Last week's Nanog meeting was in Miami, with Terremark as sponsor, again. (Terremark is about to be acquired by Verizon).
Matt Petach took detailed, almost verbatim, notes for much of the meeting - like me, he was listening and watching the video stream, not attending in person.
The highlights, from my viewpoint :
Distributed Denial of Service attacks are ramping up again, according to the Arbor security report. There was at least one attack (DNS reflection) at 100Gbps during 2010.
400Gbps (16 x 25G streams) is the next target beyond 100Ge
The second Arbor talk was about mobile networks - after the RAN and the backhaul, the traffic from them runs over the same infrastructure supported by Nanog operators, but there is very little
cross over in interest. 75% of mobile operators say security is poor, bad, or non-existent.
Jim Cowie (Renesys) ran a panel sesion in the afternoon which reviewed mobile network operations, and in particular the GPRS roaming exchange, modelled after AMSIX, and the proposals for IPX (not a Novell protocol, but a private over the top network for mobile traffic, and the issues with it for voice)
Jan Schaumann, Yahoo! described their L3 load balancer (L3DSR) development - Yahoo Mail is using it, and they will roll it out to other properties. Twitter is using this too - Yahoo will opensource
the kernel modules and iptables(8) plugin available for Free BSD and RHEL. A10, Brocade, and Citrix have built hardware support.
The silliest presentation was from Neil Farquharson, Alcatel-Lucent - his first time at Nanog, used the opening of the film Bambi (he'd been watching it with his kids) as an analogy to justify a unified control plane across IP to optics. He was encouraging people to come to a BoF at Denver.
IPv6 Deployment experiences wasn't streamed, though there are slides from Level3 (deployment experience, Hurricane Electric (deployment history, details), LACNIC (v4 runout effects), and SpaceNet AG (v6 routing table).
IPv6 day discussed - Yahoo!, and Akamai. There's a lot of brokenness .. the Day is an attempt to get a better idea of how much brokenness.
Shyam Mani, Mozilla, describes their experience in turning on DNSSEC, including the initial breakages.
BGP routing: Rob Shakir on his IETF draft which aims to reduce the impact on service availablity of session reset.
avoid sending notification in first place (avoid teardown!)
recover RIB consistency (something went wrong, but can we continue functioning?)
restart BGP hitlessly--avoid forwarding impacts while getting out of error condition
monitoring--as protocol gets more complicated, need to make sure it's still doing right thing
Route Flap dampening revived - Lixia Zhang presented her students' work - suggests decoupling flap detection, flap suppression.
Jim Cowie had a nice animation of how the Internet in Egypt disconnected - it reconnected again 5 days later.
Discussion of IPv4 runout.
Nanog 51 Agenda with links to slides and video
Last week's big ripple through the twittersphere was caused by Yuri Milner, again (his fund, DST, made a splash with its funding of Facebook and Groupon in 2009). He (personally, not as CEO of DST) and Ron Conway's SV Angel fund, are offerring $150,000 to each of the approximately 40 companies in the most recent Y-combinator group.
Lots of people wrote about it, first in breathless 'this changes everything' tones from Davos, and later in more measured 'there's lots of other fish in the pond' terms as there was more time to think through the consequences. My take is that it shows what can be done with comparatively small amounts of money by investors with the courage of their convictions. There will likely be more companies which can get their FMCA (Fast Moving Consumer App) out to market without requiring late stage or even early stage funds. Having achieved entry to Y Combinator, the startups have its contacts, its support network, to which this money adds heft. Other incubators have a model with which to approach potential funding sources.
The story on Techcrunch
Roger Ehrenberg (managing partner, IA Ventures) blog
Brandon Watson (Microsoft, ex Soros Private Equity) comments
Digital Sky Technologies background
Personal interviews with YC people
This time of year is when the weather in the valley makes it one of the most attractive places to be. We've just had more than 10 days of mild sunshine, with no frost, no rain, and day time highs around 16C. Roses are blooming, pear and early flowering cherries are in blossom, and there are ripe raspberries on the new bush in my garden.
New York, on the other hand, had 0.5m of snow during the week. Bejing looks to be cold and sunny. London has been somewhat warmer, but all those cities require coats, hats, gloves and heavy shoes - here it's been shorts and sandals weather, at least for youngsters in the middle of the day. Business goes on - in Menlo Park and Palo Alto roads were congested, traffic at a dead slow. 101, that sensitive barometer of business activity, was heavily congested in the SE direction but running at the limit going NW from San Jose to Mountain View .. both of those samples taken at around 5.30pm.
On the business front, Musemantik (from Edinburgh) are sharing a booth at the Games Development Conference taking place in San Francisco at the end of February. Contact me if you are interested in meeting with the founders to learn more about tools which formalize and computationally control real-time emotion enhancement, using new or existing music, to make a more intense experience for game players.
