I'm not on Tribe, but I recently read an article about the back end architecture growth of myspace and am working my way through a book on the history of eBay. Plus I'm working on the back end of the company I work for now to make it more robust. :) So my guess is that what is happening is exactly my nightmare.
Anyhow what most sites have happen is they have an initial attempt which is really a great proof of concept. If it catches on then you start to grow out the back end core, but at the same time you have to keep adding features and bug fixes. So now you are trying to do 2, or 3 major tasks. And all of it requires time and money. And brains. So then you keep duct taping things together until you can hopefully get enough money to hire the brains you need.
Meanwhile if things are going well then your users are starting to overrun your site -- this there is this huge hiccup while people rush to launch a major revision of the site that on the front side looks the same but on the back is vastly changed.
If you spend your resources keeping the current site up then you never get the revision done, so then the current version starts to come apart -- longer outages. If you don't spend your resources getting your upgrades done then you wind up with a totally unmaintainable architecture that you cannot manage and doesn't allow you to grow features (see very large job site that is quickly coming apart -- inside word is the platform is the same as it was in '98 and they are screwed).
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
no subject
Date: 2007-10-17 09:12 pm (UTC)Anyhow what most sites have happen is they have an initial attempt which is really a great proof of concept. If it catches on then you start to grow out the back end core, but at the same time you have to keep adding features and bug fixes. So now you are trying to do 2, or 3 major tasks. And all of it requires time and money. And brains. So then you keep duct taping things together until you can hopefully get enough money to hire the brains you need.
Meanwhile if things are going well then your users are starting to overrun your site -- this there is this huge hiccup while people rush to launch a major revision of the site that on the front side looks the same but on the back is vastly changed.
If you spend your resources keeping the current site up then you never get the revision done, so then the current version starts to come apart -- longer outages. If you don't spend your resources getting your upgrades done then you wind up with a totally unmaintainable architecture that you cannot manage and doesn't allow you to grow features (see very large job site that is quickly coming apart -- inside word is the platform is the same as it was in '98 and they are screwed).
Or it could be someone unplugged the server. :)