iseff, By Ian Sefferman

TechCrunch / PopSugar Party

2008-02-20T20:30:00.004-08:00

So, this could be a decision I regret, but this morning I made some plans to head to LA in April. Coincidentally, this evening TechCrunch announced their latest party, Geek Goes Chic, in Hollywood while I'm there. Since that doesn't happen often, I decided to pick up a ticket.

This could turn out to be the douchiest thing I've done in a long time, or it could actually turn out to be fairly cool. I'm hoping since it's not in Silicon Valley that it'll be a little better.

If you're going to be there, and you're worried about the same thing as I am, shoot me off an email -- iseff@iseff.com -- and we'll hang out.

Google, Facebook, Microsoft, and OpenSocial

2007-11-02T07:24:00.000-07:00

I'll be up front: I don't get social networking. That said, I like watching the progress of what I call the web-ization of social networking (things want to be free, things want to be open, etc).

But perhaps the most interesting thing I've seen so far is this quote from Brandee Barker of Facebook:

Despite reports, Facebook has still not been briefed on OpenSocial.

You mean to tell me that after weeks of negotiations on a hundreds of millions of dollar deal, Google never told Facebook about its plans?

My guesses as to what happened:

Google had been planning this as a contingency to losing the Facebook deal. If the FB deal went through, they may not have released this so soon.
Or, perhaps, just perhaps, Google took the more-evil route. They knew Microsoft had more to lose (they really need to spur growth to their advertising business and show a big win), so they intentionally bid up the valuation on FB with no interest in actually winning, let MSFT close at a more expensive price, and then steal their thunder just a week later.

Just thinking out loud here..

Facebook IPO: 12 Months

2007-10-17T23:01:00.000-07:00

A long, long time ago (March, 2004), the tech-world was abuzz about a startup in breakout mode named Google. Rumors were floating about if and when the company would go public when an SEC loophole was found. It was only a short time later (April, 2004) that Google did file for an IPO which set the tech world on fire.

Today the geeksters are all abuzz about another startup in breakout mode, named Facebook. The same hype circle is flowing and people are now speculating on its IPO. Let me put that to rest and predict that it will happen within 12 months.

Why 12 months? Let's revisit Google. A primary motivator for Google going public when it did was the SEC loophole. The rule is going to hit Facebook in the same way:

Under securities law, U.S. companies that have $10 million of total assets and more than 500 stockholders must file a Form 10 with the Securities and Exchange Commission.

Tonight at the Web 2.0 Summit, Mark Zuckerberg stated, according to SAI:

Facebook has 300 employees (all with equity) and will have 700 in a year.

One can only imagine that at least half of the next 400 employees will also have equity. The $10 million in assets is practically a given.

Given this, it seems that Facebook will file by this time next year. At the latest, it will file April, 2009 (assuming no acquisitions) because the rule states that if your fiscal year ends December 31 and you are above the thresholds, you must file by April 30.

Stop Backing Up!

2007-09-25T10:01:00.000-07:00

This is a post I've been wanting to write for a while, but the Mozy deal just brought it back to my attention. I'm surprised that all the talk of "the web as the platform" hasn't led to this already.

Mozy, Carbonite, and others are all great for backing up your hard disk. Ultimately, though, backing up is dead. It's no secret by now that we're moving our lives online. In a few short years we'll be producing all our documents, spreadsheets, etc on the web. We'll be happily collaborating and sharing on the web.

Knowing this, I can't understand why all these online storage companies are so focused on providing consumers backup opportunities. There's no point. If all our documents are online and stored by online file systems, they'll be backed up automatically. And available. All over the world, from any net-connected computer.

There's a level of trust which needs to be obtained of course. But, I generally trust Flickr, Yahoo, Google, Openomy, etc. I know that once my files are stored, they're safe. They'll be replicated across many hard drives. Many machines. Many data centers throughout the world.

Once we've killed backing up in favor for keeping our files in the cloud, the next step will be keeping those files out of silos. But that's for another post...

