- In all of our talk of conferences, we forgot to mention that MMS 2010 is next week! Hopefully there’ll be some fun Microsoft IT Management news.
- RabbitMQ bought by SpringSource/VMWare – the re-rise of queues. See James’ piece on the topic too.
- Hotel philly cheese steaks and fake Ruebens.
- Amazon has a lot of middleware.
- Coté really liked one of cloud presentations at #phillyete last week. Also, he gave a keynote on on developers getting cloud computing there.
- Some Think Dynamics and Tivoli memory-lane.
- People still building old-school, multi-tiered apps seems crazy to John. The “HMS Multi-tier”.
- CloudCampToronto – John does his first lightening talk.
- We talk about cloud and other conference ROI for vendors, like John at OpsCode.
- RightScale and the new type of tech CEO, accessible and brand-building.
- John sees Enomaly demos while in Toronto.
- John’s 2010 Cloudie Awards plans – see his 2009 ones.
- Gartner’s Cameron Haight goes dev/ops.
- iPad? Coté’s review. Rackspace got something on it.
- GWOS and Eucalyptus partner for cloud management – they’re looking for beta testers.
- Coté will be on a Zenoss sponsored virtual panel next week talking about managing the cloud.
- John has a lot of speaking stuff: OSCON, Dublin conference Epicenter
- We cap off the episode with airline status chit-chat and the sweet life of being Executive Platinum. Our man Tarus is a recent lifetime gold member on American, which Coté longs for – only 206,442 more miles to go!
Also, if you haven’t subscribed to it already, you should check out the ITSM Weekly The Podcast – it’s chock full of good old fashioned IT Management, which I personally adore.
(As always, I don’t fully check out and correct transcript for this show, so if we’re saying something nutty below, be sure to check the audio before you think we’ve gone off our meds.)
Michael Coté: Well, hello everybody; it’s the 15th of April, 2010. I think it’s like tax time or something who knows. Anyway this is the IT Management and Cloud Podcast, Episode Number 70. We’ve just five episodes away from three quarters as it where, which is pretty existing. As always, this is one of your co-hosts Michael Coté, available at peopleoverprocess.com. And I’m joined by the ever present also co-host.
John Willis: Hi! I am John Willis from antidistablishmentareaism.com or sometimes honorbikkerbilllatudiakabus.com.
Michael Coté: Well, those sound like domain names that would be available.
John Willis: They might be, but I guarantee your — the guy who has got to do the transcribing for this is going to be awful, it’s going to be driving him nuts.
Michael Coté: Have you been enjoying the transcriptions’ John?
John Willis: Yeah a couple of them. Some of them are like, you know, what was the one, the one I read was, when we did the one in Austin right. So I started off as Michael, John, Michael, John Person 1, Person —
Michael Coté: It was Speaker 3 and Speaker 2.
John Willis: Well, I somewhere about eight minutes in to it I devolved into Speaker 2 or 3.
Michael Coté: Yeah, we have actually a really good transcription service for you, they’re like Indian offshore I think nowadays. We do a lot of sponsored podcasts and stuff, so we do transcriptions for that. I do transcriptions for this show and other shows and unlike the ones where we are sponsored, we actually look over and correct everything right, with these ones like I just pretty much wrap some HTML around it and throw it up there. So, they get to be a little wacky after a while, but I think it’s — there’s been several people who have liked the transcriptions.
John Willis: Now it’s a good thing to do, it’s totally cool.
Michael Coté: Yeah, it’s pretty good stuff. So, I was talking before the podcast, before we get in to the IT Management and Cloud stuff how, I had an excellent Reuben sandwich this morning and it was made with turkey pastrami. As John has — it’s merely a shadow of an idea. It’s sort of like the Plato’s cave thing where there’s three levels.
So, you’re trapped in a cave and someone has strapped you into — they didn’t have chairs back in the ancient Greek times or whatever, and someone has strapped you to the wall and what you are seeing is someone is doing a shadow play where they’ve got actual forms and they’re casting fires to make shadows on the wall. Most people, John, this is how they’ve lived their life; they’re seeing shadows of the truth.
John Willis: That’s right.
Michael Coté: At some point, you can turn your head and you realize that shadows are being made and some other people live that way. But eventually you get loosened of your bonds and you go outside of the cave and there is this pond of, they didn’t use the word enlightenment that’s more of an Easter thing, but there is sort of this pond of true knowledge and you start to see the true forms and shapes, the logos. And you know this Ruben is kind of like shadows playing on the wall.
John Willis: Right, the true form has corn, beef, sauerkraut you know it’s stacked so high. So, but you know what, I was looking up on this, you know, we were talking about where did this come from. So this is the history part of the IT Management Podcast. It was actually the, the person who basically was the founder or who made the first Rueben is Reuben Kulakofsky. He was Lithuanian-born grocer from Omaha, Nebraska, wait a minute, that’s, well I weren’t expecting to see that next. Well anyway there you go.
Michael Coté: That’s right, so, you know what’s been going on John? I think it’s been about a week-and-a-half or something like that since we’ve recorded. I was at a conference last week, so I didn’t get a chance to record then. So what’s been up?
John Willis: So I have done a little traveling and so I am kind of looking at, where do you want start, do you want start with cloud, you want to start with big news?
Michael Coté: Let’s, what’s the big news?
John Willis: RabbitMQ.
Michael Coté: Oh! That’s right.
John Willis: Purchased by the SpringSource guys.
Michael Coté: Yeah, the VMware SpringSource people got. So what you know I use the, I’m the first to go with, with my take on something, what, what are you hearing in field and out there what’s going on?
John Willis: You know to I talked to Alexis Richardson, you know I used to do a bunch of podcast in my Cloud Café with him when, he was at CohesiveFT and so he had given me a call, you know kind of let me know, this was up, you know. It’s kind of funny, he gave me the, you know hey, email big news and I’m like, you know, big news for him. So, but not for me. But yeah you know I mean, he was, you know I mean it means that RabbitMQ and the use of kind of Erlang based and RabbitMQ is alive and well.
A company like VMware sees this as a possible glue; I think it’s extremely exciting. You know I think those technologies are awesome right, they are turning out. You know and I think when a company like VMware invests in buying that technology — that it’s — I mean I knew it was real deal. But it’s, it’s in the lot, a lot of people know it’s real deal, but I think that commercial world might not have known it’s a good deal.
Michael Coté: And it’s, they are an AMPQ system, right?
John Willis: That’s right, that’s right, yeah.
Michael Coté: Which is, which is basically a, I mean it’s a, I’m, sort of at lack for words because it’s kind of such an old concept that keeps getting kind of respond, but it’s basically a queuing system. So an event system and most recently we would call these Enterprise Service Buses or ESBs.
John Willis: That’s right.
Michael Coté: But basically it’s just you have, you have an architecture of your software where you want to send events from Point A to Point B and Point C and all these different points and there is different ways you might be doing it. But you want a varying degree of guaranteed delivery or un-guaranteeing, you got all these levers that’s like the robustness of it, the guarantee that you will get it and then versus the performance required and the programming models, all these different things. But, but there is kind of this new generation of AMPQ based ones. RabbitMQ being one of the more popular ones that are just basically a queue that you can have and like I was at the emerging, I mean that’s a pretty fair [?] summary of what you said, I mean —
John Willis: Yeah, yeah it’s excellent, yeah, it’s good. It’s fine.
Michael Coté: It’s interesting that you bring that up because you know the conference I was at was the Emerging Technologies for the Enterprise conference put on by Chariot Solutions. It’s basically a Java and kind of enterprise consultancy shop, which, which I didn’t know them extremely well before I got involved in doing these conference with them. But they have a, they have a pretty good podcast that goes over this interesting technical stuff and they seem like a good bunch of people and they are based in the Philadelphia area. So I went to Philadelphia and I only had a hotel cheese steak. I very much so failed in the area of cuisine, I am getting a real cheese steak, apparently you’re supposed to [?].
John Willis: A hotel, oh man.
Michael Coté: Yeah it was tough times.
John Willis: Turkey Reubens and hotel.
Michael Coté: But, this is, this is the thing about —
John Willis: Philly cheese steaks.
Michael Coté: I finally —
John Willis: This will not stand, this will not stand.
Michael Coté: Yeah, I finally got in all of my friends here in Austin understand, they used to all be envious that I travel a lot and they are like, so nowadays they joke with me, because I have joked with them but they are like, “Oh so you are going to Philadelphia, well enjoy the hotel.”
