Yes, this is to scale.
Crazy.
Cross-posted from the Spanning Backup Blog
After months of beta testing, Spanning Backup for Google Apps has launched. Since the beta began, we've backed up over 100 million calendar events, contacts, and docs for our users. That's 100 million fewer things they have to worry about.
Now the ability to back up Google Calendar, Contacts, and Docs is commercially available at spanningbackup.com.
Data loss and corruption is a serious problem for Google Apps users. Browse through the Google Apps help forums and you'll find hundreds of posts from users who have lost their data and need help. The usual reply, when one is given, is, "Sorry, your data is gone for good." If those users had been using Spanning Backup, they'd be able to restore their data with just a couple of clicks.
Clay Spinuzzi is an associate professor at the University of Texas and an early user of Spanning Backup. Here's what he has so say:
I've been using Google Docs almost daily since 2006, and I'm totally sold on cloud-based word processing. But how about cloud-based backups? For years, I had to safeguard my thousands of documents, spreadsheets, and presentations by manually downloading them once a week - and I always worried that I would miss something. But with Spanning Backup, I never have to worry about backups. Backup is dependable, browsing files is easy, and restoring a file is transparent. It just works.
Spanning Backup is a pure cloud-based application built on Amazon Web Services (EC2, S3, SQS, EBS, CloudFront, and CloudWatch) and is designed to scale to millions of users. It uses a number of Google API's (Calendar, Contacts, DocList, Docs, Spreadsheets, Provisioning, AuthSub, and OpenID Federated Login) and is natively integrated with the Google Apps Marketplace to support domain-wide provisioning and Google Universal Navigation.
The service is available for individuals at spanningbackup.com or for entire Google Apps domains on the Google Apps Marketplace at spanningbackup.com/marketplace.
After a 30-day free trial period, a subscription costs $3.95/month or $39.95/year per user. Educational, non-profit, and volume discounts are available. Spanning Sync customers get an automatic $10 discount off their first year's subscription.
Spanning Sync, Inc., based in Austin, Texas, was founded in 2006 by Charlie Wood and Larry Hendricks. The company's eponymous product
For more information please contact info@spanningbackup.com.
For years Vinnie Mirchandani has chronicled all kinds of innovation on his his must-read blog New Florence. New Renaissance. Now he's taken his examination of technology and how it affects business to a whole new level with his new book The New Polymath: Profiles in Compound-Technology Innovations.
New Polymath is a breathless run-through of the technological forces that are defining the new rules of business and the companies that are successfully harnessing those forces to succeed. (If that sentence hurt your brain, you might want to get a good night's sleep before diving into New Polymath.) While many business authors stretch what should have been an article into a book, Vinnie compresses what could have been a collection of business books into a single volume.
Vinnie's passion for innovation and sweeping knowledge of what's happening right now make New Polymath a fascinating read.
Disclosure: Vinnie is a friend of mine, and mentions my company in his book.
In my last post on this topic I said, "What I want is a way to shoot 720p video on my iPhone and upload it directly to the cloud without any loss of quality." Well, I found it.
Pixelpipe is a free iPhone app and online service whose latest version lets you upload full 720p videos (up to 200MB) from your iPhone 4 to a variety of online services including YouTube. It even features queued uploads that work in the background, so your phone isn't tied up while you're pushing all those bits.
In my testing so far it's worked flawlessly and is exactly what I was looking for. It also performs a bunch of other functions, like uploading photos and routing your videos and photos to various social media channels, but I haven't looked into those yet.
This morning I hopped on a jetBlue flight from Austin to San Francisco to attend Google I/O. I usually avoid industry conferences, but I make it a point to attend this one. I find it has a supernormal signal-to-noise ratio. Here's what I expect to get out of this year's event.
Exposure
We're getting close to the commercial release of Spanning Backup, so it's time to start directing people's attention toward it. (We recently performed our two-millionth backup. Time to light this candle!) Toward that end, here's a quick video showing what it's all about:
Information
We work at the leading edge of Google's API's, so there's often information we need that's not documented anywhere. What's the maximum payload size for an HTTP POST to the Google Docs API? How do you enable compression when talking to the Google Calendar API? Is Google planning on implementing contact sharing, or would it be a good thing for us to do? The answers to these questions are at I/O, not online.
