Good question.

That you are considering a local perspective tells me that there is local aspects of your plans that are likely to be of important to your users.
Using separate domains could be called a domain strategy to make your information seem more relevant and appeal to a sort of tribalism.

Would you be better setting up a football site with directories with clubs, or having separate football club sites. You would surely get a more passionate following on the indiviual sites. A separate site (with more targetted domain) will feel more relevant.

Could it be that seo/pr etc has obscured our vision of what the user would respond to best. Multi site technology not quite there yet to make it easy.

Well what I´m thinking is travel related. Sort of "things to see and do" around the world. Like a top ten list. The idea is a little more original but this explains it pretty well.

I could make one site where I would list the top ten things for each country.
Top ten attractions in England, France, Germany etc.

And I could make another site where I go into more detail for a particular country. F.eks. Top ten things to do in London, Manchester, Glasgow, Dublin etc. Even top ten things to do in particular areas of a city.

Either I make the first site for the world in general and then seperate sites for individual countries. Or place a directory under the mother site for each.

There are 243 entities in the world that can be called a country. Since it would take years to do do every country in detail it seems like a catch all site would be flawed, not offering the same amount of info on every country.
If they were seperated however it would not matter but then my marketing efforts would be spread among many sites (or at least two). If somebody would link to a detailed/local site the bigger site would not benefit like it would if it was a directory instead.

You are right that SEO tactics does skew ones view of what would benefit the user and the site best. From the users end it would seem more appropriate if there was a seperate site for each country. However competition in travel related websites being like it is I feel I must take this into consideration.

Has nobody else given this thought? I admit that I´m shamelessly bumping this question but I find it hard to believe that ... well read above.
There has been much hype in regards to finding a niche and profiting from it, I don´t know how many ebooks for sale (50$ plus) I´ve come across that preach that. But if you look at the biggest players; google (save blogspot perhaps), yahoo and msn they are doing the opposite. They keep all their niches under one fold. Is that a lesson to all of us? If for nothing else, if more people stuck to fewer sites it might result in fewer crappy (made for adsense like) sites.
Or what? The bottom line is visitors, has any had the experience or given this good thought?

What I thought you had in mind had a totally local component that was to be supported by visits from locals.

I'm tempted to say your plans sound more like a Lonely Planet or Roughguide scenario with people
being attracted to your site by your particular take on things. Top Ten attractions . In that case one site with you unique point of view where they fan out to a local level might be best. However when you start attracting the locals for top ten in Minor City then maybe you could consider separate local domains for a more relevant feel.

If you have one supa-site, then all your incoming links are coming to the same domain, all your visitors are too - allowing them to cross effortlessly from sector to sector, and making a mailing list, forum and blog much more viable, more quickly. All your marketing is for one domain, thus reducing dilution of effort. And you can build your brand and profitmfrom word of mouth.

so, in most cases, if at all possible, put all those eggs in one very big and fast growing basket.

Exceptions: if you do need to satisfy local markets (as opposed to local subject matter!), then a local domain and local hosting will be useful. If you are working in more than one language, separate sites are probably a good idea.

And [and this is NOT directed at you personally!] if one were to use risky methods, then one dare not put all one's eggs in one basket.

I hope that's the kind of thing you were expecting?

look at the travel SERPs today - huge sites outrank small local ones unless the small local ones are really really cool.

if you are one man band, you'd be better off picking just one city and concentrating on it only. otherwise it will be really difficul for you.

it is really easy to make a top10 tourist attractions list for each popular city in the world. i do not think it is the best idea unless you want to drill down and concentrate on details.

so my opinion is: get one domain for one city or country and start working on this one.

if you have a dozen of people working for you full time, go for a large site, but prepare for no profits for 2-3 years.

Thanks for the replies. I know a bunch of only top 10 list is not original at all and I wouldnt build such a site if there was not more too it. As it is I´m hoping to attract links from certain types of educators, .edu links are supposed to be good for ranking and that is my edge (SEO´ing again, and I don´t even know how to read a SERP or even how to get one ;) that is if youre talking something more complicated that googling for a keyword).

Taking the advice offered here I think I´m going to limit myself to two sites. One lonelyplanet/roughguide like that has travel anywhere in the world as a theme, another similar but very local about my home town. Then I´ll just have to wait and see.

It´s like launching a new website is similar to fishing through ice, throw in the bait and then wait by the hole (computer screen) for a year or so.