There is a cliche that many people often share about history, and how it is written by the victors. The conquerors across the world, for the most part, are the ones who transcribed the history for others to read. We have terms like "revisionist history" to account for the fact that we realize the truth may be quite a bit grayer than those historical accounts we read lead us to believe. And it is often the sole dissenting voice that points us toward what the true history may be – some in between version where the good guys are perhaps not quite so good, and the bad guys not so bad.

It is easy to think of these shades of history while traveling in Greece (where I spent the past week), because the evidence of this is all around. From the city of Acropolis high over Athens to the softly smoldering volcanic remains in the caldera of Santorini. What was once religion is now called mythology, a sign perhaps that our religion of today may befall the same fate. Amongst those ruins of temples and palaces, there is the beautifully frustrating knowledge that we will probably never know what these really looked like or how these people truly lived.

The history today will be different. With technology and social media, we have the effortless ability to capture our individual truth and experience in minute detail and save it on shared servers for the world to access hundreds or thousands of years from now. Ironically, this fact may make the study of history that much more complicated, as historians in the future will have many versions of truth to study and contrast. Rather than piecing this history together through buried bone fragments and stones with the rare written account as they have done for many years, they will do it through compiling and sorting data, analyzing imagery and watching video.

Combined with global historical initiatives such as the Internet Archive and Google Earth, the portrait of our world as it stands today will be far more complete and multi-experienced for historians of the future than any other age that has come before. For those of us who write blogs or upload photos, the scope of our actions are easy to forget or minimalize. But we are the new historians of our time and our content will one day be history – and probably for more than just ourselves. It’s a humbling to imagine your own words on this scale … particularly when you think of who could be reading them a long time from now.

Comments

But I always assumed that companies would have at least one server keeping things up, just in case Amazon went down. Or just because.

I was wrong.

Last night Mogulus’s CEO, Max Haot, was here at my house to film something fun for my show. Mogulus is the company that, yesterday, provided the live video for Om Malik’s NewTeeVee conference. It was so good I stayed home and watched almost the whole day on the NewTeeVee channel. But more on that when we get the video up.

At one point Max seemed like he was joking around with me when he told me “we don’t own a single server.”

I asked him FOUR more times to make sure I heard him right. I even got incredulous with him at one point saying something like “what the f*** do you mean you don’t own a server?” and “you mean not a single bit of your Web site comes from servers that aren’t owned by Amazon?”

He nicely and calmly explained that, yes, every server the company owns is actually running on Amazon’s S3 and EC2 services.

The world has changed. Now ANYONE can build an Internet company and get it up to scale. No more spending nights inside data centers trying to keep servers running.

Let’s go over to Mike Arrington’s CrunchBase and do some research. They pulled in $1.2 million in funding. Yet they don’t own a SINGLE server!

They have about 15,000 people already creating live video channels. They have one of the most innovative Web sites I’ve ever seen.

But they don’t own a server.

How else has the world changed? Where the hell is Microsoft in this whole business? How did Microsoft screw this up so badly? Let’s get this straight. Amazon used to be a book store. Now they are hosting virualized servers for Internet companies. So much for having billions of dollars in the bank like Microsoft does, some of the smartest people in the world working in your research arms and having “monopoly” market share in operating systems.

Heheh, maybe now Amazon can use some of the new money that they’ll be earning from these startups to buy some decent PR. According to Read/Write Web Amazon needs the help in that department.

Oh, back to Max. One tip he gave us is that when using Amazon’s services you have to design your systems with the assumption that they will never be up and running. What he means by that is services are “volatile” and can go up and down without notice. So, he’s designed his systems to survive that. He told me that it meant his engineering teams had to be quite disciplined in designing their architecture.

How many other Internet companies are out there that are “serverless?”

Comments

This results in eight versions of Server 2008, three with Hyper-V, three without, and two specialized types that don’t have Hyper-V:

• Windows Server 2008 Standard: $999 (with five Client Access Licenses, or CALs)
• Windows Server 2008 Enterprise: $3,999 (with 25 CALs)
• Windows Server 2008 Datacenter: $2,999 (per processor)
• Windows Server 2008 for Itanium-based Systems: $2,999 (per processor)
• Windows Web Server 2008: $469
• Windows Server 2008 Standard without Hyper-V: $971 (with five CALs)
• Windows Server 2008 Enterprise without Hyper-V: $3,971 (with 25 CALs)
• Windows Server 2008 Datacenter without Hyper-V: $2,971 (per processor)

The difference in price between the Hyper-V and non-Hyper-V versions is $28, and you can buy Hyper-V Server by itself for that same price.