Relationships
It's important to remember that there are real people creating all of these ecosystems, companies, products, and features. And even though better online communications tools exist now than ever before, face-to-face conversation (often in a conference hallway or a hotel bar) is the most powerful way to establish and strengthen relationships with those people.
Zeitgeist
Finally, I want to get a sense of where Google and its ecosystem are going. From where I sit it looks like Google is emphasizing Android over the web, a strategy I would consider dangerous and ill-advised. I want to see if I get the same sense at the show.
Oh Yeah, and a Party!
Be sure to join us and our friends at RedMonk for beers at House of Shields Wednesday night. The first $500 of the bar tab is on Spanning!
But even if you can't make it Wednesday night, please come by the Spanning booth in the Developer Sandbox area and say hello. We have some very exclusive laptop stockers for you. :-)
A couple of years ago I bought a Brenthaven Trek laptop backpack from Amazon. I loved it. But after 18 months of heavy use, one of the zipper tracks broke. Since a similar problem with a $400 Tumi bag wasn't covered by the warranty I assumed this wouldn't be either.
I started shopping for other laptop cases, and mentioned my broken Brenthaven bag on Twitter. Minutes later, someone from Brenthaven contacted me and suggested I send the bag back to get a replacement. As it turns out, they have a lifetime guarantee on their products and this kind of problem is covered. (Take that, Tumi!) I hadn't registered for it when I bought the bag, but that didn't make a difference. I contacted their service department, got a return authorization number, and sent it in.
That was two weeks ago today. This morning FedEx delivered a brand new Trek backpack to my door. It's even a newer model, with better zipper pulls and tracks.
I have to say I'm delighted with both Brenthaven's product and their customer service. The backpack itself is very high-quality, fits my 15" MacBook Pro like a glove, and has just the right number and type of compartments. Also, after carrying an older Targus laptop bag for the last two weeks—and feeling every minute of it in my lower back—I'll never go back to carrying a traditional over-the-shoulder case again.
Aside from that, the fact that they're actively listening for customers who need help is more than just icing on the cake; it's turned me into a customer for life. The next time I need anything Brenthaven makes, like maybe an iPad sleeve, I'll look no further.
I just ran the SunSpider JavaScript Benchmark on my iPad and my (2.66GHz Unibody) MacBook Pro. The raw results are here and here, respectively. Here's the comparison:
TEST COMPARISON FROM TO DETAILS
=============================================================================
** TOTAL **: 26.7x as fast 10999.0ms +/- 0.7% 412.0ms +/- 1.0% significant
=============================================================================
3d: 38.7x as fast 1995.2ms +/- 1.7% 51.6ms +/- 1.3% significant
cube: 39.9x as fast 622.2ms +/- 3.5% 15.6ms +/- 4.4% significant
morph: 41.5x as fast 796.8ms +/- 0.9% 19.2ms +/- 5.4% significant
raytrace: 34.3x as fast 576.2ms +/- 1.5% 16.8ms +/- 3.3% significant
access: 41.4x as fast 1621.0ms +/- 2.5% 39.2ms +/- 2.7% significant
binary-trees: 20.4x as fast 114.2ms +/- 2.9% 5.6ms +/- 12.2% significant
fannkuch: 37.5x as fast 562.0ms +/- 3.8% 15.0ms +/- 5.9% significant
nbody: 75.2x as fast 751.8ms +/- 2.9% 10.0ms +/- 8.8% significant
nsieve: 22.4x as fast 193.0ms +/- 1.8% 8.6ms +/- 12.9% significant
bitops: 39.6x as fast 895.8ms +/- 4.4% 22.6ms +/- 9.2% significant
3bit-bits-in-byte: 52.2x as fast 167.0ms +/- 2.2% 3.2ms +/- 17.4% significant
bits-in-byte: 22.1x as fast 159.0ms +/- 3.1% 7.2ms +/- 14.4% significant
bitwise-and: 33.7x as fast 154.8ms +/- 27.5% 4.6ms +/- 14.8% significant
nsieve-bits: 54.6x as fast 415.0ms +/- 0.5% 7.6ms +/- 14.6% significant
controlflow: 33.6x as fast 134.6ms +/- 1.8% 4.0ms +/- 22.0% significant
recursive: 33.6x as fast 134.6ms +/- 1.8% 4.0ms +/- 22.0% significant
crypto: 31.3x as fast 777.4ms +/- 0.5% 24.8ms +/- 4.2% significant
aes: 17.4x as fast 229.6ms +/- 1.2% 13.2ms +/- 10.3% significant
md5: 41.8x as fast 267.8ms +/- 1.7% 6.4ms +/- 10.6% significant
sha1: 53.8x as fast 280.0ms +/- 2.5% 5.2ms +/- 10.7% significant
date: 12.5x as fast 739.8ms +/- 1.4% 59.0ms +/- 4.7% significant
format-tofte: 12.3x as fast 331.0ms +/- 1.1% 27.0ms +/- 6.5% significant
format-xparb: 12.8x as fast 408.8ms +/- 1.8% 32.0ms +/- 9.1% significant
math: 44.1x as fast 1543.4ms +/- 0.2% 35.0ms +/- 5.6% significant
cordic: 55.6x as fast 544.6ms +/- 0.6% 9.8ms +/- 5.7% significant
partial-sums: 32.9x as fast 579.0ms +/- 0.5% 17.6ms +/- 9.5% significant
spectral-norm: 55.2x as fast 419.8ms +/- 0.5% 7.6ms +/- 9.0% significant
regexp: 47.6x as fast 989.8ms +/- 0.7% 20.8ms +/- 7.8% significant
dna: 47.6x as fast 989.8ms +/- 0.7% 20.8ms +/- 7.8% significant
string: 14.9x as fast 2302.0ms +/- 1.4% 155.0ms +/- 1.5% significant
base64: 20.4x as fast 334.8ms +/- 0.9% 16.4ms +/- 6.8% significant
fasta: 15.1x as fast 452.8ms +/- 2.4% 30.0ms +/- 2.9% significant
tagcloud: 11.7x as fast 385.2ms +/- 4.3% 33.0ms +/- 0.0% significant
unpack-code: 13.5x as fast 620.2ms +/- 1.2% 45.8ms +/- 3.5% significant
validate-input: 17.1x as fast 509.0ms +/- 1.1% 29.8ms +/- 3.5% significant
It's one thing not to be able to run Flash apps. But JavaScript performance like this effectively means the iPad can't run complex JavaScript apps either. Interesting.
Update: Tim Anderson reports that even his $325 netbook runs the SunSpider benchmark 4.5x faster than an iPad. This (plus, of course, Flash support) could turn out to be a distinct competitive advantage for Google Chrome OS netbooks, which will surely be JavaScript speed demons.
Update 2: Iñaki Rodríguez ran the benchmark on his Acer 1810T netbook ($612 at Amazon) running Google Chrome. The results were 13.3x as fast as an iPad. This could turn out to be a BFD.
Update 3: According to the benchmark, the iPad runs JavaScript 1.52x as fast as an iPhone 3GS.
Compare this image (click for larger image):
with this one (click for larger image), created by Trey Ratcliff:
Which way do you see the world? There's a powerful lesson there.
These days I, uh, tweet a lot more than I, er, blog. Here's what I tweet most about, according to tweet cloud:
This is a paper I wrote for my Health Economics class. I originally shared it on Google Docs, but due to popular demand (well, from one person anyway :) I'm cross-posting it here. Comments welcome.
Charlie Wood
November 30, 2009
On November 7, 2009 the U.S. House of Representatives passed HR 3962, a comprehensive health care reform bill that included among its numerous provisions the creation of a health insurance program to be operated by the federal government, commonly called "the public option", and a mandate that most legal U.S. residents obtain health insurance.1 This paper examines how the combination of the public option and the individual mandate should, all else equal, be expected to affect health insurance premiums facing both participants and non-participants in the program and total medical expenditures.
What is the Public Option?
In order to analyze the effects of the public option it is important first to define it. The public option as described by HR 3962 is a government-operated health insurance program that will offer coverage to legal U.S. residents who are not covered by existing government programs like Medicare or Medicaid and who do not have insurance through their employers. Reimbursement rates for health care items and services will be negotiated with health care providers by the government. The premiums charged to participants will be required to cover the cost of the program itself, including startup costs, benefits, administrative costs, and a "contingency margin" of at least 90 days of estimated claims. Benefits and premiums for the various levels of coverage available under the public option will be published online. In addition, although participation in the public option itself is voluntary, HR 3962 includes a mandate that most legal U.S. residents obtain health insurance coverage.
Effects on Premiums Faced by Those Not Covered by Group Insurance
People eligible for the public option will see the most immediate, direct effect on the premiums they face for health insurance. Since they are covered by neither employer-provided plans nor government programs like Medicare or Medicaid, those people currently have two options: forgo health insurance altogether or purchase individual coverage, which is typically very expensive.
Four main factors contribute to the high premiums in the market for individual health insurance: adverse selection and a lack of economies of scale, both of which increase the insurer's costs, and the market's oligopoly structure and limited price transparency, both of which allow insurers to set prices above marginal cost. Together the public option and the individual mandate address each of these factors, which should lead to lower premiums.
First, the information asymmetry inherent in the market for health insurance can lead to adverse selection and ever-increasing premiums. People generally know more than their insurer about how much health care they're likely to consume. Relatively healthier people, that is, those who expect their health care expenditures to be less than the premiums they face, can simply opt out and not buy insurance. The remaining pool of people is relatively less healthy and will therefore incur higher average health care costs, driving up actuarially fair premiums (AFP's). Again, the healthiest people in the group opt out of insurance and the cycle repeats, driving premiums ever higher. The individual mandate included in HR 3962 prevents this cycle by requiring most legal residents to carry qualifying health insurance, solving the problem of adverse selection and substantially lowering AFP's.
Second, the cost of insuring individuals is significantly greater than that of insuring groups due in large part to the economies of scale enjoyed by group insurers. Loading fees for individual policies ranges from 60%-80% of benefits, while those of groups of more than 1,000 are only 5%-8% of benefits.2 The public option is a group plan expected to cover millions of people, and should therefore enjoy significant economies of scale.
Third, the health insurance industry is not a perfectly competitive market, as evidenced by the fact that insurers are able to generate and sustain profits.3 In fact, individual health insurance markets are typically oligopolies dominated by their top three firms.4 As oligopolists, private insurers set prices above marginal cost (PO in Figure 1) and provide an economically inefficient quantity of insurance (QO).

Since the public option is required to (and presumably, only allowed to) break even, it acts as a competitive firm, setting premiums equal to marginal cost and thereby lowering premiums faced by consumers from PO to PC.
Finally, since information about benefits and premiums for the public option will be easily available via an online exchange established by HR 3962, consumers will enjoy increased price transparency, leading to more competition and lower premiums.
One factor that would make the public option more expensive for some people than it would otherwise be is the requirement that it not vary premiums to reflect differences in enrollees' health. While such a community rating system benefits sicker participants, to cover its costs it does so at the expense of those who are healthier, since they would pay lower premiums if their good health could be taken into account as with experience rating. However, some risk-adjustment mechanisms for the public option are specified and allow premiums for the oldest participants to be twice those of the youngest, helping to mitigate the issue.
By solving the problem of adverse selection with an individual mandate, taking advantage of economies of scale, setting prices equal to marginal cost, and increasing price transparency, the public option is expected to offer premiums that are lower than those currently facing people not covered by group insurance.
Effects on Premiums Faced by Those Covered by Group Insurance
Two other groups will be eligible for the public option in addition to individuals not currently covered by group insurance: employees of small businesses that choose to participate in the public option and workers with employer-based coverage whose existing premiums exceed 12 percent of their income.
Even though the public option will enjoy greater economies of scale than do private plans for small groups, it will still be smaller overall than existing large insurers and will, according to the Congressional Budget Office, "probably engage in less management of utilization by its enrollees and attract a less healthy pool of enrollees," and will, "typically have premiums that are somewhat higher than the average premiums for the private plans."5 Higher premiums will provide no downward pressure on premiums for existing employer-provided plans.
