Science Library Pad

meta: in which I make a claim

Just a posting to try yet again to claim the 2nd URL that Technorati sees for my blog

Technorati Profile

In other news, Technorati now presents a single search box, to search for tags specifically you'll have to go to advanced search.

UPDATE: It's working for both TypePad blog URLs now.  An interesting reminder about basing any decisions on a particular Technorati ranking result.

Science Library Pad      

Rank: 74,734 (84 links from 63 sites)

• URL: http://scilib.typepad.com/science_librar...

Science Library Pad

Rank: 56,727 (174 links from 82 sites)

• URL: http://scilib.typepad.com

Previously:
July 6, 2006  Technorati rank, updating, and multiple blog URLs

Everything is Miscellaneous: a review

Everything is Miscellaneous could be, at a very crude level, described as "Folksonomy explained for librarians".

This is to me the core of the first 100 pages, and a topic of substantial value.  Although I don't think Weinberger ever states this explicitly, our schooling and our concepts of the world help to shape our understanding of what is possible.  I am not a librarian, but I have to assume, at least until recently, that library school is a great deal about the physicality of books, their need to be classified so that they can be shelved in their one-and-only-one location.  The lessons one learns from that kind of thinking, the implicit understanding of "information" and how it can be used, can lead one to have a set of assumptions that are so fundamental they are probably almost invisible, difficult to expose to self-examination.

Weinberger explores the reasons for the classification systems used by libraries, helping to make explicit the choices and trade-offs involved.  Then he leads one into the possibilities opened up by the unconstrained world of electronic information.

After this strong start, I found the book drifted a bit, touching on, well, various miscellaneous topics.  There is a useful section on the importance of standardized, universal, unique identifiers, a fair bit of The Usual Discussion of The Usual Suspects (WikipediaFlickrDelicious etc.), a brief pondering of the popular topic "will we filter ourselves into a reflective bubble of self-reinforcing beliefs",  then quite a bit of philosophizing about the nature of knowledge.

The last element left me a bit wary, as it edges into relativism, knowledge as an individual construct.  Yes, it is true that there are many areas where things are "fuzzy".  What is this book about?  Well, it's to some extent "about" whatever the percentages in the tag clouds associated with it say

It's "information organization internet folksonomy tagging classification".  But it is nevertheless a book, a physical object, published in 2007 and located in my house.  I think there is a lot of danger in overstating the flexible nature of understanding; nature itself is not so understanding.  All the belief in the world will not get a rocket into space, only a very precise agreement of your design with the real physical constraints of the universe will do so.

There was a good, albeit brief section where he discusses the challenges of the Semantic Web, I would summarize his conclusions as "too big and ambitious to ever work, but whatever they come up with will be a useful addition to the web of knowledge".

The book has a few minor errors when it touches on technology, for example it says you can enter a book into LibraryThing by uploading a photo of the barcode, perhaps this is some confusion with Delicious Library or with the operation of the CueCat scanners.  It also says you can look up libraries at LibraryLookup dot com, the actual URL for Jon Udell's tool is http://weblog.infoworld.com/udell/LibraryLookup/

Librarians, library bloggers, and cataloging concepts are all well-represented here, I learned some interesting things about Dewey (he was obsessed with the Metric system, and had originally specified metric dimensions for the ubiquitous "3x5" card). 

If you're already familiar with the concept of folksonomy, or quite willing to accept that a book might be reached through many different paths of full-text search, blog entries, tags and catalogue metadata, this book may not add a lot for you.  But I know that there are still those out there who don't quite get it, rejecting full-text search and yearning for some nostalgic yesteryear of controlled OPAC Glory Days.  So if you're a librarian who is open to, but doesn't quite get this whole folksonomy thing, or if you're a cataloger who thinks full-text search is the devil's work, this would be a good book for you.

The one area I would have liked to have seen addressed - if only in the form of a pointer to other references, is an explanation of the fundamentals of how web pages and web servers work.  This fundament is what all the other linkages are built upon, and a lack of understanding of this technology leads, in my experience, to many fundamental misconceptions about link resolvers and the web in general.

I find myself wishing I could give two ratings on LibraryThing, with different semantic context.  For people who already know about folksonomy, or who have read similar books, I would say three stars.  For librarians, catalogers, businesspeople and others who want to understand how the digital world of unlimited categories and tags differs from the physical world of one object, one location, I would say four stars.