Feature Or Bug

2007-08-26T21:08:00.000-07:00

Back in May, I bought a fun domain name: featureorbug.com.

After debating for a few days, I decided I was going to make it a tumblelog for all things I enjoy-- technology, code, some business, and even some finance. I mostly post quotes and links of things I find around the web.

I've told a few people about this site already, so there's at least a couple readers, but I think it's time to open it up to the public.

So, if you're interested in programming, technology, business, etc, definitely subscribe to the RSS feed. There's a lot of gems in there with many more to come.

Feature Or Bug

P.S. Feature Or Bug is made possible by the great tumblr software.

On Mahalo (or, Where's the sweet spot for the fat head without the long tail?)

2007-08-18T20:55:00.000-07:00

Recently, I've been thinking a lot about Mahalo.

I think Mahalo has great results. Especially for non-techies. I asked my friend yesterday whether she'd rather look at the results for Mercedes Benz S-Class at Mahalo or at Google. She instantly wanted Mahalo. Me too.

For the fat head of results that Calacanis is aiming for, he's going to get great results. There's no question a human touch will help.

The issue is the average user won't know the distinction between the fat head and long tail. Moreover, I don't think they'll care enough to use two search engines for different purposes. Google provides alright results for my example above, it's just not as pretty and easy to understand as Mahalo. But Mahalo doesn't provide any useful results for Openomy, while Google does. Sure, Mahalo is trying to use Google here, but their "More results from our friends at Google" thing just doesn't cut it from the user experience (performance, grouped results, etc).

I'm trying hard to find out where the sweet spot is for Mahalo, then. Jason wants to ensure his service is used by the mainstream web users, but are those users really going to be able to know when to which search engine? Are they even going to want to use two search engines when one can provide average results for all terms, but the other doesn't even provide any for half of what they search for? Even the average user dips into the long tail quite often.

I think Mahalo can do well even if they don't figure this out. Search is such a large market that even the smallest success is pretty good money. But it won't be a Google killer, that's for sure. I don't know what the answer to this is yet, but it seems likely that an innovative solution in their upcoming platform could help them bridge the gap and hit their sweet spot.

Facebook RSS scares me

2007-08-14T16:04:00.000-07:00

Facebook RSS feeds really scare me!

Hey Dale Lane, it's really great to know about your friends info.

At the very least, couldn't Facebook use a robots.txt file here? At least make someone work for it a little bit.

This is fantastic news for Facebook developers...

2007-07-23T23:08:00.001-07:00

...even if I think the stats may be a bit dubious and not-all-that-meaningful.

Surprise: Facebook apps may help grow home sites:

Slide, Rockyou and HotorNot, three companies with the largest number of users on Facebook, are showing continued traffic growth on their own sites.

Yahoo! leaks

2007-05-20T01:06:00.000-07:00

Have you noticed that Yahoo! seems to have acquisition and other product rumors leaked more often (and with more detail) than other companies such as Google? Note the latest talks about Bebo.

Is this true or revisionist history on my part? If so, why is that?

New tumblelog: http://iseff.tumblr.com

2007-04-22T16:19:00.000-07:00

My presence on the web is ever-growing, so I wanted to make it easier for people to ~~stalk~~ follow me. Tumblelogs seem like a good way of doing that and tumblr is a nice, clean, free service.

I've put together a tumblelog available at iseff.tumblr.com which puts together all the feeds I could think of about myself:

One feature request I have for tumblr is to have the ability for setting where I want to auto-discovered feed to point to. That is, I'd like to wrap the iseff.tumblr.com RSS feed with FeedBurner and tell tumblr that so that I can accurately track subscribers.

Target 2045 Retirement

2007-04-16T22:08:00.000-07:00

As part of my 401(k), I'm invested in a "target retirement" fund. The general idea is rather than worrying about what to be invested in, just choose when you'd like to retire and the fund management takes care of the rest. Sounds nice, right?

