Marcelo Calbucci

 

   I just published over the Seattle 2.0 blog the new Seattle Startup Index for the month of job and added a new trick to the bag, the SSI50 -- the average ranking of the top 50 Seattle Startups.

 

   Sampa stayed at #25 on the list, but our Alexa Ranking moved up 2,374 positions. We would have taken over Mixxer, but their Alexa Ranking also improved, which is good news to all of us.

 

   It is good to see the companies of my friends moving up, like the new Should Do This from Josh Petersen (Robot Co-Op) which is a new service that is not even released yet, but already starts at #69; WishPot from Max Ciccotosto moving up 11 positions (top 3 movers of the month!) to #61; BlueDot from Mohit Srivastava moving up 1 position to #7; Twango from Philip Charmichael, Serena Glover and Jim Laurel moving up 2 positions to # 21.

 

 

 

 

    A few weeks ago I've met Ksenia Oustiougova which is building a company to help kids learn through mobile devices.

 

    Anyway, I went to her website http://lilipip.com and for a minute I thought I had mistyped the URL because it showed a GoDaddy "parked domain" page. You know, one of those they try to make money out of people that bought a domain but didn't put up a website yet?

 

    So I double checked her business card and the address was correct, just missing the "www".

 

    Heck, GoDaddy didn't configure her DNS properly and they are "hijacking" part of her traffic. You don't have to understand what I'm saying, just believe me, it's wrong and they will have some lame excuse that Ksenia forgot to change something or misconfigured her DNS. It's BS. DNS can be quite complicated for non-technical people, and even for people that are technical.

 

    For the record, her company address is www.lilipip.com (don't forget the www)

 

 

 

 

    nPost is another startup from a friend of mine, Nathan Kaiser. What do they do? From their site:

 

We are dedicated to entrepreneurship and startups, with a startup job board, entrepreneur interviews, and networking events.

 

    Their last event (last week) had more than 100 attendees. That is pretty good. If you are an entrepreneur you probably want to check this site.

 

    Nathan and I have been discussing about how can we work together (re: Seattle 2.0) to increase even more the networking between entrepreneurs in Seattle. Stay tuned.

 

    I'm not a power blogger, I'm not on the A-list (not even on the B-list), but this blog has a few hundred readers, so why not use it to help friends.

 

    That said, I'll start to blog more often about friends products and startups.

 

    To start, check out Max Ciccotosto (ex-MSFT) new product: WishPot, a better way to create Wish lists from online and *offline* stores.

 

    They are getting some great press, including Mashable!

   

 

    This is a pretty cool concept: Co-working.

 

    The gist: People working on their own project on a shared office space.

 

    If I'd be starting Sampa today or if I had the means to jumpstart I would pursue something more like that. A place where eager entrepreneurs-developers can come, have an office space, share resources and jump start their business.

 

    I have no doubt that quite a few people would enjoy doing that. One of the things that I missed the most at the begining of Sampa was somebody to bounce ideas of. Somebody to have lunch. Somebody that cheers with you on your successes and motivates you on the down turns.

 

 

 

    In February I started counting how many pages Sampa sites had on the Google index. All sites are out of three domains: *.sampasite.com, *.inthesphere.com and *.brainuse.com.

 

    The first time I tracked, all three had less than 6,000 total pages indexed on Google. If I'm not incorrect, back then we had just about 8,000 sites, and we know that each site certainly has more than 1 page on average, more like 3 or 4. So Google wasn't doing a good job at indexing Sampa sites.

 

    Over the next few months we added multiple features that could improve the number of indexed pages. First, we added Google Search. It was a pretty good integration and made it for a very interesting scenario on how to leverage Google Search and integrate it seamlessly with a site.

 

    Then, we added Site Maps. And the third thing that we added was the Sampa directory. Before, there was no place on the web that would point to Sampa sites so how on earth would Google find those?

 

     At first, the number of pages in the Google Index was increasing at about 10% per week, from 5,969 to 6,434 to 7,012, etc. Then on the end of May, we had an increase of 50% in a single week, from 19,232 to 29,020. Two weeks later we had an increase of 83% from 28,980 to 53,010.

 

     Currently, we have 66,430 indexed pages on Google, which is an astounding 1,110% growth over the last 16 weeks.

 

     The key here is that we passed some threshold on Google's algorithm that tells it to index more of our content. Great for us.

 

 

    During the Keiretsu Forum last week I had the opportunity of meeting Max Shapiro. Max is an experienced person on the startup world.

