 So Vick is a good friend of mine and we're here together at the Cloud Era office. Vick as an entrepreneur was previously at Yahoo and then Google. Now he's working on a startup. Vick Singh is in the house. Vick, we first met when you were at Yahoo. Any questions? You wrote Yahoo boss on a weekend. No, is that delicious? You did some work on Yahoo boss, right? You're known for some of that work, right? Yeah, so we had a team of quite a few talented people and we had a mission to try to open up search and that turned into Yahoo boss and it was a fun project to be a part of. Cool. Well, for the folks who don't know out there, I've known Vick for a couple months now. We've been kind of sitting next to each other here in the Cloud Era labs office and SiliconANGLE labs office and Vick's working on a super secret startup, which we won't talk about, but it's very data focused, big data, the big trend. I know you're not going to talk about that, but I just want to ask you just what's your observations on data right now? I mean, we were at the Stratoconference and obviously data, a lot of people are partying with data these days in terms of playing with it and using data and the social media business has been starving for ROI. We're seeing now people using data in social media. You're seeing enterprises go in and look at big data from a pure analytics standpoint to actually competitive advantage. You've worked at Yahoo and so did Amar Awadallah, the co-founder of Cloud Era and those big companies have had the big data problem and they've kind of lived through it. What are you seeing right now in the world in terms of, I know you're taking around coding away, but without talking about your company, just in general market trends around data, what are you seeing? Yeah, I think the big data problems, I mean, I think the way people think of it is it's this problem that is confined to just the big companies, right? Like if you're Microsoft or Yahoo or Google and you're building an entire web search index from scratch, you're the ones who have the big data problems. I think that's not necessarily true anymore. I think there's a lot of companies, even like medium-sized companies and startups who are trying to tackle problems that are overlapping with some of the problems that are faced when building solutions for big data. For example, if you want to build kind of artificial intelligence machine learning models over even not even that large of a data set, it requires a considerable amount of horsepower, hardware to be able to do it and then as you get more and more data and you want to be able to model that type of data, then you start running into a lot of the problems very quickly that you run into at some level at a search company and so I think that's becoming more of a norm rather than an exception is trying to be able to manage data. But I still think there's another flip side to this as well and I think it's not like this is a surprise. I think a lot of companies who've been burned in the past in building solutions for big data have tried to scale out too quickly. What do you mean by that? Keep an example. Like what do you mean like scale out too quickly in terms of like the sprawl of data or just coding or scaling? Well a lot of times the way their data gets generated is typically from user behavior or customer behavior. So let's think of a consumer application and the only way where you get data in some of these cases is if you have a lot of consumer activity, user activity. And so you know when you think about like you know hey we took my SQL database and we took you know like some like basic kind of programming stack and put it together, oh it's not going to scale for more than like 100,000 users or something like that. So it's hard to throw really fancy solutions to solving these problems very early on. They start using the SQL database system and they start using all these different types of things. They overbuild basically. In a way they overbuild and almost foreclose their capabilities right? Yeah I think you know some of these solutions are a lot easier to use now. But I think a lot you should only do it if you have those practices. Like if you're not like an expert about how this new data store is going to work. But you're told that this data store will scale to you know like a bazillion users. That's not going to get enough reason to use that. Especially before you have one user. That doesn't make sense. You should rather earn the right to scale right? Like get destroyed. You know build a platform that you know gets destroyed. So play around. Do some R&D and let's see how the data kind of reacts if you will. Kind of like let it kind of morph. And use pieces of technology that you're familiar with. Like I think the most important thing is understanding your staff as opposed to embracing things because people told you to use it. Like a vendor you mean. Yeah I think like for example like you know a lot of people give like things like my SQL hard time. But you know a Postgres SQL database. It's tried and true over time and people are using it. And it's very understandable. Facebook is a great example. People talk about Facebook. You know you have a lot of friends that work there and we know people that work there. The classic it won't scale argument have we heard over and over again. But they keep on working at it. They essentially writing their own operating system. In a way or I don't think they're writing codes that make it work right? Yeah I mean I think you know for it's kind of a flip of a coin. Where you either you can build a really nice architecture early on. But