John Willis: Yeah, now they know it, that’s funny.
Michael Coté: That’s right, you know, like they realize it that when you travel a lot you basically sight-see a bunch of hotels that’s where it went bad.
John Willis: That’s right.
Michael Coté: Any ways getting back to topic, so there were actually quite a few interesting talks of people in the financial industry and other people who were using AMPQ based things, whether they are using Rabbit or other stuff. And you know it’s interesting that Q as in queuing especially in sort of a cloud architecture, distributed architecture kind of coming back because that’s, those are really the type of architectures when you have a grid or a cloud or a distributed or a sloppy or a whatever, where queuing kind of starts to become interesting.
I sat in on a Jeff Barr presentation. I don’t, I actually don’t think I have ever seen one of his talks and it was really, a really good, it did exactly what I wanted. I wanted to be like the lazy analyst and I figured I am going to see Jeff Barr Amazon, Evangelist and I basically going to get an overview of everything that Amazon web services has to offer and sure enough that’s exactly what I got.
So it’s like, that guy gives a great, if you want to know everything that Amazon does in web services, he is pretty good at it. And I didn’t realize that they, you know, we probably even talked about it, but I haven’t gotten into the detail that he even got into about that. They have a queuing service as well, we did talk about this.
John Willis: Oh yeah, yeah, that’s [?].
Michael Coté: But it’s kind of interesting when you take an hour and sit down like and you know Jeff Barr walks you through everything, like you realize at this point, they pretty much have every piece of middleware you would need to build an application. I mean there is various thing, they don’t have a UI later, they lack that, but they have like they have got a database, they have got everything, which is, and it got me wondering like, I wonder if, I wonder how many people if any like, like what, if they are like 80% of our staff are Amazon Web Services, like, who is treating AWS kind of like the way you would treat, you know WebSphere like your complete sort of application stack or something like that.
John Willis: Yeah I know, I mean you know this is, I just, I was actually at CloudCamp in Toronto last week and this debate came up and I think somebody was arguing that infrastructure is a service and this and that and that you really need a pass, and in some ways you could argue that infrastructure is a service.
The Amazon Web Services is becoming past like but the truth of the matter is, I’m going to sneeze excuse me, so the truth of the matter is that Amazon as it stands right now is really a full service infrastructure as a service, offering and because the argument was well we still — you really couldn’t still build an app without a [?]. The truth of the matter is that with a little bit of help or a little bit of understanding you can walk in from ground zero go to Amazon you can get your the elastic load balancing saying which gives your basic — your stack for doing basically a LAMP stack or you know or some type of stack.
You got the relational database service right there, virtual data storage or whatever it is which is basically MySQL as of service, at SQS for any queuing technologies. I mean you literally and in fact you know Segway, they just announced a couple of new features within the last week or so. They’ve got sticky sessions for ELB, which was, you know, the ELB, elastic load balancing.
So, now you got sticky sessions. And then they also added simple notification services, which is more like TIMCO like offering, you know it’s kind of a message broadcasting.
Michael Coté: Alright.
John Willis: So I mean, it shouldn’t rise, you are actually right, I mean you could, armed with like a couple of cheat sheets you could build an application infrastructure, prior to it, you have to developer on your right side, you could tell the developer, say here I have got some codes and you could arguably build a pretty non-complex application infrastructure and I think you could probably do it a day if you had the right cheat sheets.
Michael Coté: Yeah.
John Willis: Maybe at worst case a couple of days, but think about three years ago what it would have taken to build an elastic load balanced stack that has queuing built in, and master slit, well I guess they don’t have the master slit database stuff, but as the relational database back-end set up. I mean, that’s months. Maybe it’s a high paid and it’s actually usually not one sys admin., it’s a kind of a sys admin for this, a sys admin for that.
Sometimes I guess in the web world, it is one guy, but I mean they have reduced the ability to get – people talk about how the cloud has reduced the ability to kind of get resources really fast, but and when you start talking about what they have done with all the different resources, I mean you basically, I mean that’s all right, let’s all make a plug with something like a chef or puppet. I mean you abstract those things and literally you can build infrastructure in minutes.
Michael Coté: Yeah it makes me think, we should gather up some sponsors to have some sort of contest, that’s like and we come up with some sort of enterprise grade application and we say like we are going to give you like 24 or 48 hours and the person who writes the most code on their own, the least amount of code on their own, right and runs everything in the cloud is sort of the winner, right.
And that the point is like now you got all this — we usually talk about the infrastructure operations angle of cloud computing, which is all fantastic, no problem with that. But it’d be interesting to kind of see for an application developer like taking this aspect of all the middleware that’s out there, like how far you could actually go with implementing something right.
John Willis: Yeah, now, well you know they used have this conference call, The Enterprise Summit, it was over 10 years ago and it was really kind of just, it kind of died in the Internet probably I don’t know why, maybe that something to do with people and the word enterprise was not sexy back then, but that’s exactly what they do, it would be enterprise conference where they would talk about the latest, greatest stuff going on in distributed computing mostly around the enterprise, a lot of sessions on DNTF and systems management.
Then one of the days they would have this kind of all-day bake off and in the morning they would give, if any vendor could come and play they would kind of unseal the task and it would be okay to create a system that does this, this and this, it has a help desk, it has this and this and –
Michael Coté: Oh! Yeah.
John Willis: And at the end of the day they would have a panel of judges that would go and they’d pick a winner and IBM had them won a couple of years a row and then back it would be HP, IBM, CN, they would be and it would be awesome because I mean if they did like, you can, they’d rope off their areas, but you can watch, and then at the end they do all like, show what they have done and it’d be real stuff. I mean –
Michael Coté: Yeah it’d be sort of like the Iron Chef of cloud development, right. I don’t know, I mean so, if there is, I mean if there is anyone out there who wants to help kick in like probably just be like a few thousand dollars, most of it for prize money right, and then we could just come up with I think we would need like one type of application or something like that, right and we would just kind of specify that and it would just of normal traditional boring sort of application. And the point would be like the people who figure out, I mean you obviously it has to be a working running version of whatever this application is, but the people who figure out how to use the most cloud stuff.
John Willis: Yeah.
Michael Coté: Because the whole point is to figure out how much, never mind the operation stuff, but to figure out how much you can rely on the cloud for your application development, not just deploying stuff, but that would be –
John Willis: Yeah that would be very cool, you know if I could see, you know actually if we did a little leg work on it, we could probably get some big sponsorship money and have a really cool contest.
Michael Coté: Sure.
John Willis: But, yeah, yeah, so that’s cool.
Michael Coté: I got some clients in my couch I can put towards it.
John Willis: There you go. I have got, I have some Bulgarian pennies and dollars that we can have.
Michael Coté: I got some Italian bucks left over too.
John Willis: Oh! Man there you go so.
Michael Coté: So you know also since we are on the topic of the Philly ETE the Emerging Tech and the Enterprise I am going to try to make it to that conference again next year, because it was really a good conference I liked it a lot and it was, it was very, it was a user conference, alright.
So it’s all practitioners and people who do things and most conferences have a bunch of users at them, but it also wasn’t centered around one particular community or brand or vendor. It was a whole lot of, it was a whole lot of clojure talk or clojure which you know it’s just a, any ways it not really a development podcast, but it’s just a, an exciting way of doing functional programming which the kids love and at least the Java kids love it and probably some other ones.
But there was also a lot of interest in cloud talk and there was one talk that I want to call in particular that I really liked about these guys, I am probably going to say their names wrong, but this guys Chris Cera and David Brussin and they work at Vuzit and Monetate, I don’t know how to say this, but I’ll put a link to it and I kind of walked in about 10 minutes into their talk.
It was a very bare bones, it is kind of like, what do they call, enterprise cloud computing pitfalls, puzzles and great rewards. And it was one of these great things, like hey we have been using this whacky cloud stuff for a year now and here is like all the weird stuff that happened and like things that were interesting and it was a very sort of maximum aphoristic kind of thing. Well each slide things like, expect failure all the time or we never reboot systems anymore and things like that and hopefully there will be a recording of that. I don’t know if there was recording of it, but I am going try to get them on this podcast.
So just kind of reprise presentation and talk about it more, because it was a, you can kind of check out the slides. But it was really good and a lot of what they called about centered around, you know a lot of it was the pitfalls right, like, it was basically the situation within a new technology like if something is going to go wrong, if something can go wrong, it’s going to go wrong not only the way you think it will but in weird unexpected ways.