For me, I am waiting for the next generation of tools that exploit even more information about and within books.  As an example, I'm guessing that very few tools would be able to tell you that both in this book

It was Jean-Baptise Lamarck--unjustly remembered primarily for being wrong about how giraffes got long necks--who not only sorted out Linnaeus's worms but changed the basic shape of Linnaeus's tree.  As Gould recounts the story, Lamarck loved invertebrates so much that when he was almost fifty, he was appointed professor of insects and worms at the Muséum National d'Histoire Naturelle.

and in Paris to the Moon

The entrance to the paleontology museum at the Jardin des Plantes is graced by a statue of Lamarck, with the engraving "The Father of Evolution," in giant letters, on its pedestal.  Darwin, on the other hand, is nowhere in sight.

two books which could not have more different topics, we yet converge on a location in Paris, a location where I happened to be just weeks ago.  When we start to develop tools that help us make these kinds of unexpected connections, I will know that we are finally starting to mobilize the information in the containers our libraries have so carefully catalogued.

UPDATE 2007-05-27: There is a good video of Weinberger presenting about his book at Google.  Via Thingology.  ENDUPDATE

Misc: I read this book because it was mentioned in the LibraryThing blog.  The dedication of this book is "To the librarians".  There is a review in the ALA Techsource blog.  The book has a blog at http://www.everythingismiscellaneous.com/

ISBN-13: 978-0805080438

Wikis, blogues, folksonomie et autres

Links from a presentation given by Élisabeth Lavigueur in Quebec

Liens de la présentation du 9 février 2007

Amapedia - Amazon Product Wiki next generation

I'm not sure this effort is going anywhere, but what I find interesting is the amount of content that Amazon is able to get customers to contribute for free.

http://amapedia.amazon.com/

via Ten Thousand Year Blog

Previously:
November 26, 2005  Behold the power of Wikizon: Amazon adds ProductWiki

building on data: Yahoo TagMaps

Q. What are TagMaps?

A. TagMaps are a new way to visualize text on geographic maps. TagMaps can be used to communicate key characteristics of location-based data in an easy-to-understand way.

...

Q. How can I contribute to World Explorer?

A. You may have contributed already! If you have a public, geotagged image on Flickr, then it was part of our computation. If you added tags to the photo, these tags were also analyized - and your photo might appear in the visualization (look for it!). We update our tag maps every few weeks, so newly uploaded photos could influence the next version of World Explorer.

http://tagmaps.research.yahoo.com/

via Flickr Groups - Geotagging Flickr

presentation on library services 2.0

You know it's strange, I don't really browse videos on YouTube or photos on Flickr, but I do find all sorts of stuff of interest on SlideShare.  I guess because I am an information junkie.

I found a great presentation Services 2.0 dans les bibliothèques vers des bibliothèques 2.0 ? that covers the various aspects of integrating your library into web browsers and into other web sites, as well as improving your own library site.

The author of the presentation blogs at BibliObsession.net.  It's in French.  Be warned there are some possible NSFW images there.

(If for some reason you're interested in the discovery chain, it was via Mediapolis who favorited my Web 2.0 presentation, to the presentation Blogs et wiki : outils et phenomenes - it was third in the "related slideshows".)

my Web 2.0 presentation to InfoConference

Here's a presentation I did on a few Web 2.0ish topics, with LibraryThing as an exemplar

http://www.slideshare.net/scilib/web-20

You may want to click on the "full" icon (lower right of the presentation) in order to get the fullscreen view.

Interestingly, SlideShare uses Amazon S3 for storage.

The presentation was given to a CISTI internal event, our yearly InfoConference for information specialists.  It was originally done in Keynote 3, exported to PowerPoint for the purposes of SlideShare import.  In uploading it, I was as usual reminded of my difficulty in tracking as-presented files.  By the time it's gone through various drafts on various local and network directories, circulated in email a few times, and passed through a laptop and USB drive or two, it's a wonder I can still find the file at all, let alone know which version I actually presented.  This one has been very slightly redacted to remove some internal info.