you could get caught up in becoming an infrastructure company versus building a product that people are going to use or whatever your business case is. And so in the case of Facebook I don't think the beginning stages of how they built their company was a serious problem. Like using my SQL using my PHP. That's totally fine. But it's better to have a problem where you could you know have enough money and revenue and users kind of see money exactly to go and reinvest and rebuild that infrastructure. Now could they have rebuilt that at different points of time. That's more strategic. But I think in terms of that they did right I mean. Yeah I think there's certain questions about whether they could have done it earlier before. Twitter is more of an obvious fail in the market where everyone is exposed to the failure of their scale. The growth was massive. There was only their fault. They built what Ruby front-end and kind of an easy packed back end. And then just keep up with it. I don't know too much about their infrastructure but he knows. Their needs I mean they successfully come out of it. I mean Twitter has come out of it. Yeah and you know they have some really good folks they brought in who I think are going to do a great job with that. But I think like I rather have that problem than the opposite problem. Which is no customers. You have no company and it's scalable. Fantastic infrastructure that is not battle tested. What other trends are you seeing out there. So the scale out was one. So there's that and I think the other one is more intelligence around the data. So I think a lot of people go into the data, data is king, data is king and then explaining why that is is sometimes missing in that story. And so I think it's up to, I think startups and medium sized companies to try to deliver the reasons as to why data is king. You know provide killer applications that literally party on the data and derive insights from that data to explain things that would have otherwise been difficult to do through a manual process. If there's companies that could do that in a way that would otherwise or delivers killer consumer applications, then those are I think the future of the big data and of world. I think that's going to be to keep the data. I've always said the developer kits the data now and I wrote a post in 2009 this hot trend. I mean literally developing with the data. Being a developer like coding with the data. What do you think about, I mean you're younger than I am by a few years. There's kind of a new breed of data there. When I did my first company 13 years ago it was kind of like huge investment to start up and everyone kind of knows that stick now, why common is handing out free money to people and there's all this kind of angel money. You can literally get started for 50K and 150K or whatever you need. It's a lot cheaper. What is the mindset of the young entrepreneur right now? When I say young entrepreneur, I mean like the technical entrepreneur, people coming, you went to MIT, this MIT Stanford, you got Georgia Tech, Carnegie Mellon, Princeton, these big schools northeastern in Boston, you know these big comp side programs. What's the mindset, I mean no fear, are they attacking the problems differently? Well one correction I didn't go to MIT but I was fortunate to receive an award from them. You got an award from MIT. But I was at Berkeley and that's actually where I met my co-founding team and it's hard to generalize and know exactly what's in the minds or heads of folks who are starting companies these days. But I think it does feel more realistic, right? People aren't like putting bubble mentality. You guys have seen the bubble from college right? And so that seeing the carnage it's not a real profligate kind of environment for these young entrepreneurs. They're pretty cautious. I think there are more people who are at least in my like my network, my close network, who are more willing to do a startup now than before. And I think that's mainly because people know now that you don't need that much capital that you need before. So the terms are a lot better these days. People can build things in the cloud and those costs come down considerably. And I think that's encouraging people to go do it. And I think when you have, when all your other friends are doing startups and you're the one who's not the grass is always green on the other side. It's hard to do a startup. It's not easy. It's not entirely you know, there's definitely a lot of pros to being a larger or medium sized organization. But I think that not knowing what's there it captivates people. And the fact that they can own and control something I think is the something that's very difficult. What's an experience that you could share with folks out there as an entrepreneur that you didn't expect? Good and bad. Things that, because you came into the startup you've had a good network of friends. You went EIR at BC from Sutter Hill great people. You certainly had certain expectations. What things didn't you see that was a surprise for you, both good and bad that you could share? Yeah, I mean we're still very early. So I think Is it the freedom? Freedom? Yeah. So we haven't had that many crazy ups and downs yet. And I think they're bound to happen. But based on currently what's been happening I think the thing in terms of good is how I guess we were very fortunate because getting the team together, getting people to support us, being able to get access to great resources like this, I think came rather easy to us and we were just very fortunate to be at the right place at the right time. And your