 

    Max has a company that provides a brilliant service. PeopleConnect helps startups find executives and other professionals that are willing to work for equity for a period of time. According to the service description on their site:

 

"...This is an unique win-win solution to the 'Catch 22' of not having funding in place to develop your team, and being told by potential investors that you need a more developed team..."

 

    You've got to agree with me this is a fantastic service for startups.

 

 

 

    Yesterday we presented at the Keiretsu Forum in Seattle in the morning and Bellevue in the afternoon. For those that don't know, Keiretsu is a for-profit organization of (angel) investors. Each month the screening committee at Keiretsu selects 4 companies to present at the full forum.

 

    The presentation went great. The three other companies were also good, and I'm pretty sure that two of them will get some funding. The third one had just too many hurdles even for high-risk-tolerance investor.

 

    The most amazing story was about Earth Class Mail, that presented a few months ago and was giving an update. Not only they raised all the money they planned to raise, but the've got oversubscribed by 3x (3 times more investors that they could accomodate) *and* they've negotiated a term sheet with two local VCs for another boatload of money.

 

    For Sampa, everything went quite smooth. Our presentation was very polished. We've got some tough questions from the investors. Tougher than you would expect for a company trying to raise a seed round.

 

    The results of presenting at Keiretsu will take 6-10 weeks to know, because we now have to go through their process that includes creating a due diligence committee, doing the due diligence, negotiating terms and closing.

 

    Since we are trying to raise not a lot of money (mid-six figures), and we have about half of that "soft circled", we'll only need a few more angels to participate to close the round, and I'm very optimistic about it.

 

    In just a few hours we'll be presenting at the Keiretsu Forum meeting in Seattle, and on the afternoon we'll be presenting in Bellevue.

 

    Of all the angel groups, Keiretsu has been hailed as the most effective. To begin with, they are a for-profit organization, which helps on the professionalism side, but the core aspect, IMHO, that makes Keiretsu so different is that the Angels there are very active and really interested in investing, not only listening.

 

    The downside is the long due diligence process of about 9 weeks. This is great to give the angel investors more confidence on their investment, but it is not so great for the entrepreneur, since you are taking angel money (low amounts), but going through a VC-like process.

 

    On the 10,000 foot view, the due diligence might not be a big thing if at the end of it you get out with your round complete (raise all money you wanted). But it can be tough if no one invest.

 

    One of the things that I learned on the last 7 years of working on Web Services is that you either have an awesome strategy to avoid downtime on your service or you have a poor service (lots of downtime), there is not much in between.

 

    There is scheduled downtime, like when you are upgrading the service, or the  network router is being replaced, and there is un-scheduled downtime, when the disk is full, or the network router failed, or the software has a memory leak and it reached the breaking point.

 

    Un-scheduled downtime you address with redundancy on your servers and network gear, good testing and good quality server software overall.

 

    Scheduled downtime is a matter of good planning and great execution.

 

    In the last twelve months, we have some un-scheduled downtimes, mostly related to a blackout in Seattle in December and other network related issues. We had just one partial downtime (only some sites were unavailable) when we rolled out a new DNS server.

 

    Scheduled downtimes we have at least 3 times a week. That is because we upgrade our software furiously. Yesterday, we deployed new features for Facebook. Today we deployed even more features for Facebook, and so on.

 

    I've been trying to improve the scheduled downtime period to a minimum. It used to take 3-5 minutes of downtime, than I did a fix to make it just under 2 minutes, but the number of sites started to grow so much that even with the fix it took about 3-4 minutes.

 

    Today I did a new fix to how we deploy things, and the results were spectacular: Under 25 seconds!!!

 

    This is fast enough that a visitor browsing the site will never see an error page, and, using a uniform distribution of requests, would average to a wait of just 12 seconds to display a page. Not bad. Heck, some services take 12 seconds to display a page even without downtime!

 

 

    Today, while debugging some Sampa for Facebook issue I had Outlook 2007 crash twice. The second time I immediately realized what was going on. I found a buffer overflow that caused Outlook 2007 to crash. The only thing somebody needs to do is to open an email message for it to happen.

    I'll try to create an shorter repro message to see what the real problem is. This could be very interesting.

    I found similar issues on the past with IE and Firefox. I reported them to Secunia, and they were not exploitable. This one is different, though.

 

    Big, big news...

 

    Sampa Announces a new product: Sampa for Facebook.

 

    And, we are featured on TechCrunch as all: Sampa Brings Personalized Pages to Facebook.

 

 

 

    I can't believe it is Friday. The last 10 days have been pretty crazy around here. I can't recall a week that I worked as much as I did this week on Sampa. Yesterday I put a bit more than 16 hours of work.

 

   When I worked at Microsoft things like that would happen 3-4 times a year, but at Sampa it never happens because we have a steady development pace.

