As many of my readers know, I’ve got a small record label, in partnership with my band-mates from Soul Patch. The genesis of Toothless Monkey Music (the name is another story entirely) starts when I set up my first “real” recording studio in a sound-proofed, detached two-car garage at my former home in Portola Valley, CA. It is on an idyllic creek-side spot, and many of the albums on the label were recorded and mixed in that studio.

One day, back around 1999 or 2000, when the studio was not-quite-fully operation, some friends dropped by, and a phenomenal luthier from the Santa Cruz area named Fred Carlson was with them. Fred builds some amazing instruments, starting with guitars, but ending with sympitars, harp guitars and other one-of-a-kind works that defy categorization. Check out his site at Beyond the Trees.

Anyway, Fred sat down and performed a song, which I am assuming is called Rattlesnake, but I don’t know for sure. We had one microphone set up hanging above him that captured the performance. My band-mate Nick Peters dug it up from his archives, and I did a little bit of mastering to add some stereo imaging, some gain and to bring out the vocals a bit, since the original recording wasn’t as good as it could have been — the performance was great, but I was a very green audio engineer at the time. It is a fun, quirky song, and well worth a listen. Enjoy!

A tip of the hat to my friend Dennise for sending this one to me, and another tip of the hat to the folks at Topatoco.com who created this fine technology / science cheat sheet, which would be quite useful in rebuilding modern science and technology if one were suddenly thrown back in time or we all wind up back in the stone age through pure human folly, something I place higher odds on these days…

My partners and I at Foundry Group have been involved with Eric Norlin for the past few years, and he has put on two great Defrag Conferences in Denver over the past two years (the inspiration for which came out of our work on our Implicit Web investment theme), and now we’ve decided to add another one to the mix: The Glue Conference, which will be held on May 12th and 13th in Denver.

The idea for the Glue Conference came out of our work at Foundry Group on our Glue investment theme and brainstorming with Eric about how we could take the basic notion of the web as a platform and dig in at a fairly technical level into the problems, challenges and opportunities that arise when one assumes the web and the cloud as a given and as the fundamental platform going forward.

Glue’s got a great agenda, including keynotes from the likes of Mitch Kapor (Lotus founder), Bob Frankston (VisiCalc creator) and Josh Elman (Facebook platform).

And, as far as conferences go, it is a bargain, only $495 for two days, so go ahead and register here. The Foundry Group partners will be out in force, and Seth and Brad (and I’m sure Jason and I will join in the fun at some point) will even be sitting down to listen to pitches from entrepreneurs during the conference.

Today Netflix announced that they delivered their two billionth DVD, an impressive milestone. This brought to mind something we learned in the early days of Excite, which was to never underestimate the bandwidth of physical storage media sent via UPS, the USPS or FedEx. When we opened our second datacenter circa 1996 (on the east coast in one of AOL’s datacenters) we quickly found out that it was faster, cheaper and more reliable to simply FedEx overnight an archived backup tape copy of our search index to the east coast mirror site than to transfer the files over the internet.

I’ve always thought that Netflix’s business was a brilliant bet that the bandwidth and quality of a rental experience powered by DVDs sent via USPS was going to be cheaper and exceed the capabilities of on-demand via the internet for much longer than people were expecting, and of course this turned out to be true. And now that internet tech is finally catching up to the low tech method of shipping atoms full of bits around the country, I think Netflix has done a brilliant job with their internet strategy and distribution partnerships, which should enable a graceful (and still longer-term than people expect) transition from postal delivery to internet delivery.

So when I saw the announcement of the two billionth DVD delivered, I decided to do a quick and dirty back-of-the-envelope calculation of how much data Netflix has delivered to customers in the roughly 10 years since their subscription service launched. (Apologies in advance to all the sticklers out there who might point out the imprecision of this calculation since I’m using factors of 1,000 instead of 1,024 to measure my gigabytes and petabytes.) Here goes:

A DVD has a max capacity of 4.7 gigabytes. Since Netflix also ships TV shows and the like, which don’t fill a DVD to capacity, let’s assume that the average DVD contains 4 GB of data. So, that means they’ve delivered eight billion gigabytes, or eight million terabytes, or eight thousand petabytes, which boils down to an average of 800 petabytes per year over a ten year period. Multiply that by 8 bits per byte and divide by 31,536,000 seconds per year, and you get 202,942,669,000 bits per second, or a sustained ten-year average bandwidth of 200 gigabits per second.

Of course, that’s an average spread evenly over 10 years, and today’s outgoing bandwidth from Netflix via the USPS is many times higher, given the ramp from zero DVDs shipped in the early years and given the fact that Netflix is now shipping Bluray discs, which hold 50GB vs. the 4.7GB on a traditional DVD. And, finally, given they are actually delivering movies via the net, they are now using actual net bandwidth instead of theoretically derived, USPS-enabled bandwidth, though I’m sure their actual bandwidth consumption is still dwarfed by the discs in the mail.

But still, I was surprised by the 200 Gbps number and had to check my calculations a few times to make sure I was right. And it is conceivable that that means that their theoretical bandwidth today could be in the terabit per second range. Golly.

So, the best corporate April Fool’s joke I’ve seen today comes from Antares Audio Technologies, makers of the ubiquitous Auto-Tune pitch correcting audio plugin, which is used extensively throughout the recording industry, though some producers and vocalists will deny it. There’s a great article in the New Yorker about the impact Auto-Tune has had on popular music.

Today they announced the upcoming release of Direct Mind Access (DMA) Composition Technology, a faux product that claims to enable composers to “think” their compositions directly into a computer, provided they’ve undergone the requisite cranial implantation of the iLobe USB device.

I’m amused that this is the second fake mind-reading device I’ve encountered in as many days (see my post from this morning for details on the other one). I’m also amused because while these two products are spoofs, there’s clearly a meme around mind-machine interfaces picking up steam out there, which fits quite nicely into Foundry Group’s HCI theme, and is likely influenced by the fact that there are real companies in the market today (including companies like Emotiv, NeuroSky and Foundry Group portfolio company EmSense) building real technology and real products based on EEG technology.