Like their big example of that was like you got all these APIs that used be cloud stuff and that stuff is just not going to work a lot. Like sometimes it’s just APIs aren’t going to work and then you are kind of like, what do I do now, right and so you know there is things like that to worry about. And the other thing which you know will be kind of more near and dear to yourself John and your employer and other people listening to this podcast is that actually they did say a lot of interesting stuff about how it changes the dynamic or release management and configuration management they do, right.
So they don’t do a lot of, like I said they don’t really reboot systems to fix them. They just kind of re-provision them or something like that. I didn’t quite understand the distinction then, I was one of the part, maybe that was before I was listening really closely and then also there is more of continuos deployment that they’re kind of doing and there is more. They don’t really do database migrations so much. I mean there was, it was interesting to think about the way deploying to the cloud and the different automation, and of course they are sort of like you have to automate everything or you are going to die was a big point they were making.
And the consequence of that was that a lot of the, the sort of upgrading and migration and kind of flexing with things didn’t really exist for them, because once you kind of, and I am kind of reading between the lines when we talk to them, I ought to get some more, but it seemed like, once you separate the data out from your application a lot more and if the application goes wrong, you basically just blow it away and rebuild it at some point.
So there’s not a lot of troubleshooting that you can do in which and I have made a call back to this OPsCamp Austin a little. There was this kind of little just side conversation that Mark Cathcart who is at Dell was having with some people and he was kind of trying to get to this point where a lot of people who are doing cloud computing, they don’t really care about detailed root cause analysis anymore, right. They are more interested in raising the scope of the node where something is going wrong and just destroying that node and getting a new node to be really abstract in the talk, but anyways.
John Willis: Yeah, well I mean I think the kind of Holy Grail just leads back to what you are talking earlier about queuing systems. I think that people, when you get into the cloud you have an opportunity to kind of completely refactor and do it completely different, because the your traditional tiered stack or even more specific the idea of a server being of importance in an architecture, you can destroy that concept.
I need the idea of like what appears to all, I kind of laugh at bMotion. bMotion in some ways to me is like, it’s this old cloudy way, this way to keep a hold on the old school because bMotion is something that should be useless, right?
Michael Coté: Yeah.
John Willis: I mean, you know, it’s designed a kind of a message plus architecture where, you know, work is queued, you know, helpers, you know, start more helpers, or pull things off the Q to do the work, servers die, they just, you know, the Q doesn’t get the, you know, for whatever, the request doesn’t get completely processed. It’s still about [?] in the Q another one process, you know what I mean?
Michael Cote: Yeah. I mean what you are getting at it is you have to think most of your application has to be stateless, which is —
John Willis: Right and probably decouple from anything, yeah and so that’s the huge opportunity that when you go into the cloud and you start thinking about, okay I have an opportunity to write an application from scratch.
I have this opportunity now to building in a way that, then you start agreeing with I mean cloud in one sense is also, you know, like I mean the whole Amazon Web Services or what Mark Cathcart says is absolutely true, like when you are starting in the cloud I don’t care anything below the OS.
The truth of the matter is if you are doing it right, you don’t even care about the OS anymore right, because that’s just a, just an operating system that will just either be, you know, provisioned you know, the kind of throw it out the window test or the, or it’s going to be just, you know, something in a large architectural Q manage system that’s just completely [?].
Michael Cote: Yeah, yeah that’s interesting.
John Willis: So, yeah, so I think all of that, the problem is we have, you know, it’s funny I was at, I was at IBM. So I was going to give a shout at, actually I might get them in trouble, but so maybe I should tie these two together. I was going to give a shout out a guy Andrew Trossman I met him at IBM. I met him at IBM Pulse last year and then I met him again this year and then I met him at CloudCamp, Toronto and the guy is brilliant, he is the guy; he was one of the original founders of Think Dynamics. And so and he has worked for IBM ever since- –
Michael Cote: That was the — like SaaS hosted stuff right?
John Willis: No, what it was, it was actually the, when IBM, Think Dynamics did provisioning and orchestration. And so they, they IBM bought that technology and kind of tried to replace the existing Tivoli configuration management stuff with it. They really have, they haven’t succeeded completely in some areas.
So I have had a lot of talks since we have, you know, a common history as I know both worlds very well Think Dynamics and the Tivoli world and he is in this kind of think tank cloud team he is brilliant guy. So I was wondering, I was like, you know, why does IBM do this, or how come they don’t do this, or why do they do that, you know.
Since, I go around and I preach the technologies in the way things should be done, so this is the part right where I might get them into trouble, but I don’t think so. So I’m in a session at Pulse later, and they’re showing this new configuration management product that they made in IBM primarily for desktop configuration management, not virtualization old school configuration management, but built on a new architecture and fully scalable up to the, you know, the million of devices.
So I asked, “You guys going through the architecture and it’s a traditional tiered stack,” and I’m thinking to myself why would anybody starting from scratch today, who has any knowledge of what’s going on built a tiered infrastructure? I mean if you think about all the problems we’ve had with monitoring software with the tiered structure, right, it always is that the two structures, you know, connect some server, then you have kind of mid tier servers, then you have some level of clients or agents, right. And then, and the problem that you always have with those structures is, I mean they work okay pre-cloud because you didn’t have an unlimited or API literally to just add resources.
So you are always going through this growth pain of adding more, you know, horizontal middle tier servers. And at some point then it got two big even for the clusters, so you had to build a second, you know, server, you know hub spoke infrastructure. And, then we look at what’s going on cloud, it’s like you’ll be out of mind knowing that you could, that horizontal tier could be completely a [?].
So why I even bother writing an architecture like that where we know that resources or certain resources are so available whether they’re private clouds, or virtualization, or a cloud, you know, knowing what we know about all the mistakes of a tiered structure. Sure enough I asked the guy that in the session, I was like “You are IBM, I mean you guys are the one that kind of coined SOA, I mean well, you know, is it SOA the core of a non tiered structure?” And he just “Well, yeah you are right, you know, so [?] about it. He is like, “Man I wish I was a fly on a wall, [?] because I tell you guys over and over.
It’s like — I mean it’s just — I think that anybody who is stacking in the tier structure got to go, they are like that legacy [?] I was talking about, you know, you better think about refactoring your architecture because, you know, that old school architecture, and here is the point along with it, back to the point and that is even, you know, great developers which IBM has are still have this ingrained idea of how to do development and they wind up making that same mistake over and over because that’s what they do and do.
It’s not just IBM, it’s other people who are developing new products too that, they just need to, you know, kind of break the mold, step out of the box whatever you want to call it. And say okay there is, there is the potential of a new way to make so much more sense in a world that’s elastic, you know.
Michael Cote: You want to get off the HMS multitier.
John Willis: There you go.
Michael Cote: Yeah, no good.
John Willis: I love it the HMS multitier I got [?]. I love the old people’s quotes and tweeting this is my, this is what I do these days.
Michael Cote: There you go, there you go.
John Willis: So the CloudCamp was fun.
Michael Cote: Yeah, yeah, that was in, where was it you said it was?
John Willis: Toronto, Toronto.
Michael Cote: Toronto how is that?
John Willis: Great, that was cool, so I had some fun so —
Michael Cote: That that’s like the only part of North America where they know how to pronounce my name right in Quebec or Quebec or Quebec or whatever they say up there in the land of poutine.
John Willis: Right, but actually Toronto, Montreal is a little better I think.
Michael Cote: Well yeah, but I, you known embarrassing point here I always get my Canadian geography run wrong, I pretty much all I know is Vancouver is on the west side.
John Willis: [?] I think that’s where, yeah there you go; you know that because of the Olympics or no?
Michael Cote: It’s just because I know plenty, because I’ve made the mistake many times, so I’ve learnt the hard way.
John Willis: There you go, well I think there is a, I don’t know for sure, but I know there was, a while back there is the whole Toronto, Quebec Montreal thing going where Toronto was more kind of, you know, British English type and they were real upset about being forced to actually have French and English on all the signs and all that stuff. So, [?] but anyway what I was going to say is, so I did my first lightening talk for OpsCamp, so I did that.
Michael Cote: Oh! Yeah how did that go?
