1

I am in the process of creating a youtube type site and would love to get some help regarding the setup.

Here is what I have planned so far:

  1. Use my current hosting: which is a dedicated virtual
  2. Upon user upload: use ffmpeg to convert and resize videos
  3. Videos will upload straight to Amazon S3 where they will be redistributed using Amazon CloudFront (which provides the media streaming)

My Questions:

  1. Will this setup work? If not, what am I missing?
  2. Is this an inexpensive route? (as in, is it a fairly decent setup?)
  3. If successful, will I lose my butt in overhead? There is always the gap between 'getting popular' and 'popular enough to make money with ads'
  4. What will I be looking at in startup costs?

I hope I posted this question on the correct site.

6
  • There is so much about this that is a bad idea. The single most important thing that you're missing is a time machine to roll back to about 5-10 years ago.
    – gWaldo
    Oct 9, 2010 at 22:03
  • Not to discourage your efforts, but you should look up how much youtube is costing google on a daily basis and then compare that with the per-gigabyte going rates for bandwidth.
    – Greg
    Oct 9, 2010 at 22:05
  • gWaldo, could you please elaborate. I've been building websites for about three years now and have never ventured into the realm of online videos. If it's a bad idea, I won't waste my efforts. Oct 9, 2010 at 22:14
  • Oh, and it wouldn't be all videos in general. just one particular type ( not smut ). Oct 9, 2010 at 22:19
  • 1) it's a pretty saturated market. Vimeo is the only such site that I can think of besides YouTube. 2) you're not fruit to do something new or original; you described your own site as being a 'YouTube clone'. That would be fine if you wanted to build something like this either out of curiosity (to see if you could) or just for your personal use, but 3) you want to monetize. I'm fairly certain that the YouTube founders weren't able to turn a profit, and it's a loss leader for Google. You haven't described what would differentiate your site from those existing successful predecessors.
    – gWaldo
    Oct 9, 2010 at 22:47

2 Answers 2

1

Will it technically work? Decent software architecure assumed, yes. Starting small is and inexpensive is the right idea. BUT if you talk about economic relevant scale, here are the drawbacks:

  • depending an you needs might "hosting from the cloud" not be enogh
  • this is a very tight market: you start on the premise of trying to make money off the web while having to pay a service making money off the web. That means that you have to make more money than Amazon S3 per traffic to stay in the game. Thats the hard part. If your idea supports this, congrats.
1
  • wow, an answer that didn't include some snide remark. thanks Oct 10, 2010 at 3:04
0

What's dedicated virtual? Isn't it one or the other?

As for your questions;

  1. Yes, scalability.
  2. Almost too much.
  3. Yes, very much so.
  4. Comparitively little.
1
  • My expertise has been websites, not servers. I thought that one of the nice things about AWS was scalability. Oct 10, 2010 at 0:19

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .