Top 10 posts of 2007

 

    There still 5 days left this year, but the 360 days before today pretty much defined what was popular on this blog. The top 10 most read posts on this blog were (a lot of those are from 2006):

1. Outlook 2007: Slow as a turtle?
2. Amazon SimpleDB: What nobody is talking about
3. 14 Tips to Speed Up Your Web Pages
4. VTech shooting: Caught by surprise. Twice!
5. List of Seattle Startups by Alexa Rank
6. How much does it cost to found a Microsoft-based startup
7. Amazon's Start-up Project: Best event of the year
8. Top 7 mistakes at Sampa, a Startup tale
9. C# tip: Optimizing ArrayList usage
10. Download WHOIS database

 

    The one thing to notice on this list is there are a lot of "random posts" but they get picked up by a popular blogger or rank very high on Google for certain queries so they are read a lot.

 

 

Top Editorial Chosen Posts of 2007...

 

    This is the list I'd consider the most interesting of my writings for 2007...

 

...in the realm of startups and entrepreneurship...

 

1. The big misunderstanding of the Long Tail
2. Ready to join a startup? 10 things you need to know
3. Iron Chef and Web 2.0
4. How to raise angel investor money?
5. Please, define "viral"
6. Investors have no imagination
7. Business Plan vs. Executive Summary
8. Ignite Seattle: How not to give a talk
9. Redfin Dilemma: Is it a database or a search engine?

 

... and in the realm of web-tech...

 

1. 13 tips to your email not end up on the Junk Folder
2. How to confuse your users?
3. 7-steps to website nirvana
4. 69% to 79% to 85%
5. The support cost of a Web 2.0 user
6. 1,100% increase in the number of indexed pages on Google
7. The case for full-text feeds

 

The Crunchies nominees have been announced

 

    The Crunchies -- the award for tech startups organized by TechCrunch, GigaOm, Read/WriteWeb and VentureBeat -- have finally announced the five finalists for each category. No word yet from the official hosts, but below is the list of nominees. For reactions check out TechMeme.

 

 

• Best Technology Innovation/Achievement: Earthmine, Like, Move Networks, Twine, Viewdle
  
• Best bootstrapped start-up: FriendFeed, PoliticalBase, ProductWiki, Techmeme, UpNext
  
• Best new gadget/device: iPhone, Kindle, Ooma, Pleo, Wii
  
• Best business model: Glam Media, Imeem, Prosper, Weatherbill, Zazzle
  
• Best design: Etsy, Jackson Fish Market, Netvibes, SmugMug, Songza
  
• Best enterprise start-up: 37Signals, Attributor, EditGrid, Ribbit, Zoho
  
• Best consumer start-up: 1800-FREE-411, 23andMe, LinkedIn, Meebo, Zillow
  
• Best mobile start-up: AdMob, Fring, Loopt, Shozu, Twitter
  
• Best international start-up: Atlassian, Gizmoz, MusicShake, Netvibes, Openads
  
• Best user-generated content site: Digg, Facebook, Geni, Instructables, Yelp
  
• Best video site: Aniboom, Hulu, Joost, Justin.tv, Tokbox
  
• Best clean tech start-up: A123Systems, Ausra, Gridpoint, NanoSolar, Tesla Motors
  
• Best user of viral marketing: Flixter, iLike, iminlikewithyou, RockYou, StumbleUpon
  
• Best time sink site: College Humor, Duels, Kdice, Kongregate, Pandora
  
• Most likely to make the world a better place: Causes, DonorsChoose, ZeroFootprint, Kiva, One Laptop Per Child
  
• Most likely to succeed: Kayak, Mint, Slide, Wordpress, Zivity
  
• Best start-up founder: Reid Hoffman (LinkedIn), Max Levchin (Slide), Kevin Rose (Digg), Evan Williams (Twitter), Mark Zuckerberg (Facebook)
  
• Best start-up CEO: Gina Bianchini (Ning), Dick Costolo (Feedburner), Toni Schneider (Wordpress), Rob Solomon (Sidestep), Lance Takoda (Rock You)
  
• Best new start-up of 2007: Hulu, iMedix, Joost, Ribbit, Tumblr
  
• Best overall: Digg, Facebook, GrandCentral, Twitter, Zillow

 

Do you remember "The Crunchies"? The real story behind an award...