John Willis: It went really well and I was going to say is that, you know, the Canadians are I think the nicest people on a planet and if you are going do your first presentation somewhere, you want to do it in Canada because they just, they just really nice people. So even if you do screw up, you know, they kind of, I don’t think I screwed up, but you know, they just people are just so nice there, but it went really well I mean it was my operations as code kind of being theme that, you know, trying to —
Michael Cote: So, you now, both of us are pretty biased about CloudCamp having, I mean you’ve done, I’ve only helped to organize one and you have helped to organize many and whatever, but now that you are kind of more so on the winder side of things if you will, right. So what, I mean do you, do you feel like CloudCamp is worth the investment of a lightening talk or just showing up as far as the business side of stuff or like what, how does it play into like the business needs that you have, beyond your own, you know, your own like personal like of CloudCamp stuff?
John Willis: You know you are troublemaker Cote, is it Cote or Cote? So all right, you know, so like I’m always honest, so yeah, it’s a tough call. I mean there’s some companies just do it over and over and at one point the course of getting your face in front of customers, I mean a lightening talk can get $1,000. I mean, you know, prepared to pay in 5 grand and getting a booth at some large conference if you look at it that way —
Michael Cote: I saw a, to be a platinum sponsor of some virtual conference I saw the other day a virtual conference.
John Willis: Virtual conference.
Michael Cote: It was like $30000 and I mean it’s just like, I mean they had some like guaranteed amount of leads and things like that, so always just kind of crazy.
John Willis: I don’t know, I think that the old school, I talked to a couple of people, you know, so when I took this job I went on, try to talk to, many people would have, would let me ask them questions about and [?] going to be responsible for spending budgets on these kind of, what is your, you know, what is, and you know, and nice thing about most people it’s just awesome, even Tara Spalding of GroundWork [?] you should have spit in my face, she gave me great advice and, you know, this Mark Hinkle gives tons of great advices just a lot of cool people.
I think there is this kind of ground root, you know, this get back to the roots of, you know, not going to the big conference spending tons of money where you could get into like the bar camps and I think that, so on the positive side of, you know, my response to your question is that I think CloudCamp is a really good way to get in front of people and, you know, the thing is you get on the expert panel, you’re going to be in the open session.
So you really do get a lot of mileage out of your, out of your topic. So get your topic over there, people get to hear it. If there’s not enough people that raise their hand for experts, you know, I’ll raise my hand and say, okay I’ll jump up there. And then you’ve got open sessions and then you like what I’ll do is I’ll post an open session on a generic topic like dev/ops or, you know, operations, infrastructure, cloud and configuration management. Then you get a smaller group and they are the people that are real interested.
So for, you know, if you are like the Dine, I was going to a shout-out for the Dine guys, the Dine Inc. you know, they’ve got the Dine DNS stuff. They’re always doing these Dine [?], they’re at every CloudCamp, they’re great guys right.
So they’re just saying, they do every CloudCamp and, you know, and they will get on to the panel, they will talked about, you know, they’ve got, they do hosted DNS, they sell the story of kind of Global DNS Load Balancing it’s a great story. And then, you know, then they’ve got breakout session and the people who are interested in that, so I think that, you know, for $1,000 you can’t beat that.
Now on the downside of the CloudCamp is, the problem you have is, you know, you can go to a CloudCamp spend a $1,000 and if you look at like the registration [?] like which one you should go to but you might get like, you know, it might be a 120 people there and it might be a 100 people that are brand new, newbie cloud guys and all they’re coming there for is to learn what is a cloud. And, you know, and so that’s, you know, think that’s where it gets little dangerous as a vendor sponsoring this, I have had that happen at least once.
Michael Coté: Yeah, so what would you think if they like, if they like videotape your lightening talk and put that up on the website?
John Willis: Oh, I’d I love it, yeah I think, you know, I think I’m actually doing it again in New York. I’m going to a Cloud Expo next week. That’s a 2 week road trip, it’s kind of cool. I’m at Cloud Expo next week, next week the New York there’s a CloudCamp one of those days and then RightScale is having user group thing, so I’m going to go to that, so that’s going to be lot of fun.
Michael Coté: Yeah that does sound enjoyable.
John Willis: And then I’m going to hook up with Mark Hinkle and we’re going to go up to this Northwest Linux user conference thing.
Michael Coté: Oh, yeah, yeah now that they — the Mark Hinkle and Matt Ray it’s enough, so I always tell —
John Willis: Yes, I’ll be out with those guys and then we’re doing our first public training class that following week, so I’m just going to get on road Sunday and then –
Michael Coté: Everyone gets a spatula.
John Willis: Yeah, yes right. But yes so yes so I’m going to do, I definitely want to, I’m going to probably videotape mine this time myself if somebody up there could be useful. It’s a lot of fun, you know.
Michael Coté: Yeah, you know, I mean I asked because I get asked this a lot right, like people whether their clients or just people I know they’re always interested in, you know, it’s a problem getting, it’s a problem getting ROI on conferences nowadays Return On Investment, so it’s a hot topic among people on the winder side.
I think your impression kind of sums up what I hear and kind of my own beliefs and understanding is that it is, you know, it’s not the kind of thing where you’re going to end up with like 500 like well ordered leads in a excel spreadsheet or something like that right, which can be extremely valuable and extremely unvaluable depending on who they are and how they use it and everything right. And so but it is I do sort of feel like these especially with CloudCamp like it’s kind of I mean it is like just showing up it’s just being part of the community, if you will.
If you want to sponsor it even whether it’s at the $500 or a $1,000 level or whatever like that’s a way to be even more part of the community and really, you know, what I really think it is john is basically it’s good brand advertising right like it may not be extremely efficient transactional advertising for a vendor right like to your point there’s a 100, 120 people there right.
So, it may not be a big bang for your buck as far as like exactly tracking the process of like giving – finding customers, giving them knowledge, telling them why they’re confusing, why they need to give you money right which is like, the cloud, you know, it’s a cynical version of tech marketing is like create a mess, tell people they have that mess and then get them to pay you to clean up the mess is, you know, a very cynical way of looking at things but pretty crystal clear.
Well any ways but well a lot of what you’re doing is just sort of brand advertising right, so and, you know, you can see that, you know, RightScale, for example, since you bring, they have like a very strong brand I think in this area. And a lot of it, you know, they were like, I guess they’ve been around two or three years now but when they were first starting out, like no one really knew who these crazy RightScale people was and then Michael Crandell right their CEO like I remember that guy was, and he is still is to some degree, like that guy was everywhere.
He was just like built up, they built up really good brand awareness and then in addition to the actual hard work of writing code that does stuff and winning customers like he threw it together those two things and I think you, he build up a pretty solid thing.
John Willis: No, you know, I already talk about that too and then the thing about Mike Crandell is that he was so accessible too I mean, you know, I was, I talking lot to Jesse our CEO is, you know, very popular, you know, very smart. And I’m trying to get to say that, you know, there’s this kind of new breed of CEO and they just, they make you feel like you’re the most important person when you’re talking to them. You know, and, you know, and that’s part of, I mean, you know, part of it is like because they do it, you know, but part of it is they are actually kind of nice guys, you know, too.
Michael Coté: And it’s kind of depressing that that’s not the default way for Techworld executives to be.
John Willis: No, no [?].
Michael Coté: You’re exactly right.
John Willis: The idea of CEO being accessible, you know, even to the point where, you know, it’s like you’re just kicking, you know, kicking a soccer ball or bunch of things but, you know, in fact there was another guy I wanted to shout-out which is kind of something interesting is the SOASTA guys right, the CloudTest guys.
I’m going to buy the same kind of, you know, if somebody should like write a book or a article about the new kind of CEO because these guys have this kind of DNA that’s not like the old school. They are just friendly as all get out, they’re accessible. They make you feel good when you’re talking to them. It doesn’t feel dis-genuine, you know, you feel like it’s genuine and, you know, they’re interested and hear what you have to say.
They’re kind of power guys, you know, and it’s an interesting, I think the companies that don’t get that, you know, that just goes a long way I mean RightScale has done a lot of things right, excuse that kind of fun, but I mean a part of that fabric is Mike Crandell and, you know, and how that played out and I see the same thing happening [?] to know this Tom from SOASTA but I was going to mention that SOASTA actually announced, you know, so they do the cloud testing and they got all these great stories where they do, you know, they just blast tests and they can emulate 300,000 users in a 22 hour period for rollouts and because they’re using the cloud as the backend, he was at Amazon primarily. I wanted to say I didn’t know I was talking to those guys recently is they have always had a really deep kind of whole lab analytical back end, analytics back end.
