Family Learning Center

Posts Tagged ‘Servers’

Facebook May Begin Allowing Developers To Store User Data For More Than 24 Hours

Facebook’s f8 conference is shaping up to have quite a few improvements in store for developers, and we think we’ve come across another one: a change to Facebook’s data retention policy. Yesterday, Facebook employee Monica Keller (who left MySpace to join the company last month), took part in a conversation on Twitter that seemed to indicate that developers may no longer have to delete user data. The possible change came to light after Gnip CEO Eric Marcoullier gently chided Keller about developers being unable to store any user data, to which she responded, “come to f8!”.

Since that tweet, we’ve heard further whispers about a change to Facebook’s 24 hour policy retention, and that Facebook is already briefing developers on the upcoming changes.

So why does this matter? Facebook has historically been quite restrictive with regard to what developers are allowed to do with user data; in particular, it only allows developers using Facebook Connect to store user data for 24 hours before they have to delete it, or ping Facebook’s servers for a refresh. There are some exceptions to that rule (you can get a sense for them here), but the majority of the ‘meaty’ content can’t be stored by developers. As a result, applications have to constantly connect with Facebook’s servers, which multiple developers we’ve spoken with say is very  inefficient.

If Facebook does extend that 24 hour window, developers would be able to do things that generally require locally stored data, like batch processing (this isn’t feasible now because the application would have to make an API call for each user). Likewise, if these applications had the data stored locally they would be able to boost load speeds because they wouldn’t have to  wait for a call to Facebook’s servers after their applications loaded.

Of course, such a change would also have privacy implications (we’ll wait until we have all the details before we analyze those).  That said, we’ve also heard that many developers simply ignore some of Facebook’s data retention policies, in part because Facebook has a very hard time enforcing them. So it’s unclear just how big an impact this would have on the applications being developed.

Read the whole story…

Plane Crash Kills Three Tesla Employees, Takes Out Power In Palo Alto

A small Cessna twin-engine plane crashed in East Palo Alto today, killing three Tesla Motors employees and taking out power in parts of the city. The plane took off from Palo Alto airport and reportedly hit some electrical lines. The names of those who were on board have not been released. The plane was registered to Doug Bourn, a senior engineer at Tesla. It is not clear if he was the “high ranking official at Tesla” piloting the plane or if he was even on it.

Nobody on the ground was injured, but two homes caught on fire, including one used as a day care center. Fortunately, the day care center was empty at the time of the crash.

Eric Savitz at Barron’s has been blogging updates about the accident all day. It appears to be responsible for a widespread power outage in Palo Alto, which affected Facebook’s offices and many startups. For example, power is still out at TechCrunch HQ in Palo Alto (although our site is still up because our servers are not there).

The plane was headed to Los Angeles and was an older Cessna 310 manufactured in 1976.

Read the whole story…

feature: The Ars Technica Guide to I/O Virtualization

Virtualization is a key enabling technology for the modern datacenter. Without virtualization, tricks like load balancing and multitenancy wouldn’t be available from datacenters that use commodity x86 hardware to supply the on-demand compute cycles and networked storage that powers the current generation of cloud-based web applications.

Even though it has been used pervasively in datacenters for the past few years, virtualization isn’t standing still. Rather, the technology is still evolving, and with the launch of I/O virtualization support from Intel and AMD it’s poised to reach new levels of performance and flexibility. Our past virtualization coverage looked at the basics of what virtualization is, and how processors are virtualized. The current installment will take a close look at how I/O virtualization is used to boost the performance of individual servers by better virtualizing parts of the machine besides the CPU.

Read the whole story…

Openmoko brings improved keypad / scrolling and math equations to WikiReader

It sort of blows our mind that OpenMoko has managed to sell more than a smattering of its $99 WikiReaders (you know, considering just how limited in scope this thing is), but evidently there’s a huge demand out there for improvements. The company has just rolled out its Spring 2010 update, which adds math equations, fluid scrolling, and enhanced navigation. Reportedly, the latter comes in the form of an improved touchscreen keypad, which is said to make “finding Wikipedia entries easier and faster and makes enjoying one of the largest reference archives known to mankind even better.” We won’t bother pointing out the gross exaggerations in that there comment, but at any rate, existing users can download the update today free of charge. Don’t all hurry over at once and crash the servers, okay?

Openmoko brings improved keypad / scrolling and math equations to WikiReader originally appeared on Engadget on Tue, 09 Feb 2010 16:06:00 EST. Please see our terms for use of feeds.

Permalink   |  Business Wire  | Email this | Comments
Read the whole story…

Amazon Says No To Blippy

Blippy, the Twitter-like service that lets users publish the details of all their purchases, is just a couple of months old. But it already got Stephen Colbert’s attention (thumbs up). And now it has Amazon’s too (thumbs down).

Cofounder Philip Kaplan first mentioned that Amazon had turned off Blippy’s access to the service on an episode of TWiST with Jason Calacanis. I spoke to Kaplan tonight about Amazon’s reaction to Blippy.

He says they didn’t block Blippy, but simply insisted that the service stop pulling user purchase data, and erase all historical data they had already collected. They were also summoned to Seattle to speak with a “high ranking executive” of the company. Blippy complied (with both the summons and the demand to stop accessing user data).

Kaplan is soft stepping around the Amazon issue, and is hoping to come to some agreement with the company to allow them to access data in the future. He says “We believe our users feel strongly, as we do, that it is their right to access and use their data however they want. We’re optimistic that Amazon will come to the same conclusion.”

None of the other thirteen companies Blippy takes data from have complained, Kaplan says. And he notes that users must actually request data to be collected before Blippy begins to do that.

There is certainly an issue with how Blippy collects data – by storing user credentials on their own servers. But Blippy says they use APIs to log users in when available. And that may be the issue Amazon has with Blippy.

But it doesn’t explain why they’re insisting Blippy delete historical data that’s already been collected. The users have given their permission, and in fact have indicated that they want this data to be moved to Blippy. And it is the users’ data, after all. Amazon would be smart to remember that.

Perhaps, and I’m speculating here, there’s a reason Amazon doesn’t want all this data published. They’ve engaged in variable pricing in the past to test the demand curve, for example. They certainly may be using it again.

Read the whole story…

FMyLife Starts Clamping Down On Its API, Has Some Developers Saying FML

FMyLife’s developer community has a new reason to visit the site this week: to complain about the restrictions the company has recently started to enforce on its API. From now on, FMyLife requires all applications that feature advertising or that have a price tag (e.g. on the App Store) to send 50% of their revenues back to FMyLife. Apps that are available for free, sans advertising, will be able to operate as usual.

For those who haven’t been introduced to the FMyLife phenomenon, the site invites users to leave brief personal stories that generally end in catastrophe or extreme embarrassment. All of the stories conclude with “FML” (or F*** My Life), which has become a catchphrase outside of the site as well. It’s a great Schadenfreude fix, and you may even wind up feeling some empathy for your peers (or not). The site, and the third party applications it has spawned, have proven to be extremely popular.

Now, FMyLife disallowed paid applications and advertising when its API launched in February 2009, but the company has been inconsistent about enforcing those rules. Some developers have offered their applications with advertising for some time. And FMyLife has even approved the use of advertising and premium versions in some cases, without anticipating just how popular these applications could become. As it turns out, some of these applications have turned into big businesses in their own right, and some have proven to be drains on FMyLife’s servers. Rather than kill off all applications that are monetizing the service, FMyLife has decided it wants a cut.

Here’s how FMyLife co-founder Didier Guedj is describing the changes to the policy:

The Fmylife API was created to spread FML stories on the internet for free, in the spirit of sharing. However, in recent months, several developers made a very big business by selling FML’s applications or by advertising on it. This has led us to change our policy:
1) Access to the Fmylife API remains free for those using it for non-business purposes.
2) We will now ask that those who are using Fmylife for profit share their revenue at a fair 50/50.

