How I would handle this would depend upon the performance of the ecommerce website and which entrance paths via the website convert higher.

You could easily instruct search engines not to index the search results page by adding the following in your robots.txt:-

Disallow: /search.php?search=*

But is there a real likelihood of duplicate matching content with your actual category pages? It's unlikely in all honesty - but depending on your website content and product range, I suppose possible.

If many visits to your website arrive via indexed search result pages, I would be inclined to leave them indexed however and implement measures to ensure that they won't be flagged as duplicate content.

Ways to handle this depend on your ecommerce provider and it's capabilities sometimes but more often that not, is just a case of ensuring there is plenty of unique content on your category pages (as there should be) and there is no chance of other pages of your website hindering their ranking potential then.

It amazes me that there is still discussion about dofollow social bookmarking websites, article spinning, automated backlink sourcing and the like.

Focus your attentions on other efforts that matter, other efforts that will prove genuine, legitimate and natural.

Sure, if you feel a social bookmarking websites such as Mister Wong will drive targeted traffic to your website, go for it. The likelihood is, it won't. Social bookmarking has had it's day really, few of them are really great traffic drivers and even those that are, the value of the traffic is questionable.

I wouldn't recommend wasting your time on sourcing dofollow links only, it's not natural, it only indicates to a search engine that you're trying to game their system. A handful of relevant nofollow links that send traffic is going to prove far more useful than a handful of dofollow links from websites that no-one ever visits.

There are plenty of automated tools on the market to produce what you're after, but expect to get penalised further down the line.

The allow directives need to come before the disallow directives for the same directory/file paths. (I have never personally tested this although it makes logical sense to instruct a robot to access one particular path within a directory structure before it sees that it is blocked from crawling that directory).

For example:-

Allow: /profiles

Disallow: /s2/profiles/me

Allow: /s2/profiles

Allow: /s2/photos

Allow: /s2/static

Disallow: /s2

As per how Google have formatted their robots.txt.