If you were reading here last month, you’d have noticed Microsoft registering tens of domains related to Hyper-V, advance notice that this announcement was coming.

Comments

Tag:

Jonathan Kauffman, General Manager for Enterprise Search with Microsoft, wrote at the product’s blog of its capabilities. He cited several features like relevancy tuning and security-trimmed search results as being part of the product.

It will also support the OpenSearch standard, for federating searches across products. Kauffman acknowledged federation support from EMC, Cognos, HP, Business Objects, SAS, and OpenText. Connectors for those products should arrive with the full release of Search Server Express 2008 early next year.

Though Kauffman claimed corporate passion about giving away useful tools as a motivator for making Search Server Express 2008 available for free, there’s probably a couple of other reasons that apply. Competing with the IBM/Yahoo product would be a classic Microsoft tactic.

Search Server Express 2008 also serves as a gateway product to more advanced, for-pay solutions, just as IBM Omnifind Yahoo Edition does for IBM. In Microsoft’s case, it leads to Search Server and then to Office SharePoint Server, the latter product likely ensuring plenty of Microsoft products in the enterprise accompanying it.

Delivering a capable enterprise search product for free also gives Microsoft an edge against its hated rival, Google. The dominant search company sells its Google Search Appliance in a couple of form factors for business, as well as offering the free Google Desktop for Enterprise for local PC searches.

Quick Overview

Gmail has very effective spam filters that we can utilize to filter our non Gmail email. There are many ways to set up filtering with Gmail with Cpanel. One way to do this is to forward our mail to our Gmail address. Then we can respond using our shiny new Gmail account and have all our email filtered. If you want to continue to use your old email address though this doesn’t work well.

A second way to do this is to forward all email to Gmail and allow Gmail to send email as your original domain. Allowing another server to email on your domains behalf is called “relaying” and done wrong can be a source of spam. Smartly Google requires a special server setup that tells them they have permission to send email on your domains behalf. The setup is called SMTP AUTH and isn’t standard with Cpanel so it was out for me.

A third way is to set up our email to handled by Google. In order to do this you sign up for Google Apps and set up your dns records accordingly. I think this option is great for regular family sites or beginners but for my business. If you don’t have or are uncomfortable using shell access then this option might be best for you. I chose not to use this though because don’t want to be left high and dry if Gmail encounters any problems.

Lastly you can set up a series of rules at Gmail and on your server to filter your email. The first rule is set on your server and forwards your email to your Gmail account if its not been filtered yet. The second rule is setup at Gmail to forward email to your email server if its for your domain. The last rule is on your server that checks if the emails been to Gmail yet and then routes it to the appropriate mail box. This option works great but requires shell access to your server and can be tricky to setup with Cpanel. This is the option I chose to use.

How I Setup Gmail Filtering with Cpanel

I chose the series of rules option but I needed to modify it a bit to work with Cpanel. You can not add the rules using the Cpanel mail filter in the control panel because you can’t add all the necessary rules. Since thats the case you need to edit the filter file manually on the server. On a regular *Nix server with Exim as the mail server you would put a file called “.forward” in the root of your home directory. The “.forward” file would have all the rules you exim to process when it handles your mail. On a Cpanel server this doesn’t work though. There is a directory with files for filters for all the domains on a Cpanel machine. The directory is called “/etc/vfilters/” and the file name will be your domain name. With VI I opened and edited the file adding the rules noted here. Then I added the rule at Gmail to forward my mail back to my server.

Here is an example of my setup:

In the file “/etc/vfilters/web-professor.net”:

# Exim filterif not first_delivery and error_message then finish endifif $message_headers: does not contain "X-Forwarded-For: myGmailAccount@gmail.com"	then		deliver "myGmailAccount@gmail.com"	else		deliver "admin@web-professor.net"	endifendif

In my Gmail account I went to the Settings page and then to the Filters tab. I set a filter that causes Gmail to forward all email addressed to a the domain “web-professor.net”, and keep a copy in Gmail’s inbox. To do it yourself just go to the advanced filters and enter your domain in the “to” field (no need for wildcards). Then select forward for the action and fill in the email address.

The Results

For the first couple of hours Gmail let through most spam but after marking the bad stuff off and on it catches most of it.  For me that means it stopped ~2000 spam mails  and let ~5 through this morning. To contrast these results with my previous spam filtering  I was receiving ~80 spams messages a day. With some more training I am confident that those numbers will get even better.

Postscript

I hope this helps you set up Gmail filtering for yourself. I am sure there are other ways to do this this is just how I solved it. Also not that the solution I posted sends all the email back to one email address. That might not work for you if not then take a peek at the exim filter documentation. You should be able to come up with more complex filters and rules to forward email to more than one email address.