Last week Peter Thiel (ex PayPal) talked at the NextGen conference about the investment perspective he uses for the Founders Fund.
He's interested in things which can make a big difference - if you are starting a company, you are taking so much risk that you might as well try to do something big. Pick a contrarian field, rather than dive into the current bubble. They are interested in artificial intelligence, robotics, space, and next generation bio-technology, among other things. They look for talented people passionate about changing the world. Only some thinking should be outsourced to others. Companies with a complex sales cycle are underestimated.
All of this is encouraging for someone who has made a habit of looking out for the ideas and technologies which are likely to have long term fundamental effects, as an indicator for where to invest time and attention. It feels very different from the 'chase the current meme' which appears to drive many investors' decisions.
Talking of current themes, Brad Feld said today he's seen several proposals for 'the Quora of X' - people taking advantage of the meme for curated Q&A to use it as an explanation for their latest idea.
Quora is one of the better Question and Answer forums - I joined it last August. To get started, one can follow Facebook friends, Twitter followers, and/or topics. Anything one writes is public - so over time, answers accumulate into something that looks a lot like a blog. There's some thoughtful design keeping the pages simple in appearance, while linking back to the original questions to preserve context. There's an accumulation of reputation, too, as logged-in readers can vote up and thank for answers. So far, it's not obvious what the business model will be (and the question has been asked).
Robert Scoble, who tracks interesting technology trends and write about them, recently discovered Quora, and is enthusiastic - when Robert is enthusiastic, several thousand of his followers know all about it, and quite a number of them joined Quora too. It seems to have survived that onslaught.
Roger Ehrenberg, at IAVentures, has been writing long form answers in the Venture Capital and Startups topics - a couple of days ago he asked (on Twitter) whether he should cross post his answers from Quora to his own blog. Quora makes it easy to send a post to Twitter pointing to a new Quora answer. Adam Lasnick, who's a Google Webmaster, has a good list of pros and cons.
Like him, I conclude that the visibility, associated possibility of helping more people, and ease of use outweighs the disadvantages - like having much less visibility into who sees your writing. You note, however, that I'm writing this as a blog post on my own site - it's not a question, and some of the audience for my blog hasn't (yet) found Quora ...
Stackoverflow is a longer established Q & A site, much more tightly focused for programmers - reputations built there are an obvious target for recruiters, as Robert Scoble pointed out yesterday. Hunch is also driven by questions, and gives answers - initially it looked like a trivia game, but has become more obviously a way to recommend products by matching people's preference patterns.
All examples of crowd sourcing - building a site where the users of the site build the content and the value. Facebook are past masters at this (and Goldman Sachs agrees, given its $375m ($75m to Digital Sky) investment). We are being their digital peasants. Time to build our own castle ?
My purpose for attending SF MusicTech Summit was to explore a different part of the technology business. My background and experience is mostly in the plumbing of the Internet, in the switches and valves which connect the bit streams, both to make them go faster, and in filtering out the evil bits. The music business, like many other application areas, assumes that all of that infrastructure exists, so that their customers can get any bits they can explain that they want. To them the network really is like electricity and water - available for a small fee, and they need pay no attention to how it is constructed, since their bottlenecks are not the network nor anything in the computing infrastructure. Their challenges are finding people who want to find them - competing for attention, which is a scarce commodity.
Hardware Demos :
Yobble showed a device which connected a movement sensor in a guitar pick to an iPhone - making it possible to get an 'air guitar' effect, either modifications to existing music or freeform new music - using 'buttons' on the iPhone for note selection. Target market is people who have iPhones and enjoy Rockband.
HRT played another device - this one connects to an iPhone on one side and a good quality entertainment system on the other. The embedded Digital to Audio converter bypasses the DAC in the iPhone and produces audiophile quality sound from a box about 6" x 2 1/2" x 1" ; there's a USB version too, prices about $199.
The Developer Platform panel discussed what it takes to get your app noticed - attend Music Hack Days, and/or engage with Appbistro - Ryan Merket, formerly of developer relations at Facebook, was articulate and passionate about the potential for music apps on Facebook.
Most interesting was the last panel, What Music Companies need from Startups, chaired by Ian Rogers. Ethan Kaplan (Warner) and Aaron Foreman (Universal) were there from the big labels - they were quite specific. They want an app, app keys, some Python or PHP - and a clear demonstration of the technology behind the proposed solution, along with a demonstration of how it adds value to what the label is doing, and a way to measure that it is really doing what it claims. Don't expect to get a hearing unless you have a test plan ! Ian Hogarth (Songkick) and Rachel Masters (Red Magnet) were all singing the same story - it's all about the musicians and the fans. It all sounded a little forced - the music is a product, like many other results of creative talent.
