Tuesday, November 29, 2011

BigData - Society (!) edition

As members of our society, we generate a ton of data.  Some of this ends up as governmental (tax information, census), some of it ends up as administrative (land records, traffic accidents...  yes, a fine distinction, but still a relevant one), and some of it is, well, just data (rainfall).

Over the last few years or so, there have been a number of attempts to make this data public - typically under some form of open data initiative. The general idea, of course, is that once people can start messing with this data, they can come up with all sorts of relevant and fun uses.  For an example, check out what the good people in California have done, and in particularly, the remarkably cool Explore California mashup...






 While their heart may be in the right place, politics usually ends up causing chaos - and when I say politics, I actually mean national security.  You'd be surprised at how much stuff is considered 'dangerous'.
Rainfall!  Terrorists could use it for an attack when there is a storm!  Think of the children!
Movie plot threats can be very conducive to shutting down access to data - and it almost always is.
That said, Britain's latest attempt at re-packaging open data into a new 'Conservative' initiative seems to be somewhat plausible, and hopefully, one that'll succeed.
Some of the initiatives include (hat-tip Bobbie Johnson at GigaOm)

  • Increased health dataA raft of new initiatives including: better links between data sets for individual patients; new data release services to push information about healthcare from the NHS — Britain’s enormous public health system — to researchers and industry.
  • Increased transport dataFree availability of a range of real-time data covering the running of the rail network and bus infrastructure, plus regular data releases on local highway and road congestion
  • Access to weather dataBritain’s weather forecasting unit, the Met Office, will open up more of its public weather service data for free, in open formats, in what the government says will be “the largest volume of high quality weather data and information made available by a national meteorological organisation anywhere in the world”.
  • More housing dataThe national Land Registry, which keeps track of sales and land ownership, will release monthly data on residential home sales, including prices paid.
    The key, of course, is to see how much of this actually makes it through the system...
  • Monday, November 28, 2011

    Facebook and BigData

    A nifty study by Facebook on how connected we actually are.  In a nutshell
    1. 69 billion friendships.  (Don't get into the Is This Really A Social Graph argument.  Please)
    2. The six degrees of Kevin Bacon, are now effectively Five degrees of Kevin Bacon.  They're also getting shorter, and could maybe end up as Four degrees of Kevin Bacon
    3. Most of the connections are weak ties, i.e. ones where we are barely/casually/dis-interestedly connected.  We might care about them under certain circumstances, but mostly we don't.
    The tricky part here is the weak ties bit.  Its clear that network effects are in play with weak ties, i.e., the more people that are connected weakly, the more likely that some (significant) subset of these people will care about the same thing, hence pushing their strong ties into caring about it too, hence magnifying the cause (egypt!  w00t!).

    Fun stuff!

    VoIP Geek - Thats me (cough, cough)

    I stumbled upon the iFusion phone dock while searching for something completely different (isn't that the way it always works?  Its exactly like the NVX 610, just not as spectacularly expensive.
    Anyhow, its basically a 'desktop dock' for the iPhone, which basically answers the question
    Why have a separate desk-phone, and a mobile phone?  You should just have your mobile phone, and if you happen to be near your desk, you can take your mobile calls on your desk phone.
    Ok, I know, you're thinking to yourself "So?  Who cares?".  A lot of people, as it turns out. You get better voice quality, and its more comfortable to boot (unless you're one of those people who likes to walk around with a headset that is :-) )

    Anyhow, I'm now google-ing for an Android version...

    WebRTC - coming to Chrome

    Ok, thats really not all that new or interesting, but it is relevant. I discussed WebRTC before, and its pretty obvious that embedding it in Chrome is a pretty simple way to provide P2P video capabilities easily and natively to most apps.

    On a related note, anybody note the entertainment around TechCrunch and Paul Kinlan?  He was at a conference where he mentioned that webRTC could allow browsers to 'do' OnLive without plugins - this somehow got translated to Google is building a competitor to OnLive!.  More details here...
    I love our blogging world...

    Tuesday, November 22, 2011

    Nokia S60 (sucketh?)

    3 days into a 10 day trip in Italy, armed with nothing but a nokia symbian fone (ok, i have my laptop too, but thats a different issue). 
    Net result - every time I go by a TIM/Wind store, its a struggle to not go in and just by a bloody samsung SII. 
    Its just sheer force of will (and desire to complete this experiment) that is preventing me from hurling this phone w/ muttered imprecations...