Well, if 2045 is 38 years away, why are they investing in:

Vanguard Total Stock Market Index Fund (71.9%)
Vanguard European Stock Index Fund (10.4%)
Vanguard Total Bond Market Index Fund (10.2%)
Vanguard Pacific Stock Index Fund (4.8%)
Vanguard Emerging markets Stock Index Fund (2.7%)

Index funds!? >70% to a stock market index fund?! It seems to me that with 38 years to make money, I'd like to way outperform the market. Here's more or less what I would do if I were managing a Target 2045 Retirement fund:

Stem-cell research start-ups (20%)
Software start-ups (20%)
Nanotechnology start-ups (20%)
Cancer research start-ups (20%)
International small-cap growth fund (20%)

If I lose money in the short-term on any of these, no big deal; I still have 30+ years to make it back. But the upside potential is huge.

And here's what I'd do if I were managing a Target 2012 Retirement fund:

Vanguard Total Stock Market Index Fund (71.9%)
Vanguard European Stock Index Fund (10.4%)
Vanguard Total Bond Market Index Fund (10.2%)
Vanguard Pacific Stock Index Fund (4.8%)
Vanguard Emerging markets Stock Index Fund (2.7%)

Look familiar? With this scheme, I'm highly unlikely to lose a lot of money in the next few years, but I'm also highly unlikely to kill the market.

What am I missing?

Is Haskell going to be the next 'it' language?

2007-03-31T02:12:00.000-07:00

Is Haskell going to supplant Ruby as the Next Big Language in the next couple of years? I have a gut feeling it just might happen. Things I've noticed:

I've randomly found two other developers wanting to learn Haskell in the past week.
New mailing lists are being created at haskell.org, including web-devel@haskell.org.
Smart people, like Joshua Schachter, are talking about Haskell.

I just get the feeling there's something rumbling around. I'm very interested to see what happens and what it takes to make a language popular.

Learning is the most important thing

2007-03-27T20:20:00.000-07:00

Maurice wrote a great introductory post today about us beginning to write the next version of the Openomy APIs in Haskell. In it he mentions that we're keeping track of how many times we say a variant of "we're fucked." We've said it many, many times so far.

One could easily say that we're being irrational for developing a web service which will see millions of hits per week in an "unproven" (whatever that means) framework. Or for using a language we're just learning. Whatever.

To us, it's all about the We're Fucked Count. The one thing we always strive for while developing Openomy is to keep learning. It's just not fun if we're not learning. And if it's not fun, we're not going to produce anything great.

The question then becomes, "How do we know when we're learning?" We like data a lot; making subjective statements isn't something I like to do. The We're Fucked Count makes sure we don't have to. If we can keep the WFC high, we can be relatively sure that we're continuously learning.

I think learning should be priority #1 at any organization. The WFC helps us to make sure we're achieving that goal. Follow Maurice's blog for the continuing series on building a Haskell web service, which I'm sure will provide us lots of excitement.

I love my Pearl more everyday..

2007-03-12T22:34:00.000-07:00

I've been a BlackBerry user for a few years now, but when I saw the Pearl announced a few months ago, it was love at first sight. I preordered it from T-Mobile.com only to find out they wouldn't be sending it until a few days after the release day. Like any good geek, I couldn't wait that long so instead I went and picked up a second one on the release day at the store. I've been in love ever since (and am a huge evangelist, persuading at least 3 others to buy one as well).

The thing I love most about the Pearl is that I find new uses for it everyday. For example:

The ability to browse web pages inside the Amazon firewall.
The ability to SSH into any box in the world, inside or outside of the Amazon firewall. Seriously, SSH! I can admin anything I need from my cell phone! I mean, really.
Google Maps Mobile. Has helped me find my way many, many times. And adding traffic to it only made it better.
Yahoo! Go. This is a new one, but it's totally great. I'm mainly using it for uploading cameraphone images directly to Flickr.
The speed at which email is pushed to the device. It boggles my mind that it's faster to receive email on my BlackBerry than it is to wait for it to show up in Outlook, but it's true.

