Librari.an.droid

Android: Budget Droid Pro

2009-10-21T08:35:00.000-07:00

Budget Droid Professional is now available on the Android Market under Applications and Finance. It includes charts, weekly, bi-weekly, and monthly budgets, password protection, and much more.

Fedora: Commons Pool

2009-10-06T10:44:00.000-07:00

If you're ever installing Fedora and it crashes during start, citing (in the fedora.log):

java.lang.NoClassDefFoundError: org/apache/commons/collections/CursorableLinkedList

You'll need to install the Apache Commons Pool:

commons.apache.org

Extract the archive and copy the jars to your Tomcat libs directory.

Restart Tomcat and you should be good to go... unless you're having additional issues.

Android: My Next Phone

2009-09-26T04:12:00.000-07:00

I do believe I've found my next phone.

Hopefully it's GSM and works (without issue) on T-Mobile.

M&Ms: Survival of the Fittest

2009-09-21T11:40:00.000-07:00

Someone posted this today on a forum I hang out at:

Whenever I get a package of plain M&Ms, I make it my duty to continue the strength and robustness of the candy as a species. To this end, I hold M&M duels.

Taking two candies between my thumb and forefinger, I apply pressure, squeezing them together until one of them breaks and splinters. That is the "loser," and I eat the inferior one immediately. The winner gets to go another round.

I have found that, in general, the brown and red M&Ms are tougher, and the newer blue ones are genetically inferior. I have hypothesized that the blue M&Ms as a race cannot survive long in the intense theater of competition that is the modern candy and snack-food world.

Occasionally I will get a mutation, a candy that is misshapen, or pointier, or flatter than the rest. Almost invariably this proves to be a weakness, but on very rare occasions it gives the candy extra strength. In this way, the species continues to adapt to its environment.

When I reach the end of the pack, I am left with one M&M, the strongest of the herd. Since it would make no sense to eat this one as well, I pack it neatly in an envelope and send it to M&M Mars, A Division of Mars, Inc., Hackettstown, NJ 17840-1503 U.S.A., along with a 3x5 card reading, "Please use this M&M for breeding purposes."

This week they wrote back to thank me, and sent me a coupon for a free 1/2 pound bag of plain M&Ms. I consider this "grant money." I have set aside the weekend for a grand tournament. From a field of hundreds, we will discover the True Champion.

There can be only one.

Source

It made me laugh, and so I share it with you all.

Android: Budget Droid 2.0 Update

2009-09-01T05:50:00.001-07:00

Progress is going well on Budget Droid 2.0. I expect to have it released late September or early October.

I'm hoping to have it translated into several languages, and really, much of the time it will take to finish the app hinges on that.

Android: Budget Droid 2.0

2009-08-15T16:15:00.000-07:00

As of today, I've officially started work on Budget Droid 2.0. When I wrote Budget Droid 1.0, I had very little experience with both Android and Java, meaning that there are quite a few things that I now realize I could have done much better. Plus, I made both the database and the code very inflexible, meaning that when it came to writing updates, I found myself really working uphill. As a result, Budget Droid 2.0 will be built on a completely different code base. Plus, I only have evenings and weekends to work on this (I'm not a company... just a solo developer), meaning that this certainly won't be out next week. In fact, as far as release date is concerned, the only answer is that I don't have a release date. I want to do it right (though some time mid-fall is my goal).

At present, I'm just working on my feature list. Luckily, thanks to user input, I have a very good idea of what a lot of these features will be.

I'll try and post more information here as I continue to work on this project. Feel free to post suggestions in this thread. There's no guarantee that I'll be able to implement everything users have suggested, but I hope this version will be a significant improvement over the previous.

Android: Two New Apps

2009-08-09T09:50:00.000-07:00

It's now possible to download limited versions of my Wine Diary and Beer Journal applications from the Android Market. These are already available for $.99, but the limited versions exclude the ability to upload recommendations (you can view only) and include ads.

More information:

Beer Journal

Wine Diary

Android: Android Network Awards

2009-08-05T16:29:00.001-07:00

So I found out recently that several Android sites have gotten together to sponsor the Android Network Awards, which I thought was pretty cool. Then I found out that one of my applications was nominated, which is even cooler.

