Fargot Password? / Help


Why the "NoSQL" moniker is useful - for now

There's been a lot of crowing as of late from folks that wish the whole NoSQL "movement" (where "NoSQL" means "not only SQL" - but gets misinterpreted a LOT) had been named something else. Since the whole movement is essentially engineers at this point, I thought I'd weigh in (I'm just a lowly marketing guy, so reader beware). The NoSQL moniker is, frankly, about as good as it gets for an early adopter technology. Why? Because it's polarizing. Of course, the engineers hate it -- it doesn't *accurately* describe anything about what's going on there (in case you haven't heard, engineers love accuracy and precision). However, when a new technology is gaining ground, marketing matters. And when marketing matters, the BEST thing you can have is a "storyline" that writes itself for the journalists, analysts, etc that will help to distribute these ideas inside of enterprises. "NoSQL" does just that. It's got an implied conflict in the name, and the single easiest story to "write" is one of conflict (see, oh, I dunno, ALL of the mainstream media outlets). Folks on both sides of the debate should be embracing this name for the time being. Reason: the name provides a context that will get attention, and early on attention equals adoption (pilots, etc). I'd contend that the name will fade - especially as "NoSQL" constructs get integrated into SQL data models. Now, the zealots will always want the divide. But the truth is that the divide will disappear over time as the technology matures, and in five years time, we'll all look back at "when we used to call that stuff NoSQL." In the meantime, engineers -- embrace it. Yes, the moniker isn't accurate. Yes, it feeds the flames of inaccuracy. Yes, it's an upstart naming convention that begins by shooting an arrow at the "old man" of SQL. That's all fine and dandy. It's actually *exactly* what you want. (Don't believe me? Go study Oracle's early ad campaigns -- which were brilliant -- for proof.)

Michael Stonebraker to Keynote Gluecon

I told we still had some tricks up our sleeve. ;-) I'm very very pleased to announce that Michael Stonebraker, creator of Ingres, Postgres and countless other relational database innovations, has signed on to deliver a keynote at Gluecon. I was poking around Mike's bio a bit this morning, and found this great quote on the IEEE site: "Virtually all business computing applications - from accounting, inventory and shipping to point- of-sale, online commerce, human resources and computer-aided design - manage data using relational database systems that can be traced to his work." And that's really not an overstatement. Mike Stonebraker's work and research has been foundational to the multi-billion dollars per year business that is databases. He's a legend, and we're honored to have him join us. We're still talking through exactly what Mike will be speaking about, but I have no doubt it'll be amazing. And if you're keeping track at home, that now means that Gluecon will open with two giant talks about data models: Eric Brewer (of CAP theorem fame) on "wide area data management," followed by Mike Stonebraker. I think it's an ideal starting point before we delve into cloud platform providers, integration in the cloud, NoSQL examples, Identity protocols (SAML, OAuth, etc) and the rest of the awesomesauce that is Gluecon Day 1. Take advantage of that early bird pricing, and register now.

Early Bird Expiring

I've written about the value. I've pointed to the agenda. I've blogged about the keynotes we're lining up. And now we're coming up on early bird pricing expiring. Bottom line: The agenda is not complete yet. I'm still working on *multiple* surprises that I'm not able to speak about publicly yet, but I assure you, it only gets better from here (and "here" is pretty damn good, if I do say so myself). Glue is planting the flag: If you develop in or for the cloud, or you're going to be (hint: every developer -- in the enterprise and out -- is going to be), then Gluecon is the place for you. As I said to a vendor on the phone the other day, we're not debating the "economics of the cloud," we're not even debating whether or not the move will happen, and we're certainly not wasting time on fluff. At Gluecon, we're getting down to brass tacks -- and quick. From the apache cloud stack to SAML to Riak to Heroku to OAuth to provisioning to legal issues to SLAs to API management to API terms of service to major cloud platform providers (and so much more) -- the sheer breadth of topics we're covering makes Glue worth it, and when you add in the depth of the content -- well, come judge for yourself. If you're coming from outside of Colorado, you'll be able to attend Glue (and get airfare and get hotel) for less than the cost of just registering for other "cloud" events (even after they offer you 40% off of their "standard price"). And if you live in Colorado and work on software, then missing out on Gluecon would just be silly (yes, silly). The early bird price is $525 (that covers 2 full days of conference with meals provided, the evening reception with open bar, and the hackathon), and after next Friday, April 2nd, it's going away. Don't end up costing yourself money, and don't lose the opportunity to meet with, hack with, and interact with some of the brightest minds in software. Register now.

