Author: barce

  • Blogging Every Day For Six Months Starts Now

    I’ve decided to blog every day for 6 months starting now to see if it will get me 10,000 uniques a month. That’s blogging everyday until February 18, 2011.

    In this blog post, I just want to share a few things that make blogging different from writing.

    I want to share these things because of this piece of wisdom: Blogging is not writing.

    • Each sentence must have a popular Google Ad Word, like blogging or Google. 🙂
    • Keep things between 50 and 150 key words.
    • Engage folks via social media.

    Writing is really about engaging people, really moving them. Blogging is a subset of writing that has to take into account search engines and making words and the code that underlies them very friendly to the search engines.

  • Internet Speeds in Santa Monica and Westwood Suck

    Internet speeds in Santa Monica and Westwood suck when you compare them to Kyrgyztan (14 mbp/s) or Latvia (24 mbp/s). Never mind that these areas in LA have economies that dwarf these two places.

    The average speed in these areas seems to be at about 1.5 mbp/s in Westwood and 10 mbp/s in Santa Monica.

    1.5 mbp/s is simply not acceptable in Westwood. 10 mbp/s is simply too slow in a town that prides itself on being the movie capitol of the world, especially for someone like James Cameron who has his offices there. Using a messenger to ship a DVD is still faster than using the bandwidth available there.

    If West LA is to have a world class tech scene, we must have more bandwidth. More bandwidth is what allows a place like Silicon Valley to develop technologies first. A good example is AJAX, which sucks when most folks are on a modem. However, leveraging faster than average bandwidth allowed folks in Silicon Valley to learn this crucial technology / technique first.

  • 4 Ways to Avoid Foursquare Fatigue

    As a social media expert, I’ve noticed a severe decline in Foursquare check-ins by innovators and early adopters.

    This phenomena, which I’m calling Foursquare Fatigue, can be avoided by following these 4 tips.

    1. Use two phones on different carriers. We all know that AT&T’s 3G blows on the iPhone, so carry a spare phone like a Blackberry running Verizon. That way you get all your bases covered. Can’t check-in with the iPhone? Check-in with the Blackberry. Now you’re stylin’.

    2. You don’t have to check in all the time. Ya, getting the mayorship for Starbucks is huge and forces you to check-in 5 times a day, but after that 5th cup o’joe there’s a diminishing point of return. Why not save money, not check-in. In fact, don’t go to Starbucks.

    3. Check-in with Yelp. Ya, I know. It’s pretty pointless because you don’t get the Foursquare deals, but there’s been less of a reliability issue with Yelp where check-ins are currently faster and less buggy than Foursquare.

    4. I have Stalin, who said, “No person; no problem,” to thank for this one. The discovery I made is this: No Foursquare, no foursquare fatigue.” Logic is magic.

  • How To Pitch At Capitalize

    Last night I got to check out an event in Cambridge called Capitalize hosted at the SCVNGR offices. The audience gets to see a company pitch to VCs. Last night’s company, Peekaboo Mobile, pitched to two investors from Dart. They were asking for $550,000 to $770,000.

    Basically, the investors wanted to see 3 things:

    1. a great solution to a small problem involving a very specific customer, e.g. an app focused on moms that tells moms where stroller friendly businesses are,
    2. a message that shows how all parties involved are going to make lots of money, e.g. tossing the phrase, “We are going to make you lots of money and this is how,” works wonders, and
    3. a solid set of numbers showing cost to acquire customer and ROI.

    Peekaboo didn’t have these 3 things so they weren’t written a check outright, but they’re still making money. Also it takes a lot of courage to stand in front of a crowd and get scrutinized by VC.

  • How to Get a Good Tech Job

    I love this quote from Mark Suster:

    “Finding the best jobs takes a lot of commitment to taking many different networking meetings with executives, recruiters, entrepreneurs, VC’s, investment bankers, etc. The best jobs (as you know) are found through personal connections. The best jobs are the ones that have not already been put on a job board. The best jobs are the ones that certainly haven’t gone out to an executive recruiter. The reason these are the “best” jobs for you is that once it goes to an executive recruiter there will be a stack of 100 prospective recruits, 20 amazingly qualified resumes that will have phone or in-person interviews with the recruiter of which the company will meet 5-6. So unless your last job is a mirror image of your next then good luck with those odds.”