 

    About a month ago, four of the most popular technology blogs on the whole world (TechCrunch, GigaOm, Read/WriteWeb and VentureBeat) decided to get together and create an award called The Crunchies.

 

    From December 2nd to December 12 they asked users to nominate companies and people for 20 different categories of award. According to their website, the day after the nominations closed they would pick the top 5 of each category and open it for public voting for a month. That would be on December 13th, a Thursday.

 

    Well, that day came and went and nothing was announced.

 

    About 4 days later they actually updated the website and said the chosen companies/people would be revealed and voting would start on Monday, December 17. That came and went and then they said it would be on Wednesday, the 20th (yesterday) -- which also came and went.

 

    So far, there has been absolutely no public statement from any of the organizers (Michael Arrington, Om Malik, Richard McManus and Matt Marshall) and there is a long list of very frustrated entrepreneurs and award-winners-wannabe on the Crunchies website (some very funny comments, some very pathetic ones)

 

    Anyway, I think people are a bit desperate for information, so I thought I would give 5 different explanations for the delay:

 

Reason for Delay #1:

The most nominated companies/people were PayPerPost, Uncov, BlogNation and Sam Sethi.

 

Reason for Delay #2:

They forgot to save the nominations into a database or file and now are making up companies and names.

 

Reason for Delay #3:

They had to actually visit the nominated companies to see what they actually did.

 

Reason for Delay #4:

There was so many unknown entrepreneurs/CEO nominated they had to do background check on them. God forbid they gave an award to a sex offender.

 

Reason for Dealy #5:

The Statuetes they promised to deliver during the award ceremony turn out to be too expensive, and now they are looking for more sponsors before they commit to anything.

 

 

 

   

Why people still use IE 6?

 

    It has been a while since Microsoft released Internet Explorer 7. They even made it part of a Critical Update on Windows Update. However, 37.3% of Sampa site owners are still using IE 6. This month we just crossed the line where we have more site owners running IE 7 than IE 6, but still IE 6 should be into the single digit market share by now and it's not!

 

    This annoys us mostly because there quite a bit of backward code that we have to write just to make sure things work on IE6. The two biggest issues are the many bugs around the "boxing model" and "positioning" and the lack of support for PNG Alpha channels.

 

    However, last week we started a more aggressive strategy to get our users out of IE6. First of all, every time a new user signs up to Sampa, if they were using IE6 during the sign up process they'll get an extra email message telling them about IE6 and suggesting they should upgrade to IE7. The second thing is when they sign in to edit their website, they'll receive a message telling them they are running an older browser (they can dismiss that message forever).

 

    We hope to see a significant drop of IE6 users over the next few weeks or so.

 

    I wish other Web 2.0 startups would be doing something similar. All of us can only benefit by accelerating the migration to IE 7 (and FF2) and hopefully can completely drop support for IE6 in a year or so.

 

What is a good backup/sync solution?

 

    Just to clarify the title, I'm not looking for a backup product, mostly because I don't think it will do what I need.

 

The problem:

    Every night, around 1AM each front-end server backups every single file associated with a site, which include images, configuration files, data files, log files, etc. Each server on average has between 2-3 million files to be copied every night.

 

    Currently, I wrote my own backup solution which checks file timestamp and syncs two folders on separate disks.

 

    For example, for "marcelo.sampasite.com" there is the site folder on drive C and a backup folder on drive D. Every night, the C folder will synchronize with D folder. The key here is that it doesn't take the site offline so every file copy locks the site for a few miliseconds.

 

 

What I need?

 

    My stupid file-sync algorithm is becoming too slow. It takes many hours to sync each server hard disk. I need something that is faster, optimized for Windows, and, ideally, would be able to integrate into my app so that it can copy files while we have open handles to it.

 

    I'm looking at calling Robocopy, but I'm not sure that is optimum. I'm also wondering if Windows Shadow Copy can help with that. Anyone has experience doing this?

 

Amazon SimpleDB. What nobody is talking about?

 

    Amazon Web Services just released their new service: SimpleDB. This is a pretty brilliant idea and people are taking notice.

 

    But there are two pieces that bloggers are not paying attention to, or they didn't realize (after all, most of them are not developers)

 

    I felt the service was so interesting I checked the API and how it works and bam! I was hit in the head.

 

 

Did anyone say X.500?

 

    My first realization is that it's not a Database, it's a Directory Service!

 

    Ok, most people (even developers) would not know what a Directory Service is even if you hit them with an Active Directory book on the head. Anyway, if I remember correctly of my years on Exchange Server (98-99) while working on the Active Directory integration, a Directory Service had a few peculiarities that differentiated it from a traditional database.

 

    First of, each object (this is what a "record" is called on a Directory Service) can contain different attributes and the schema can be changed on the fly (a bit more complicated than that).

 

    The next interesting aspect is that a single attribute (field) can have multiple values, just like the Amazon SimpleDB! This means if I define attribute "UsedBy" I can set the values to "Realtors" and "Brokers". On traditional relational databases you'd need 3 tables to do something like this.

 

    Finally, a Directory Service allows a hierarchy of objects, meaning instead of Tables you have nodes (which are container objects) and objects hang out of those nodes. Oh oh, SimpleDB doesn't have that, so all my theory goes down the drain.... Not really, they provide a thing called "Domain" which, if you want to (but you don't), can be used as a hierarchy.

 

 

And the best application for SimpleDB will be...

 

    Calling SimpleDB a database or a directory service doesn't change what it can do or what people can do with it, it's just a convention. What matters are the nice products that will come out of that, and IMHO, one of the most interesting one will be...

 

    A search engine!

 

    What? Somebody will built a search engine on top of SimpleDB to compete with Google? Nah! Somebody -- lots of body, actually -- will be able to built their own site search service on top of SimpleDB.

 

    Imagine that Redfin is not a gazillion-dollar VC-backed startup. They are just getting started and want to index all listings from MLS to do a kind of search that you cannot do directly to the MLS database. They can put all that data into SimpleDB (the flexible schema is a huge plus) and not have to worry about having Terabytes of data on their own database. Do you know how much it costs in time and money to maintain a Terabyte database? A lot. There is backup, there is perf issues, there is hardware redundancy, etc.

 

    The only thing missing from the SimpleDB API to provide some serious "site search" capability is a way to rank attributes when doing a query.

 

 

How to handle bad press in a Blog World?

 

    I wrote a few times before about this topic and this week there was a great example that can be used to illustrate the problem.

 

    First of all, if you are ever attacked publicly in a blog, forum, website, wiki or whatever, you must respond! You must respond in as many always as you can, as fast as you can and make sure that every single word you write reflects the facts of the situation.

 

    Case in point, Sam Sethi and Michael Arrington spat. This is a long thing so I won't go in detail, but you can read Sam attacking Michael and Michael's response.

 

     It's a bit clear that Sam is a problematic person (just being politically correct here). You might think that for a case like this you just ignore and let it die, after all, what's the value of a wacko saying bad things about you? No much, right? Wrong.

 

    Most people on the Internet, when they are looking up for stuff about you, will not do background check on every person that said something about you. If you don't react to something like this, you are letting just one voice be heard. So, speak up and tell your side of the story. Just be as fact-based as possible, because everything, I mean *everything* you write on a blog post will *never* be erased from the Internet. Search Engines and Blog Search Engines will cache it and you don't have control over it.

 

    And the right way of responding is to write your own blog post on your own blog, so nobody can remove it, and, also leave a comment on the original blog post attacking you. Odds are that comment will be deleted, but if it's not, better for you.

 

 

Is the Bubble exploding? Is sanity coming back?

 

    Just on the last week or so Edgeio and BlogNation have hit the floor. And now there is rumors that OmniDrive is facing the same end result.

 

    I got to be honest with you all. Every time I hear a company has gone out of business I feel a bit of relief. First, because Sampa has outsurvived them. Second, and more importantly, because it brings a bit of sanity back into the Web 2.0 world.

