My name is Sam Deskin and I am the founder of samfind. My experience in making samfind has been long and circuitous. But who ever said developing a website would be easy? samfind has gone through iteration after iteration, which seems normal, but it has also gone through developer after developer and long delays to get to where it seems like it is finally arriving.

It started in 2006 - I started looking for a way to build my concept with out spending too much money. Developers in America who I contacted were ... expensive. Typically, I am against outsourcing work that can be done by Americans. But were my ideals worth more than my idea? I determined that it would simply not get built if I built it in America.

The only option I saw available to me without breaking the bank was to find some competent Indian programmers and have them build samfind. I contacted Indian programmers through getafreelancer.com by placing an ad with my general description. I chose the three top-rated programmers and spoke with them. At this level, the programmers are not individuals, but organizations with hundreds of programmers. That should have been a warning because any rating associated with the programmer's account was likely the result of their best programmer's skills.

Ultimately, I settled on a company called iNavigators. Sandeep Palod was my new best friend. We spend hour after hour and sleepless night after night (I was going sleepless, not Sandeep) going over my design specs. He seemed genuinely interested. He listened and asked questions as if he was taking notes. After about a week, he told me about the new and improved pricing that would have to be implemented because samfind was much more complicated than he first thought. After having spent a week on the phone with him - I relented. The price was still a fraction of development costs in the US. Development of samfind would begin.

I sent my first payment only to find out that my new best friend Sandeep was not the developer - he was the sales guy. Everything that I told Sandeep was for not. I had to repeat this process with the actual developers - for the next week. Hmph. They listened to some of what I had to tell them but were eager to start developing. For months they developed and we had late night calls having them fix issues that were coming up. You know small things, like the homepage being 300+ kilobytes and the database having the same data repeated time after time. Even I knew that there was something wrong with this picture.

For a while I figured that I could manage them into becoming competent developers. I hired consultants here who could tell me what to tell them to do to clean up the database. I even toiled around the homepage myself and took out about 200 kilobytes of excess code - through trial and error. I am not a developer.

Toward the end of a few months the website seemed to work - it was a little slow and had some weird quirks, but those small things added to its character, right?

In their defense, some aspects of samfind evolved as we were developing them, but these evolutions were oftentimes because they could not figure out how to do things the way the specs were written. As the final stretch came around, the developers became increasingly distant. Grumblings for extra money because of the amount of time they had spent on the development were subtly hinted at. Although they had told me it would take about a month to develop by the second month I was so invested in them, that I could not simply drop them. By the third month I was still holding on. By the fourth month - my eyes were black from lack of sleep and I was quite irritable, so I am told.

About the fourth month of development I rationalized to myself that after they were done, I would get a local developer to clean it up. I even started doing some publicity when we were contacted about our startup that Tech Address somehow heard of. To keep within some semblance of a budget I hired a local university grad student who seemed competent. By coincidence he was also Indian, but he was local. He was in fact very good - but the website that he had to work with was such a mess that it was practically unsalvageable. I wouldn't find this out until an Indian wedding and 6 months of my life had passed by.

It took me a little while to get out of the dumps before I could pull myself together to get a new developer who had the time to fix it. I decided that it was not worth trying to get it developed out of the US because it would simply not happen anyway, so I placed an ad in craigslist and got hundreds of responses. Some were clearly incompetent - even I could see it. Others were hopelessly expensive. Still others suggested that since I had such a clear idea of what I wanted that I should try to fix it myself - that it would only take me about a year to learn how to be a programmer and then I would be able to build it.

I finally settled on Robb, a programmer in Washington state who seemed to know what he was doing. He humored me for a while and tried to fix the mess that I had, but constantly let me know that if I really wanted this thing to work, I would have to rebuild. Eventually - I gave in. This was a hard pill to swallow. I had spent untold thousands and - at this point - more importantly, over a year of my life trying to get the mess of code that was samfind to market. I had gone through false starts and written marketing plans assuming it would launch when I was told it would. I even had a launch party and gave away free t-shirts.

In November 2007 we began redeveloping - scrapping most of samfind as it was built - Robb built an entirely new database and about 98% of the front end. The most important things that he salvaged were some of the open source ajax and nearly nothing of what the previous developers had done. At this point, we were building at break neck speed. After the initial development got to a point where Robb felt like it was done, I rehired the local Indian grad student who has since graduated from grad school and is now working on samfind part-time.

The private beta will be starting in April with the public launch soon after.

Let's hope for the best.

sam