The main points of the presentation were:

1. it is difficult to predict the future
2. Web 2.0 is very new (about 5 years old at most)
3. we can extract some basic characteristics of Web 2.0 applications, as demonstrated by LibraryThing

http://www.librarything.com/

These characteristics are:
- a personal or person-centric nature ("your library")
- user created/generated/submitted content ("add books")
- novel, informal description and discovery tools ("tags" / folksonomy)
- recommendations generated based on human-generated content and relationships, or human intentionality ("pssst")
- social networking ("ratings, reviews, shared books")
- open, active participation (discussion groups, easy inward linking)

4. In addition to text, audio and video, there is a new type of data being added to the web: geographic information.
4a. As a secondary note, the patterns of the (now over 6.5 million) geographically tagged photos tell us something about the profile of active web users - the Flickr map of photos

http://flickr.com/map

to me bears a striking resemblance to the classic "earth at night" satellite view, and I have also seen this pattern in Google Analytics maps of web traffic.  It seems that it is mainly in the developed world, in Canada, the USA, Europe and Australia, that people have the free time and wealth to be able to afford to participate in these kinds of peer production activities.

5. Web 2.0 is also opening up science to the public - I give four examples of projects that invite public participation.
5a. As a secondary note, these projects have a relationship to the Amazon Mechanical Turk concept - it turns out that there are many tasks for which a crowd of humans is dramatically better than machines (artificial intelligence).  Humans are particularly good at visual recognition tasks.  There is a general Web 2.0 idea I think, that people (if you have a big enough crowd) are better in some sense at producing content, relationships, and certain types of analysis than machines.  Machines are better at some classes of meta-analysis, e.g. deriving recommendations by data mining Amazon user intentionality.

6. Blogging is growing tremendously (in fact, all Web 2.0 graphs seem to have a sharp upward curve).  That means a lot of junk, but also a subset of valuable content, including serious scientific content.

Although not covered in the presentation, I also think Second Life and World of Warcraft (WoW) tell you a lot about the kinds of efforts that people are willing to put into online activities.  To me WoW indicates that people will actually tolerate a very high level of complexity and tremendous time investment, if the (perceived) reward is very high.  (WoW has over 7.5 million paying subscribers, far outstripping other online 3D environments.)

Slashdot bookmarks and tags

Slashdot has a bookmarking and tagging system, but it's a bit awkward.
Once logged in, you can bookmark manually at

http://slashdot.org/bookmark.pl

They also have bookmarklets to do the same thing.
You can add tags, but onewordonly, no "multiple word" tags.
To edit a bookmark, the only way I could find was to go to Bookmarks->Recent and click on the "bookmark" link next to the one you want to change.

http://slashdot.org/bookmark.pl/?op=showbookmarks&recent=1

You can view your tags at http://slashdot.org/my/tags but you can't actually do anything with them yet.
There is some cross-integration between bookmarks, your journal, and story submission.

You can even get yourself a little "Slashdot Me" badge.

   Slashdot It

There's a FeedFlare to SUBMIT a Slashdot article, but I couldn't find one to bookmark an article, so I just made one.

Download slashdot-bookmark.xml

Flickr Uploadr resize loses metadata

Summary: If you want to preserve EXIF-GPS and IPTC metadata in your photos when uploading to Flickr, do not use the auto-resize feature in Flickr Uploadr.

Details:

Picasa allows you to embed some useful metadata directly in your photo files.  These metadata include IPTC keywords, IPTC caption, and EXIF-GPS location information.  Flickr has the capability to automatically detect these metadata, saving you from having to retype or re-create them (keywords are turned into tags).

I noticed some odd behavior with the Flickr Uploadr, so I decided to do a systematic test.

Using Picasa 2.5 + Google Earth
* geocode image
* IPTC tag image
* IPTC caption image

Windows Uploadr 2.3 (and 2.5)
Mac Uploadr 2.2

Scenarios
1. Windows Uploadr (already resized)
2. Windows Uploadr (auto resize to 800)
3. Windows direct web upload
4. Mac Uploadr (already resized)
5. Mac Uploadr (auto resize to 800)
6. Mac direct web upload

1 IPTC ok, EXIF-GPS ok
2 IPTC no, EXIF-GPS ok *
3 IPTC ok, EXIF-GPS ok
4 IPTC ok, EXIF-GPS ok
5 IPTC ok, EXIF-GPS no *
6 IPTC ok, EXIF-GPS ok

So if you resize in Windows Uploadr, it will lose your IPTC tags and keywords, while, just for variety, if you resize in Mac Uploadr, it will lose your EXIF-GPS.

UPDATE 2007-08-09: Incredibly, Flickr Windows Uploadr 2.5 still loses IPTC metadata on resize.  ENDUPDATE

UPDATE 2008-09-28: Flickr Windows Uploadr 3.0.5 doesn't have this issue, IPTC metadata is preserved on resize; they have fixed the problem.  ENDUPDATE

I used an image of the London Millennium Wheel.
First I added IPTC keywords using the little-known CTRL-K option in Picasa.
The only drawback is you can't have multi-word keywords.

I also added a caption in Picasa.
(Minor disclaimer: I initially spelled Millennium wrong and fixed it after.)

Last but not least, I geocoded the photo using the Picasa-Google Earth integration (you can read more about this capability in a previous posting).

With the file now geocoded and IPTC tagged and captioned, I verified it in IrfanView.  (I also used IrfanView to do a resize of the photo to 800x600 for some of the upload tests.) As you can see from the IrfanView photo information inspection, the image now contains both EXIF-GPS and IPTC keywords.

Now, when it all works, it's great.  The photo arrives at Flickr and the metadata are automatically read and filled in for you.

But when you let Flickr Uploadr resize the image on Windows, goodbye IPTC info

On the Mac, just for variety, Uploadr resize keeps the IPTC, but loses the EXIF-GPS.

I hope you've found this report useful, and I hope it leads to Flickr fixing the Uploadr immediately.

Here's the photo itself (the #1 upload, with all the metadata intact)

meta: advanced TypePad posting

TypePad posting has a lot more capabilities than I ever realized.  Including split postings (introduction plus continuation) and now Technorati tags in addition to categories (note: categories are also considered tags by Technorati).

To access these features, even if you have the lowest-end Technorati account (like me), just go into the posting screen, click "Customize the display of this page" at the bottom, and select either Advanced or the particular Custom features you want.

conference tag goodness with HitchHikr

HitchHikr (by David Warlick) brings together a couple different conference aspects in one place.
Mainly, letting you specify a conference tag (allowing me to retire my unusued wiki List of Library Conference Tags).

As far as I can tell, it basically provides a central conference inventory, and then uses all the tags indicated to query Technorati for postings and Flickr images.

Here's what he has to say about choosing tags in the FAQ:

First of all, I think that you need two tags for a given conference event, assuming that it is an annual or otherwise regularly held event. One tag should apply to the conference as an organization or concept. For instance, for the North Carolina Educational Technology Conference, a general tag would be ncetc. This tag would bring up all blog, podcasts, and flickr images associated with the conference regardless of the session, or any actual occurrence.

The second tag would be built on the general tag, focusing on a specific conference event or date. Typically, for an annual conference, this would require adding the two digit year. The next North Carolina Educational Technology Conference would be tagged ncetc06.

Finally, tagging blog entries and photos with specific conference presentations or features can be accomplished by adding an acronym for the feature or the speaker at the event. For instance, I might ask attendees who blog my presentations to tag them with ncetc, ncetc06, and ncetc06warlick. I, as a presenter, would need to visit Hitchhikr.com and add that tag to the conference.

I've been thinking about this for a while.
Personally, I think the tools should be mature enough that we can combine tags.  Taking his example, we end up with both

ncetc ncetc06

To me, if our tools were good, we'd just do

ncetc 2006

and search on the intersection.

I'm also anti-two-digit.  As a CS person, particularly since we just turned the century mark, it bugs me to see two digits when I know this will break in a hundred years.  Also, 05 could mean "the 5th conference", not 2005, as I mentioned in a previous posting.

I do very much like the idea of adding the first presenter's name as a standard way of tagging presentations.  And put this at the end of your presentation, so people know what to use.

In case you're wondering why I have such an interest in conference blogging, this blog started in November 2004 as a more professional (e.g. less strewn with cat photos) hosting space for my presentation notes from Internet Librarian 2004.

Wishlist:
It would be nice if you could automatically add conferences to the calendar tool of your choice (Google Calendar, Outlook, iCal, etc.)

HitchHikr found via Disruptive Library Technology Jester.

