We’ve had a series of “sector” events for the two broad sectors — next-generation infrastructure and digital media — where we focus our investments. We’ve had three annual Video Summit get-togethers and we’re coming up on our 2nd annual StorageFest next week (January 10th in NY, to be exact).
StorageFest 1 was an informal get-together of some 30 storage business executives, entrepreneurs, OEM heavy hitters, investors, and analysts. It was out in Palo Alto piggy-backed on Greg Duplessie’s ExecEvent last winter.
This year we’re getting north of 60 guests in NYC. What people get out of a meeting like this is intimacy and unfettered discussion. At SF2 we’re going to talk about two big topic areas:
Mind of the Customer. We’ve got several storage leaders from big customer organizations coming to SF2, and we’re going to have a free-flowing discussion about what customers actually want (as opposed to what the rest of us in the ecosystem often think they want). Curtis Preston will be proud of us!
Big Data. We want to separate hype from reality. What are the actual Big Data projects going on now and in the next few quarters/years?
Looking forward to seeing everyone. And stay tuned to #storagefest.
An anonymous guest chef (AGC) asked the Crummy Cook to help him make salmon burgers over the New Years weekend, and they were terrific.
The recipe came from the Frugal Foodie Cookbook whose authors, as it happens, are college friends of Mara’s. AGC had been given their cookbook by Mara and liked this recipe.
Here is how the patties looked before heat transformed them. It was not quite grilling weather, so we baked them in a 400 degree oven and put them on buns with guacamole (store-bought, I’m sorry to say). Results are here:
AGC was right. They are great, and reasonably healthy too. Thank you, AGC, for your inspiration and a kick in the butt to get back in the kitchen.
I may be missing something — in fact, in today’s info-overload world, I’m sure I’m missing many things — but I haven’t seen a good discussion of use cases for a smartphone beyond calendar and email.
Here are a few I’ve thought about. Would appreciate feedback about others:
“Recalculating route”… The other morning I was driving my rental car to BOS when my faithful Droid Bionic said “Droid”. I managed to read the text message (at considerable danger to myself and others, of course) and found out that my USAir Shuttle flight to NYC had been cancelled. What to do? I ended up going to the near-ish Amtrak station and parking outside while I figured things out on my iPad (bigger form factor). I looked up the Amtrak train schedule, thought I didn’t have enough time to return the rental car to the station instead of the airport and catch the train, so I looked up Delta Shuttle schedules, USAir schedules, and decided to stick with USAir in hopes of catching the next flight. I should have been able to do this all while still driving using voice and audio, or at least had some higher-level app or app-style workflow to make this an easier project. This kind of mid-course correction over multiple data sources while in motion is a key smartphone use case. If Siri could do this I’d jump ship.
Multi-app workflows It’s reasonably hard on most mobile platforms (smartphone or tablet) to read an attachment to an email, which has got to be the most rudimentary kind of multi-app workflow. My password manager RoboForm (and its cousin LastPass as well, as I recall) doesn’t work well on my Droid or iPad, and needs basically copy and paste do do what happens automatically on my PC. Imagine transferring information back and forth between a web site and a spreadsheet, working with a calendar and a map at the same time, etc etc, and you can see what’s missing from the mobile client.
Integrated Messaging On neither of my platforms can you view in the same “inbox” your email, text messages, and various IM and social-networking client outputs. Sad, but true. I even remember fondly the Blackberry integration of at least texts and emails. One place with a message timeline, like what Google is moving towards online.
A facility for musicians to virtually jam on the Intnernet. “Licks with Friends” or the like.
Not sure where things are at with the platform this would require, but I think we’re probably still a ways.
What’s the latency below which you are all playing in rhythm? Actually, half that latency I guess, since I have to react to your cue and play something back. That’s what we would need.
Plus it would be a nice-to-have to see the rest of the group as we play, although I’d settle for audio-only in R1.
I was talking with our senior partner Art Marks this morning about tech predictions for 2012, and he made a couple of interesting observations.
Back in the day Art used to run GE’s time-sharing service. At the time, some of his main headaches were:
Finding points of presence, on-ramps to the GE network. This involved painful and protracted negotiations with postal and telephone authorities, who viewed time-sharing as both a nuisance and a competitive disruption
Training users how to use the computer. Time-sharing terminals were unintuitive, inflexible, and unforgiving. Users had to supply all of these.
