Given that Sun's been such a promoter of cloud computing for so long, you would think this was a no-brainer. But it's latest actions with MySQL suggest it may be going in a different direction, and that's unfortunate. Update: Marten Mickos of MySQL/Sun responds in the comments, and Glyn Moody points us to an interview he recently conducted with Mickos suggesting that Mickos is thinking along similar lines about cloud computing.

Filed Under: cloud computing, databases, mysql, web platform
Companies: mysql, sun

Cloud Computing Comes In For Some Nasty Weather

from the outages-weren't-in-the-forecast... dept

I've definitely been a fan of Amazon's web services offerings. By providing cheap, scalable and robust processing power and storage (and, as of recently, a simple database), Amazon has started to singlehandedly change the economics of doing certain things online. It's become quite common for startups to build their platform entirely on Amazon's systems -- and given the stability of Amazon itself, it always seemed like a good idea, especially following some high profile outages at popular data centers. However, this morning, that "robust" part got called into question as Amazon's S3 service had a major outage, taking down a variety of startups and services (and, apparently, plenty of images on various websites). While some are suggesting this shows the weakness of "cloud computing," it's not necessarily all that different than those earlier outages where you had a data center go down. Given this and the recent Blackberry outage, we're again learning that robustness isn't just nice to have -- for many services it's really a need to have, and we're not quite there yet.

Filed Under: cloud computing, robustness, s3, uptime
Companies: amazon