So they have — and it was kind of interesting because they didn’t really pronounce that so much, you know, the early [?] their product was cloud testing. But right from the get-go they based in the idea that they were going to snap in with all the different tools. And so they just recently announced this kind of cloud test analytics where they and they’ve kind of had it all long but now they’ve kind of brought it now from kind of tip of the iceberg type thing. They were showing really what their deep back end is all about.
So not only doing the test, they actually like tap into like IBM Tivoli or Nimsoft or different tools and they will actually go ahead and throw that all into OLAP and they are actually be BI type, performance management BI guys that, you know, at their core I mean so that gets really, really interesting.
Michael Coté: So what are they like doing with the analytics?
John Willis: Well they can, you know, so they take all those performance data and they data warehouse it. You know, so you run this test, so it isn’t just, “oh yeah it broke.” “Good, you go fix it,” you know, what I mean. It’s like “it broke and it looks like, you know, based on this, you know, this, you know, these cubes or the OLAP reports that it looks like, you know, you started breaking here, you know, this buffer got backed up,” you know, what I mean? So, you know, just basically kind of doing what performance monitoring tools should have been, you know, doing all long which is kind of aggregating and looking for, you know, patterns of, you know, that’s why I always —
Michael Coté: I mean do they just do that for a customer or do they start to aggregate stuff over other people?
John Willis: No, they don’t, as I said the [?] I know you’re always thinking about that so yeah. No it’s a great idea it’s funny whenever I hear you, I just [?] holy grail but —
Michael Coté: Yeah, yeah you know it.
John Willis: One repository in the sky that basically allows us to basically look cross history, cross company and might say what’s good, what’s –
Michael Coté: Yeah, yeah it’s like dummy eye right?
John Willis: Yeah, yeah, yeah.
Michael Coté: It’s, you know, sort of like self aware intelligent thing but it’s just bring for us like pattern recommendation.
John Willis: Yeah I don’t think they do that but —
Michael Coté: Yeah, that would be a really difficult problem to solve when it comes to QAing stuff.
John Willis: Typically for theirs because as far as I can tell they’re not encroaching on kind of the Zenoss marketing or anything like that. In fact, you know, what they’re saying is, you know, a test by itself is great but a test where you can then analyze in gory detail what happened with tons of metrics and then also, you know, using state of the art data warehousing techniques to do that analysis becomes a much more powerful test right.
Michael Coté: Alright.
John Willis: In fact, I think part of them say that; I mean they’re trying to going after the whole Mercury Interactive which was the HP story right from a cloud angle, you know. So I mean that’s ultimately what the whole Mercury was all about, was kind of, you know, running test but then storing back data, you know, so anyway, so that’s cool.
So I’m doing all that and then one of the thing I also want to say about the CloudCamp I was actually, got to meet, I’ve met Reuven Cohen a bunch of times but actually I went over to his office and hung out with those a guys for a little bit.
Michael Coté: Oh yeah because he is on up there in the Canada, the —
John Willis: [?], so they invited me over to their office and I got a quick demo of their Enomaly and –
Michael Coté: Yeah, I saw they had some announcement about some super secured something using Intel Hardware, cloud and infrastructure.
John Willis: Well that’s a thing, you know, it’s about, you know, people, you know, a lot of people think, you know, what is, I always see people talk about, you know, the Enomaly or I’ve never seen it work those sort of, they showed me bunch of places where it’s running and actually doing really well within, they’ve kind of collaboration with Intel and in Asia they’ve got some really interesting customers. So it’s, you know, for those people that are and myself was, I weren’t sure myself but I mean it’s real stuff and it works. You know, it’s a true private cloud they, sell what they call a White Label, Private White Label Cloud and so they sell it to other people that want to have a cloud, kind of like what and then had we talked about 3Tera and CA yet or we talked that out?
Michael Coté: Yeah, yeah. We did a little a bit, yeah.
John Willis: Okay that’s right so.
Michael Coté: You’ve got the, you’ve got the 3Tera Cassatt, Nimsoft and Oblicore.
John Willis: Oh yeah, yeah, announcement.
Michael Coté: Which is all going together but, you know, if there’s –
John Willis: Pretty interesting stuff that’s how, we did that remark yeah so that, so yeah so Enomaly ECP. I mean I think I mean they’ve got some stuff going there and one other thing I was going to mention to is, so you have by stupid cloudiest thing that I’ve done last two years.
Michael Coté: Oh yeah. This year chef takes top price.
John Willis: No, no, no, no, no. okay, no Michael. No, so no. In fact I’m glad that was a great lead into this. So Dave Nelson the Cloud king right? And so he approached me and said, “hey, you know, I love this cloudy thing why don’t we make it something real and big?” You know, and I’m like, yeah what the heck I mean it’s usually just a pain in the ass thing [?] I do, you know, in the later days of December because I figure I did when last year, you know.
So what we’re going to do first cut is, we’re going to announce it next week at the CloudCamp and we’re going to have it run through CloudCamp and we’re going to make it, you know, this first year we’re going to do make a quasi official, we’re going to select the team of people to be on a panel in fact that you are one of the guys I was going to invite if you wanted to be on this kind of committee, we’ll pick a committee and we’ll get, I think there’s a couple of other guys, well known cloud guys who’ll have a lot of committee of Judith Hurwitz is — so you Judith, I think, I don’t, can everybody talk to [?] possibly and his, and maybe Mark or somebody like that and we’ll get all on a team or a committee and then we’re going to go ahead and officially state the categories. And as a committee we’ll pick the like two or three in each category and then we’ll put up on a website so people can vote.
Michael Coté: Oh that sounds wonderful.
John Willis: Yeah, so, so it’s still, we still want to maintain the kind of whacking us like have all these weighted categories and have [?] and then maybe next year even the committee will get voted in or something I don’t know.
Michael Coté: Well, use some of that couch money to make one of the [?] crystal award things.
John Willis: They, well it could be better than that because they, you know —
Michael Coté: You can send in one of those giant checks.
John Willis: Well, Dave is the sponsoring machine right so his idea to have a big old banquet sometime next year at one or two large events, you know, maybe the cloud.
Michael Coté: That’s a good idea.
John Willis: You know, actually give people their awards and just do it real funny.
Michael Coté: Well, I’m looking forward to the 2010 Cloudy Awards.
John Willis: Yeah, so I mean I mean you’re, you know, so you might have, I don’t know you might have to stay what did they say for the category of best analyst you might have to–
Michael Coté: I’ll have to recuse myself.
John Willis: That’s right you recuse yourself for that category.
Michael Coté: I think that’s what the people in black robes are supposed to do.
John Willis: That’s right so. Because I think, you know, you’re always in the top running for a best analyst in my book, so —
Michael Coté: Yeah, well, you know, so what are you going to do?
John Willis: What are you going to do?
Michael Coté: So I’m trying to figure out a transition to, you know, when I noticed speaking of analysts, did you see that there was this post by Gartner’s Cameron Haight where he’s kind of like getting towards, you know, leaning towards the whole dev/ops sort of thing. And I’ll put a link to it in the show notes as always. But it’s from I guess it was from last week the 7th of April according to the date I’m looking at. And, you know, it was a nice summing up of like there’s this crazy stuff going on here.
He linked it to Damon and [?] over at DTO, the dev2ops blog and [?] and even to the IT Skeptic and also to the Agile executive where I do a very small amount I basically help one of my old bosses who is real [?] mostly runs that and we do a little bit of dev/ops stuff over there but it’s exciting, more people getting in to this business.
John Willis: Yeah, so I mean the thing is this that, you know, I know Gartner is involved in this. So it’s I mean it’s interesting I mean it brings awareness that’s positive and I should look at it in the — you know, part of my, put on my — now I work for vendor hat and it’s like great, you know, take off my now I’ve, you know, when I didn’t work for a vendor hat, and, you know, it’s like I’m already seeing companies starting listing themselves it’s “Hey, you know, we got a dev/ops too, you know, here is our dev/ops,” you know, and I’m, you know, it’s like cloud, you know, what I mean, you know, so that preludes the, you know, I mean there’s all these arguments about people are already arguing about dev/ops, you know, and it’s like, “What’s the argument? It’s a concept,” it’s you know, it’s, you know, it’s a high level kind of theme or a spectrum that, you know, of, you know, how we should attack a problem and what type of, you know, ways we should we go about this.