founding team has been pretty tight. You guys work together to have the roommates. I think that's really important to know who you're working with and they're the smartest people I've ever worked with so it always makes it easier. But I think there's that and the in terms of the negative is trying to you know it's something that a few folks have always told me but a lot of times a lot of people go in and they feel like they can have everything figured out on day one. And as much as you want to believe that you have something that is a fully proofed idea, you really have to go and test it and it's kind of like the scientific method if you're most likely your hypothesis is wrong so you have to go and test and figure out why and then re... No matter how smart you are you just got to keep on testing it to the thesis, right? And so for us we started out with a very humble proposition and we tested it out and then we tried to incorporate the feedback and make it stronger, stronger, stronger and there's a certain point where you have to kind of figure out are you close enough to the truth and do you think you're in a big enough market where you could go and exercise it and leverage all the capital that you've gained in terms of people and talent and money to go after is that something that you could spend like 5, 10 years doing. I think that is, especially when you have a team and other people are depending on you, you want to make sure you make that decision in our way. Closer to the truth that's really really good advice because that's really what you got to do is you got to test it and you don't have a big team so you didn't really take a lot of funding and you guys were very kind of lean so it's not the big resource from the big mother ship you don't have the cafeteria the main the biggest the biggest resource that you don't have a lot of this time and so that's the thing that we want to kind of make sure we leverage properly and so there's always a clock ticking so you kind of have to manage that well. What do you think, I know you're going to get back to coding away but I really appreciate you coming on the Cube and by the way Vic is the inspiration behind Silicon Academy which is going to be coming online hopefully next quarter which is going to be more of an educational nonprofit and we were talking about doing something similar and that idea came up and URI was available so Vic credited with that inspiration. What was our, you and I were talking about so you were there so that's the history of that. I think it's a cool, I really think it's something interesting there and I think what the Khan Academy was Khan Academy is fantastic it's microeducation microVC, microeducation Absolutely. I think something like that more information delivered through that medium where you have just like short clips that provide a significant amount of value that can be geared towards different like verticals or anything else and in our case we know we have access to a bunch of folks who understand a lot about technology we have a lot of access to people even to the venture industry I think if you could provide those in a way where people of all ages or some kind of selected ages can digest it in an easier format the better because this stuff is super hard to read in a book. You guys have books in your desk I know the programming books, everyone has books and when you have a problem you need inspiration you kind of go to a chapter, you kind of zip open a chapter and say okay, what's that code that guy wrote, oh and you get inspiration read a quick study and you're out in an out five minute look up and you're gone so that kind of quick digestible topic based table of contents like YouTube video we're going to do a Silicon Academy coming online next quarter I hope, some folks may kick in some cash, if you're interested email me and send us money we'll get it online faster so we can do some stuff too, I know you wanted to volunteer there but for the folks out there that are kicking the tires everyone's always working on a start with the next big idea what are the challenging areas that you see people really working on the news that Facebook is going to X Facebook is going to back in AI some AI work and obviously AI has been an academic paradigm for years and grounded in big theory but now with the web AI looks promising and I'm a big believer of it but there's these big kind of technical engineering software challenges that are emerging whether it's data, scale out I know one of your co-founders was getting masters of PhD at MIT he took a little break and we were talking about scaling out and all these things what are the key areas out there that computer scientists are trying to crack the code on that you think are really interesting that you could talk about or identify trends is it new algorithms, is it new filters I was talking about CORE earlier about some of the challenges that they have around people rank and contextual relevance any big ideas you see the way we even kind of think about it is like you could go at it and try to solve some really interesting problem within your algorithm but I think it's like in the context of something and at least for us the way we think about it because we're very much in the data space we're very much in the AI space so it hits on a lot of the themes that you're hitting on but I think what we're interested in is we want to use these tools for the purposes of making something really important, simpler to use and I think if you strive for simplicity there's an incredible amount of complexity in order to build a solution that can deliver that and so I