Manipulating web sites

[vvlad]

Hi everybody,

I have an ethical aspect I would like a piece of advice about.

I work for a large company with a huge intranet. We use internally a search engine to index and search for documents. Besides this, many people also use dedicated internet web sites to search for technical terms.

A colleague had an idea on how to encourage as many people as possible to use our internal search. He’s thinking about “manipulating” the response that comes from external search engines, for instance by dynamically inserting in the results page that the external search returns a link saying “You can also find the keyword you have searched for with our Intranet search …, etc”.

I hope you got the point. I’m not talking about the technical possibilities to achieve this, I already see here a lot of problems.

What I’m saying is: We shouldn’t even try to do it, in my opinion this is not ethical, not fair and perhaps not legal also. What do you folks think?


vlad

 

[grande]

Aside from using a Greasemonkey script, I'm not even sure how you would do that...

-------------------------
Matt Grande
C# Master.
Ruby on Rails Admirer.
ActionScript Student.
JavaScript Hate-Monger.

 

[sleipnir214]

The legality of all this depends on where all this is taking place, I suppose. Employers in the U.S. have a lot more control over what goes on on the network than do their counterparts in the E.U.

In terms of ethics goes, I really don't see a problem so long as you're not limiting query results without telling anyone. Internet search directory aggregators do something like what you're proposing. All you're talking about doing is doing a custom proxy of some search engine's output.




Want to ask the best questions? Read Eric S. Raymond's essay "How To Ask Questions The Smart Way". TANSTAAFL!

 

[dwarfthrower]

I don't see any problem with it. There's no reason why a company shouldn't seek to limit the bandwidth consumed by searching outside the company for things that can be found within.

Putting a note at the top of their search results pointing them to the company intranet sounds like an excellent idea to me.

[sub]Never be afraid to share your dreams with the world.
There's nothing the world loves more than the taste of really sweet dreams.[/sub]
Webflo

 

[gbaughma]

I guess I really don't see it as an ethical issue either... and as far as the technical ability, it would be pretty easy to do....



Just my 2¢
-Cole's Law: Shredded cabbage

--Greg

http://parallel.tzo.com

 

[vvlad]

Thanks for your input, although it contradicts my thoughts. Personally, if I had a web site I wouldn't like other people rebuilding it with external content in their own interest. Maybe I'm too old fashioned.

To clarify one point: the headquarters are in the E.U., but the company is spread all over the world, including U.S.A.


vlad

 

[sleipnir214]

I think you'll find that most people will live with it, so long as they know it's going on and you don't filter their query results.

I'd check with E.U. legal authorities before doing this. The E.U. laws side much more heavily with users over privacy and content, even over employers' networks.




Want to ask the best questions? Read Eric S. Raymond's essay "How To Ask Questions The Smart Way". TANSTAAFL!

 

[franklin97355]

In our school district we have a proxy server that filters the web sites that are accessable and also caches some of the more popular sites to reduce traffic and limit access by students.

The answer is "42"

 

[jsteph]

In another thread here someone posted a very interesting link that talks about this very thing:

http://youtube.com/watch?v=H9_mG_yAoTQ

It talks about how ISP's inject ads, etc, into sites, and possible technologies to defeat it.
--Jim

 

[vvlad]

Thanks Jim, very interesting video.

Finally we will not implement this idea, I’m glad to see that other people share the same opinion.



vlad

 

[lionelhill]

vvlad, just a thought, but I'm wondering if the original approach was, in any case, the wrong solution to a very real problem.

I notice that where I work, if I want to find an internet page (note: inter) produced in-house, it's much easier to find it using Google than using the in-house search. I'm also wary of in-house searching and databases on the grounds that they only find things someone in-house already knows about. The big wide world might have better ideas, and Google is good at finding them.

On the other hand, Google cannot find the brilliant idea someone down the corridor put on the intranet.

Perhaps this is a social problem that needs a combination of educational and technical answers. First, you need to make sure that your intranet search is as good as your boss thinks it is. Do users find it easy, or, when they're searching for something written by a chap called King, do they find themselves inundated by pages written by people who's hobby is drinKing? If they're looking for staff details on a lady called Lea, will it find the entire cLEAning department?

Once your intranet document resource is as easy to use as Google (tall order!), you need to get at least some key users singing its praises. You need to educate people that it will get them useful results, and fast.

Not an easy task, but if a few people stand up in meetings and say "well, actually Joe in Manchester solved a similar problem three years ago, and I found his solution/comments on our intranet" people will eventually start to notice the value.

Good luck!

 

[gbaughma]

lionelhill:

Actually, google has a version that can be used by corporations for intranets.

You can install the spider on your internal servers, and have it crawl intranet sites. Pretty slick.

It wouldn't work well for my situation, unfortunately, because almost *everything* on our intranet is dynamically generated.



Just my 2¢
-Cole's Law: Shredded cabbage

--Greg

http://parallel.tzo.com

 

[vvlad]

lionelhill

The problem you mentioned is well known in the company, we are struggling with it for years. As you said, it's not an easy task.

Our internal search is not as good as Google, but it is acceptable (anyway, it does NOT find cLEAning when searching for Lea) and we are working on improving it.

The idea behind was not to improve our search with this new approach, it was just a new idea.

gbaughma
We already tried Google for intranet, unfortunately it did not work for us.

It's just a black box that one can not manipulate in any way, and we have a lot of search applications with non-standard metadata.

Also, the Google approach to rank the sites based on links does not seem to work in the intranet.


vlad

 

[lionelhill]

Thanks for that!
Li