    Okay, that’s pretty general. What’s a specific example of getting the best job?

    A good example is starting out small. Get a reputation for doing quick tech jobs on weekends. This can be anywhere from:

    • writing a basic facebook app, or wordpress plug-in
    • fixing bugs that you are sure you can fix
    • addressing server issues that you are sure you can fix

    After a string a little successes like these, you can start taking on bigger jobs.

    Recruiters at this point will start knocking at your door. It is not worth it. In my experience, they will fill a position that hasn’t been filled in a long time and with good reason. Also, the fact that your prospective place of work is going through a recruiting agency means 3 things: they are ramping up with hiring and are being quick to hire and slow to fire. There are good reasons not to do this. Another reason is that, they simply cannot recognize talent. If they cannot recognize talent, then they cannot understand you. Misunderstanding is the source of a lot of work grief.

    The 3rd thing that it means is that you are walking into a den of Vampires that will take all your soul and talent and feed off of it. You can find the worst Vampires in large corporations and they are usually project managers, or middle management. It is one thing to let a contractor just get the job done, and another thing to take what they do and keep the glory for yourself. These glory hounds, these vampires should be avoided at all costs. They will burn you out.

    At this point, be sure to be paying your taxes. You should have a nice little consultancy going. Be disciplined. Don’t be afraid to say, “No.” If something doesn’t feel right, e.g. you sign a contract for PHP work but are doing .NET, then drop them. Yes, drop them. Any person with business sense will understand bait and switch is not the way to think long term.

    Two choices will present themselves:

    1. Continue freelancing, or
    2. Focus on one company, e.g. being a co-founder or lead

    The first choice makes sense if you are shooting for the 4 hour work week. The second choice makes sense if you are hoping to turn a company in a few years.

    If you are going for the 2nd choice, then a few things are key:

    • You need to court this company if they’re a hot commodity. This means hanging out with the devs, the marketers, top-level folks. If you’re really good, they’ll just ask you to go for the interview process. But the best jobs are never listed…. therefore…
    • Use the process to understand the company’s true business needs. From those needs list the ways that you can contribute to the company.
    • From that list, you’ve created your position.

    The best job ever is something you create. It is not some laundry list of things someone you don’t know has given you. Your best job ever is your destiny, your meaning of life. Do not settle for less.

  • Installing Snowflake

    I forked the code that Ryan King wrote for Twitter called snowflake.

    Snowflake works by guaranteeing a unique user id based on a few simple guarantees: time, machine id and sequence number.

    My fork has a fix for the ruby client allowing you to query multiple servers.

    Here are the prerequisites:

    The install process is fairly straightforward:

    1. Clone the repo: git clone http://github.com/twitter/snowflake.git
    2. Edit project/build.properties to use sbt.version=0.7.4
    3. sbt update
    4. sbt compile
    5. sbt test
    6. Copy one of the configs into /etc: cp config/development2.conf /etc/unknown.conf Or you can also just use ‘sbt run -f config/development2.conf’.
    7. Edit the conf and add a unique worker ID: worker_id = 1
    8. In the conf: skip_sanity_checks = false
    9. sbt run

    When you run the client like so:

    RUBYLIB=./target/gen-rb ./src/scripts/client_test.rb 1 “192.168.0.103:7610,192.168.0.109:7610” test-on

    You should get something like this:

    557299556602089472 test-on 1
  • Twitter: Thoughts That Are Hard to Fit Into 140 Characters

    What are the limits of expressing thoughts in Twitter?

    Here’s a powerful but inefficient (when run) thought that can be expressed on Twitter, a quick sort in Erlang in 126 characters.

    qsort([]) -> [];
    qsort([Pivot|T]) ->
    qsort([X || X <- T, X < Pivot]) ++ [Pivot] ++ qsort([X || X <- T, X >= Pivot]).

    A lot of Perl one-liners can fit into a tweet – powerful and useful ones.

    Haikus can be expressed in a tweet.