A Startup Guy and His Tale of Recovery

I'm a startup guy. I didn't know that until I was already neck-deep in a startup, but it's true. I often get asked why Glue's sponsors are so dominated by startups. Partially that's because guys like Techweb (producers of Cloud Connect) have "national account" reps for the big guys; folks that can pick up a phone and talk to the VP recently put in charge of the new cloud computing initiative at big software vendor X. That's not me. Where they can land the big 50, 100, 150 thousand dollar sponsorships, it takes me months just to get put on Oracle's budget planning for 2011 (ie, next year). And that's cool. I'm not complaining. Because the other side of why Glue's sponsors are dominated by startups is that I'm a startup guy, and Glue is a startup, and really -- I just love the energy of startups. All of that said, I was walking the dog (our golden) this morning, reflecting a bit on the 5 years since I left the last software startup I was involved in. And I thought I'd blog a bit about it. It's not exactly "on topic" for Gluecon, but I figured what the f&#ck -- I'm a startup guy. Let me start at the ending-- namely, what I have I done since leaving Ping Identity (the startup in this story; disclosure - I still have equity in Ping). In July 2005 (when I left), I began working as a contractor for IDG (the company that bought Digital ID World -- the identity management conference I helped start), wherein I worked on shows like Syndicate (an RSS/feed show that was ahead of it's time and went under) and LinuxWorld. I also started (inside of IDG) a show called SaaScon, which now runs under the Computerworld brand. After leaving my IDG contractor position, I started Defrag (now entering year 4 of existence) with my old biz partner Phil Becker and the guys from Foundry Group. That went so well that we started Glue. And Glue's gone so well that we're looking at a third show (Blur), and I've got one other fourth, super top-secret project in the hopper that I'm not discussing - yet. Sounds like a lot, doesn't it? Here's the thing: From July 2005 until January 2010, I really took it easy. I was very careful not to take on too much. When we started Defrag, it felt exhausting. When we added Glue, it felt exhausting. And then, what seemed like very suddenly, in December of 2009, I started waking up in the morning with this overwhelming sense that I wanted to (needed to?) do more. An energy level was coming back that I hadn't experienced since 2003-04. I've since doubled my workload, and I feel great. And it's only upon reflection that I now know why. I met Andre Durand in early 2002 (I think it was January 18th, or something like that). We'd been introduced by Doc Searls (of Linux Journal and Cluetrain Manifesto fame). Doc told me there was this guy who started Jabber that was doing some "security" thing for the internet, and I should meet him. I did. Andre went on to tell me about his vision for Digital Identity -- which sounded suspiciously like Single Sign On to me. I thanked him, yawned and went home. I woke up a couple of evenings later in the middle of the night in a cold sweat. I couldn't STOP thinking about identity and the internet. I saw it in everything. It had completely consumed my thoughts. So, I did what any sensible startup guy (didn't know it yet, remember?) does -- I called Andre up and asked if I could work on this with him -- for free. He, of course, said yes, and we were off. We worked *constantly*. Round the clock. We didn't even have any idea what we were doing, we were just obsessed. In a few weeks, there was actually the semblance of a company idea that Andre thought he could raise money around, and seeing that I was basically not sleeping (in lieu of working), he offered to pay me *something* from the small angel raise he had done. With one caveat -- the money ran out in October (it was march or april). That was the wall - so we set off to raise money. And we did. And then we did again (a Series A). And then we got offices, and employees, and policies, and an executive team, and (oh shit) CUSTOMERS, and revenue and -- it was amazing. But from 2002 until July of 2005, when I stepped away, Ping Identity was my complete and total obsession. I used to have a contest with the main guy that worked for me (Steve Nakata) about who could show up to the office earlier. When it was all said and done, Steve and I were regularly meeting each other at the door between 5:30-6:00am -- with a smile on our face. Ninety hour work weeks were common-place and "life balance" was for people who didn't really have the dedication we did. What I never realized in that intoxicating, high energy, hyper-aggressive (somewhere in there I bought a doberman and a sports car) atmosphere was exactly how exhausted I was. When we sold Digital ID World in 2005, Ping was raising its C round. Things had gotten pretty bureaucratic, and my "screaming in the hallways" style wasn't as....appropriate. (Yes, I really did scream in the hallways, and throw things in my office, and just generally cause havoc.) I left, and my then girlfriend and I moved to a beach in Florida. At first after leaving, I really thought something was physically wrong with me. I felt like I was coming down with the world's worst hangover. I couldn't sleep. I didn't eat. I really enjoyed laying on the beach and cocktails. After about six months, I realized it for what it was -- some kind of post-startup depression, which I think was inspired by pure exhaustion. I was still working at the time, but at nowhere near the pace I'd held for years (nor the intensity). Oh - and I found myself *constantly* frustrated by working with people inside of a large organization -- it was like these people were there for their *salary* or something (not the "vision" of what we were doing). From 2005 until 2009, I slowly added work components back into my life. But the whole time there was this sense that my tank was always running on fumes. After the first Defrag, I literally slept for 2-3 WEEKS. It took everything out of me. By comparison, now when I finish a conference, I sleep for a solid day and plan to take a vacation for the 7-10 days following that (which is much different than sleeping for 2 weeks). In any case, my startup exhaustion appears to have lasted about 3.5-4 years. I don't know if that's common. I don't know if I'm an odd bird. But it is what it is. I do know this --- suddenly, I'm coming out of that exhaustion, and it feels like someone lit a fire in my belly. I'm starting all of these new work things, and I can feel that old vibe: the more work I take on, the more energized I get. Pure, unadulterated, nitro-like passion. Am I repeating the same mistake? Eh -- first off, I never saw it as a mistake. Secondly, I'd like to think I've learned a lot about how to manage my own rhythms, needs, etc. And, frankly, I'm older - I just wear out faster. But still, I'm a startup guy. Nothing gets me up in the morning like thinking I get to work on things with startups; to build a new "industry"; to break some new ground. And, to be fair, I know a TON of people that perform those jobs inside of big companies like Cisco, Facebook, Microsoft, Oracle. In the end, I know this: Startup guys (and I mean "guys" to include "gals") feel this nearly unquenchable obsessive fire around their startups. It can lead to all kinds of errors (failing too slow), and personal harm, and general badness. It can also be the most satisfying thing you ever can do in your work-life. I have some life-balance now. I'm married. I'm trying to lose some weight. I walk my dogs around the island. I still go out for cocktails and drive a sports car. But I'm a startup guy. The other day, I was at a hospital picking up some X-rays for my wife, and someone asked me if I was an employee there. I almost took offense. Seriously? Me? Work here? Really? Uhhhh...that's not what I do. I'm a startup guy. Anyway -- that's my little morning reflection on being a startup guy and my tale of recovery. Cheers. ;-) P.S. I'd be remiss if I didn't remind everyone that early bird pricing for Glue expires on April 2nd -- REGISTER NOW.