    Saturday, November 19, 2011

    DotCloud - now w/ CouchDB *and* Redis *and* MongoDB

    One of my favorite cloud stacks just got a wee bit favoriter - DotCloud just added Redis & MongoDB to its stack.  DotCloud (like Heroku) give users a bit more of an inside view as to what is happening at the infrastructure layer, in keeping with my belief that the future (for now at least) is something-aaS.


    And, if you feel like testing this, do remember that the only diff 'tween the Free and Pay versions is support for multiple data-centers and an SLA.  Makes testing a *lot* easier, ne c'est pas?

    VoIP eating into Mobile Revenues

    To summarize - people are using "other" services on their phones over non-carrier networks (read wi-fi instead of data, IM instead of SMS, Skype instead of calling, etc.).



    The study, which was carried out on behalf of Mavenir by mobile(SQUARED), found that a third of operators believe operator traffic from messaging, voice and video calling will decline between 11% and 20% over the next 5-10 years. Another 20% of operators expect even steeper declines in the 31% to 40% range.
    “The findings confirm what we have found in discussions with the mobile industry,” Mavenir VP Shubh Agarwal said in a statement. “This is one of the primary reasons the industry is currently moving towards an all-IP converged core network accelerated by the deployment of LTE technology. By allowing users to place high definition voice and video calls, chat, share content, and discover new services as part of a globally connected framework, operators can retain and even grow their share of customer communication spend.”

    From BGR

    Whither SAP?

    Barb Darrow at GigaOM assembles the case that SAP should 'go cloud'.

    "...others said SAP has a cloud deficit that it should address.  (given SAP’s installed base of ERP users, it needs to) “take the Software-as-a-Service path to the cloud.”  “They need to build, buy or partner on SaaS ERP apps , or closely associated business apps as services to then move toward cloud values,” 

    I fear this may be a bit of a cart-before-the-horse situation.  Cloud isn't the solution to all problems, it is part of the solution, and it really depends on what the problem-space is, i.e., you might have SaaS, IaaS, PaaS, or the new trend something-aaS.

    Based on this view, I'd say SAP's big gap is that they need to expose themselves a lot more to others, i.e., work on (drastically) simplifying and pushing out their APIs,, and building out an eco-system where their clients and partners can share tools, techniques, and best-practices.

    In many ways, most SAP implementations are already a cloud, just a highly private one at that.  Simply grafting on a lower-end ERP package (e.g. NetSuite) isn't quite going to do much fo them, other than help them be buzz-word friendly.

    On the other hand, working from the inside out, i.e., using their existing systems to build out a global system and network of services, well, there is quite the future, ne c'est pas?

    Nerds! Nerds! Nerds!

    Friday, November 18, 2011

    All Apple All The Time


    Everything Apple
    Really? This is news? What next? Macintoshs to include Tint Control? (Bloom county reference, for those who care...)
    Apple wants to make it easier for customers in its fastest growing market to get their hands on software, which is why it started accepting Chinese yuan for App Store downloads on Friday. The change i...

    Mobile Development in the Enterprise

    An IBM survey of IT Professionals (yeah, yeah, whatever those are) shows that (my comments in italics)
    • 75% are working on some type of Mobility solution
      • I'm not surprised that there is that much mobility work going on.  Given the prevalence of smartphones in the world, we'd be dumb not to be focusing on Mobility
    • 70% are doing Android (iOS - 49%, Windows7 - 35%)
      • I wonder how this survey was conducted.  Pretty much everybody I know is working on both.  The key is, which one comes first, and which one fast-follows.
    • In the cloud world, people are currently working on virtualization and storage, but moving to new applications
      • I'm somewhat surprised at people still working on virtualization and storage.  Seems like most of the work in this area is done, or has gotten highly localized and/or feature-specific, as in 'adding storage to apps'. (see /bin)
    • Most of the work involves either extending current functionality to mobile apps (30%), or building new 'professional' capabilities (60%)
      • No shocker here.  The interpretation issue here is the extremely nebulous area between a mobile app that extends a current product, and a 'new' feature.  I suspect most people would prefer to say they are working on something new, as compared to saying they are working on current features.
    • Entertainment/Gaming is way down on the list of things they are working on - at the bottom, in fact, with the percentages in the teens.
      • I guess this depends on how one defines 'Entertainment', and 'Gaming'.  They are full-fledged industries in themselves, and I know plenty of people in the 'Entertainment' business who are merrily, and happily building out mobile apps.  And how does one paint gamification?  Again, I guess this depends on how one defines 'IT Professional'
    • India is right on top of 'Social Business', with 57% doing something in this arena.
      • w00t!

