 Wednesday, August 2, 2006
Tuesday, June 20, 2006
| Data deluge meets data services # |
|
As we in IST: Data Services (DS) are gearing up to scope the problems that we will be taking on, it's helpful to learn about what our colleagues in the data services arena are pondering and doing. This weekend, I listened to Clifford Lynch's talk The Data Deluge Hits Campus. Today, I came across a new article The Chronicle: 6/23/2006: Lost in a Sea of Science Data in The Chronicle of Higher Education, which is background material for the Data Deluge online colloquium on Thursday, June 22 at 11am PST/2pm EST. Prominent among the people mentioned are:
-
James M. Caruthers Professor, Purdue University Chemical Engineering
-
Clifford Lynch
-
Susan Gibbons, Associate Dean of Libraries, University of Rochester
-
Steven Beckwith, Professor of Physics and Astronomy, Johns Hopkins University
-
James L. Mullins, Dean of Libraries, Purdue University
-
D. Scott Brandt, Professor of Library Science and Associate Dean of Libraries, Purdue University
-
Chris L. Greer, an advsior on cyberinfrastructure to the NSF
-
Sylvie M. Brouder, Dept. of Agronomy, Purdue Univ.
I'm going to ponder some questions in preparation for Thursday's talk.
Posted by Raymond Yee on 6/20/06; 3:51:01 PM
from the Unclassified dept.
Discuss
|
|
| Project management course and books # |
|
Two weeks ago, I took a course taught by Alex Walton called "Effective Project Management" (http://ls.berkeley.edu/mail/webnet/2006/0021.html). The course was sponsored by the Technology Program Office on campus. I found the course immensely rewarding and useful. Even though I knew a lot about the theory of project management and have managed many projects on a more informal scale, I felt empowered by the practical tips that were taught. I was reminded once again about how important it is to be part of a community that learns together to construct a common language and way of working. When I read about project management in the past and applied those practices in isolation from my immediate coworkers, I found it difficult to sustain those practices. My hope is that those of us on campus who have gone through the training can now help each other and our colleagues who have not been through the course to better manage our projects. Perhaps the Business Process Analysis Working Group will provide the much needed vehicle for long term sustainability of good project management at Berkeley.
While I figure out whether and how to share the course materials and know-how with colleagues, I'm looking at other books that might be helpful. I just bought a used copy of Project Management: The Essential Guide to Thinking and Working Smarter (Self-Development for Success), a book that I had borrowed from interlibrary loan on a recommendation for a short book on project management Scott Berkun. (I can't find the reference to that recommendation, though.) Berkun himself just wrote the (generally) acclaimed The art of project management, which is available in electronic from to UC Berkeley folks at http://proquest.safaribooksonline.com/0596007868
Posted by Raymond Yee on 6/20/06; 11:50:28 AM
from the Unclassified dept.
Discuss
|
|
Tuesday, May 30, 2006
| Migrating from Manila to something else # |
|
Witin the next couple of months, I will be retiring the IU Technology Lodge to reflect my new role as a Data Architect in the newly formed Data Services Department in IST. I will be specializing in all forms of academic data for the campus. There's a lot happening in the organization for which I work. See One IST for all the publicly available information. I'll write more later on my new role once things get a bit better defined.
In the meantime, one of the tasks I have to take on is migrating the sites that have been on the Interactive University's Manila server. There may be a good approach that involves next to no programming by using static rendering. Other solutions might involve using the Manila-RPC interface, specifically as it has been expressed in PyManila.
Posted by Raymond Yee on 5/30/06; 6:31:14 PM
from the Unclassified dept.
Discuss
|
|
Tuesday, May 23, 2006
Thursday, May 18, 2006
| Ambient Findability # |
Ambient Findability
Originally uploaded by Raymond Yee.
Even though I first spotted Ambient Findability : What We Find Changes Who We Become in Nov 2005, I just finally got around to reading it this last week. It might be odd to hear that I've not taken the issue of findability seriously enough, especially since Google and search are on the tongues and minds of most everyone in the internet software business. Perhaps the Gather/Create/Share mantra that I've used will explain my relative neglect of searching and finding. My focus is on what to do with stuff once you find it. I've taken for granted that others will build systems to enable a user to find things. Peter Morville's book has gotten me to take a closer look at ways in which I can increase the finability of data/information/documents for which I have some direct or indirect influence.
Posted by Raymond Yee on 5/18/06; 11:22:43 AM
from the dept.