Comments

The companies will jointly deliver platform enhancements and solution accelerators focused on technology and industry deployment.

Microsoft is investing more in Commerce Server, the company’s main server for developing ecommerce services. The Commerce Server can integrate with partner applications, manage and share business data along with offering customers online-based self-service with access to inventory availability as well as order status.

"Customers can expect to benefit from the combined value of Microsoft’s expertise in building leading enterprise platform and Cactus’ industry specific solutions and deep knowledge of customer needs, product customizations, and service integration," said Jean-Yves Martineau, Founder and CTO at Cactus Commerce.

"This tightly closed loop between development and implementation will also enable Cactus to leverage the successes we’ve achieved with our enterprise customers to influence and drive the Commerce Server product roadmap going forward."

Cactus Commerce ecommerce solutions will be available through the Commerce Server Information Desk program to Microsoft partners, customers and the Microsoft sales organization.

"The agreement with Microsoft accelerates access to Cactus’ expertise and their unique software assets for commerce," said Joe VanWinkle, Vice President of Global Market Development at Avanade.

• Ebay Purchases StumbleUpon for $75 million
• FeedBurner Sold! for $100 million to Google
• Google Buys Panoramio, Photo Mapping Software
• Fox Interactive Media (NWS) Acquires Photobucket and Flektor
• Google buys GreenBorder Technologies for more security in Google Apps
• Microsoft buys aQuantive Inc, for 6 billion US dollars
• CBS buys Last.fm
• AOL Buys Mobile Ad Network
• Microsoft Buys CareerBuilder; Google Interested In SimplyHired
• AOL Buys German Ad Server Adtech
• Yahoo Fully Acquires Right Media (April 30th)
• Facebook Partners with Oodle for Classifieds

Google and Microsoft seem to be positioning themselves strategically in order to keep up with each other in the war for Internet dominance. I expect to see this trend of acquisitions continuing strong into the summer.

Do leave a comment if I have missed any major partnerships or purchases.

Comments

Tag:

Jerry Yang and I just announced at our quarterly employee all-hands that Yahoo! has committed to going carbon neutral this year. Essentially, that means we’re going to invest in greenhouse gas reduction projects around the world to neutralize Yahoo!’s impact on the environment. While doing our homework on this, we measured our carbon footprint and discovered that Yahoo! going carbon neutral is equivalent to shutting off the electricity in all San Francisco homes for a month. Or pulling nearly 25,000 cars off the road for a year.

I like that we’re able to quantify it in ways that are understandable to people, rather than talking about XXX tons of carbon emissions.

Personally, I’ve been paying a lot more attention to energy issues since last October when I wrote about replacing my home backup server with Amazon’s S3. Even though I don’t know that Amazon is able to do the job with less power than I can (I have to believe they’re more efficient than my setup), the eye-opening thing to me was the actual dollars saved from PG&E if my calculations were right.

Since that time, my home computers have been powered off much more of the time and I’ve actually seen changes in my energy bill. Plus, I replaced my dying 20+ year old furnace last December with one that’s 90+% energy efficient: the Lennox G61V Variable Speed Gas Furnace.

Again, lower bills and better for the environment.

Being that I live in Silicon Valley and work for a high-tech company, I’m glad to see that so much of the interest, funding, and innovation around greener energy is here too. I can only hope that more of the country follows along, just like they have with California’s stricter auto emissions standards.

Thanks to David, Jerry, and the Yahoo! For Good team for helping put our resources to work on this.

Comments

With all the effort to push MHz, little regard has been made for power consumption, to the point where, parts become too power inefficient and temperature management becomes a problem. Also, cost of watts surpasses that of hardware until energy cost becomes the main expense for the data center.

Luiz also talked about predicting hardware failures:

The common wisdom is that disk failures are less than 1% per year, and that temperature increases failure. For a predictive failure model, Google uses SMART (Self Monitoring Analysis and Reporting Technology). This collects signals that may detect bad media surface, bad servo components, electronic/transmission problems, and vibration.

*The talk’s title: “Watts, Faults, and Other Fascinating Dirty Words Computer Architects Can No Longer Afford to Ignore.”

[Thanks Bill! Photo by Bill, with permission.]

Comments

 UPDATE: I’m Twittering it. Curtis Wong (who is one of the smartest people I met in my Microsoft travels) just showed off Sky Server, which lets you walk around the sky, much like you can with Google Earth. This is inspiring, it’s like having the Hubble Telescope to walk around right on your desktop.

It also demonstrates how much we lost when Jim Gray sailed his boat out of the Golden Gate never to be heard from again.

Comments