Effects on Total Medical Expenditures
Total medical expenditures are determined by the prices paid for medical care and the quantity of care consumed. The public option's impact on these factors will determine its effect on total expenditures.
While larger insurers are able to negotiate lower rates with providers, the Congressional Budget Office estimates that the administrator of the public option will have limited price negotiating power, and that therefore, "The rates the public plan pays to providers would, on average, probably be comparable to the rates paid by private insurers."6 In other words, the public option will not have an appreciable effect on prices paid for medical care. Its effect on total medical expenditures will therefore be determined by how it impacts the quantity of care consumed.
Demand for certain types of medical care is elastic, and in such cases the quantity consumed depends not on the prices actually paid to providers for care, but the effective prices facing the consumer. The public plan's limited cost-sharing makes these two prices different. For instance, demand is elastic for both well care and mental health care.7 Both are covered by the public option, so program participants will face lower prices for these services (shown as P1 in Figure 2) than the prices actually paid to providers (P0). These artificially low prices will lead people to consume more of these services than they would otherwise.8
Since the prices paid to providers are expected to remain the same but the quantity consumed should increase, total medical expenditures should be expected to increase.
Conclusion: Lower Premiums for Some, Higher Total Medical Expenditures
The public option is only one of many provisions of the health care reforms being debated in the Congress, and will affect more than just health insurance premiums and total medical expenditures. Nevertheless, its expected effects on these two important facets of health economics are clear. Health insurance premiums facing people not currently covered by government- or employer-provided programs should be significantly reduced, while those for others should remain largely unchanged. Prices paid to providers for medical services should remain unchanged, but consumption should increase. As a result, total health care expenditures should be expected to increase.
1 Sweeping Health Care Plan Passes House, The New York Times, 7 November 2009
2 Charles E. Phelps, Health Economics, Fourth Edition (New York: Addison-Wesley, 2010), p. 350
3 Fortune 500 2009, Fortune Magazine, 4 May 2009
4 Private Health Insurance: Research on Competition in the Insurance Industry, United States Government Accountability Office, 31 July 2009, p.3
5 Preliminary Analysis of H.R. 3962, Congressional Budget Office, 29 October 2009, p. 6
6 Ibid
7 Phelps, pp. 131, 134
8 This overconsumption due to artificially low prices, or moral hazard, creates deadweight loss, but welfare issues are beyond the scope of this paper.
No longer content merely to drift into irrelevance, Microsoft is now actively trying to sabotage itself—and nary a tear will be shed. Now, go follow @dondodge.
BusinessWeek stole the idea, and even the headline, for this blog post. But I thought of it first. And mine's better.
Steve Jobs is the CEO of the Decade. He's a multi-billionaire. And he's the best presentation presenter in the world. So what's his secret? Actually, he has exactly three of them. And by using these secrets, you can be the best presentation presenter in the world too. Here they are:
Also, feel free to go ahead and tell people that your products are the best products in the world, just in case they don't understand why you're smiling.
Joe Wilson last night made himself instantly notorious by shouting "You lie!" during President Obama's address to a joint session of Congress. He's up for re-election.
Wilson is being challenged by Rob Miller, a retired Marine who served two tours of duty in Iraq. I just made a donation to Miller's campaign and urge you to as well.
Here's Rob Miller in his own words on why he's running for Congress:
I never thought I would run for Congress -- or feel like I had to. I have always thought of myself as a Marine. I started thinking that way not long after my father died when I was 14 years old. Since I enlisted at the age of 20, the Marine Corps has been my extended family and I couldn't imagine that would ever change.But it did after my second tour of duty in Iraq -- where I fought to maintain the peace, but also to enable Iraqis to build better schools, improved hospitals, to help create jobs and, where possible, a stable economy and civil society.
When I returned home to my wife and six year old son, I saw that the politicians in Washington were not fighting for these same things at home. Worse, the congressman who represented the second district where I live -- Rep. Joe Wilson -- was part of that elite circle of George Bush loyalists who chose to enact a radical agenda, never mind the lives and wellbeing of the constituents he was pledged to represent.
Since last night, Rob Miller's campaign has raised over $500,000. You can track his progress toward $1 million here.
Last night Joe Wilson showed a shocking lack of respect for the presidency of the United States. He should now be shown the door.