    Who buys iPads?

    From VentureBeat - Wealthy, Male, Gaming, Pet Owners.  Really.

    Thursday, November 17, 2011

    Cinque Terre - Vernazza


    It may be overly touristy, and somewhat hokey, but it is a beautiful area, and this is a disaster....
    Thirty-two years ago, I met two American college girls while hitchhiking in Switzerland. They were studying in Florence, and I asked them their favorite place in Italy. They surprised me by naming a place I had never heard of before: Cinque Terre. Curious, I headed south and discovered a humble string of five villages along Italy's Riviera coast with almost no tourism...and, it seemed, almost no contact with the modern world.

    High Speed Network as Basic Infrastructure - A Success Story

    (in Hungary, of course).
    Its a town in Hungary where they basically ran Fibre everywhere, as a basic infrastructure play.  It worked quite remarkably well.  The video is somewhat advert-ish, but the essential benefits do come forth.  Broadly speaking
    • Its keeping the young 'uns in town after they graduate, since they are 'connected' - its also stimulating startups
    • Its also keeping older people (60+) more 'connected'.  (Makes sense, as anyone with kids knows, Facetime/Skype has *huge* uptake amongst the previous generation)
    • Its green, and lower cost, since it makes telecommuting actually feasible and realistic
    • Its good for industry, helping out with supply-chain management, logistics, and ERP
    • Communication is, of course, dramatically simpler, easier, and lower cost
    • Its a boon to the media/advertising industry, since its trivial to move data
    • Retail benefits, with the ease of melding the web and Brick&Mortar
    • Its remarkably cheap - it costs basically nothing to dig a trench and put fibre in.  Again, something that we know here - it costs nothing to put in fibre in rural areas..
    Fascinating to see this in action...

    Phones that 'employees' have

    BGR summarizes the result of an iPass study on phones in the workplace
    Money quotes...

    “BlackBerry has not really fallen from the top spot in so much as other smartphones have grown faster,”
    "Android...21%'
    "8% of mobile workers said they intend to purchase an iPhone in 2012, 11.2% are planning to buy an Android device and 3.6% want a Windows Phone. Only 2.3% of workers have plans to buy a BlackBerry next year."


    Kahneman (Thinking Fast & Slow)

    From the world of Behavioral Economics - Kahneman's book is both fascinating and depressing.  Depressing, because it is never entertaining to realize that we are not the rational beings that we (and all of the Chicago School doofi) believe ourselves to be.

    Go buy it, and read it..  Thinking Fast and Slow...

    p.s.  For those who are unaware, Kahneman & Tversky are - effectively - the parents of Behavioral Economics.  They collaborated on prospect theory, which focuses on how humans handle decisions involving risk (in the real world, i.e., these aren't the optimal decisions that rational beings - whatever those are - would make, but ones that actually get made by humans...).  Tversky died in 1996, and Kahneman received the nobel prize for their work in 2002.

    Rooting the Kindle Fire

    For those who care...


    Kindle Fire just started shipping yesterday and a super easy one click root / jailbreak solution is already available. We went through the steps, were able to reproduce it and duplicate the end result and we also streamlined it a bit especially for the ones that do not excel in computer skills so here it is for your Kindle Fire rooting / jailbreaking pleasure.
    What is the root process bringing to the table? The possibility to install Google Android Marketplace (remember, by default Kindle Fire only comes with Amazon Marketplace), sideloading books on Kindle Fire is another enticing possibility as well. (more...)

    Safety First!


    Speaking as a frequent traveller, I feel much safer now...
    After promising an independent evaluation of X-ray body scanners, the head of the TSA now says he’ll put it off pending an inspector general report on the machines.

    Italy's Money Supply (Hint. Not looking good)

    Hat Tip - Tyler Cowen


    Free Speech!