Elbows, Expo Floors, and Value

It's no secret that tech conferences can be very expensive to attend -- some much more so than others. There are events with similar dates to Glue (different topic area) that have early bird prices set at more than 2.5x Gluecon's early bird, and "regular" admission prices set at more than 4x Gluecon's regular price. Similarly, there are some notable Cloud events that are roughly 3x the price of what it costs to attend Gluecon. Not that I'm exactly comparing apples to apples. After all, the "common practice" with some events is to price their registration at $1895 (or $2995), and then offer 40% discounts ("secretly") up to a week before the event. Similarly, a lot of events "gain size" by offering free "expo passes" -- a tactic which certainly brings more "floor traffic" (though the quality of that traffic is always the hotly debated topic). Lastly, to be fair, I should note that while Gluecon is 3 tracks wide over 2 days, many other cloud events are 3 or 4 days long with 4-6 tracks of content -- so the argument is certainly there to be made (I guess) that the prices are in-line. In fact, I don't begrudge other conference organizers charging whatever they want for whatever they offer (although I do think 3k for 2 days of startup launches is a bit ridiculous). In all honesty, "value" probably has to measured along multiple fronts when it comes to attending a conference: 1. Topic/Content: There's a rough mix of speakers, topics, agenda items that should (frankly) excite you when you look at it. 2. Networking events/Interaction: The primary reason anyone goes to a conference is to interact (face to face) with either potential business prospects/partners/providers or like-minded individuals. 3. Quality of Event: Production value does matter. Not in "staging" (although, that's cool), so much as in wifi (good wifi - yes, it can be done), power-strips, etc. 4. Intimacy: This is the one "value axis" that I think gets missed so often -- and maybe this is just my personal preference - but I'd much rather have a great interaction setting with 300-400 people, than be in a mob scene with 800, 900, 1000's (or multiples of thousands). Yea, there's an excitement in the crowd -- but it's also incredibly hard to bump into that one guy that you'd really like to have lunch with when it's all elbows and expo floors. All of which leads me to what I'm trying to do in bringing Gluecon to life: 1. Topically: My sense after last year (post-show surveys, etc) was that there is a real thirst for technical/developer-oriented content/agenda items. My personal motto with Glue this year has been "go deeper," and I think we're starting to get there. Some of the recent additions to both breakout sessions and the keynoters have me very excited. I think Glue will bring *depth* of content, as well as an amazing cross-section of topics that you simply will not find elsewhere. 2. Networking/Interaction: I just consider this a given. We build interaction into the DNA of the event. We don't want an "audience," we want participants. It almost sounds cheesy, but it's true. 3. Quality of Event: Anyone who has ever asked me logistical/operational questions about tech conferences knows how I am about this stuff. Fanatical. 4. Intimacy: Again, this is where I like to concentrate my energies. I don't wanna "throw a party" or justify a boondoggle, I want to help *build* something. Something substantive. The only way that I know how to do this is intimacy. The result: 2 days of 3 tracks of what I think will be incredible content; where the "audience" contributes more than the presenters; community gets built around things to do; and intimacy rules the day. We feed you. We give you drinks (even the adult variety). We take care of you for 2 days. Oh - and throw a Cloud Camp in front of those 2 days, and a Hackathon at the end of one of them. And then I can circle back to price: $525 early bird (expires April 2nd). $472.50 if you use the discount code "twit1" (expires April 2nd). That means that if you're coming in from out of state you can get airfare, hotel and registration for the cost of just registering for one of these other cloud events (even at their 40% off rate). Now, let's talk value... [sidenote: if you live in Colorado and are in software development and you're NOT planning to come to Gluecon, you are quite simply missing out on a huge opportunity in your own backyard. You do not need to go to the Valley to interact with some of the leading developers in and for the cloud. They're coming to you. The only question is if you will choose to join them.]