 

    There are just too many bad companies, and the noise level is just too high to get good press, good investors and good employees. Every time you go to meet with an investor they ask if you are a competitor of XYZ, and you never heard of XYZ before. And, to make matters even worse, XYZ is not even close to what you do, but all that they think is that there might be a similarity, thus it's *crowded*.

 

    Sampa will outsurvive most Web 2.0 companies because we are not built in a feature idea. We are full business with a full product. We are not built with 10,000 lines of code outsourced to some second class development company. We've built a tremendous amount of IP. And we are not running our business like it's 1999. We spend, but spend wisely. And, finally, the most important thing of them all. We proved that people are desperate for products like ours and that *our* product is sticky.

 

    Yes, there is the revenue issue, but we are doing well on that front and it can only get better from here.

 

How much does it cost to found a Microsoft-based startup?

 

    There is a lot of debate about running Linux or Windows on the enterprise. Usually, each side throws lines like "Linux is not really free" (because of Total Cost of Ownership), "Windows is too unsecure", "Linux is slow to fix vulnerabilities", "Windows is too expensive", yada, yada, yada.

 

    But what if you are starting a new tech-company? In other words, a lot of the points of the discussions above are null. All the IT will be done in-house by experts (your team), the systems will be tight-secure because you know how to disable services and close the unneeded ports. Etc.

 

    Ok, so if security, reliability and functionality are not an issue, what is it? For me it ends up being two simple things: Knowledge and Price.

 

    Knowledge is simple. If you know Windows, use Windows. If you know Linux, use Linux. If you know both...

 

 

Microsoft's Price Sucks for Startups!

 

    I envy the LAMP guys, not because of the technology which I think is slightly behind Microsft's (VS + C# + .NET), but because they have way less concerns about pricing and licensing.

 

    Let's hypothesize that you are starting your own web-based technology company. You and 2 other friends get together and make the big decision to do this. Two of you are developers that will work on the product, one will be the business guy. You guys didn't make the promised millions of dollars in stock options so it's not like you don't have to worry about paying the bills.

 

    If you are a web-based tech company, this is the minimum (debatable) hardware and software requirements for you:

 

• 1 Server running Windows Small Business Server 2003 for Email, SharePoint, Domain Controller, etc.
• 1 Server running W2K3 Server as the front-end for the web-based app.
• 1 Server running W2K3 + SQL 2005 as the backend
• 2 Desktop machines for Development, with VS 2005 Pro, Office 2007, Windows XP
• 1 Laptop for the business guy, with Office 2007 + XP

 

    A few assumptions about this scenario is that Source Control will be done in one of the existing servers and through some Open Source option (like SubVersion or CVS), because Microsoft Source Control solution is just outrageously expensive, you'll be buying your servers without software and that you are using a co-location facility.

 

    How much will all that software cost? $7,775 according to Amazon.com's price.

 

    That is for software alone! This is a pretty hefty price tag for a bootstrapped startup.

 

    The problem is that Microsoft prices its product mostly for Enterprise consumption. Do you think Boeing cares if the $5,000-dollar server they just bought comes with a $900 Windows Server license? Maintaining that server alone will cost an extra $5,000 over the course of 3 years, so the price of a Windows Server is less than 10% of the TCO of it.

 

    But startups don't think like that. 3 years? We might not be in business so I can't use that kind of math. I wish Microsoft would have some kind of special discount or package for startups and don't tell me about the Microsoft Empower for ISVs because it doesn't apply for web-based applications.

 

 

A Challenge for Entrepreneurs, Founders, Startups and CEOs

 

    We announced on the Sampa blog our new "Charity feature" and I explained what drove me to create that. Now, I'll throw a challenge to every Web 2.0 Startup that is a Social Media / User Generated Content service: Can you do the same? Or better?

 

    Here is a short list of companies that I think could easily implement something like that:

 

• iLike -- Hadi?
• Zillow -- Rich?
• 43 Things/43 Places -- Josh?
• Wetpaint -- Ben?
• Faves -- Mohit/Rob?
• Shelfari -- Josh?
• Popshops -- Jessie?
• MyTypes -- Vipin?
• Cozi -- Robbie?
• Others Online -- Jordan?
• Wishpot -- Max?