Discuss
|
|
Tuesday, April 25, 2006
Thursday, April 20, 2006
Wednesday, April 12, 2006
Thursday, March 30, 2006
| figuring out the future of geotagging now that geobloggers.com is officially gone # |
|
It's official: Geobloggers changes direction and no longer functions as a mapping service of geotagged flickr photos. It was great service while it lasted. I certainly miss it. The old geobloggers.com aggregated information about geotagged photos and made that information available in its interface (auto-refreshing google maps) and through an API. It made effective use of google maps.
In response to the announcement, I posted a question:
I’m wondering whether you would be willing to donate your code for geobloggers.com as a mapping service or put it out as open source so that others in the Flickr community might step up to the plate and take it over.
Anyone out there willing to create a service to replace what geobloggers.com did? N ot a small task, and maybe one that will ultimately be provided by Flickr. There are other mapping services of geotagged flickr images -- but none that I know of that aggregates the geotagging information and making it available via an API.
(I just posted essentially this writeup on the GeoTagging Flickr group to see what others think.)
Posted by Raymond Yee on 3/30/06; 4:44:29 PM
from the Web Technology dept.
Discuss (1 response)
|
|
| Social Bookmarking: Some of My Own Technical Next Steps # |
For the first followup to Social Bookmarking: A First Take, I planned to describe how to make social bookmarking more of a seamless experience for myself. I wrote:
I've become an avid user of del.icio.us and potentially many other systems. I'd to be able to integrate the reference storage, discovery, sharing with my writing and teaching processes. I'll write more tomorrow about how I in particular would like social bookmarking to fit in a scheme of seamless use and reuse of digital content.
Two days ago, as I began to tally the features I wanted in an "ideal" social bookmarking/bibliographic system, I saw that pursuing such an ideal would be too large of a task to complete in a day or two. Today, I switch gears to identify a small project with a minimalist set of features that I can implement quickly. The next programming task to improve my own del.icio.us is to add a UI to my Ecco->del.icio.us script so that I incorporate richer methods of tagging my bookmarks and downloading references from del.icio.us. The user interface would be in wxPython, allowing for easier integration down the line with Scholar's Box and Chandler. (I sketched a preliminary UI design: my desk at work; sketch of a delicious uploader). Specifically, this UI will have the following features:
-
I will be able to select text in various contexts (Ecco outline and HTML to start with), the URLs will be extracted and presented in the interface for tagging en masse as well as a single form to fill in titles, descriptions, and notes.
-
In the future, I'd like to bring up the previous tags that I've used and make it easy to apply those tags to selected references.
-
On the flip side, I'd like to do (within this interface or with another one) searches on del.icio.us and bring them back in a number of formats. It would be nice, for instance, to be able to quickly format a reference as HTML or wiki markup or OpenDocument so that I can paste the reference in various places.
-
Developing this interface also gives me an opportunity to explore the potential of incorporating autotagging systems such as Tagyu and tagthe.net.
Looking at del.icio.us the last couple of days convinced me that there is greater activity and potential around del.icio.us itself (let alone the entire area of social bookmarking) than I had previously understood. Here I list some questions and areas for further exploration:
-
I think that I will ultimately need my own object store in addition to flickr, del.icio.us, Refworks to keep track of all my objects and references. Alternatively, can I just make del.icio.us into that central database? How far can del.icio.us be extended? Specifically, I've been analyzing whether we can shoehorn bibliographic metadata into del.icio.us while still honoring the spirit of the application. In other words, if I want to cite a book in del.icio.us and want to stick in things like author, title, isbn, etc, how should I do so? In terms of the mechanics, I've been thinking about geotagging in Flickr as a precedent (See Flickr: GeoTagging Flickr). Indeed, Rev Dan Catt (of the geotagging fame) wrote about this type of tagging in geobloggers: Advanced Tagging and TripleTags. (Others have had a similar notion: others have this idea: Sci-Fi Hi-Fi » Blog Archive » Meta Tags: The Poor Man’s RDF?). So I should go ahead and stick in 'bibliotagged' and things like biblio:author=William_Shakespeare into delicious? The specific tags are not clear; is the fundamental concept ok?
-
Such shoehorning of metadata comes from a desire for to have an an online database of references. I wish one of Refworks, CiteULike, or Connotea could fit the bill -- but none has a public API as far as I can tell. (Connotea seems to have a currently undocumented API: SourceForge.net: connotea-discuss) -- though Connotea does not seem to have a lot of places to stick extra data in. CiteULike has a lots of useful fields but no API. Its future is under discussion. See REST API for CiteUlike?: "Any API should make it possible to use a service like CiteULike to completely replace applications like Endnote."; RestfulApi - CiteULike - Trac and (CiteULike-discuss) Project status
-
I've started to look through the a long list of tools developed around del.icio.us: Complete Tools Collection. I was surprised to find only a few things that I wanted to follow up with. Firefox/del.icio.us synchronization deserves a closer look: Dietrich Ayala | Foxylicious - Firefox and del.icio.us bookmark integration. I will have to try Cocoalicious: A Cocoa del.icio.us Client for Mac OS X -- since it seems to be a good desktop tool. When it comes time to mirror del.icio.us to a local database, I can take a look at MySQLicious - del.icio.us to MySQL Mirroring [ nanovivid.