A Snapshot of the Agenda

I was just doing some work on the Gluecon agenda (the unpublished part), and noticed one "snapshot" (one afternoon track). Not to toot our own horn too much, but I was struck by the strength of the topics: The Apache Cloud Stack: an end to end overview (Hadoop, Hbase, Zookeeper, etc) Application Platform as a Service (e.g., Heroku) vs. enhanced cloud system infrastructure (e.g., Engine Yard) Why another databse? Drizzle explained. Inside MongoDB: the Internals of an Open-Source NoSQL Database Memcached: Fast Distributed State for your Site's Glue What struck me was the nice, chunky, technical nature of the talks. When I set out to do Glue's agenda this year, my motto was "go deeper" -- I knew coming out of last year's show that the real thirst was for technical, developer-oriented stuff. I'm not a developer, but I was determined to find out what matters, and dig in. And when I see a snapshot of the agenda that looks like this, I know that we're getting there. ;-) Gluecon: Developing in and for the Cloud. Join us.

Eric Brewer to open Gluecon

I've been working on getting this to happen (and a BIG thanks to Justin Sheehy at Basho for his help) for a while now, so I'm overjoyed to be able to announce that Professor Eric Brewer will be delivering the opening keynote at Gluecon. In typical Gluecon fashion, we're not settling for your standard "cloud computing" keynoter (i.e., a large vendor telling you about how wonderful their large vendor-ness stuff will be). Nope, we're going deep. In fact, as I dug into what the "glue space" was all about, I was always looking for that person(s) who really spearheaded the thought in the space. And then I found this post by Julian Browne. In it, Julian compares Professor Brewer's July 2000 talk on the "Principles of Distributed Computing" to the first show the Sex Pistols ever played (which was the defining moment for punk rock). A little further digging around Brewer's CAP Theorem (and its relationship to NoSQL), and his time co-founding Inktomi, and I was sold. This was my guy. The guy that saw the "cloud" vision a decade ago. The guy that inspired an entire wave of distributed computing. A guy that probably still hasn't received nearly the credit that's due to him. So, yes, I'm overjoyed to have him bring his unique perspective to Gluecon. Professor Brewer has asked me not to post a title for the session yet, but early indications are that it will be focused on "wide-area data management." Of course, combine that with Doug Crockford (creator of JSON) being on-stage right after him, and well....let's just say we're gonna blow things outta the water. I hope you'll make the choice to not miss this truly special event.

The Value is in the Glue

One the recurring themes I keep seeing is that the adoption of cloud computing is being driven by our current sucky economic circumstances. I don't disagree that the macro-economy is a *contributing* factor in cloud adoption, but I really don't think it's the primary driving factor. Rather, I think cloud adoption is part of a much larger, much longer IT cycle. Anyone remember Application Service Providers from the late 90's? ASPs were a big deal. They were gonna change the way business did business. And then the dotcom bubble burst. And ASPs quickly hit the dustbin of history. Well, kind of. emerged from those ashes, and somewhere between 2002 and 2004, we saw the first real ASP success story happen. Could it be argued that this was all dependent on the macro-economy? Yea, I suppose so. I mean - dotcom bust, recession, emerge from recession in the 99 to 2002 timeframe is the exact timeframe for the founding and early growth of But it seems to me that there were bigger forces at work (no pun intended). People, as in non-IT folks, began realizing that they could just purchase "seats" on Gartner has called it the "consumerization" of IT - a huge sweeping change, wherein mainstream technology adoption had "normal people" (non-IT pros) thinking they could just "download" something (or browse to something) to accomplish what they wanted. By 2005, IT departments everywhere had started to wake up to the beginning of this nightmare, where software as a service was being purchased "haphazardly" by lines of business level personnel. In other words, the purchasing and maintenance of enterprise software had been decentralized. Of course, that wave of decentralization is still sweeping across IT. But it was in that context that "cloud computing" rose to prominence within IT departments. Which is to say that it wasn't simply a "we can save money" or "screw CAPEX" decision. It was a reflection of a decentralization of the IT department that has been in the offing for over 10 years (in force). The "cloud" is the natural evolution of the decentralization of the complex system of enterprise IT. That evolution drives a commoditization of every point on the "value chain" -- i.e., it drives value further and further up the chain. The value in IT used to be in the hardware, the mainframe. Then it was the OS. Then the platform and database. Commoditization -- which is really just the cost of computing cycles dropping over time (while increasing in efficiency), Moore's law -- drives value further and further up the stack. As that drives us toward the application level, moving IT departments into "the cloud" was a completely rational occurrence. How far will it go? Farther than any of us think it will, I'm sure. I remember a hallway conversation I had with Jamie Lewis (then CEO of Burton Group; now Gartner) back in 2004. I raised the idea of "identity as a service." He said he didn't think enterprises would ever put their identity data on someone else's server; too risky. Of course, I countered with getting people to put *customer* data on someone else's server. Now, I don't know that we're fully to "identity as a service" yet; but the movement seems inevitable at this point (and it's not that Jamie was "wrong," so much as that neither of us could imagine the cloud world that was coming). Cloud Computing is not simply being driven by economic factors. Cloud Computing's rise is part of a much larger cycle -- one that's been happening for decades. In that context, it's probably instructive that we keep our eye focused on where the value will land --- not in servers, or storage, or processing cycles, or even in applications themselves. Just as with email, the internet, and cell phones -- the VALUE is in the connection. The value lives in the space between the servers, storage, apps, people, data and networks. The value is in the glue.

