Archive for the ‘Speculation’ Category

Anti-spam and Basic Compliance are now a Commodity

The Big News this week is Google/Postini’s dramatic price drop for hosted email security and basic compliance services.

Appliance and software-based anti-spam vendors will also feel the heat. With Google’s $3 per user per year pricing, the TCO tilts toward SaaS as the most economical (and perhaps reliable) way to get rid of spam. Does it really make sense these days to own and maintain a hardware appliance or software-based spam filter?

Who really knows if Google is taking a loss at these new lower prices to get more traction. They seem willing to seed the market for a long-term gain. And since Google is all about capturing as much data about everything, capturing data about spam and compliance feeds their voracious appetite for analytics.

One potential sting Google could feel is their privacy history. Will Google really resist the temptation not to analyze content information for future metrics, and eventually revenue? Competitors could force Google to “prove” there is no content leakage between the spam filtering side of the business and the search business.


Fixing Email, and SPAM, for good

It seems most of my recent posts have been about email, messaging and Web 2.0, instead of Sonian’s hosted archiving theme. This post continues the trend, and yes there is a method to the madness. Let’s dust off the crystal ball and take a look ahead… to a future where our electronic communications (email) are reliable, secure and spam free.

ZDNet writer David Berlind posted An Open Letter to Email Vendors on his blog expressing (justified) frustration with the current anti-SPAM solutions. He, like many of us, have implemented a rube-goldberg-esque approach to keeping our mailboxes spam-free. The thinking behind this approach is to layer a couple of different anti-spam technologies from different vendors, working at gateway, server and desktop levels, to get the best possible filtering protection. But in reality it’s not that simple or effective. Multiple quarantines and block and white lists create an administrative headache almost equivalent to the problem being solved in the first place. Maybe a possible explanation for our proclivity to use layers comes from our history with virus protection. Before SPAM was a real problem, email anti-virus was the big issue. And with virus control, a layered multi-vendor approach DOES work best.

The net net of it all is we shouldn’t have to tolerate this SPAM madness that’s purely a result of our own inaction and inability to come together as a unified group to solve the core deficiencies that plague SMTP-based email. The SMTP protocol has served us well (sort of), but it’s time for a top-down, bottom-up reset. SMTP was created with the premise that people who access the Internet would be doing good and honorable activities. Unfortunately, as is true in other segments of our society, it only takes a few bad people to take advantage of a system designed with the assumption that people are good. (Think airport security check points for a painful corollary.)

SMTP-based email is failing us, and as a result, we’re starting to see users find creative communication channels to get around all this “brokenness.” High school, college and twenty-something professionals are using IM, SMS and social network internal messaging systems like Facebook to communicate with each other. With the latter, email notifications are used to signal when a message is waiting, but the actual back and forth correspondence occurs within the protected social network boundary.

If the messaging industry can’t work together to solve SMTP problems in a coordinated manner, we will end up with a hodge-podge of different silo’d communication systems that provide rich functionality within themselves, but basic interoperability between each other.

My future for email wish list:

  • I get to choose my favorite email and calendar client technology – let’s call this the “super client.”
  • My super client is paired with a secure, reliable, hosted storage silo and processing engine. We’ll call this the “super server.” The super server is my “proxy” on the Internet.
  • Behind the scenes authentication and encryption ensure that I am the author of my own messages, and the system cant’ be spoofed.
  • A protocol that looks a lot like RSS allows me to “syndicate” my outbox. Instead of pushing messages from one server to another like SMTP does now, the RSS-like magic protocol notifies other peoples’ super servers I have shared content with them, and their super server retrieves the content. Like-wise their super servers notify my proxy content is waiting for me, and I can retrieve it as I like.
  • As the author of the content I retain ultimate control of my content’s final disposition.
  • This system allows me to create content to share with one person, or a group, (think email) or the entire world (think blog.)
  • The super client is where I live (it could be a web app or a local application, it doesn’t matter) – The super client is where I create my content (like this blog post,) or an email message to a friend. Or an instant message. They are all just forms of one to one or one to many communications.

Please share your ideas on how to fix email. We need massive, creative brainstorming to solve the problems.


10 Web 2.0 Apps CIOs Personally Use

Sonian is promoting the concept that mid-sized organizations can benefit most from a hosted archive service that embraces “web 2.0″ usability traits; such as dynamic AJAXy web UI’s that are intuitive and require no training, frictionless sign up and free trials, simplified work flow that leverages tagging and data categorization, and maintenance free (for the adopting customer) infrastructure.

I have been closely watching the up-take in web 2.0 applications and services being used in more and larger organizations. And others are starting to study this trend as well.

Over the past month major IT analysts like Gartner and Forrester have surveyed the attitudes and trends IT executives have about web 2.0 technologies.