-
Is it time to look at del.icio.us competitors that also have APIs? I've been curious about RawSugar (though I must admit that I don't quite grok it) and Simpy - tagging, social bookmarking and personal search engine.
Posted by Raymond Yee on 3/30/06; 2:10:58 PM
from the Web Technology dept.
Discuss
|
|
Monday, March 27, 2006
| Social Bookmarking: A First Take # |
|
Let me quote what I wrote in my statement of research and development interests (Seamless Use and Reuse of Digital Content and Services By Scholars):
The long-term goal of the Scholar’s Box is to facilitate the authoring of compound, multimedia scholarly documents. In the short term, we focus first on the creation of single-content type digital collections of bibliographic citations and images. We are attempting to maximize reusability of the collections so that the collections can move fluidly among library catalogues, bibliographic data managers, learning management systems, weblogs, and office suites. We will answer the questions: How reusable can one make these collections? Will the benefits of the collections be worth the effort?
I eventually want to create a full project plan around his area of "Highly Reusable Bibliographic and Image Collections." Today, I have a more limited aim: reporting on some recent concrete work that I've been doing around academic social bookmarking and its relationship to bibliographic management.
What is social bookmarking? According to "7 Things you should know about....social bookmarking", "Social bookmarking is the practice of saving bookmarks to a public Web site and 'tagging' them with keywords." Last week, I taught a section on "social bookmarking" to my Mixing And Remixing Information class. I sought to give students some background on social bookmarking, talked about the motivation for using such systems, walked through some specific examples (primarily, del.icio.us), and then highlighted what one can do with the del.icio.us API.
Social bookmarking is an area in flux. There is a helpful chart comparing features of 19 systems (pdf) (linked from roxomatic - Social bookmarks' review - version 3.5) Another list shows Another list shows 70+ social bookmarking sites. I gave up on trying to figure out everything that's happening on this front. I certainly didn't look at 70+ or even 19, or even get to services that I would have liked to try before talking to my class. In many ways, for general social bookmarking, my stance takes the form of two questions: 1) do I want to use any social bookmarking system at all? 2) if so, why should I not just use Idel.icio.us. I just highlighted what I thought useful. Which systems did I look at?
-
http://del.icio.us is the grand-daddy of social bookmarking sites, the site that kicked the whole folksonomic craziness. A good number of people already using the system. Though my first entry on delicious was from January 2004, I didn't use the system much because it seemed to be too much work for most bookmarks. In many cases, I just want to have a place to store my links so that I can find them later. I am willing to drop things into a system without putting any tags, in the hope that I will be able to write a script to crawl through my old links, create an index so that that I can do full text search over the collection of links. I took a first step towardssuch a system by writing a Python script to upload locate URLs in a selected EccoPro outline and upload them to del.icio.us. That script, combined with the fact that posting to del.icio.us now seems more fluid with a Firefox del.icio.us. Now that I have the option of just dumping my links into del.icio.us without having to tagged them, I now feel confident to use the system as a whole.
-
Yahoo!'s My Web 2.0: I'm interested in Yahoo's social bookmarking system because Yahoo is a big company (with tons of users), that is adament about getting heavily in the folksonomic space (by buying del.icio.us and flickr for instance). Yahoo's MyWeb2.0 certainly has features I like, including the ability to save web pages. But I've not been able to parlay what I thought should be the chief advantage of MyWeb, namely, that so many people I already know, have Yahoo accounts, into real advantages. (I'm just too hesistant right now about inviting friends and family because I am not convinced that sharing bookmarks with my friends and family is useful to them yet.) I heard from students who are deep in the world of Yahoo researchers that MyWeb2.0 is finding good use in Yahoo Research. (I'm waiting for the full API for MyWeb 2.0 to come out to see what can be done to make MyWeb 2.0 more useful for me.
-
http://unalog.com is the creation of my friend Daniel Chudnov: It's a place where a lot of members of the code4lib community gather URLs. I've yet to used it extensively (see unalog: Person (raymondyee)), but I'd like to enable bookmarks that I capture to flow appropriately to code4lib and other communities of which I am a part.
-
academic bookmarking site: There are two academically oriented social bookmarking systems that deserve further exploration. One is Connotea, run by Nature, specializing in the scientifc literature. (Connotea: about: "Connotea takes this concept and adds extra features to tailor it to the needs of scientists.") I've started to adding some bookmarks to my account ( Connotea: rdhyee's bookmarks). There is currently no API for Connotea though there's some discussion of creating one. The other system that I've been exploring is Citeulike. CiteULike: rdhyee's library is my library of references. I've put a few items in the library, but not enough to get a feel for how the two systems work.
I'm deeply interested in the interoperability/integration possibilities of each of these systems. None of these systems will satisfy all my needs for managing bookmarks and references. I want to make sure that whatever system I use, I have a way of getting data out of it. Ideally, I'd like to be able to use all of these bookmarking systems seamlessly. For instance, drop a reference in one system and have all my references synchronized appropriately to other systems.
What Next?
I've become an avid user of del.icio.us and potentially many other systems. I'd to be able to integrate the reference storage, discovery, sharing with my writing and teaching processes. I'll write more tomorrow about how I in particular would like social bookmarking to fit in a scheme of seamless use and reuse of digital content.
Later on, I will take up the question of whether, why, and how social bookmarking systems can be supported at a place like UC Berkeley. Is social bookmarking ready for wide deployment? How can we figure that out?
Posted by Raymond Yee on 3/27/06; 4:18:31 PM
from the Web Technology dept.
Discuss
|
|
Tuesday, March 21, 2006
Tuesday, March 14, 2006
Wednesday, February 15, 2006
Tuesday, February 14, 2006
Thursday, February 9, 2006
| I paid money for PHP Hacks # |
|
I bought a copy of PHP Hacks : Tips & Tools For Creating Dynamic Websites (Hacks) even though as a UC Berkeley person, I have access to an electronic version of the entire book online at safari.oreilly.com. I had stayed away from learning PHP until this semester because I didn't feel I had room for yet another programming language in my toolkit, which was already under-exercised. Working through many PHP-based examples in Flickr Hacks convinced me that I needed to know PHP better, if for no other reason than to help my class through PHP.
I got to hand it to the O'Reilly folks for getting me to fork over money for essentially the same information in a different package. I considered buying the PHP in a Nutshell (In a Nutshell) but decided that I am fine with treating it as a reference book for which electronic access alone is adequate. In contrast, I like sitting in an easy chair or curling up in bed with books such as PHP Hacks so that I can soak up the ideas and allow myself to imagine the possibilities of the technology.
Posted by Raymond Yee on 2/9/06; 10:50:38 AM
from the Unclassified dept.
Discuss
|
|
Tuesday, February 7, 2006
| Notelets for 2006.02.06 (Dinner with Dave Winer?) # |
|
I hope to hear today whether my proposal to ALA | ACRL/CNI/EDUCAUSE Joint Virtual Conference was accepted. Notices are supposed to be issued today. I wonder whether that means only acceptance notices.
Laura and I plan to join in on Dave Winer's Berkeley Blogger's Dinner. Anyone else?:
Reminder: The second Berkeley Blogger's Dinner is Thursday at 7PM. We have room for lots more people this time. If you're going to be there, please post a note, let the networking begin now. You don't have to be a blogger to come, people from Berkeley and from the University are especialy welcome, as are people from the Yahoo office, which is just a couple of blocks away. We make these inclusive, the more the merrier, let's have fun and all that. Bring someone with. It's around the corner from BART, so we're an easy 20 minutes from downtown San Francisco.
Posted by Raymond Yee on 2/7/06; 1:34:50 PM
from the Unclassified dept.
Discuss
|
|
Saturday, February 4, 2006
| Documenting the Scholar's Box code # |
|
I have printed out all the code for the Scholar's Box and sitting here writing documentation to future users and developers of the Scholar's Box but also take stock of where we are and we need to go. There's a lot of logic to be understood, documented, and highlighted. Tom Schirmer has built some real nuggets:
-
a responsive GUI derived from the threading of time-consuming tasks
-
a drag and drop system among collections
-
metadata conversions among items coming from a variety of sources
-
a plugin search architecture
Posted by Raymond Yee on 2/4/06; 8:58:35 AM
from the Unclassified dept.
Discuss
|
|
|
|
| August 2006 |
| Sun |
Mon |
Tue |
Wed |
Thu |
Fri |
Sat |
| |
1 |
2 |
3 |
4 |
5 |
| 6 |
7 |
8 |
9 |
10 |
11 |
12 |
| 13 |
14 |
15 |
16 |
17 |
18 |
19 |
| 20 |
21 |
22 |
23 |
24 |
25 |
26 |
| 27 |
28 |
29 |
30 |
31 |
|
Jun
Sep
|