    Basically, if your service allows a user to create a profile, a webpage, a blog, a wiki or any other UGC, just give them the choice to pick a charity. It's that simple.

 

 

Integrating Charity Giving into your Business Model

 

    We just released a new feature on Sampa that allows users to easily select a charity to display on every page on their site, with a link direct to the charities donate page.

 

    Since I started Sampa I thought on ways that I could make donations part of our business. The problem is that we don't have any hard cash to donate. Our revenue is so small that even if we donated 10% of it, it wouldn't make a difference. We also don't produce anything that we could donate in kind and our service is already free, so charities can use it anyway.

 

    However, just a couple of weeks ago I had an epiphany.

 

    We could point hundreds of thousands of web pages to a charity and have a dramatic impact on their PageRank as well as referral from a Sampa sites. If we assume a low CTR like 0.5% we could generate more than 100 clicks per day to charities. The "web value" those clicks can be the equivalent of thousands of dollars per month if you are buying a CPC or CPM campaign.

 

    But you might challenge me and say that telling users to add a link to a charity is just a 3-4 step on any web service. We removed most of those steps and made it a single step during the sign up process. Now, when you sign up for Sampa you have a immediate opportunity to choose a charity. Even if you are super lazy you can do it.

 

 

Vote/Nominate me for The Crunchies

    Yes, it's a serious award and in case you think I deserve it, click on the image below and put Marcelo Calbucci under "Best start-up founder"

 

    This is this first phase of the award where the top 5 people/companies with the highest number of nominations will compete on a second round of voting.

Amazon Start-Up Challenge Award

 

    Yesterday I went to the AWS Start-Up Challenge Award dinner. It was at the W Hotel in Seattle (very nice), and you can read more about the companies and the event on John Cook's blog and Brier Dudley's.

 

    The winning company was Ooyla, which was founded by 3 ex-Googlers. They do have a very cool product around monetization of video content and they use pretty much everything Amazon Web Services has to offer.

 

    I've also met the CEO of WeoGeo. Very smart technical guy, but as any smart technical guy goes too much into the technology talk versus the benefit for his customers (or the world) about his product.

 

    Also met Scott Rafer, one of the co-foundes of Mashery, a product to help web-based companies that want to expose APIs to developers deal with API Keys, registration, tracking, limits, etc. I think if Sampa had an API I would seriously consider something like that.

 

    If I look at an event like this and compare it to something Microsoft would have done (sorry, I have to compare because they are both local companies and I worked at Microsoft), Microsoft would have done a 3,000-person event in Las Vegas with the cheap price of $995/person and it would include a 3-day event with seminars, tech-talks, hands-on sessions on how to use Microsoft technology.

 

    Amazon's style? It's pretty much a grassroot event. Get the CEO/CTO of small startups and get them engaged and talking. A few of those will become CEO/CTO of Fortune 500 companies and all built on top of Amazon Web Services.

 

Microsoft will win the Online Office battle

 

    There has been dozens of startups fighting to be the official "Office Online", or a part of it, like "Online spreadsheets", "Online Word Processor", "Online Presentation", etc.

 

    Finally, I'm convinced that Microsoft will dominate the online market the same way it dominates the offline market.

 

    I came to that conclusion after thinking about technology leap that people need to take to transition from one product to the next. Going from Word for DOS to Word for Windows was a natural transition. The same way it is each new version of Office. You continue to use the keyboard and mouse, you click on the shortcut on the Start menu, an empty document shows you and you start typing.

 

    The leap from Offline to Online is not that small. There are many different steps, like opening the browser and going to a link, signing in, etc. Then, there are the perception aspects, like you don't own that document, that it's not safe, etc.

 

    The moment Microsoft releases an extension for Word that allows you to store documents online as well as offline (like Sharepoint for the rest of us), it will start an inevitable trend of dominating that space. That is such a short leap for a user to make because he still has his Word icon to click, and is just a matter of having a synchronization service to seamlessly look like the document is local and a backup/version is online.