    Wednesday, November 16, 2011

    300 Billion! (Italian Bond Edition)

    Yup, thats the number.  300 Billion.  The amount Italy needs to roll over next year.  Who d'you think is going to buy all those bonds?  Especially with everyone (banks in particular) desperately trying to get rid of the ones that they already have?

    <cough>ECB</cough>

    Monitoring your 'Cloud Stuff' (Enterprise Edition)

    ServiceMesh seems to be on a bit of a roll nowadays.  They just scored $15M from Ignition to (one presumes) expand their offerings and their geographical footprint.  Mind you, the latter is somewhat interesting since they exist to help enterprises control/manage/track their Cloud Services, which are - kinda - geography neutral :-)

    Regardless, this is a space that historically companies like BMC and HP have lived in.  They have been somewhat slow to adapt, though BMC has been making relevant noises vis-a-vis their Cloud LifeCycle Management line.  ServiceMesh's 'secret sauce' is that they are actively pushing their Agility Platform, a one-stop shop that (supposedly) brings the benefits of 'Agile' across an enterprise's IT services. (At this point, my snark sensor kicks in, wanting to point out that it might be easier to teach an elephant to tango, but hey, whatever).  Still, it does look kinda nifty - largely because they have nice APIs.
    Probably something I need to keep an eye out for....

    Shatner + Turkey + Fryer =:= Something

    Seriously Shatner is approaching Norris-ian proportions (or, for my Tollywood brethren, Rajnikant-ian)


    Tuesday, November 15, 2011

    Vasquez Always Dies

    Its always nice to suddenly discover a classic trope, especially one that has been around for a long time, but which you just hadn't noticed.  I present to you - Vasquez Always Dies
    When two Action Girls are featured in the same film, invariably the tougher, more competent, more aggressive and less feminine character will die, despite being better equipped for the situation at hand. This can range from appearance (if one woman is wearing sensible shoes, and the other is wearing high heels), personality, or other traits. (more...)

    Nepotism doesnt work


    Who knew! It turns out that the data are actually against it - Nepotism makes companies underperform...

    I use data from chief executive officer (CEO) successions to examine the impact of inherited control on firms? performance. I find that firms where incoming CEOs are related to the departing CEO, to a founder, or to a large shareholder by either blood or marriage underperform in terms of operating profitability and market-to-book ratios, relative to firms that promote unrelated CEOs. Consistent with wasteful nepotism, lower performance is prominent in firms that appoint family CEOs who did not attend ?selective? undergraduate institutions. Overall, the evidence indicates that nepotism hurts performance by limiting the scope of labor market competition (more...)

    First Amendment Issues

    Monday, November 14, 2011

    Getting past 'as a Service'

    It strikes me that, for the last ten or so years, I have been deploying Telephony as a Service.
    So?
    Note that this isn't Telephony running on Freeswitch running on AWS (Twilio), or Telephony running on Asterisk running in our datacenter (Fonality), or even Telephony running on a Huge Solaris Cluster Somewhere (Broadsoft).
    The point here is that the type of service I built doesn't neatly break down into the oh-so-neat layers that you always see in these type of diagrams


    If anything, what I've built is more like
    where the various parts are all munged into one somewhat nebulous cloud.
    "But!", I hear you start, "But!  How can you build anything to scale when its all cloud-y and nebulous-y!?"
    The easy answer, of course, is to wave my hands and say something mysterious and profound, like "Exelsior!" or, "Erlang!".
    The reality, however, is that in the fine, you will find various elements of Infrastructure, Platform, and Software, all doing Service-y things.  But, in the gross, they are not so easily de-reference-able.

    The thing is, the layers are all a-merging.  It turns out that building complex distributed telephony systems is hard.  Really hard.  To build out some of the more complex applications, you really need to have a pretty good handle on whats going on at all the levels - the servers handling the calls (infrastructure-aaS), the applications moving data around (platform-aaS), and the APIs used to actually manipulate the calls (software-aaS).
    Yes, of course, by the time this is all presented and visible to the end user in some nice-neat GUI, or better yet, API, its all taken care of and is a simple SaaS.
    Except that it isn't.
    Why not?  Because there are all sorts of fun things involved like rate-control, telephony costs, QoS, etc., that really cut across various levels of the platform.  And yes, I know, we could make oure APIs even more complicated, and end up with a scenario where the stack is truly de-referenced, except at this point it is not just getting a little ridiculous, it is pretty much making my original point, which is that the layers are a-merging, and you need to have visibility across the different layers of the stack.
    Its not just me, by the way.  Take a look at Heroku, or Elastic BeanStalk - they do almost exactly the same thing.  Its basically the growing realization on everybody's part (even Werner Vogels!) that once people start to get specialized (like us! Telephony clouds!), they need to have greater visibility into how the cloud functions, and consequently too much de-referencing can actually end up being counterproductive.
    In short, the future is about Cloud-as-a-Service (CaaS?), where the cloud is the merging of the previous as-a-Services.


    Data - Crunching vs. Visualization

    Google is opening up their engines for you to crunch on - basically, if you're data is BigTable-able, then you could run your crunch-jobs up in Google-space.  Which is all well and good, but it is getting quite apparent that the ability to crunch is really not the gating factor anymore.  Take the democratization of NoSQL (Hadoop, BigCouch, Mongo, whatever), add in a soupçon of AWS, find an erlang developer or two (so you can do real distributed development :-) ), and your ability to crunch is limited only by your wallet - and not all that much at that.

    The gating factor, of course, is Visualization.  Knowing what you are looking for is easy.  Finding something you weren't expecting is hard, and doing this quickly and easily is, well, close to impossible.  Thats where the emergent breed of BigData Visualization companies are really starting to shine.  Me, I'm quite partial to WeAreCloud (Bime from France).  They have a very usable interface, and actually make this fun.  

    PostNote:  A friend turned me on to TipTop - I need to see whats going on there...

    Phone as a Camera

    Sooo, I took my phone w/ me to Toronto (cousin's wedding), intending to use it as my camera.  I could go on about the effectiveness, functionality, pros/cons, etc., but its really quite simple.
    No Zoom =:= Deal Breaker

    Yup, thats pretty much it.  Back to the Panasonic ultra-zoom (zs10) for me...


    (I know, I know.  There is a time and a place for everything.  And it looks like I'm now going to be lugging around three cameras.  The P&S, the SLR, and the phone.  Joy.  My life is a *lot* simpler now.  Clearly)

    Felix Salmon on the Liquidity Crisis

    He captures it thusly
    What is true is that Europe is in the middle of a textbook liquidity crisis. Banks are not lending to each other — and the ECB isn’t stepping in to solve the problem. This is a serious structural issue with the way that the European monetary system was constructed: the ECB is tasked only with guarding inflation, and not with ensuring the health of the banking system. Individual national central banks are meant to do that. But they can’t print money — only the ECB can. So when there’s a liquidity crisis, no one’s able to step in and solve it.
    Note quite close to Krugman's one sentence, but close enough, eh?

    What has happened, it turns out, is that by going on the euro, Spain and Italy in effect reduced themselves to the status of third-world countries that have to borrow in someone else’s currency, with all the loss of flexibility that implies
    .

    Ericsson on the growth of Mobile Data

    Ericsson has a new report out which obsessively details the growth of Mobile Data and Traffic.  You can go read the whole thing (hey, I did!), but there are some highlights which are quite relevant


    1) People use less than 5GB/month.  If you eliminate file-sharing and video, this is pretty much a guarantee.  And yes, the 5GB includes online media.


    2) Hogs watch videos.  On the other hand, people that use more than 5GB/month are almost certainly going over by watching videos.  Which also makes sense - after all, its tough to suck up data without streaming video!

    3) Metro areas are growing. The largest growth in subscriber base is going to be in Metro areas.  Given the already high density in these areas, and the crappy existing quality, it pretty much guarantees that we're going to see more wi-fi/femto-cell implementations a-la Republic wireless


    4) No Videos during dinner.  Mobile PCs (?laptops on the road?) are largely used for work, and file sharing, while people use phones for pretty much everything - and mostly videos.  Strangely enough, Mobile PCs are used during the day fairly steadily, while phones get used pretty much at any time, except during dinner.