There was remarkably little live music actually played. There was a grand piano with a cover on pushed under the stairs beside the downstairs toilets in the hotel - after the sessions finished two people had opened it, and were playing, prompted by chord progressions on a iPhone propped on the music stand.
Other people who know much more about the music business blogged about the conference too.
Several weeks back Brad Feld (Foundry Group) and David Cohen (Techstars) started a book tour to promote their book - Brad writes about the process on his blog. One of the first events of the tour was held at Silicon Valley Bank's Palo Alto office - Brad and David talked a little about the writing exercise, answered questions, and passed out copies of the book to attendees. Last week David Cohen made a polite follow up request for reviews, so here is mine.
Brad Feld is a Managing Director at the Foundry Group (Boulder, CO), a venture capital company, and with David Cohen founded TechStars. Techstars is a 13 week mentorship-driven seed stage investment program which runs several times a year, in, now, 4 US cities, having started in Boulder. The book is a lot like a collection of blog posts - it draws on the experiences of the people who attended Techstars, and on the people who mentored and invested in them. If you've followed me @vcwatch on Twitter, the intended audience is the same - people starting companies, people operating small technology companies, people looking for investors, investors looking for teams and ideas, particularly seed-stage companies.
The contents include how to identify ideas which might turn into a product, how to tell when you should change your mind about what you are doing, finding a co-founder and a team, measuring what you are doing, fundraising, legal, and how to have a life as well as start a company. If you intend to apply to TechStars, you should read this book, since the other applicants will have, and it gives you a head start on the process.
TechStars, and this book, are as Brad acknowledged about a very specific kind of business - one where the customers can be reached via the Internet, and where it is possible to get to a point where value is demonstrated in less than 3 months. It's not clear whether the mentorship and small amount of funding model applies to businesses which require a larger scale implementation to tell whether they are viable. Biotech and cleantech businesses, for example, or anything which requires new ASICs, take much more money and much more time to show a return.
If your first books about business were 'A Passion for Excellence', Tom Peters, pub 1984, and or 'Thriving on Chaos', Tom Peters, pub 1987, and you haven't been paying attention to very small startup businesses in the last 5 years, you can catch up by reading 'Do More Faster' - and by thinking about what is implied for the dynamics of business by the kind of companies described here. The ideas are sometimes very limited, and the company lifetime short - some of the companies funded in 2007 at the first TechStars have been acquired and their founders are investing in other newer companies.
Another group who should read this book are people working for governments which would like to see the same kind of small company ecosystem described here in their jurisdiction - and they should notice that the book says nothing at all about government assistance. There's a short section on the choice of state law in the US (found the company in Delaware, not any of the other 50 states), and on picking a lawyer. There's no government support mentioned, nor required. Implicitly, it is fairly straightforward to found and operate a company in the US, and if it isn't at least as easy to do that in your jurisdiction companies will want to move away from you, diminishing tax take and the possibility that successful founders will do the trick again, while employing people who might also catch the company formation bug.
The book has a decent index, and has very few typographical errors - unusual for first time authors.
It is worth the time to read, to get perspective on the kinds of ideas which have been the kernel of new companies, and the accelerated process by which those companies fail and succeed.
Cunning Systems evaluates product and service ideas in computing and communications. If you would like to discuss an idea, contact us at email@example.com
Content Networking, or Content-centric Networking as PARC are now calling it, is an architecture which Van Jacobsen has been promoting for some years now. It's worth attention, both because of Van's reputation and because it's a very interesting answer to the question "knowing what we know now about what people want and what capabilities can be provided, how should networking and computing be structured ?"
"Content-centric networking is PARC's vision for taking the next step in data communication — a change in network architecture to make content retrieval by name, not location, the fundamental operation of the network. Our approach is to reuse and build upon successful features of TCP/IP, with the key change of replacing the machine-oriented IP model with a named content model as the basis for the central protocol that connects networks."
If networks work natively in terms of content – in other words, the name a user would express to the network to retrieve a piece of content refers in some way to that content item itself, rather than to some specific host on which the content can be found, then users can securely retrieve desired content by name, and authenticate the result regardless of where it comes from. (Refer to the paper on Securing Network Content)
Pere Monclus, with whom I used to work when at Cisco, talked at PARC a couple of weeks ago about how difficult Cisco finds it to implement support for new stuff like this.
23 Nov 2010 - Updated
As Named Data Networking, there is NSF funding for further research into CCN. The PARC technical report NDN-001 (Oct 2010) describes the project in more detail.
PARC's content-networking page Discussion between Van Jacobsen and Craig Partridge ( free,