Previously:
April 30, 2006  Postgenomic - conference review markup
April 24, 2006  Google Calendar experiments
November 07, 2005  conferences, tag collisions, the power of AND
October 26, 2005  blog and wikis for conference coverage *****
October 24, 2005  IL05 - Internet Librarian 2005 - blogging and such
September 05, 2005  Internet Librarian International 2005 in 5 weeks
July 05, 2005  more ideas for conference websites and webtech *****
June 27, 2005  Gnomedex conference
June 07, 2005  how to tell if you're at a geeky conference
May 02, 2005  in which everyone finally starts to get wikis
March 17, 2005  some thoughts on conference blogging
February 16, 2005  more on conference blogging
February 12, 2005  OLA! more grumbles about conferences
January 17, 2005  SLA blog and wifi?
November 20, 2004  very old conference notes

ZoneTag - Flickr auto geocoding for cellphone pics

Find your photos by location
ZoneTag can automatically tag your photos with the location, based on the cell tower, [near] which they were taken. The location of the cell tower is not known until our user community updates the city, state, or zip code of their photos on Flickr. ZoneTag is then able to map the cell tower to the location.

http://zonetag.research.yahoo.com/zonetag/

via TechCrunch - Zoomr

Previously:
March 12, 2006  a couple neat photo tagging and geocoding apps

a couple neat photo tagging and geocoding apps

One neat thing about having a blog is sometimes info comes to you, rather than you having to seek it out.

Via a comment on universal photo tagging - almost, I find

ITag - Photo Tagging Software

ITag is a free tool that is designed to make editing of key IPTC fields of your digital photos very easy.  The editable IPTC fields are Headline, Caption and Keywords (which effectively translates into title, description and tags).  The IPTC data can be read by other software such as IrfanView and Flickr.

Geocoding your photos is also straight forward by drag and dropping Google Earth placemarks (*.kml).  When geocoding, the data can optionally be saved as IPTC tags (Flickr style), to the EXIF fields or both.

Geocoded photos can be opened in Google Earth to see the location they were encoded with.  Photos geocoded with WWMX Location Stamper are also recognised and can be viewed in Google Earth.  Groups of photos can be selected and the photo's viewed in a single kml - like this.

There is also an option to convert GPX tracklogs to KML (Google Earth) format.

WWMX Location Stamper

checks a digital photo's embedded timestamp, correlates it to GPS track files loaded in GPX format ..., then stamps the image with latitude and longitude location information.  Don't have a GPS device?  This application can also be used to drag and drop your images onto a map in order to stamp them.

That's pretty neat.  Both apps require Windows with .NET installed.

For the Mac, a Google search turned up GPS Photo Linker.

The functionality of these apps actually raises an issue, which is

Dear Google,

Instead of (or in addition to) stuff like buying Writely, it would be nice if you would enhance your existing applications.  For example, Picasa - Google Earth integration would be a natural (UPDATE 2006-06-18: Google added this feature.  Thanks!).  Enhanced use of IPTC in Picasa and Google Image Search would also make sense.  Also, what happened to the Picasa blog?  http://picasa2.blogspot.com/ now gives 404 not found.

But also

Dear Digital Camera Makers,

Please have your cameras automatically geocode, e.g. by talking to a GPS over Bluetooth.  Or have it talk to my cellphone, which may have a variety of ways to determine where I'm located.

UPDATE 2006-04-02: Also check out my new page on how to geocode photos.

reader2 web book thing, plus coding thoughts

http://reader2.com/

• Social list of books - share your reading experience with other people.
• Recommendation system - find new books to read in the categories of your interest.
• Folksonomy - everybody can use his/her own keywords to categorize books.
• Faceted classification - predefined types of keywords for authors, reading statuses.
• Friend finder - easy discover people with similar reading tastes.
• Virtual community - discuss books and authors with other readers.
• Exportable and trackable - via RSS feeds and JavaScript snippets.

Am I permitted to add...

• does what normal humans actually expect their public library websites to do

Now, that being said, don't expect miracles,  The site only reports "34499 total books", and the revenue model and plan (or as rather gleefully described on the site "global world takeover project") appears to be mainly around driving traffic to Amazon.com  However, it does also link to Wikipedia Book Sources, which should land users at your library (you've set your library up in Book Sources, right?)

I like the status display of books people are current reading.

I also found the creator's reference to Rent A Coder in the site's blog interesting.
See, you don't have to be a coder.  Just have ideas, and rent the rest.

As I've said before, my recommendation is for libraries to get together, build a common front-end layer, then build your local community while transparently using Amazon's data.  (By transparently I mean: user enters review on your site, it also gets copied to Amazon; conversely, if a user in your area puts a review directly on Amazon, it should show up on your site.)  The reason to do this is Amazon has the network advantage - they have way more info than you will probably be able to generate in your relatively smaller local library community.  You should be able to easily switch modes: "what has everyone on Amazon tagged this book" vs "what tags have local users put on this book".