And it already gets — when it’s got to this point where people already going to try to, you know, I can’t, you know, like people said what a hypocrite because I, you know, I mean I do try to commercialize Opscode as part of a fabric for things that, we have dev/ops but you won’t see me say, “we’re the dev/ops product,” right. I mean there is, there are things that I think are fabric of this concept of dev/ops, you know, and to be honest with you actually I’ve been getting kind of beat down on this argument and the dev/ops people are going to hate this.
So, when everybody likes Harleys, I like to go buy a Kawasaki I’m just wondering one of those kind of guy, right. So now that everybody likes dev/ops, I’m starting to think that that it’s less about dev/ops, and it’s more about how do you make if I, there’s a great article this morning in InfoQ I was just half way reading before — and I got called off on something else but it’s how do I — the truth of the matter is in my mind, I’ve been in operations my whole life right.
So I have a perspective to be done. And dev/ops in some ways is an apology of operations but we have a lot to be, you know, we’ve got a lot of things we need to apologize about right we suck, you know, what I mean, sorry folks, you know, the development world in the last 15 years has gotten very mature. You’ve got agile methodology, you have, you use sophisticated tools for development, you use collaborative techniques right. It’s never single minded focus. And in operations we’re in most cases still, you know, Bob scripts wherever Bob keeps them and when Bob dies all hell breaks loose, you know.
There’s no operations, you know, manifesto, all right. I mean there’s no, you know, people talk about creating a dev/ops manifesto but truth of matter is development has their manifesto, operations doesn’t. Development has it’s pretty clean methodology or different methodologies for doing their, what they did 15 years ago is much better today, and they use tools, and operations in many cases are still, you know, single focused, you know, individually focused non process driven, not methodical, you know, Bob script, Bob’s directories, you know.
I think that what really needs to happen is and this is my [?] that operations is to pony up and supply dev what they needed to get their job done. And I think it’s more about I got hammered on a tweet where I said, you know, maybe it’s not dev/ops maybe it’s self service operations for developers.
Michael Coté: Yeah, yeah and you know we’ve talked about fine about this a little bit especially in reference to some of the stuffs at Pulse where, you know, operations has another difference between provisioning a server right or infrastructure and then getting it fully configured so developing can start working with it, which is just one angle on that. You know, this is like, this sort of topic is something that I whenever I talk with people about cloud for an extended period of time I try to bring up and especially like the cloud keynote I gave it that Philly ETE thing like it was one of the points that, you know, and I was speaking to a developer crowd.
So I was being more on the developer side and saying, you know, you got to, you got to get to be friends with the operations people right. Either you got to get rid of all of them and do it all yourself and wake up at 3 a.m. and you know enjoy your life after that or you got to get friendly with them at some point and I think it’s, but it’s bidirectional, right I mean —
John Willis: Right you know, I think it’s that operations has to give them the tools that they need to be able to do the things, so that you know, I love the story of what JPMorgan Chase did with Tivoli and I probably told this five or six times over the last, how long have we been doing this now two years?.
Michael Coté: Something like that.
John Willis: Okay, right and so but it’s a great story, and I think it fits here very well and that, at one point everybody in JPMorgan Chase hated Tivoli. They paid ridiculous amount of money for it. So JPMorgan chase runs as multiples like 17 different business units. Investment banking, merchandise and just all different divisions that are almost silo companies, right?
Michael Coté: Yeah.
John Willis: So there was one team that brought in Tivoli and you know, Plan A was to get all the different business units to write and do their own Tivoli monitoring and all that stuff, and that never worked. You know they were so busy, the last thing they want to do is learn a product and a new technology that wasn’t part of their core focus.
So then Plan B was, you know, this kind of glass house federated group would actually go ahead and make the choices for them, that never worked, right. Because then you got the, you know, “why did you wake me up at three in the morning and tell me that an, you know a file system that has Oracle databases on it is 90% full.” “Well it’s a 100 gig and it’s supposed to be 90%.”
So what they were one of the few very successful Tivoli monitoring companies that actually got full value out of the product and what they did is, they wrote, they built the wall between Tivoli and the business units. And so what they did, they implemented a self service implementation where the people in business units could basically fill in the blanks of everything you want to do.
The word Tivoli didn’t matter and they got exactly what they wanted. And you know what, at the end of the day, the business unit, the developers and the quasi sys admins in the business units were happy with this help and for the first time they were monitoring what they wanted to monitor, they were, you know, they had control, they knew you know, if they got an alert, they knew where it was, they knew where it came from.
The operators were just as happy because they didn’t, you know, they got woken up, if they got called at three in the morning by one of the guys in the business unit saying, “Why the hell did you send me alert in three in the morning?” He’ll say, “Dude, look what you have put in the self service portal. You said if this is over 80 call you at three in the morning.” You know, “Leave me alone; I am going back to bed,” you know.
They were — operations focus on giving them the ability to do their job and that was their job, their job was make sure the infrastructure was up and the tools that they are provided or the abstraction, the execution of the abstraction that they have developed always worked and never failed.
I think that to me, that’s what operations needs to do, operations needs to man up and give development, not that it is dev/ops and you know, I mean I think dev/ops is great idea but I think before you can talk about true dev/ops, I think you have to figure out that operations has got you know not, I know every once a while I get somebody to tweet up and be furious with me about something I said on a podcast.
Our good friend from Oracle was really upset about something I said last time, I don’t remember, but for the people who are doing it right, I am not talking about you, all right, because there are companies that get this right. But more people don’t get it right, and I am talking about the people who don’t get it right. They need to — I think operations needs to man up and get — put something on the table in form of allowing developers to go as far as they need to go and today it’s about allowing them to control the destiny of their deployment, right?
Michael Coté: Yeah, yeah.
John Willis: I think in the future though —
Michael Coté: The way I put it and the talk I gave is that the, you know, and again this is coming the from the development angle, but the point being that operations needs to enable this and embrace it is that you know if you are a developer, it sort of like your responsibility for the code sort of ends once you build the final, the image or whatever right, sort of like, you stop being responsible for the code and you give it over to operations people.
If you think about it it’s kind of not very, it’s pretty irresponsible to be that way. I mean if to be out like highfalutin if you are like a craftsman right, you should be interested in the way people are using your code and responsible for it and everything. And it seems like, I mean I think on the development side, people are not used to think and in the main stream and in the whole of the development world, they are not really used to thinking that way, right. That, they would actually be responsible for the application and production. To your point right, like operations doesn’t exactly facilitate that and make it easier because they don’t trust developers. I mean both of them don’t trust each other to do their job. Essentially or mess with their job in any way.
So operations might give production requirements to developers and developers think it’s all crazy and then application people might like want to, like talk to ops people, like “you should set it up this way and these are the kind of things you should do and everything.” And any ways it seems like there should be, there is, there used to be a lot more sharing of responsibility of the application you are supporting.
John Willis: I think that I am not firm believer of live and die by this but I mean I think that operations to be successful has to build on abstraction layer for the business unit owners, developers whatever you want to call them to be able to, and in fact if we circle really back to the Amazon story early, in a lot of ways Amazon is fulfilling that kind of destiny, right.
In other words they have built so many features now that a developer can actually manage an operational infrastructure without operations. Well that’s what to be, so it’s really that, you know if operations wants to succeed you know they are going to have to pony up and, and do because their competition is places like Amazon and cloud and things that you know it’s not about the competition, it’s about server replacement. It’s about the end, the end game, right and so you know people talk about, “Well we are going to be private cloud,” well private cloud would be a massive failure as well, because if all you have done is given them the ability to just do the same thing they are doing over and over but you’re still not helping them manage the op.
I mean it’s about the abilities. I got that from, from Damon, in fact I called ‘Damian’ on our last one, he is furious. So Damon Edwards of DTO Solutions you know he calls them the abilities, right. You know the manageability, the availability. And so I mean today operations just gets in the table of manageability of things like configuration and deployment and continuous integration and all that.
Then there are some shining stars there. But really at the end of the day I mean IBM 15 or 12 years ago created a DMTF, this isn’t the first time I have said this on this podcast. A DMTF specification called AMS, it was called Application Management Specification where a, this is exactly the problem they tried to solve. You know it failed for many reasons but they tried to put in place an industry standard, that application developers could speck out the manageability of their application and it not only included inventory and configuration management it also included monitoring and event management.
This is 12 years ago, IBM saw this, you know, that you, they already were way ahead of the game in understanding the dev/ops problem. They were actually trying to solve it as an industry standard you know 12 years. So we are talking about dev/ops today. You know IBM had a blue print as part of DMTF, explicitly designed to solve the problem we are talking about that’s red hot today at Gartner. In fact the guys that were around at Gartner back then they probably talked about it are all dead, you know. So there.