In a post I wrote last February titled End of the Honeymoon for Apple and Google, I said:
Celebrity marriages are hard to maintain and often end in very messy, very public divorces. But for the moment at least, Apple and Google are still together. Eric Schmidt still sits on Apple's board, but now recuses himself from mobile phone discussions at board meetings. Presumably he'll need to start doing the same when the subjects of online productivity applications—or photo applications for that matter—come up.
The news this morning: Dr. Eric Schmidt Resigns from Apple's Board of Directors:
"Unfortunately, as Google enters more of Apple's core businesses, with Android and now Chrome OS, Eric's effectiveness as an Apple Board member will be significantly diminished, since he will have to recuse himself from even larger portions of our meetings due to potential conflicts of interest. Therefore, we have mutually decided that now is the right time for Eric to resign his position on Apple's Board." —Steve Jobs
This marks a sad turning point in the relationship between the world's two most innovative companies. I'm sure Jennifer Aniston, er, Steve Ballmer is pleased.
From the AT&T site:
iPhone OS 3.0 Software Apple® iPhone™ users, get started with Wi-FiIf you purchased an iPhone 3G S or updated your original iPhone or iPhone 3G with the latest Apple iPhone OS 3.0 software here's how to access AT&T Wi-Fi Services:
- Activate Wi-Fi from the settings icon on your iPhone.
- Once you are in an AT&T Wi-Fi Hot Spot, select "attwifi" from the list of available networks.
- Every time you enter an AT&T Wi-Fi Hot Spot, you will be automatically and securely logged into the AT&T Wi-Fi network.
- Enjoy surfing at broadband speeds.
Finally, a reason to go to Starbucks.
And before you ask, Macworld explains why this is a big deal:
Last year, AT&T granted its iPhone-using subscribers free access to what’s now 20,000 McDonald’s, Starbucks, airports, and other locations in AT&T’s U.S. hotspot network. But the process of logging in was tedious. At the hotspot, you had to enter your phone number, receive a free SMS, and then click a link to gain access. (App developer Devicescape made this process much easier via its Easy Wi-Fi apps.) iPhone OS 3.0 automates that process, not just for AT&T but for non-U.S. carriers who include Wi-Fi access in their plans.As soon as you’re in range of an AT&T hotspot, the iPhone logs in and switches to the network. AT&T was so excited about dumping its terrible old method of logging in that it even issued a press release. The carrier says that there were 4 million smartphone connections at its hotspots in the first quarter of 2009—and I’d bet that most of those were iPhones. (Here’s AT&T’s motivation in making it easy to use its Wi-Fi network: Every byte of data that moves from the cellular network to Wi-Fi reduces cell network congestion, making the overloaded cell network function better for everyone else.)
From Wikipedia:
In the United States, Flag Day is celebrated on June 14. It commemorates the adoption of the flag of the United States, which happened that day by resolution of the Second Continental Congress in 1777.In 1916, President Woodrow Wilson issued a proclamation that officially established June 14 as Flag Day; in August 1949, National Flag Day was established by an Act of Congress.
Here's a Fermi question for you: How many Googlers does it take to run Google?
More specifically, what's the minimum number of people required to run Google's money-making operations? Anyone care to hazard a guess?
(My estimate: the AdWords sales team, a few of these guys at each data center, and an accountant. Oh, and the Google Apps group. :)
According to published rumors the next iPhone OS supports tethering. But how important is this feature?
Initially I thought, "Awesome! Now my Mac can access the Internet even when there's no Wifi! I'll use this all the time!" But when I tried to think of how often that happens I came up empty.
The only places I find myself without Wifi are actually places that do have it but charge for it: usually Starbuck's, airports, and hotels. And at a rumored $15/month for tethering, it would be cheaper just to pay for the WiFi.
At our cabin in Colorado we have great Wifi but no cell coverage. Same with most convention centers. (Have you ever tried to make a call from the bowels of Moscone South?)
So while Apple and AT&T have been figuring out the details of tethering, WiFi has become more ubiquitous than cell coverage—at least for me.
Back up Google Calendar, Contacts, and Docs to the cloud with Spanning Backup. Spanning Sync customers save $10.