EA and SOA info overload, courtesy of John Gøtze

A couple resources with an overwhelming amount of information:

Gotzeblogged Blogging eGovernment - from Enterprise Architecture to eDemocracy

GotzeTagged

GotzeTagged is an open content linksharing service with thematic, categorised and annotated links on a variety of subjects.

In one word, GotzeTagged is a personomy, a one-person folksonomy/tagsonomy. The personomy belongs to me, John Gotze ...

The portal has been augmented by the use of various services, such as del.icio.us, MSN News, Yahoo News, and Amazon.

He has many links gathered for Service-Oriented Architecture, Enterprise Architecture, and related topics.  As well as indicated above, the individual pages for each tag pull info from many sources including Technorati, delicious, MSN Search, Yahoo News and Amazon.

Dr. John Gøtze's bio reads in part

I preach and teach e-government, enterprise architecture, standardization, openization, governance, digital leadership, strategic planning, communities of practice and much more.

...

I'm also a non-tenured associate professor at the Department of Informatics at Copenhagen Business School, where I teach a new course on enterprise architecture and run an EA master's class at the IT-University of Copenhagen.

That's kind of cool, because CBS is where I attended Info Grid 2005.

Amazon does tagging

Alan Taylor and Anil Dash report that Amazon is rolling out tags in a big way. In a discussion board message, Amazonian Blake Scholl announced that 50% of Amazon customers will see the tagging features at this time.

from Read/Write Web: Amazon Does Tagging

Another potentially huge surge in user-entered metadata.
Interesting to ponder how to integrate it.
Do you have your own separate tag cloud for your catalogue?  Do you say, "on Amazon, people tagged this book with..."?  So many folksonomies, so little time...

conferences, tag collisions, the power of AND

The California Library Association conference is going on (November 4-7, 2005).

Logical tags would be the usual suspects: CLA2005 or CLA05.

Only problem: the Canadian Library Association conference was June 15-18, 2005.

Logical tags were: CLA2005 or CLA05.

It's going to be in Ottawa next year and I was certainly planning on tagging it CLA2006.

Hmm, problem.
In fact, when Librarian In Black wrote that she was blogging CLA, I was like err, wasn't that conference months ago?

So even in this small field of ours, the simple folksonomy may not be handling things so well.

(As a very minor aside, the toplevel domain for the entire country of Canada is .ca, which also happens to be the mid-level domain for California.  For example, California government is www.ca.gov, Canadian government is www.gc.ca  Don't ask me why they made the country codes only two letters.)

Plus which, this single tag thing is not going to scale.  Want to see all the postings for five years of conferences?  I guess you have to search ILI2005 ILI2006 ILI2007 ILI2008 ILI2009 ILI2010.  Or maybe you search ILI05 etc.  But wait, is ILI05 the 5th ILI conference, or the conference in 2005?

Can Technorati do boolean tags (it's not clear to me) and if it can, shouldn't we be tagging with two tags ILI, 2005 so we could search e.g. ILI AND 2005?

And using a conference tag registry to prevent collisions?
Hmm, if only someone had made one.
(I might move it to LibSuccess, that seems a more logical location.)

Who shall tag the taggers?

Supr.c.ilio.us is a tag site for tagging tag sites.  Is that meta enough for you?

via Tag Tuesday

Technorati Blog Finder

Technorati Blog Finder is a new venture into the blog directory scene.

Good thing about this is the blog categories are assigned by the user (up to 20)…the system already suggests 20 categories based from the most frequent categories you use on your blog, but as mentioned you can change these to your liking.

You can search for a category or browse the cloud, the ranking (for the list) or weighting (for the cloud) are still based on the usual most incoming links, so they are like topic based 100 lists.

Otherwise sort by recently updated (this is definitely needed as a blog with heaps of incoming links may stay on top even though they haven’t blogged in a while), or alphabetically (great for those on the bottom to be seen, unless you don’t have many incoming links and your blog starts with “Z”, well then you can’t be helped as of yet).

via Library clips

UPDATE: I have tagged SLP with

Tagged                 Weblogs, Science, Searching, Conference, RSS Feed Tools, Wiki, Digital Library, Research Tools, OPAC, Service Oriented Architecture, Folksonomy, Web Services, Technology Foresight, Firefox Extensions, XML, E-science, Open Access, OAI, Publishing, Library

I may clean these up a bit.  A few of them were suggested by Technorati but I filled in the rest myself.
                        