Michael Coté: There you go. Way to bring it back to the Gartner topic.
John Willis: There you go.
Michael Coté: Yeah, yeah now that you know all of the major vendors have talked about this dream of getting developers to instrument things, to put it in you know yet an another way right and —
John Willis: I’d love to see.
Michael Coté: Hopefully this dev/ops thing will get towards that.
John Willis: You know someday we need to get your buddy on, the guy who is the DMTF guy that works with VM, VMware now.
Michael Coté: Winston Bumpus.
John Willis: Yeah let’s get him on; because he’d know about this, right, he’s being doing it ever. And I’d like to talk to him about, you know whatever happened, why don’t we you know, we are all talking about dev/ops, now is the time to like resurrect that idea.
Michael Coté: That’s a good idea John.
John Willis: It happened here on the IT Management Podcast. The genesis of the resurrection of the AMS standard as part of DMTF that will solve the dev/ops problem started today.
Michael Coté: That’s right, dust off your SOA architecture books folks.
John Willis: That’s right.
Michael Coté: Because there is a whole bunch of that coming.
John Willis: All right, well let’s get it on, let’s talk about. We have been talking about getting him on a podcast and then…
Michael Coté: Yeah, all that.
John Willis: Plus we talk about the dev/ops, the OVF all how this stuff all, it’s time to actually resurrect this concept.
Michael Coté: So John here I am, I am on my iPhone, you have heard of the iPhone.
John Willis: Yes. I have, I certainly have, I already have.
Michael Coté: I am opening up the things to do application and I am writing in ‘get Bumpus.’ But you know while I am doing this on the iPhone have you used that, that wacky iPad thing?
John Willis: No have you? Oh! You have.
Michael Coté: I’ve used it for two days John and I’ve got to tell you I haven’t stop thinking about it since.
John Willis: Oh! No, my wife let me buy one.
Michael Cote: She won’t let you?
John Willis: No.
Michael Cote: You know what wives are, they are the biggest preventer for us wasting money on awesome gadgets.
John Willis: There you go.
Michael Cote: Why are they trying to save us money?
John Willis: It’s easier for me to get a new requirement in an Opscode through a justification process than it is to actually get new technology in my house.
Michael Cote: I mean, I mean I could be taking my cart full of gadgets to the poor house, if it weren’t from my wife Kim, I would own every, every single stupid gadget out there.
John Willis: Now we need to get Bumpus to work on this one too here.
Michael Cote: But yeah, you know, there really is, you know, there is a whole lot of abstract cloud I’ve had related stuff, but you know, I wrote a review of it and I said it was pretty awesome, but it’s a little expensive. It really is, you know, it’s a nice device and when I was at that conference last week, there were a few people who were iPading it as it were, you know, only bringing the iPad.
John Willis: You know, you want to be, you definitely want to be, I mean I you know, I am not really, you know, you’ve seen me in person I’m not a well dressed guy. I don’t look great but, but it would be fun to be — get bumped up the first class and on my next flight and be sitting there, to be the first guy in first class with your iPad, “Oh! Hey well you got an iPad hey oh.”
Michael Cote: What’s going on with that? Anyway we would [?] miss except to mention that Rackspace had some sort of —
John Willis: Oh yeah actually at the Cloud Connect one of the guys from Rackspace was showing me the kind of prototype of that application, very cool, I mean the iPad is very exciting for I think potential, you know, manageability or ability tools, you know, you start thinking about the, we’ve talked about this before this the idea that you really can’t get kind of a dashboard or and really see some, you know, things that are going on in your enterprise and stuff.
Michael Cote: Yeah, I know it’s a, it’s a, you know it’s a good device I think I don’t know, I mean I have to see what happens, I mean if I were to buy one I think I would wait to spend even more money and buy the ones that got 3D in it. So it’s not just wireless bound.
John Willis: Yeah, that would be kind of cool.
Michael Cote: It just, yeah, it just like, you know, I am completely off topic here, but really like it really is a really efficient way to read a lot of stuff really quickly, which is exactly the criticism that it gets, that it’s only good for reading, but man I feel like I feel like the, when I had it and I had also had a lot of calls to wake up early in the morning and a lot of sleepless nights, but I feel like I really caught up on all of my, my Google Reader stuff, which is kind of amazing like nowadays like I read through everything and I feel like I finally caught up on it.
John Willis: That’s pretty cool yeah.
Michael Cote: I think part of it was like was like, you know, the mobile interface for Google Reader on the iPad is really good.
John Willis: Yeah, that’s right.
Michael Cote: So, anyway they were a few little items I wanted to, I wanted to mention. I’m sure you got a few things left, but you know, we forget to or admitted to mention this last time, but our friends at Reductive Labs have very smartly renamed themselves Puppet Labs.
John Willis: Yeah, I saw that pretty cool.
Michael Cote: Yeah, so they got now you don’t have to because it’s use to be I would always be, and I’m sure this happened to you, you would be like, you know, Reductive Labs, the people who make puppet.
John Willis: Yeah, right, right.
Michael Cote: So there was that; so what else oh and also, you know, the announcement round up at the end there was a, last week GroundWork and Eucalyptus had a partnership announcement where they, they GroundWork and Eucalyptus have gotten together to do some sort of monitoring and management stuff for private cloud installs. And they were actually, you know, in addition to announcing their partnership, they were, they were trolling and I mean that in a good way for, for beta testers for that set up. So if you sort of interested in beta testing they’re definitely looking for people to beta test using GoundWork and Eucalyptus together as far as private cloud.
John Willis: I may go to a ball game with some of those guys, your Simon buddy there?
Michael Cote: Oh! Yeah.
John Willis: So I am going to be out there for the OpsCamp that’s right.
Michael Cote: Simon Bennett right.
John Willis: Yeah, yeah so I was watching there was a baseball game the other day [?] San Francisco it was just, that stadium out there it’s just awesome I’ve been there.
Michael Cote: You know, I’ve been to their office.
John Willis: It’s right across the street yeah I know.
Michael Cote: I forget, I guess I’ve been only there once or something, but there is a place up, up there and I think it’s called Dogpatch I don’t anything about San Francisco, but yeah you are right, it’s down by the stadium. I think they are really close to the Twitter offices and there is a place called 21st Amendment and I think they actually have a Twitter Beer. So you should, you should definitely go up there and try to get the Twitter Beer.
John Willis: So I just tweeted hey man that place looks like kind of, I mean just watching that, seeing that ball field and those things man it just a great place to catch a day game and you just say when you come up here I’m like I’m going to be up here and may for the OpsCamp, you know.
Michael Cote: There you go.
John Willis: There you go.
Michael Cote: Then let’s see — I think the only other thing I was going to point out is I actually signed up for a little seven day trial, you know, GigaOM Pro they have this like paywall around some of the research content that they have and they had like a infrastructure cloud quarterly review and I was like hey I’ll check that out and so I signed up for a little seven day trial, which I need to go cancel before they charge me it’s like $79.
John Willis: Alright now since you put me on the spot earlier, so that’s a good invest or not for a guy like me?
Michael Cote: Well, I think, I’ll tell you, I’ll tell you, I don’t think I’m going to sign up for it because I sort of already read through all this stuff right, I mean it’s kind of my job. So it’s, but I do have to say like looking through all the other research that they have that their quarterly review is things like infrastructure like I forget the name of the dude who does their infrastructure stuff, but man it is really comprehensive like I was reading through, they’ve only done it for four or five quarters and it really is like you read it through it and you like oh! Yeah that happened, that happened that happened and it really is they round up everything and so. So it’s actually.
John Willis: I’m actually going to —
Michael Cote: Like thinking about if this isn’t what I do as a job partly right, I mean it really would be, it would be really easy to go into there and, you know, pay your $79 a year, which is a rebate or discount or whatever.
I think you could kind of, you know, just like stick you head in that corner every now and then and get an update of what’s happening. The other content they had, it was kind of interesting too, I mean they had some stuff that was like, you know, a lot of stuff they have is kind of like beginner to lower level intermediate stuff, right.
So, how should you be using social media and things like that and that, you know, they also have a lot of, you know, a lot of the, the background of GigaOM stuff is a lot of a telco broadband stuff like I think [?] use to cover a broadband and telco stuff and then like they have an Austin based persons Stacey Higginbotham who I met and very nice person and she does a lot of coverage of, you know, telcos and stuff.