Finding and integrating applications for the GE service in order to incent users to sign up and to differentiate GE from other service providers.
Fast forward to today’s network:
With wireless connectivity shared between cellular radios and WiFi, points of presence are almost ubiquitous. Most carriers accept the fact that they must carry both “open system” and proprietary traffic, although there’s still some fear and loathing about that.
Users today train the computer how they want to communicate with it. OK, not quite, but we’re getting there. New services like Siri on the iPhone, voice search on Android phones, and location-aware services on all platforms, work with the user and attempt to be intuitive, flexible, and forgiving.
Instead of a few huge applications which are bears to maintain and integrate, users today use a multiplicity of small single-purpose apps which are increasingly mashing up dynamically to produce larger-scale results.
This slide — a bit tongue-in-cheek, but fundamentally serious — tried to slice history into three eras, one pre-dating Gutenberg, one from Gutenberg to now, and one going forward. The one going forward uses video a lot of places where we use text today.
Not implausible. But what will go along with text? I think sometimes the Enlightenment was predicated on following long text arguments rather than “short-form” video. We’re certainly going to lose our privacy in the coming era. Will we lose the whole Enlightenment sense of “self” as well?
I’ve now installed a few magazine and book apps on my Android phone and my iPad, and I’m struck by how many of them attempt to reproduce the look and feel and interactive style of their print forebears (like the New Yorker app, for example). (Same applies to catalog apps as well, like the Ikea catalog app.)
Result? A clunky product that has a huge throw weight over the network, takes a long time to download, is hard to cache, slow to navigate, and not a pleasure to use even though individual pages are often a pleasure to look at.
Haven’t the teams that build these apps read Nicholas Carr’s The Shallows? The Internet wants to grow our brains differently from the Gutenberg era. We need to react quickly to patterns of information, not delve deeply into a logical stream. We need to interact with the controls of our information rendering machine, not move our lips while we read words.
I’m being a bit arch, but there is a point here. If you want a media property — app, site, whatever — that draws a big audience (and I assume these vendors do), then build something that works with the grain of the underlying medium, not something that recalls a lost era.
The Internet of Things is one of those predictions that is always pending: this coming year will surely be the Year of the Internet of Things.
It will happen some time (although it’s neck and neck which will happen more slowly: the Internet of Things and the related advent of IPv6). But the emerging flood of devices will need better interfaces than they have today.
I have a few “Internet of Things” things in my life: a home automation controller, my home router, our networked printer, our home NAS, our TiVos, Most of these devices have a Port 80 browser interface, and they are almost all… terrible.
Slow to load, buggy, clunky, non-responsive (you click on a button and it doesn’t give you feedback that it got clicked). Ugly. Prone to crash or seize.
Reasons are straightforward: these are proprietary browsers, written in haste (by third parties, I should imagine), with no competitive pressure to improve or revise. They are evidence that competition is basically a force for good.
Is there a standard or standards to be set here to make the market larger, to make it worth somebody’s while to do the “non-computer edge device” browser once and for all, and right?
Cutter Consortium was kind enough to publish an article of mine — “Disrupting the Disruptors” — in their IT Journal, a special issue on “Creative Destruction” in the IT industry.
Here’s a link to a complimentary copy of that issue of the magazine.
Cutter is a huge panel of IT experts who will give you disinterested advice on most IT questions.
My article was on how incumbents might be able to defend themselves against disruptive technology business threats. A quick read, IMHO…
A recurrent meme at the Hadoop World conference last week was the idea that part of “Big Data” or even “the heart” of Big Data is “bringing the computation to the data.”
At first I thought that the main impact of this — beyond the very real observation that “shared-as-little-as-possible” architectures are great for scaling data processing — was poetic: it was sort of a democratization of compute power, or liberating compute power from the dark satanic mills of Oracle or the like.
But there appear to be architectural implications as well. A stateless or practically stateless approach to data weakens any hope of transactional integrity, for example. If you coordinate enough to be certain that everything will be undo-able, you’ll never get anywhere on your data. You need probabalistic assurances, not logical ones.
Also, new approaches will be needed for security and storage in an architecture where a vast universe of data/computation nodes coordinate. Maybe there are startups looking at this today, but would love to hear of anything interesting going on in these areas.