think striving to make processes that are manual or striving to make processes that are very complex and simplifying them in a way where you get 80% of the value that you got before perhaps even much better by utilizing more data, by utilizing automated techniques like intelligence I think that to me seems to be a common theme I see in some of these CloudEra has a hot thing right now with HBase we're kind of playing around with it a little bit but things like HBase and Hadoo any particular technology is popping out of the woodwork that you can see entrepreneurs maybe getting behind and drafting behind and behind every open source movement whether it's Apache or there's always been this ecosystem in the lift behind it do you see any things? I mean we're talking about folks about Cassandra and the NowSQL stuff going on any trends there? it's still kind of for us at least we use the platforms that we know how to use and so some of these things are still kind of game bedded and so we're letting that kind of run out but the things that they're promising are fantastic and they would be huge to have in our platform I think in terms of things that I know well my friends and colleagues know a lot more about some of these other platforms that you're mentioning but I think if I look at something like Lucine for example for search that for us has been actually quite pivotal in a lot of our work and so we've built a lot of stuff around it to make it work we use things like protocol buffers and serialization formats to make it easy to communicate between different languages and to provide services really easily we look at things like even some of these machine learning libraries that are out there are a lot of really good open source libraries but they're all fragmented, they're all over the place and there's some efforts to open source projects that provide some of these libraries and I think there's going to be more improvements in making those lots of use. So stitching those together in terms of kind of that big software framework has been hot obviously it's bringing source results to VMware so those kinds of integration plays? Yeah I think these open source projects are going to become more mature and I think they're going to get more adoption because I think some of these things are still very early and I think a lot of it is due to the fact that these applications in the world have not yet been created I think a lot of these kind of data intelligent data where applications haven't been created but the moment that they start to kind of take off these platforms are going to really kind of take off as well with those because they all are all connected and obviously processing technology has increased and machine learning is hot outside of your work that you're working on the startup and the data stuff outside of that what's really exciting you about tech these days is it simply the platform what's getting you most excited out there that you're seeing mobile, is it the social is it the abundance of data, is there anything you can say that's really one of the coolest trends we've seen is it cloud, is it... Yeah I mean for me I've always kind of operate at a lower level and try to I want to be like a player where we could obviously have angles and all those things where whether you're helping cloud services whether you're helping mobile but provide some layer where you could provide this intelligence to these other things that's how I kind of always have seen myself in that world but I think in terms of general trends that look really interesting I think obviously mobile is going to be really really interesting The wives are calling Yeah that's Linda Honey I'm live, look at angles.tv Do you want to know what's for dinner tonight? Yeah I think he knows something like mobile I think there's a lot left to do there I think something that's interesting is you know I just recently got a smartphone so I hadn't had a smartphone before I always had a phone that was really really bad a flip phone that could do nothing except make phone calls and so once I got like an android phone I started using it and I had like I had a browser and I could do all this great stuff and saw applications it was really fantastic I mean the main thing I use it for is email but I mean... Do you use the speech recognition as phenomenal? Yeah it's getting a lot better it's really a swipe is really cool it's really nice I think though the one thing with these kind of I mean my take is and there's some folks who agree and there's a lot of folks who disagree but like these application marketplaces where you go and install these apps it kind of is like the emphasis of what we learn from the web in some ways like I think of these application marketplaces is kind of like AOL right like you go I was just going to ask you about AOL I was just going to ask you about AOL It's a big content trend What do you think about it? You know you basically take whatever is popular and you download and install it but the nice thing about the web is when you have things like you know links between things and you could have search engines which can try to find fairness in terms of determining which ones are relevant and which ones aren't then you can rank the content so that people can find it based on the need so that's why I like the web because the web is the most part is very democratic you offer these things to kind of take off on their own people link to things that are good and these search engines can go and search and rank over it very intelligently but in the application marketplace world you're really bound to the out market and