So they have a pretty interesting and I don’t really cover that space at all, but they have interesting rap ups of stuff in the telco and broadband section and, you know, it’s kind of interesting poking around there. So I think and, you know, the other thing is like really it’s actually really cheap. It’s like, like it’s half off now, I don’t know why I am talking so long about this, but you know, I’m always fascinated about actually paying for content right?
So I think, I think if you are someone whose job wasn’t to follow this kind of stuff and, you know, their stuff is pretty worth paying attention to. Now I would say if you paid attention to my more or less daily like link collection if you kind of get a sense of the same sort of things happening, but you know, it’s interesting stuff, so, so good on them for that infrastructure review was fun reading. And I think that’s about all that I was going to mention you got any other last minute items to jimmy in there.
John Willis: Jimmy in there how do you jimmy in something.
Michael Cote: I don’t know you just talk really loud, that usually works.
John Willis: No, we have covered everything on my list so.
Michael Cote: Yeah, well, you know, I don’t have any conferences coming up sound like you had a worldwide tour there. I actually have a — I’m going to be on like a virtual panel with one of the guys from Zenoss next week talking about managing, you know, tips for managing the cloud how to put a —
John Willis: That would be cool.
Michael Cote: I think it’s actually during the same time that the Cloud Expo thing is, but —
John Willis: Yeah, that’s a busy week up there man, they got so much stuff going on. The I actually, I am going to, I’m speaking, I’m going to be speak, I got a couple of cool speaking games I’m going to be at Oscon. So I am going to be doing some really cool are you going to go Oscon this year?
Michael Cote: Probably not Steven O’Grady usually goes there.
John Willis: Yeah I knew Steven is there so and then so, but then I also got invited, first week of June to go to or second week of June to go Dublin to speak at this, it’s a Java conference over there.
Michael Cote: Yeah.
John Willis: Let me pull it up here it’s eyepiece, epicenter or something like that. Hold on.
Michael Cote: Man you got to tell these people to invite me to these things I need more international travel John.
John Willis: Yeah, no I mean.
Michael Cote: That Executive Platinum status doesn’t get itself that’s all.
John Willis: Yeah really I don’t know some people laugh and say it’s a big deal, but this is like the first year I got the gold status in March.
Michael Cote: Wait, wait you know I — this is the first time I’ve had Executive Platinum this year, so nice.
John Willis: I think this.
Michael Cote: Just like your whole another category of key person when you have Executive —
John Willis: Yeah, I am going to
Michael Cote: You know what happened?
John Willis: What?
Michael Coté: The desk agent actually smiled at mean and was very nice, it was amazing.
John Willis: There you go — it’s a, what’s Executive Platinum?
Michael Cote: That’s top tier Americans.
John Willis: But it’s at 100,000.
Michael Cote: Yeah, yeah it’s the highest level John.
John Willis: So last year, I am so pissed, I was, so on Delta what is it, 70,000 miles, 75,000 miles. I was like 735, you know, so I missed it by like about 1,500 miles you know.
Michael Coté: Yeah, that’s when you’re like, hey honey you want to go to Tokyo?
John Willis: Well you know it was December.
Michael Coté: Yeah.
John Willis: But —
Michael Coté: And also you have kids.
John Willis: Yeah that’s right, but, yeah, I know I think this year I’m definitely going to make Platinum but you know that’s not really a good thing.
Michael Coté: Yeah, yeah, yeah.
John Willis: You know, of course kids are like —
Michael Coté: Now, I always joke that’s it’s that that Executive Platinum is otherwise known as divorce status.
John Willis: Yeah, really so — I think I was trying to hire and he said to me, he said, yeah John, he says I am looking to — so well so what you thought about travel and he is like, I think gold would probably be okay and platinum I might get a [?] I am not definitely not looking for job for Executive Platinum.
Michael Coté: Yeah, yeah.
John Willis: So we use that as your kind of —
Michael Coté: No, it’s true although on the other hand I love to say when my wife Kim travels with me she, she doesn’t enjoy quite a lot.
John Willis: Well, that’s the upside of those things.
Michael Coté: Especially when you travel. Some people find this talk fascinating, but whatever, when we travel international and you have at least mid tier status on airlines you get really good, you get really good lounge access and everything which is, it’s nice.
John Willis: Well, no I mean, you know for me, not you know this, theirs is a big thing about status is about being on the board early.
Michael Coté: Yeah, yeah.
John Willis: A lot of airports you can take the short cut entrance.
Michael Coté: Oh! Yeah.
John Willis: You know when you are, when you are going to, if you need to check in luggage you can always check in on the gold, you know the gold platinum line which, you know, sometimes and I can say how many times I would have missed my flight and you know and I was in Vegas, coming back from Vegas. It was like for some reason the airport lounge were like wrapping around the building and then they had a special line for platinum and I walked right through, you know, so…
Michael Coté: Yeah, yeah, I always forget to take it back, because that’s the line I need to look up, because it says first class. And usually that means that it’s any ways, it usually…
John Willis: I always, whenever I just want I go right through, so those thing, all those things on Delta man, the Delta awesome because I don’t say this as an American but they know [?] like a week before the flight and you don’t even have, you don’t rush through anything.
Michael Coté: Now if you have executive platinum they, I don’t know if it’s a week, but it’s like that, they give you auto upgrades instead of you having to go ask credits for to whatever and they’ll just do it and it’s, I love that executive platinum status, very nice.
John Willis: Yeah but you, when you loose it you are going to be —
Michael Coté: Yeah, well I’ll get busted down to platinum or if my family is lucky I’ll get busted down to nothing.
John Willis: Yeah, when you hit down to gold they start spitting on you again and…
Michael Coté: Well, you know, I know a tourist over at [?] mess him and I are like American Airline air status [?] buddies and we were emailing with each other when he was tracking. Once you get a million miles on American, you get lifetime gold and we were like I’m counting down the miles, you got better [?] sometimes so that was —
John Willis: Yeah I wish [?] because I don’t have —
Michael Coté: It will be an exciting event.
John Willis: I don’t’ have that because I keep moving and then I keep winding up, I probably got like 600,000 and 700,000 on three different airlines.
Michael Coté: Oh! John you got it. You are in Atlanta.
John Willis: [?] Delta was —
Michael Coté: That’s what I am saying is like just always Delta, I always fly American.
John Willis: When I was in North Carolina it was American and I mean that was American was I lived in North Carolina for [?] and —
Michael Coté: Yeah.
John Willis: It used to be their hub.
Michael Coté: Yeah, I hear you.
John Willis: So listen to this, so I’m going to be speaking with this, its Epicenter right, so it’s Dublin in June.
Michael Coté: Epicenter.
John Willis: So it’s pretty cool and it’s going to be and the venue pretty, pretty cool and it’s actually a famous university there I’m trying to get — here is the venue.
Michael Coté: Trinity.
John Willis: Yeah, Trinity College yeah ain’t that pretty cool.
Michael Coté: Yeah that will be fun, Dublin is a nice, I haven’t been there forever.
John Willis: I haven’t been neither.
Michael Coté: Yeah, well that will, that will be good times right there.
John Willis: Absolutely.
Michael Coté: So you know, you get a little bit of airline information and a whole bunch of IT [?].
John Willis: That’s right.
Michael Coté: But you know I don’t think there has been a whole lot of just good old fashion IT management news of late, although you know I have been taking a lot service desk people who are sort of recloudising themselves. So that’s interesting but you know there you go, that’s the week in IT Management & Cloud.
John Willis: We should, you know I am trying to get these [?] they gave a great presentation.
Michael Coté: Oh! Yeah I love to talk to them.
John Willis: Yeah, so they did a presentation, was one of the best cloud presentations I have ever seen and, what they talked about is, how they are integrating their ITIL process into their kind of three tier cloud strategy and it’s just, it’s great start. You know I’ll see you and [?]. They can at least talk about what they talked about [?] so.
Michael Coté: Yeah that’d be perfect.
John Willis: I’ll try to, that’d be cool. But let’s get Bumpus; I want to do the Bumpus thing.
Michael Coté: All right, well it’s my to-do application now.
John Willis: All right.
Michael Coté: Once it involves the iPhone it generally gets done.
John Willis: There you go.
Michael Coté: All right well, let’s see everyone next time.
John Willis: Take care.
Disclosure: many RedMonk clients were mentioned, including John’s outfit, OpsCode. See the RedMonk client list for clients.