You could split Google in Search/Chrome/Android/Cloud, and that would help in reducing they monopoly, but if there is one thing I could change it would be to decouple ranking, filtering and targeting from the rest, and open them up.

There should be multiple rankers on top of the index, users should be able to choose and create new rankers. Same for filtering - what gets hidden/censured, put the power back to the users, let them select their filter lists. Instead of one true ranking and filtering, let users customise it as they like.

That's a neat idea, and it might help - but it's not the kind of problem governments would chose to solve.

Or put another way, governments would chose to intervene by helping to create more 'competition' - which creates product variations approximating true market needs - as opposed to trying to 'legislate features'. 'Feature lists today' but something else tommorow. If the market is healthy, it will move in the direction of value, is the idea.

Governments can also help by promoting and supporting standards.

I always thought YouTube should be the first thing that is split from Google.

Spidering and building an index is relatively easy. It's not the barrier to creating a search engine, and I don't think you'd find that Microsoft's index is materially smaller than Google's. The hard part is figuring out how to turn that content into relevant results.

Only 2 companies in the US have an independent English based index with the contents of the entire web. Granted, the sheer volume of data is a barrier to making the index but removing that, only 4 US companies have crawled the entire internet. I'm going to have to disagree with you on that one. To write a crawler capable of the scale and timeliness to crawl the entire web in a week or two requires some pretty solid engineering. I don't however disagree that building a good search is also difficult.

Google and Microsoft were not the first web spiders.

A famous example, Yahoo!, didn't walk away from search and partner with Microsoft because of the difficulty of building an index. They did it because it was going to cost billions per year to try to keep up with Google in producing results.

I'm not arguing there's no work to do in building an index, but the problems of crawling and indexing can be solved by cash. They're a moat against small challengers, but not against well capitalized ones. Ranking and filtering require lots of research and tuning. This is the moat against even the well capitalized.

Put another way: do you really disagree that Google would still easily dominate search based on result quality even if small startups got access to their index data?