And, today I found a new feature that kicks ass. In less than 10 minutes -- following the directions posted here -- I set up my Pearl to act as a modem for my PowerBook. If that's not cool enough, it does it through Bluetooth, which gives me all the more geek cred. I've heard rumors of people seeing ~150kbps with this, but I'm seeing closer to ~50-60kbps speeds, which is fine for me.

As an aside: I've been preaching for a while that rather than make online applications work offline as well, just invest in making net access available everywhere. Literally everywhere. This is a step in the right direction for me.

In conclusion, I <3 U, BlackBerry. Don't ever leave me.

Capistrano and Java

2007-03-07T06:29:00.000-08:00

I'm sure a lot of you Rails developers who read my blog are well aware of Capistrano. For the non-Rails developers, here's a two sentence intro: Capistrano is automated deployment. Fill in some configuration about which boxes perform which tasks (things like which are the DB servers, which are the web servers, which are the application servers, etc) and then simply type "cap deploy" from a command-line every time you want to release a new version. Needless to say, it's very slick and almost 100% necessary for any application that runs on more than 2 machines.

Capistrano is definitely Rails tailored. It makes a lot of assumptions (in true Rails philosophy fashion) that make it dead simple to use for your new Rails app. However, there's nothing really constricting about Cap itself that forces it to be Rails only. Recently, Maurice and I did our first Java project for Openomy. We knew we'd be deploying to more than one box, so we decided we'd try out Cap for deploying a Java service to multiple boxes. I figured I'd write up some of what we learned and what we'd like to change in the future.

(From here on out, I assume basic working knowledge of Cap.)

How It's Implemented

1. First, we created some new tasks (overriding some of the defaults):


set :application, "file_service"
set :is_test, ""

task :build_and_run do
  run <<-CMD
    cd #{current_path} &&                                                       
    env JAVA_HOME=#{java_home} #{ant} dist &&                                   
    scripts/start-fileservice #{is_test}                                        
  CMD
end

task :deploy do
  update
  build_and_run
end

task :deploy_test do
  set :application, "file_service_test"
  set :is_test, "test"

  deploy
end

2. Next, we wrote the scripts/start-fileservice script which starts up our Jetty housed service. There are about 50 lines in this Perl script but I'm not going to go into them because I'm sure you can figure out how to start up Jetty in an automated fashion. Basically, we keep a pidfile and use the standard jetty-start.jar to start up and restart.

3. Note that the "#{ant} dist" line runs ant and builds our dist target, which creates our Jar file.

4. Now, if we want to start up our test service, we simply run "cap deploy_test". Otherwise, if we want to start up our production service, we run "cap deploy". That's basically all there is to it.

Things We Assumed

1. We assumed that Perl, Java, and Ant were all installed on each of the application boxes.

2. Further, we assumed that they existed in exactly the same location on each box.

3. We similarly had to make the same assumptions that Cap makes (that subversion exists on each machine, for example).

Things I'd Like To Change

1. Instead of having to pre-install each box with certain software, I'd really like to be able to just acquire a new box with just an OS (and subversion) on it and run "cap deploy" and have everything automagically work. To do that would require me to have JDK and Ant and Perl all within my subversion repository previously built and then use those to run my scripts and services.

2. More realistically, I'd rather not have a dependency on Ant. The primary reason we use Ant is to define our classpath correctly for us. It'd take a lot more engineering to have cap learn about our classpath and then set that correctly before calling our startup script. Add on to that the fact that we really only need to build this once and deploy it everywhere (rather than building everywhere) and Ant seems a little abused in this regard.

Note that neither #1 nor #2 are particularly important at a scale of <25 boxes, in my opinion. Once above that range, though, they start to become very important.

3. As you can see from our tasks, our rollback procedures could use some work, though the standard rollback tasks built in by cap are moderately sufficient for our uses.