    The answer to the question, “What form of body language do most FBI interrogators consider to be the most telling?” can be answered in a tweet.

    A marriage proposal can be answered in a tweet.

    You can propose the concept of a hash tag in a tweet:

    hashtag proposal

    However, there are many thoughts that seem to be difficult to fit into a tweet:

    • The Pythagorean Theorem and one of its many proofs
    • Anselm’s Ontological Proof for God’s Existence
    • Merge Sort in Ruby
    • Merge Sort in PHP
    • Why you should or shouldn’t outsource
    • What qualities make a great tech hire
    • Well-thought out political proofs
    • How to subtly tell someone something in an indirect way with the only others knowing being those in the know
    • A legally-binding, work contract – It would be amazing if you could!
    • The mechanism for how DNA works

    Twitter encourages the laconic expression of thought which means plenty of affirmations, aphorisms, insults, congratulations, and reminders that can display any combination of sharp wit, pointed humor, and succinctness of expression. The mot juste becomes very important with the constraint of 140 characters.

  • URL Shorteners: tinyurl.com bit.ly and seductive.me

    There’s a pretty useful spreadsheet comparing different URL shorteners here:
    http://spreadsheets.google.com/pub?key=pApF4slh39ZkqUOoZQSo8bg .

    Tinyurl just really shortens your URL and doesn’t provide any other data. A great feature is the preview option that allows you to preview a link so you don’t get Rick-rolled.

    Bit.ly is my favorite service. The features I like are

    • its API,
    • its analytics, and
    • its Pro feature which allows you to make a branded URL shortener

    I wanted to figure out just how much effort it was to code a URL so I coded up my own MVC based on Rasmus’ article on making one in PHP, and added URL shortening code to it. You can get the URL shortener I wrote on Github.

    Here are a few things that I noticed once I put this code on Seductive.me:

    1. Somebody found a bad, infinite loop that I accidentally put into the code
    2. URL Shortener services that offer analytics have to sort out “bots” from real user clicks, and I’m not sure they are correct 100% of the time
    3. Using CRC32 as my hashing mechanism for shortening URLs will cause a link collision after 65,536 URLs shortened.
    4. Seductive.me is a lame url for a URL shortener which is why I got 3vi.be the next day.
  • Frequent Insights; Less Focus

    The problem with professionalism is that there’s only one way to get a job done right. For example, if you’re a DBA working from a relational background there’s no way that you could ever professionally suggest a NoSQL solution to a problem.

    The great thing about professionalism is that you can get paid for your detachment to reality and by following “The Process.” The process is something that’s hard to argue with, has authority, and won’t ruin your career. But the process can ruin a business.

    A good example of the process ruining a business is the current shakedown in advertising.

    There used to be a process where creative output was cheap and what was treasured was the global reach of a global agency.

    By following this process of client close, account executive direction and creative output through various channels, advertising turned a blind eye to the rise of the Internet.

    Global companies no longer need the global reach of an agency. Global companies need social media experts who can leverage nodes of attention in form of Google, Facebook and Twitter.

    Professionally, a creative steeped in radio or print cannot suggest that the solution is something Internet-based. By the same logic a social media expert, enjoying the zenith of the social media frenzy, cannot suggest professionally a non-social solution.

    How do we escape such professional straight jackets?

    The solution seems to be that being less focused leads to more insights:

    “People who seem to have frequent insights do not do so by focusing harder on the problem, instead they have learned to switch off their thinking – to access a quieter mind on demand. Having insights involves hearing subtle signals and allowing loose connections to be made by quieting the mind – letting the brain idle with minimal electrical activity.” — David Rock

  • Streamys Tech Fail

    Last night’s streamys from a tech perspective had 2 big pieces of fail which some might’ve overlooked because of the creativity of the presenters and comic skills of Paul Scheer.

    1. The free, event based iPhone app never updated with the latest winners.

    2. The video player that showed the nominees broke down half-way through. The work around was to tell jokes and then eventually bring up the nominees on stage.

    I really wish – and I’m not alone in this -that they kept the bar open for the whole show.

    These fails show where corners are cut which is understandable in a fledgling industry where creators of content are often the monied producers, too.