P2P Foundation

Researching, documenting and promoting peer to peer practices


Featured Book

The Transforming History of Land Ownership


Book Store



Admin

Subscribe

Translate

Robin Good on the Ideal Profile of a P2P Search Tool to replace Google

photo of Michel Bauwens

Michel Bauwens
25th March 2014


Excerpted from Robin Good:

Robin also give recommendations to use already available alternative P2P search engines.

“What could be the alternative search route available to us?

How could we escape the limitations imposed by the Google search model?

Proposal:

What would happen if it was me and you, individually, the ones who selected the criteria, ranking algorithms and penalization approaches to use to determine our search results?

My assumptions:

a) To make search results more useful, while becoming more trusted and much less vulnerable to being reverse-engineered and gamed by unscrupulous marketers, I don’t think there is a need to make your search engine and your ranking algorithms secret.

b) Secrecy promotes and breeds black markets, underground work and a well-defined objective for everyone: uncover the secret. Reverse-engineer it. Game it.

c) Following the patterns we have seen at work elsewhere: from centralized to distributed; from top-down expert secrecy to crowdsourced, open-sourced and distributed co-operative participation.

Here is how I see my ideal future search engine:

Open-source

Distributed – P2P

User-controlled

Imagine if:

a) Search was similar to a free public service.

b) Users could see search ranking and filtering factors and, if they wanted, they could change them according to their own specific needs and preferences.

c) “Trusted search curators” for specific vertical information niches started to become themselves the new relevant results. The idea is that they could provide the needed “trust” and transparency to search results by co-creating curated collections on the topics in which they have already demonstrated a high level of competence.

d) An ecosystem of open-source public search algorithms, filters, aggregators and curated collections of sites and resources on specific topics emerged.

e) Content indexing became a distributed activity in the hands of we the users. With this approach, individual users contribute to index and add information into a shared database aggregating each user personal index.

In this fashion, users not ONLY would have greater control of what is actually indexed, but they would actually be creating a real search commons index – a collaborative effort by all users that is available to everyone. (An example of a distributed search engine – where peers collaborate to construct their search database – is the YaCy project. More info on this Wikipedia page).

In simple terms:

Turn the search ranking mechanism upside down by giving back control to who is searching and in need of taking decisions based on that information.

Achieve this by allowing the user to see at all times, what is under the hood and to have the option to modify it, rather than achieving this by personalizing his results univocally or by differentiating them from those of others based on history, preferences or the social graph.

Key Benefits

An ungameable system.

If everyone could individually select and rank results according either to their preferences, or by utilizing user-defined filtering pre-sets, ranking plugins done by experts and niche curators, it becomes much more difficult for anyone to game search engine results, as now there would be an infinite number of different ranking systems at work.
But unlike what Google does with personalization, the rules by which results are ranked are not secretely set by Google, but it is me and you who decide how we want to slice and dice them.

For those who wouldn’t want to bother with tweaking and setting up their individual search preferences, they could be offered to select among alternative ranking algorithms (e.g.: Pre-Panda, 1998-style, etc.), or through open-source or paid-for algo pre-sets designed by users, groups or even other search engines themselves.

Commercial search engines like Google could even license their own algo as a plugin to use by paying a monthly subscription fee. Better yet, they could offer different flavours of it tailored to different audiences, applications and with different levels of customization possible.

Finally search users could also select to fallback on the preferences set by their close network of friends (on Facebook, Twitter, etc.) or even by Google (or other search engines) itself if they wished so

If you had no “online” friends and did not want to set preferences, paradoxically you could be given an alphabetical, or chronologically indexed set of results, and then you could move on to refine and distill what you need out of it, by applying on the fly, your own criteria.

If search was a distributed mechanism, there would be no central server, and thus there would be possibility to censor or block specific content.

Reliability would be significantly increased as single points of failure have been eliminated and the search index is stored redundantly across all users in the network that opt-in to participate.

It becomes possible to index and crawl contents, databases and web sites previously not accessible.
It is finally possible to achieve a high degree of privacy and to leverage the power of crowdsourced ranking.

What You Can Do About It

If you like the future of search that I am painting, you should not be sitting there waiting for it to happen. It won’t.
This future can only take place if you stop what you have been doing until now and you start doing something different.
Here a few simple suggestions:

a) Vote with your click

The first and foremost thing you can do to change the status quo, when it comes to online search, is to vote with your clicks. If Google does not offer what is your “ideal” situation, then do not support its growth and power by increasing your use of it.

b) Use alternative search tools

Start using an alternative search engine, among the ones that I suggest later in this article. Stop using Google or at least start to use it in combination with other alternatives.

c) Curate collections of quality vetted resources

If you are a subject matter expert or just passionate about a specific topic take into serious consideration the idea of starting to curate resource collections that can offer a one-stop-shop for those looking for insight into that topic.

d) Speak up, share what you find

Speak up, share, let others know. If you find better results by using one of the alternative solutions available, or if you discover something new or valuable that can help other people search in new, effective ways, share your information via social media so that others can pick up from it and move forward.

In Conclusion:

Search as we know it today, could be a million times better, if we only decided that we wanted it to be so.
While we associate searching the Internet with Google, the future of search could be quite different from what you have been used to until now.

In my view in fact, this approach to search, will not remain the major, most common and most effective search solution available in the digital realm.

What is really best for us?

A centralized, secret and proprietary search engine driven by Wall Street or a distributed, fully transparent and open-sourced one that placed each and every user / searcher in the driver seat?

Put the choice of how to rank Internet search results results in the hand of the searchers, not in the hands of those who control both the search and advertising marketplace.

Let users index, refine, develop and improve search engine ranking algorithms by applying the filters and metrics that serve THEM best, and not only the Google stock.

Move from listing titles-URLs-descriptions to curated search results, in which “trusted search curators” will provide bundles of high-quality results, selected and organized together in new emerging formats.

If Google and the other major search engines are not willing to be transparent about how they organize, filter and rank information, how can you trust that the answers you are given do really provide you with the best option possible?

Access to information should not be based on some “social” secret recipe of what is good and what is bad, – that is taken care of by religions of this world – as there is no objective metric that can measure the different needs and information requirements of each human being.

Unless I can check it.

But unless you decide that Google should not be your only key reference for finding any kind of information that exist out there, then nothing is going to change.”

FacebookTwitterGoogle+RedditShare

Leave a Reply

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>