 

    From that, Microsoft can provide it's own web-based Word if people want to do some basic editing on the go, and then the product gets better and better until it's good enough to be used without the offline version.

 

    One big mistake startups make, IMHO, is to think that the world moves at a much faster pace than it really does. It'll take a decade for online document editing to have 50% market share of all document editing. Yet, a lot of startups assume that Microsoft will lose Office domination in 3 years because they are not going online. Hey, I have friends running Office 2000 and they are quite happy.

 

Netflix for Book: It won't work

 

    I'm not a fan of shooting ideas in the head, because we usually know less about those ideas than the people executing it, but at the same time we don't suffer from the "curse of knowledge" which gives us the right of being skeptical.

 

    This week John Cook wrote about Paperspine, a new Issaquah-based company that defines itself as "Netflix for Books". This morning on my daily routine reading blogs I found on ReadWrite/Web another company doing exactly the same thing: BookSwim.

 

    I was already going to write why I don't think Paperspine will work and now that I know they have a competitor I think the odds of success (for both) are even less.

 

    These are some of the arguments:

 

1) Pet food: Remember the Bubble 1.0 when a lot of startup thought was super smart to sell 50lb dog food and 10lb cat litter over the mail. A DVD weighs like letter, while a book weighs about hundreds of letters.

 

2) Time commitment: The time commitment to a book might go from several days to many months, while the time commitment to a DVD is capped at 2 hours (maybe 3). For me to think that I'll watch 2-3 DVDs per month and pay Netflix $9 is fairly reasonable. Now, paying $15/month for a book service, I must really be convinced that I'll read at least 1.5 books/month over the curse of my subscription. What about months that I'm busy? Or on vacation?

 

3) Book worms vs. Non-worms: I see there are two types of people. Those that read lots of books (non-technical) and those that don't. I read lots of technical books (about 30-40 per year), but very few fictions. The people that don't read that many (which is the majority) won't use the service. The people that read a lot, already figure out a cost effective way of doing that, like using the library, half-price books, sharing between friends or spending lots of time inside a B&N.

 

 

Amazon WS Start-Up Challenge Awards Dinner

 

    I'll be going to this event tonight. I'm sure there will be at least a handful of people that I know (mostly from Amazon). It should be fun.

 

    UPDATE: Apparently all the tickets are sold out.

 

3 hours to make a button in CSS

 

    It's no surprise for people that understand the Internet main presentation protocols (HTML/CSS) how hard certain things are. Not only CSS/HTML are quite limited, but there are huge differences in behavior from one browser to another and from one version of the browser to another.

 

    I just spent the last 4 hours coming up with the HTML/CSS for our new button layout. Finally, I've got to a solution that works perfectly in IE 6, IE 7 and Firefox. There is just one small issue that I couldn't fix and I don't care anymore. Enough.

 

    Our new UI design will have some new elements that will enforce our brand and make the site look cool, that is why I'm working on buttons. There are still about 4 other things that I'll need to redesign (like Dialog Boxes, Drop-down menus, etc), but the button was probably the hardest one because buttons must resize automatically according to the text/icons inside that button. Dialog Boxes and Drop-down menus have fixed size, so they are much easier to implement.

It's officially a bubble 2.0 and here is the official video

 

    Too good to not watch: http://youtube.com/watch?v=fi4fzvQ6I-o

 

    Cue: If you don't find it funny, you are not in the middle of the bubble 'cause I'm knee deep.

What it means to be the best startup founder?

 

    I don't know, but I never won an award, so in case if you feel like I deserve it, click on the image below and put my name under "Best start-up founder"

 

 

 

   

    And since you are in a nomination mood, vote for Sampa under the "Most likely to succeed" category.

 

Seattle Startup Index

 

    I just published the Seattle Startup Index on the Seattle 2.0 site.

 

    Sampa improved the Alexa Rank by 4,647 positions and the Compete Rank by 3,780 positions, yet we dropped 3 positions on the SSI list. What gives? 4 new startups have been listed above the Sampa rank (SEOmoz, ActiveRain, WidgetBucks and 3Guppies).