4. Cap really needs a web based interface. I'll probably work on this at some point though I've heard rumors that someone else has already started this and it looks good (yay!).

February 12x30 Report Card: B

2007-03-02T08:53:00.000-08:00

I figured I should start recapping my 12x30's after each month to look back and see how I did.

The goal for February was to blog every day of the month. I definitely did not achieve that. However, I still feel like I accomplished a lot this month in terms of blogging more often and being more comfortable blogging. (Also: I was out of town for 7 days this month which made blogging much more difficult.)

Overall, I give myself a B for my efforts.

Here are some stats:

Of the 28 days in February, I blogged 13 of those days. Just under 50%.
Take away the 7 days I was out of town, and I blogged over 60% of the days in the month.
Of those 13 posts, 11 of them were about technology and the other 2 were about the 12x30 program.
Previous to February, I had a total of 90 posts on this blog which began in June, 2005. So, thanks to my 12x30, I added approximately 14% more new posts in just one month.
The two most searched for stories (of these 13) were My experience with Prosper and Salesgenie.com "deliver a targeted message".

Now, to start the March 12x30. I think I'm going to switch up what I originally wrote I was going to do for March (read every day) and instead do my original plan for April: work on Openomy everyday. We have a lot of great ideas that are all about to come out very shortly so this'll be a great time to work on it daily!

Slashdotted

2007-02-28T08:17:00.000-08:00

I'm now extremely happy we made a recent code push, 'cause it looks like we're going to need it today.

Online Storage 2.0.

Google starts awesome new college course; cuts 20% time in half

2007-02-20T12:25:00.000-08:00

Okay, this is really cool. According to the Seattle P-I, Google is creating a college course at UW for, essentially, teaching students how to create massively scalable and reliable systems. I would've loved this during college (actually, I essentially did this course as an independent study type class with the awesome Svetlozar Nestorov -- also founder of Mobissimo and Ph.D. candidate with the Google Guys).

But, almost as noteworthy, the P-I writes:

Bisciglia would design the course under Google's "10 percent" program, which allows employees to use 10 percent of their work time to dream up big ideas.

Uh oh, maybe Google is finding that 20% projects take up too much time, and halved that? Oh noes! Is the free food next?

Why I'm Glad I'm Not Netflix

2007-02-15T18:29:00.000-08:00

I've had this conversation at least three times in the last couple weeks with various people and Kathy Sierra (who else?) hit the nail on the head with the post The real secret to a successful blog/book/business....

As I've told a few people, I wouldn't want to be in Netflix's shoes because for them to be wildly profitable, they have to treat their best users the worst (throttle those who rent the most). That's so backwards.

Or, as Kathy says:

The secret [to success] is simply this: you have a much better chance for success when your business model makes what's good for the users match what's good for the business, and vice-versa.

Follow Donald Trump around the country (but not the Google guys!)

2007-02-14T18:31:00.000-08:00

I'm probably way behind the times on this one, but today I found out about the site FlightAware.com.

The main jist of FlightAware is to provide you with details about all the planes in the sky right now and throughout the past. You can order by fleet (Delta, American, NetJets, FedEx, etc), by a specific plane, by airport, etc. This is really, really fun stuff.

It gets even more fun when you start combining it with Google and the public FAA plane registry.

Here's an example: Google for donald trump plane and the first result will give you the tail # (VP-BDJ) of the Trump 727.

Now, go back to FlightAware and search for that, and you'll see the flight history of Donald's plane. Looks like Donald likes to go between New York (home), Las Vegas (where he's building), Los Angeles (The Apprentice), Atlantic City (Trump Taj Mahal), and Florida (more property). We can set up alerts and follow him around now! (BTW, I wonder what he was doing in Portland recently?)

Let's try another example: The WSJ said that the Google's guys created a holding company for their new 767 named Blue City. Searching the FAA records, we see the tail # for this plane is N2767. Now we'll check the FlightAware page. This time we see:

BLUE CITY HOLDINGS LLC (OKLAHOMA CITY OK)
This flight is not available for tracking per request from the owner/operator

Apparently the Google guys don't want us sniffing around their travels. Hmm. Shady? Maybe...