The Security Glue for the Cloud

If you look across the agenda of Gluecon, you'll actually begin to notice several horizontal themes that include (but aren't limited to): APIs, NoSQL data models, connective protocols, frameworks/providers, and identity. The themes share a common thing: they (wait for it...) "glue" the cloud together (at various points in the stack). Being an old identity guy (from way back when, you know, like 2000), I've always believed that identity is the driving factor for any real "security." And so it was with great pleasure that I read this piece over on ReadWriteCloud, wherein they refer to Matthew Gardiner's presentation, "Identity as Security Glue for the Cloud." Now, Matthew's an old identity guy too, and CA's pushing big into the Cloud space, but it's his choice of that one little word that intrigues me. You see, "Glue" might sound like some cutesy little marketing name for a conference, but in reality it's not. Glue is a descriptor of what the things we talk about at Gluecon *do*. Whether it's APIs, new data models, activity streams, OAuth,, web oriented architecture, OpenID, Facebook Connect or SAML (that's the "Security Assertion Markup Language" for all of you non-identity folks out there), they all perform an essential function -- they GLUE things together (people, identities, data, apps, etc). So, it's no coincidence that folks like Ping Identity are participating in the Glue agenda. And I hope that it's no coincidence that Matthew chose that one little word for the title of his presentation. (CA, are you listening?) ;-) Coming at the cloud through the prism of "Glue" provides a unique perspective on the problems, hurdles, challenges and benefits of the "post-cloud" world. As always, I hope you'll choose to join us.

The Dynamic Points in the Cloud

I continue to work on Gluecon's agenda (I'll work on getting a new version up this week), and as I do, I always keep a glance reserved for "what the other guys are doing" (what's going on at the RSA Security Conference, for example). And you know what's becoming obvious? The "heat" in this space lies at the point of intersection. In other words, it isn't "ooh I can store something in the cloud" that's important, it's when that thing you're doing in the cloud has to move, or touch something else, or interoperate, or send/receive API calls --- the point of intersection; the act of interaction -- that is where all of the meaty, fun, gnarly stuff lies. Take, for example, the Cloud Security Alliance's announcement about a "Trusted Cloud Initiative," that's all about intersecting and interacting. Or the work going on over in the Cloud Audit group, same thing. Or the work going on around activity streams, OAuth, webfinger, or the Twitter streaming API -- all about the point of interest being the point of intersection and interaction. Call it the dynamic points in the cloud. These aren't static clouds, or simple storage clouds, or web apps that interact with nothing in the cloud. In fact, it's not the "cloud" per se that's interesting. It's the things interacting in or via the cloud. It's the layer of web infrastructure that "glues" all of this stuff together. ;-) Yea, I'm talking my own book now, but c'mon! Don't settle for the static cloud, when you could dig in, build, learn and educate at Gluecon. I hope you'll join us.