Library clips :: May

May 10, 2006

Newspaper generator

Filed under: tools

Micro Persuasion has a post on creating an image of the front page of a print newspaper by adding some text to a box and hitting submit…the image can be saved as a file…this service is called the Newspaper Snippet Generator.

How about whacking in an RSS feed, and then generating the stories for the newspaper as an e-book type result.

I think we will see something in this space soon, as people will take to creating their own newspaper…lots of fun even for non-web heads.

Some movers in this area:
RSS Star
myBroadSheet

My initial idea:
RSS Daily Newspaper

Comments (0)

View blog reactions

Print Friendly

Link Leecher : generate a list of links from a webpage

Filed under: tools, opml

Check out Link Leecher, enter a URL and it will make a list of all the hyperlink URL’s on the webpage.

eg. www.grazer.com

If I filter this for the term “api” I have limited it to only 2 items out of the initial 5

…you can also limit it by document format extension.

This is very similar to Tony Hirst’s bookmarklet, only this version outputs into an OPML URL, whereas Link Leecher outputs in a browser, text file, or CSV file for excel.

There is also a bookmarklet.

With these types of tools how do you stop it from collecting links other than story links eg. navigation links?

Also, what if you could say “generate a list for all the links on this page that have the hyperlink label “google”.

I’d also like it to temporarily mark up the actual webpage…each hyperlink in the body of the post could have a number next to it, this number corresponds to the number in the generated list, acting like a footnote. I’d like to do this in order to print it out, so when I’m reading a print version I know the URL of a given hyperlink (this can sometimes be helpful for some added context, as URL’s can sometimes contain the title and/or date of a blog post).

See my post, OPML of links on a given page, which points to a tool that can do this very thing…if Link Leecher incorporated this feature as well as OPML output, it would be a handy multi-purpose tool.

Comments (0)

View blog reactions

Print Friendly

Tagground : meta-bookmarks

Filed under: tags, search

Tagground is a type of meta-bookmark search tool, check out the FAQ.

So far they are tracking RawSugar, MyWeb2.0, and del.icio.us…more to come.

There are two ways you can search:

Search for tags by URL
- enter a URL, and you will be returned tags that have been applied to that bookmark
- if I click on one of these tags it will do the next kind of search mentioned below
- next to each tag it will list the services that used this tag for this bookmark, clicking on a service launches to that service

So if I entered the URL www.google.com, the results show heaps of tags, eg. search engine, google, internet, search, links, engine, etc.
eg. next to the tag called “engine” are the icons from the services that bookmarked the google homepage with the tag “engine”, in this case the google homepage URL has been bookmarked and tagged “engine” 10 times (in RawSugar, del.icio.us, and myweb2.0)…if I click on the del.icio.us icon it will launch to the “engine” tag page…http://del.icio.us/tag/engine.

The results are ranked by the highest amount of times a bookmark has been tagged with the same tag name…so in our example the most frequent tag (61 times) used for the google homepage (according to the 3 services) is “search engine”…there are 115 different tags used for the google homepage from these 3 services combined…librarians can have a heart attack. Actually, I think librarians are coming round to the idea of personal context for some type of content.

Here is the first 10 most popular tags for the google homepage:
1. search engine 61
2. internet 38
3. links 15
4. google 13
5. search 12
6. search engines 12
7. engine 10
8. searchengine 9
9. reference 8
10. googlebrowser 5
…this keeps going to no. 115.

Here is some maths, the google homepage has been bookmarked 153 times from just the top 10 tags.

NOTE: no’s 5 and 6 have both been used 12 times (I’m not sure how ranking works in this case).

NOTE: These figures don’t make sense, because when I look at the URL history for the Google homepage in del.icio.us, the tag “engine” has been used 408 times, whereas Tagground says it has been used only 10 times from 3 services combined.

NOTE: this del.icio.us URL http://del.icio.us/tag/search+engine wasn’t an icon for the tag “search engine” in the Tagground example mentioned above, as del.icio.us only accepts single term tags…but as you can see, the google hompage has been tagged with these 2 tags together in del.icio.us (just scroll to see). I was lucky to see a result on the 1st page, otherwise I’d be scrolling pages forever, not knowing if it would turn up…as the URL history page doesn’t show tag intersections.
At the moment, in del.icio.us and others, we can see how many times a tag was used for a given URL (via the common tags list view), but we can’t see how many times a tag intersection was used for a given URL, eg. search+engine.

I’d like to see a bookmarklet for this service, so next time I bookmark a page, I can see the most used tag, or I can see the least used end, and maybe propagate the long tail.

Or if I want to re-tag lots of my bookmarks, I can just consult this service to see what the preferred tag is across 3 services…I can already do this in del.icio.us via the URL history page, but 3 heads are better than 1.

Moving on, if you click on one of these tags it will do the next kind of search.

Search for URL by tags
- enter a tag and it will return all the bookmarks with that tag…pretty straight forward

I entered the tag “opml”, this returns a list of URL’s with the number of times each URL has been bookmarked with this tag across 3 services.
eg. the first hit is the URL grazr, it has been bookmarked and tagged “opml” 4 times, by 3 services…I find this hard to believe, in del.icio.us alone the URL for the grazr homepage has been tagged “opml” 65 times.

Here I’d like to see a list of related tags, eg. TagCentral.

More

Next I’d like to see searching for a term within the title/description field, then I would like to see searching full-text…a human indexed search engine.

Even search for a tag, then search full-text within all the bookmarks that have this tag.

For every hit in the search results, you would hover the mouse over to see a list of tags…clicking a tag will show all bookmarks with that tag from all the services (maybe sort it by service or just mixed results)…see more.

NOTE: I just noticed Simpy has now been added…via the blog.

Find a list of similar services at, Keotag and friends : tagging tools