Other fun examples: DaimlerChrysler folks sure like to travel between Detroit and Stuttgart. Surprise -- who would've guessed that Eagle, CO (by Vail) would command so many private flights?

Can you find anymore top execs planes? Or maybe just tell me who the clever person is who named the holding company for his/her plane Acme Rocket Sleds, Inc and likes to fly throughout the west coast?

Two More 12x30 Ideas

2007-02-11T12:02:00.000-08:00

(Note: I might write a second post later tonight, since this is a cop-out post, but I actually want to do this for my own memory.)

I've come up with two more ideas for my 12x30 project. Along with these goals I'll now be through May.

Openomy (April). No, Openomy is not dead. Not even close. Unfortunately, we just don't have much time to work on it these days. But, I believe if I make it a 12x30 and try to work 1.5hrs/day or launch a feature once/week, we'll see some drastic improvements (also, stay tuned, as we'll be launching a HUGE new change shortly).
Learn A New Programming Language (May). I'd like to spend some time learning a more "wacky" languages every once in a while, so I'll spend a month diving into one or more of these languages.

Amazon + Half.com Arbitrage

2007-02-10T12:18:00.000-08:00

This week I was introduced to the genius world of Half.com + Amazon arbitrage.

What's that, you ask? Let me explain.

Apparently there are a few people who have written some programs which look at Half.com (through the eBay APIs) for products at cheap prices, then automatically post those products on Amazon's Marketplace (through Amazon's APIs) for a markup.

Once the product sells on Amazon's marketplace and the developer receives notification, they automatically buy the product from the Half.com seller, telling them to ship it directly to the address of the Amazon buyer.

Of course, they also watch the Half.com listings at all times to ensure they aren't bought and if they are, they cancel their Amazon listing immediately.

This is so genius. You can easily take 30% premiums with virtually zero costs day-in and day-out. Once the program is written, you literally don't have to do any work. Allegedly, one developer is pushing ~$350,000/mo of goods this way. We looked up a couple of his listings and they were between 25%-45% markup, meaning he's likely making between $87,000-$150,000/mo. Not bad.

I'd love to see this on a larger scale, perhaps with more retailers. I'd love to see the developer keep the data and use that for predicting eBay auction prices and using eBay along with Half.com. There's a whole slew of things you could do with this type of data.

One interesting side note is that apparently in most cases the Amazon Marketplace is worth more than Half.com.

Five blogs I can't live without

2007-02-08T23:14:00.000-08:00

Five blogs I can't live without:

Did the blogosphere design the iPhone?

2007-02-07T21:54:00.000-08:00

Today's post is a short one. I just want to pose a question...

I was following some links today when I found this post at Business Logs. Mike writes:

I do believe it's a first for Apple, Inc. but Steve Jobs just delivered a keynote where a new product was announced that had absolutely everything in it that people were rumoring about. The Apple iPhone. Does it have a nice, big screen? Yes. How about Wifi and Bluetooth? Yes. Plays video flawlessly? Yes. New interface running some variant on Mac OS X? Yes. Camera? Yes.

The blogosphere was abuzz about a possible iPhone for many, many months. I can only wonder: did the blogosphere design the iPhone for Apple (and Apple just implemented it)?

Salesgenie.com "deliver a targeted message"

2007-02-06T01:23:00.000-08:00

How many things are wrong with the article, Super Bowl ads let Salesgenie.com out of bottle?

Let's start with the fact that they paid $3.7 million dollars for advertising in one game.

Or the fact that it "generated as many new trials for the product in one day as it takes to generate in an entire month," but we don't have a clue how many sales it generated.

Or, the fact that they believed this was a "targeted message [delivered] to a specific niche of workers." This was targeted!? And delivered to a "specific niche"?! The SuperBowl!?

WTF?!