    (Hat tip to GigaOm for the charts)

    Tellabs and the Smart Pipe (Game Theory edition)

    Tellabs has a long paper out ostensibly showing that the ISPs should move to a 'Smart Pipe' world, for the betterment of mankind.  I won't bore you with the details (if you feel like it, read the summary here), but in a nutshell, it says that Smart Pipes are needed to

    1. Ensure that data is transported efficiently so that capital and operating costs are minimised... 
    2. Improve user experience by matching the performance of the network to the nature of the application or service being used...
    3. Charge appropriately for use of the network...
    I'm not even going to bother getting into any of the above.
    Really.
    I mean, seriously.  Whats the point?  Its basically a flame-bait, just thrown out there

    However, I do want to use this as an opportunity to bring up a paper by Cheng, Bandyopadhyay, and Guo, talking about Net-Neutrality in terms of Game Theory*. It takes the idea of Network Optimization (or its Evil Twin, Network Throttling) back to basics, i.e., it asks the simple question Cui Bono? (For all you philistines out there, Who Benefits?).
    Consider your local ISP - Bob's Network Company.  Everybody pays pretty much the way they pay now.
    Bob comes to you, and says"Hey!  I'll prioritize your traffic, if you pay me an extra $100/month"
    If there are no bandwidth/latency issues, you're probably answer is "Meh".  After all, things are fine, why pay? (Remember, this is a rational decision on your part.
    A year from now, Bob has a lot more subscribers, bandwidth sucks sometimes, and Bob approaches you again.
    Your answer this time around is "Sure!  I Need This!".
    So far so good, right?
    The problem here is, what is Bob's Incentive?  His Motivation?
    He could take this money, and upgrade his network.  Except, the moment that  bandwidth is plentiful, nobody else (you included!) is going to pay him $100/month for prioritization!
    Ergo, Bob's Incentive, from a Game Theory perspective, is to keep the network always Just Bad Enough that people will pay for prioritization.
    The point being, the Market works, its just not necessarily working for you....


    * For a nice writeup about the article, check out Science Daily from back when, and a recent GigaOm article.





    Video Chat, and WebRTC

    You know, and I know, that Video Chat isn't taking the world over by storm, even though it is supposed to happen Any Day Now.  The lock-in factor is definitely a culprit - think of all the services out there, and how none of them inter-operate
    • Skype
    • GTalk
    • Hangout
    • Tango
    • FaceTime
    • TinyChat
    oh, the list goes on.  And none of them work with each other.  Mind you, this is even before we get into Device issues (quick!  Can your Polycom work with eyeBeam?), not to mention the horror that is 'Standards' (For some fun bed-time reading, check this out!).

    Enter WebRTC.  Its a simple standard that enables Javascript based RTC.  The API document may look complicated, but it really is not.  Much more importantly, it is simple, open, and did I mention, simple*?  
    Serge LaChapelle of Marratech fame is now over at Google moving 'em into the WebRTC world **.  Expect to see the fruits of this work to show up in G+ Hangout, not to mention Gmail, Gtalk, etc. pretty darn soon.  

    Mind you, this will almost certainly resolve the inter-operability issue (again, w/ the possible exception of Skype which has always marched to the beat of its own drum :-) ), and it will amost certainly not resolve the Why would I want to see you when I talk to you issue.  But that, as they say, is another story...

    * Ok, maybe not so simple, but trust me, given the horror of trying to do any kind of video-chat service, it is about as easy as falling off a log.

    ** A little video from Serge talking about this

    So *thats* what happened to Yap!

    Yap was one of the big names in Voice Recognition - powering a lot of the voicemail transcription market - till they suddenly went silent on 10/20/2011 (Silent Thursday?).  The rumors were rife - everything ranging from software breakdown, to Filipino strike, to (the most baffling one) Sneaky Verizon Ploy.  It turns out the truth was a lot less strange - Amazon bought 'em out to (most likely) embed in the Fire.  Its certainly an anti-Siri ploy, but I guess we'll find out about the Fire rumors Any Day Now...

    BigData - Social Analytics

    Another day, another BigData investment - this time with Kontagent getting a bunch-a money from Battery, Maverick, and Altos.  The particular angle here is that Kontagent focuses on Social Analytics, in particular, gaming.  Their secret sauce is in helping 'social gaming' folks figure out how to get their users more engrossed.  I assume this is essentially BigData merged with a healthy dose of Behavioral Psychology (shades of Kahneman & Tversky).