The recently policy changes are directly related to a conflict FMyLife has had with Enormego, a developer that built two applications for the iPhone: “F MyLife and” “F MyLife Pro”, (the free version was briefly ranked as the top application on the App Store) . Here’s how Guedj describes the situation:

Enormego created two applications for iPhones (”F-MyLife” et “F-MyLife Pro”) which generate consequential revenue (through the sale of the apps, plus advertising) because they got more than a million hits per day since April 2009. These two iPhone applications only work thanks to our website, its concept and content. It was agreed with Enormego on July that any revenue generated by these apps would be split 50/50 (no contract was drawn, just via an email exchange). Enormego has never paid to Fmylife any money, and has not replied to any of our Emails since September. After several warnings without any answers, we decided to cut off their access to the API function. Since, they have been pirating the content of our website to feed their applications. We then asked them to stop. They did, only a few days ago.

We’ve reached out to Enormego for their side of the story.

FMyLife’s motivations for implementing a more restrictive API are obvious: they want to make money off the site they’ve built. At the same time, the FMyLife service has certainly benefited from the efforts of these third party developers — you can be sure plenty of people who downloaded these mobile apps are also frequent visitors of the service’s homepage. FMyLife can do what they want with their data, but it’s clear that they should have been consistently enforcing their policies from the start. Some applications have already been discontinued (or removed FMyLife support) since the change.

All of that said, it does sound like FMyLife would be willing to work something out if a developer has already built an application using the API and objects strongly to the changes. From Guedj:

For new applications (premium or ads), we explain the new rule to their developer and they have the choice to share the future income, or they don’t have access to the API.

We’ve never disabled an application (except Enormego’s) because these developers work hard too, and we respect their work a lot. We talk with them, explain the situation and all of them have agreed easily (for now) with our new conditions.

Image via sjdvda on userlogos.

Crunch Network: CrunchGear drool over the sexiest new gadgets and hardware.

Read the whole story…

AdWhirl Spins Its Own Open Source iPhone SDK And Server

AdWhirl is a mobile advertising service that allows app developers to easily switch ad networks without being a hassle to the end users. They were acquired by the ad network AdMob back in August, but that hasn’t stopped them from doing what they do. And today, they’re making their offering even better by open sourcing the whole thing.

With the new AdWhirl iPhone SDK, developers can now customize which ad networks they’d like to use, rather than being forced to include them all. This can significantly cut down the size of app, the company writes on its blog today. But the open sourcing of the project means that developers can now also use ad networks that weren’t previously supported by the service, and can let other AdWhirl users how to do the same. AdWhirl notes that developers currently using the service won’t have to change a thing to accomodate this new SDK.

The other announcement today involves their AdWhirl Server. This the the backend infrastructure that makes the whole system work, not only serving up the ads, but allowing users to monitor statistics. AdWhirl says it has redesigned this server from the ground up and it’s now using Amazon Web Services EC2. And beginning today, the company is providing the code so that developers can run their own AdWhirl servers. The move to EC2 should allow running their own servers to make it easier to scale as well, they note.

You can find all the new code on Google code.

Earlier this month, .App/Ads launched its own open ad platform.

Crunch Network: CrunchBase the free database of technology companies, people, and investors

Read the whole story…

Week in gaming: exploit ban-happiness edition

Microsoft has begun issuing temporary bans to players taking advantage of an exploit in Modern Warfare 2, while Infinity Ward works on a patch to fix the issue. While the servers may be kept slightly cleaner for the efforts, the amount of control Microsoft holds over owners of their consoles, and the arbitrary way they are able to wield it, is troubling.

With many platforms and little oversight, clones of popular games can become a major problem for original designers. Ars speaks with Jenova Chen, creator of Flower and flOw, about his experiences of having his games cloned. The problem may not be as cut and dried as it first appears.

Read the whole story…