you know you have these apps that get put in and if you become popular then you're always popular like winners always win and you know I just feel like you have to install things and I think you'll eventually go back into the web I hope that everyone is talking about designing for mobile experience everyone is trying to design for but let's talk about the web obviously you had a lot of experience with search with Yahoo you were at Google you did work with Yahoo bosses open source kind of search results and you know you play with delicious one on your own time you came over and played with it trying to make it more real time you know the sunsetting delicious but the real time web is challenging Google has a Yahoo have an incumbent position with a CEO market and advertising the way it is with search results and Google buys Huffington Post this week and they've been on a barrage they bought TechCrunch and Gadget a bunch of blogs and they have now Huffington Post but the big thing has been dissing these content farms and there's been articles out on the web today about how machines are going to end up killing journalism so being in the machine learning and AI space you know serious need for new approaches to essentially vet spam, splogs or content farms because not that AOL is being accused of content farms although they're saying they want more page views they are investing in quality Huffington Post is a very reputable organization TechCrunch has grown to have a bunch of writers so at least AOL is putting on a good face with their content farm but is the web getting dumbed down and what's your view on all this what's needed you know it's not really getting dumbed down just more and more blogs and more content it's a search problem it's a hard problem I think spam is going to be this continual fight as long as there's economic incentives to create spam it's going to be created and so the techniques I think are always going to get better and I do think search engines search engines that will last are incentivized to deal with spam in the long run if they want to build a compelling experience for the users they can't really find these kind of give and go kind of relationships where it's okay to have some spam of the economic benefits get back I do think that these companies really do try their best to prevent spam it's just a really really hard problem and I think it's a combination of a bunch of things one is maybe some of these models need to work differently maybe some of the ways the economics work need to be redodged maybe the way humans get involved in the process like us in terms of labeling spam can be more encouraged or something of that nature like what blood goes around to do it's going to be interesting to see how that plays out my feeling is there's going to be a lot more techniques now that especially now that it's been in the public and people are complaining about this and it's becoming a problem there's kind of two things that are going to come out one of the techniques are going to be a lot better but I think the other thing is it's going to motivate applications like different types of applications applications that are more vertically interested that are socially filtered I think this might be their chance to shine because maybe these vertical you can decouple them and let yet have them integrate together and the combination of the two can create a filter whether it's gesture based friends networks closed networks because this stuff is becoming more of a problem or it's becoming a problem that is being perceived as an issue to the public that this will create more a new way for other applications to come in and provide these filters well I mean it's a problem everyone knows the noise levels huge on twitter core recently is trying to get the social media experts out which makes total sense from them in the sense that they don't want that noise it's kind of like the use net groups to comment on my facebook when I wrote about it I said yeah it's like use net when all the AOLers came in back in the day kind of like the newbies I call the people who want the free beer but you can't just kick them out you gotta make room for them you can't have a social app without having a social media component it's a challenge you don't want to boot out people who are using the application in a meaningful way to themselves and to others if they're doing it for selfish reasons that are purely economic then that's questionable but in some ways you want to embrace the social media experts because those are the guys who actually leverage your application the most those are the guys who get you virality I think you brought up a good point too the other thing about Quora that I'll comment on is that they start putting these things in place too early like moderators they could be foreclosing data that could be telling them what to do I had some things voted down on me from a guy I don't even know who had no context to who I was and what my expertise was and actually was relevant to one topic but I'm not going to complain I see Quora doing some advancements there and I'm not down on Quora I agree young team just got to get the product they got a good product I like the UI they did a good job with the UI some of the questions are fantastic get real experts answering I mean those contents I mean I could literally cut and paste and there'd be standalone blog posts it's that high quality okay with Vixing we're here in the cube fellow cloud era labs member here at siliconangle.com siliconangle.tv and soon to be Silicon Academy when we get that off the ground see you in the next video