It’s useful to define how the term web 2.0 is used here. “Web 2.0″ is a broad catch-all term to describe everything from YouTube, Digg and Del.icio.us, to Basecamp, Facebook, and Wordpress. For this post, web 2.0 means a web service that is easy to use, relatively inexpensive, and offers a focused feature set to solve a specific problem. Web 2.0 is most likely used for collaboration, project and personal/group information management. A key trait is a typical web 2.0 application can be adopted at the individual or departmental level.

We have seen this trend before: knowledge workers have a problem that needs to be solved, IT may not act quickly, so a solution is adopted “under the radar.” Email in the early 1990’s came in the enterprise at departmental levels before being deployed top-down across the organization. PDA’s and Blackberry’s followed the same “ground-up” enterprise penetration.

The survey results below show that CIO’s are starting to personally use classic web 2.0 services like wiki’s, blogs, and business-focused social networking tools. This could mean we’ll start to see a “top-down” implementation of web 2.0 in the enterprise.

Which of the following Web applications do you use personally?


N=150
Video over the web 54%
Wikis 49
Blogs 48
RSS (Really Simple Syndication) 47
Podcasts 39
Social networking (e.g., tagging, social bookmarks, community sites such as del.icio.us, LinkedIn, Technorati) 33
Expertise location and sharing 21
Mashups 13
Virtual worlds (e.g., Second Life) 12
Instant mobile updates (e.g., Twitter) 11
None of the above 11

Source: CIO Insight, August 2007

IT departments need to understand how strategically deployed web 2.0 IT services can be a good thing for the organizations they serve. Eventually end-users will find the solutions they desire (often times end-users are closest to and most knowledgeable about, their own needs.)

IT should get ahead of the web 2.0 movement.


Email circa 2012 Could Feel Like 1992 – and that's probably a good thing

Alan Leinwand over at the excellent GigaOm blog posted some thoughts about the future of data centers. His post was inspired when Intel announced their plan to have a 80-core processor by 2012.

…it is very conceivable that by 2012 we could have Intel-powered servers with 80-core processors interconnected by one-hundred-gigabits-per-second Ethernet connections. To fully utilize the processing power in these servers, they will probably run virtualization software that isolates processors to virtual run-time environments.

Whoa! 80-core processors with 100 gigabit Ethernet is an amazing amount of compute power. Just imagine the types of hosted applications that could be created to serve business needs, especially when applied to next-generation email and communication infrastructures.

Before I talk about what Email could be like in 2012, lets hark en back to email’s nascent days. Email circa 1992 was dominated by cc:Mail, MHS, DaVinci, MCI Mail, Compuserve, AT&T Mail, proprietary gateways, dial-up connections and MS DOS single-tasking email clients. During this time period Email started to mature from a novelty to a must-have communication infrastructure. Proprietary gateways and “closed” networks like MCI Mail gave way to SMTP gateways and higher speed Internet connections. The merits of SMTP and X.400 protocols were debated, and ultimately SMTP won the hearts of email administrators world-wide. The downside is SMTP’s ease of use and openness is also its Achilles heel. While this openness was initially attractive (how many of you remember what it was like to configure a closed system AT&T Gateway? – not pretty!), the SMTP protocol has brought a lot of pain and suffering with out of control spam and virus problems. It’s easy for me to write this now, with 20/20 hindsight, and identify where different choices could have had profound positive changes on our life today.

A popular saying these days is “we are where we are” – meaning we can’t do anything about what is done, but hopefully we learn from the past and chart a path for a better tomorrow.

So what does all this have to do with 80-core processor Intel chips? And why should email in the future start to look more like the past? Because we (the email industry) need to shake things up and start to lay the foundation for a new way for us as individuals, and groups, to communicate electronically. And this new way will probably look a lot like the old way. We need secure interconnected systems that guarantee message delivery, enforce identity and authentication, so we don’t have to worry about unlawful spammers and out of control email-borne viri. We need systems that treat electronic communication systems like a “utility” – meaning it’s assumed to just work all the time, like electricity and water. We need communication systems that have built-in compliance and policy controls, while at the same time preserving the privacy each of us deserves with our own personal missives. We need something like a modern version of MCI Mail (don’t throw anything at me :) .

80-core processor, 100 gigabit Ethernet, virtualized “monster” servers will be the perfect platform to sustain the “utility”-based communication systems of the future.


Google agog for GoDaddy?

eWeek posted an interesting article right before the Labor Day holiday titled – Google-GoDaddy Marriage on Tap? – speculation that Google might have an appetite for the GoDaddy domain management business. Google already uses GoDaddy and Enom to power the domain registration system behind Google Apps Premiere. Based on the recent Postini acquisition, Google has a history of “trying, liking and buying” enabling technologies and companies with smart people. Google started using Postini to protect Gmail and decided they needed to own the defense system

GoDaddy’s DNS knowledge would certainly be an asset to Google’s overall strategy of information management. Our modern-day Internet requires an excellent understanding of DNS management, and GoDaddy has obviously figured all that out.