RawSugar tagged bookmarks

RawSugar is another service that lets you tag bookmarks, search the corpus of tagged pages and such.  You can make collections on particular topics, e.g. here's one for Ajax programming resources.

category changes

My TypePad categories:

ILS has become OPAC
Cyberinfrastructure has become E-Science

Also see my Furl: OPAC, E-Science.

recommender systems

I've noticed some buzz around recommender systems.
Personally I'd like it if my blog was smart enough to automatically link to my own similar postings.

LibraryClips had a couple postings:
Related by RSS
Related articles

It seems that Waypath can provide its guess at related postings and related Amazon books (Waypath Related Stuff).
You can do this with plugin modules for various blogging platforms, or as a bookmarklet, or you could get an RSS feed and turn it into HTML.

I'm going to use my previous environmental scanning posting as an example.

http://www.waypath.com/query?type=url&x=0&y=0&key=http%3A//scilib.typepad.com/science_library_pad/2005/07/environmental_s.html

You can get best match by putting &mode=best at the end

http://rss.waypath.com/query.xml?url=http%3A%2F%2Fscilib.typepad.com%2Fscience_library_pad%2F2005%2F07%2Fenvironmental_s.html&mode=best

or most recent matches with &hours=24&mode=new

http://rss.waypath.com/query.xml?url=http%3A%2F%2Fscilib.typepad.com%2Fscience_library_pad%2F2005%2F07%2Fenvironmental_s.html&hours=24&mode=new

I can't figure out how to do "find all posts that reference this SITE" though.

Findory also provides its own related postings guesses, on a per-site basis.

Here's a page with code to see related postings inline

http://findory.com/inline/source?source=Science%20Library%20Pad&ib=1

You can also get it as an RSS feed

http://rss.findory.com/rss/Blogs/?ras=Science%20Library%20Pad

You can embed the Yahoo Y!Q related-content search into your site.  Or you can use one of the Y!Q Firefox extensions as you browse.

The frassle environment also can provide other people's categories that are related to yours.  Via Shimon Rura's blog posting Blogs and Libraries.

Jon Udell has been doing thinking about how to find postings related to your interests: More delicious collaborative filtering.

You really should be able to have a visual navigation environment:
- find me all postings that link to this one
- find me everyone who has bookmarked / categorized this post
- find me closely related postings
- show me postings related to those postings (and so on...)
- position this posting in a web "discussion thread", if applicable (e.g. this posting refers to posting X, and is then referred to by posting Y)

Any other ways to integrate recommender systems with your website or as you browse?

previously:
2005-Feb-04  Yahoo Y!Q contextual searching

spam and tagging

The unfortunate thing about spam is that it's easy for one person to generate, but it takes the entire community of non-spammers to fight it.

I get a flood of email spam to my domain and to long-established accounts, right now I have over 3000 spams since my last cleanup in May.

The spam load on my blogs so far is fairly light and easy to clean.

The ILI2005 wiki got spammed, and the IL2004, but only once each, again easy to clean once detected.
I basically have to watch the wiki change feed and check each change.

Even the RSS aggregation feed got spammed, some postings to a bunch of identical spam blogs, full of keywords, one phrase of which matched my running search.

I have also read, but not experienced, about recent Flickr spam.

So I guess my doubts about tagspam (spat?) are probably optimism, it is probably inevitable that there will be spam in the folksonomy.

delicious has rolled out a new "for:" tag, which lets you direct your bookmarks to particular individuals.  As already pointed out in the comments for this feature, this seems like a tempting venue for spam.

I don't see how we can really combat this other than by having good tools and by working as a community.  Having good anti-spam laws also helps.

Del.icio.us Toolbar for Firefox

The guys at del.icio.us have launched a “very preliminary del.icio.us firefox toolbar at http://del.icio.us/toolbar/

via loose wire: The Firefox Del.icio.us Toolbar.

litinform aggregates categories

I discovered today that Litinform.com also aggregates categories.

Here's their page for ILI2005, the Internet Librarian International 2005 tag

http://www.litinform.com/category/ili2005