 

    We prefer to have a constant stream of releases by launching new features 1-3 times per week, versus going trough a traditional development cycle that can take months to show progress.

 

    The pros of that approach is that you don't have to wait for a "train" to ship something cool. If it is cool and it is worth doing for the customers and for the business, just get it done and get it out there. Sometimes, we pause some feature development to work on something else that might be more interesting.

 

    That is what just happened last week. We had an idea for a great product. A revolutionary product. So, we stopped all feature development for Sampa and started working on this new product. We will have a beta of it shortly, and for now that is all that I can say.

 

 

    Am I the only one receiving a bazillion emails from Joe Benjamim about the New York Venture Summit?

 

    I don't ever remember having added my name to any list from Young Startup Ventures or anything like that, so, they've got my email from the website or some site list of Web 2.0 startups.

 

    I must have received more than 30 emails on the list few weeks. I wrote about this on the past. I just don't understand why are they so annoying.

 

 

 

   

 

    Believe it or not, we had all (tens of thousands) Sampa sites running from a single server.

 

    Although I bought and configured the second server a couple of months ago, I was postponing to put this server in rotation to avoid having to maintain an extra server.

 

    Because of our growth and some major features coming on the next few weeks, it was time to activate this server. It's live now.

 

    I read this morning Hillel post on his daughter ballet pictures.

 

    It is a very interesting coincidence that I share the same feelings as he did about professional photographers being stuck in a dead business model.

 

   Two weeks ago me, my wife and my son went to take some family pictures with an excellent photographer. She is a close friend of Paul (Sampa's CEO). The package included 4 prints. The session was excellent and the results were fantastic.

 

    I really didn't care much for the prints since I can do them myself, so I asked about the option of getting the hi-res pictures... To which she replied as if this was the most normal thing in the world: $75 per TIFF, or $2,500 for all of them.

 

    Over the last 7 days, Paul and I spent a bit of time discussing what I think it is broken on this business.

 

    To begin with, photographers were a different beast when there was no digital picture, which is pretty much just a decade ago. They could charge as much as they wanted for prints because it was too hard to reproduce them.

 

    The other important point is that they currently consider themselves as "manufacturers" instead of "service providers", this is, the print *is* the product, not the photo session on itself. Again, this is broken. They charge a reasonable amount on a per-hour basis, and then overcharge on a per-print basis.

 

    So, why don't they change?

 

    Let's see. Today they charge anywhere from $75 to $250 per hour of photo shoot. Then they charge between $10-$50 per print and $50-200 per digital copy. After all is said and done the value they get from you goes from $200 all the way to a few thousand dollars for a multi-hour event, like a Wedding (it can actually go well beyond $10K).

 

    Ten years ago, if you wanted a new set of prints for your mother-in-law, you'd have to order from them, so they would get an extra couple hundreds from you.

 

    Nowadays, you are way more likely to scan it, share it on a site, or just go to one of these places that will scan and print for less than $1 per picture. Yet, photographers have not realized the world has changed.

 

    Why?

 

    Here is my view on this: They are artists, and as good artists they are not very savvy on the business side. They can, and do, innovate on their art, using new tools and techniques, but they don't mess with the business model because they don't understand the consequences, so, they just copy what everybody else is doing.

 

    The problem is that the world has changed. What they considered a core piece of their work (prints) have become commodity, or, worse yet, irrelevant.

 

    A new wave of photographers, let's call them the Facebook generation, gets it. I'll predict this new generation will take a lot of photographers out of business. That will cause a major shift on the industry the same way Turbo Tax changed the accounting industry forever.

 

 

    Lisa LaMotta from Fast Company just wrote about the new Google Street View feature. Here is the first paragraph:

 

Google is bringing stalking to a whole new level. With the recent launch of Google's map feature, Street View, people are getting all shaken up about the real-time views that the Internet will allow of several cities, including New York.

 

    What is wrong with this?

 

    I'm assuming she didn't try the product and is making her own assumptions based on a popular article written on the NYT today. How do I know? Well, Google Street View is not real-time. You cannot stalk anybody. Period.

 

 

    Jason Calacanis is asking how to make the pages of his new company (Mahalo) load faster.

 

    I'm sure there are plenty of good people out there giving him good advice. I have compiled a list of 14 tips that is very comprehensive and valuable, IMHO.

 

    Besides those, I'd strongly recommend that he take a look at the "perception factor". That is, some pages load fast, but because of scripting on the client or poor HTML, it takes a few extra seconds to appear fully to the user. There is no amount of optimization you can do on the server to address that. You can ask the user to buy a faster machine, though.