    Rands on Triage

    Hat tip - Rands in Repose

    Wednesday, November 9, 2011

    The Economist loves, *loves* Berlusconi...

    Really.
    Don't believe me?
    Check it out...


    Take *that* Efficient Markets Hypothesis!

    From the NY Times, the story of how a financial advisor lost his house.
    The dumb part -->
    I should have known better. No matter how well things are going, borrowing 100 percent of the purchase price of a home is not a good idea. I shouldn’t have relied on someone else to make that calculation, let alone the guy who was making money putting me in the loan. I was a financial adviser, and I never sat down to figure out what it would take to make this work. I just wanted to believe him. And it was so easy to believe he had been right, at least at first. We loved living there. The children went to an awesome public school, and we made some great friends. I could ride my bike to Red Rocks, the wilderness area outside of town. And for a time, the real estate market erased any doubt I may have had. It just kept going up.
    The smart part

    At first, I dismissed the idea of a short sale. Late that summer, I sat down with a really close friend in Las Vegas, someone I looked up to. He cut to the heart of the matter right away: Why, he wanted to know, were we still making the payments?
    Because I have a moral obligation, I said. You pay your debts.
    He proceeded to explain that I didn’t have a moral obligation to the bank. I had a moral obligation to my family. I had a contractual obligation to the bank, along with a clear moral obligation to be honest in my dealings. 
    Bloody accurate, I must say...

    Wikets - Product Recommendations

    Sigh.  I actually see this as potentially being useful.  Which is both horrific, and tragic, given my built-in luddite-loathing for our consumer-culture ...
    Anyhow, without further ado, Wikets...
    The web is chock full of reviews, but I can see Wikets taking off because it makes the process remarkably simple, more intimately social, and actually rewarding. Users can compile lists of recommended products to share with their friends and favorite other users’ recommendations by either sharing them as a “Re-Rec” or marking them on a “Wishlist.” They can also use Wikets to ask for specific recommendations. While recommendations can only be created within the iPhone app, they can be shared over Facebook and other existing social networks and viewed by others on the web. When people buy products a user has recommended on Wikets, the user is rewarded with “points” that can eventually be converted into gift cards. (more...)

    Everything Old is New Again

    BigData used to be about scientists moving tons of data around. 
    Guess what?  Its the same thing all around again!  And its all new!  W00t!
    Ok, to be fair, its about the need to move all this data around, instead of just processing it in situ.
    He ascribes this massive amount of data to the emergence of cheap compute, better imaging and more information, and calls it a new way of doing science. “In every area of science we are generating a petabyte of data, and unless we have the equivalent of the 21st-century microscope, with faster networks and the corresponding computing, we are stuck,” Szalay said.

    Telcos and Bandwidth Charges

    Yet another article about how wireless providers need to get more money from consumers.  Money quote
    The end game for operators is to figure out how to get people to use less data or pay more for the privilege, while also steering users to content that makes users happy without straining the network. So maybe they offer unlimited Facebook and charge more for Netflix. After all, the last thing they want to do is to scare people from signing up for data plans for fear of expensive overage fees....The point here is that as operators try to change the way they charge, because the current models aren’t economically sustainable for them, consumers have to believe the deal is reasonable. That’s why I’ve been a proponent of measures such as data happy hours or other plans that incent users to use the network during non-peak times. Doing so helps manage congestion, and can still generate revenue for carriers.
    The entire article sounds like the discussion we had (and are still having!) with banks about debit-card fees.  Just because you (the company) have made a rentiers fortune in the past charging for stuff that doesn't cost you anything does not imply any sort of implied right to continue extracting said rents.  Ethereal Mind has it right - either get used to less 'free money', or build out additional bandwidth.  Hey, it seems to work out just fine in the rest of the world...

    Cassandra and de-normalization

    An article on working with graph-like data on Cassandra, which basically translates to make it all a really long row with timestamps for column-ids.
    The cost of denormalisation is duplication of data, and in this case we have duplicated each message by copying it to the timeline of every user who follows the author of that message. This means we incur a write cost when broadcasting a message, since we must insert the same message multiple times. Luckily for us, Cassandra is optimised for high write throughput (writes perform only sequential I/O) and it is this performance profile of Cassandra that allows us to trade some write speed for increased read throughput.
    It works (and how!), but its this kind of bigdata solution that gives me the jeebies...