So on August 10th, public voting begins. If you've used (and like) my Budget Droid software, then I ask for your vote. :)

Android: New Story

2009-08-01T11:02:00.000-07:00

Yesterday I uploaded a new story to the Android Market. You'll find it under Applications->Entertainment, with the title "Her Penance".

Or, maybe easier, just do a search for "Penance". I'm pretty sure it's the only application that shows up under that name.

Anyway, it's a short fantasy piece. You can read it in less than an hour. Enjoy.

Bots: Greatest User Agent String Ever

2009-07-14T10:31:00.002-07:00

Scrolling through a list of known crawler User Agents, I ran across this gem:

Wir sind die Borg (Version: 1.03, Sie wurden Assimiliert +http://www.yammba.com/suchmaschine/bot.html)

Which translates into: We are the Borg. You will be assimilated.

Classic.

TV: Warehouse 13

2009-07-10T18:55:00.000-07:00

The Playstation Store was offering the pilot episode of Warehouse 13 for free, and so I grabbed it and watched it tonight.

I'd have to say that I enjoyed it... quite a bit. It has a feel similar to Eureka or Stargate SG1, i.e., hammy humor mixed with creative sci-fi scenarios, except here instead of getting sci-fi, we're getting paranormal.

Anyway, it's a promising start.

Linux: Google Chrome OS

2009-07-08T06:35:00.000-07:00

This is crazy, but Google has just announced that they're breaking into the OS market.

Well actually it's not that crazy. We've seen this coming for a long time now, but to have the project finally formally announced is refreshing in a way. Google is probably the only company with the clout to really take the battle to Microsoft.

Sure, Red Hat and Novell have great OSes, but they just don't have that same Google gravitas.

What I'm really hoping to see with ChromeOS is some tight integration with Android, perhaps even to the point of running Android applications on the ChromeOS desktop. An Android Market client would be killer.

The overall idea of ChromeOS makes a lot of sense, considering Google's arsenal of products. Focusing on Chrome as the primary application suggests that rather than coming bundled with OpenOffice.org like most Linux distros do, ChromeOS will simply come with an icon shortcut to docs.google.com. This will make it easy to deploy the OS onto older hardware or netbooks.

However, it looks like Google is building its own alternative to X for this distro. What concerns me here is interoperability between ChromeOS applications and GTK or QT applications.

Does this mean that X would have to be ported over to ChromeOS in order to run, say, Gimp?

PS3: PlayOn

2009-06-30T12:16:00.000-07:00

I'm thinking of installing PlayOn as a possible solution for my PS3 woes now that Hulu has blocked the PS3's browser.

But there's a catch. Actually, there are a few catches.

I don't actually have a Windows box at my place. Just a bunch of Linux and a Macbook. Unfotunately PlayOn is only out for Windows. Six months ago, this would have been a deal breaker, but I'm a little less rigid than I used to be.

PlayOn costs money. Specifically, it costs $20. That's not much cash, but I'm so used to using open source software, that the notion of spending money on software is foreign to me.

Hulu cat-and-mouse. Who can say if one month down the road, Hulu won't block PlayOn again? I think I heard that Hulu has blocked them before.

Time. I'll have to pull out some dusty copy of WindowsXP and wait for the damn thing to install. Then I'll probably have to go out and grab a bunch of drivers, etc, etc, etc. I really hate installing WindowsXP.

PS3: Did Hulu just start blocking the PS3 Browser?

2009-06-26T19:22:00.000-07:00

I just watched several episodes of Stargate, and then, suddenly, I get a message informing me that my device is not supported.

I went back and dug through several previous episodes that I knew worked but got the same message each time.

Hopefully this is a fluke and tomorrow it will all be back to normal. I've been a loyal Hulu watcher for a long time now, on both my laptop and the PS3. I even fill out all their little surveys and patronize their advertisers in order to support the service. I'll be very put out if they blocked my PS3.

Meanwhile one of their big competitors, Joost, just released a PS3 interface. Guess I know where I'll start spending my time.

ZF: Something Goofy about the new Autoloader Class

2009-06-24T13:12:00.000-07:00

I don't get why I'm having so many problems working with the new resource autoloader class in Zend Framework 1.8.

I spent hours last night trying to get it to see my database models, which I finally fixed by changing my application directory structure (Documentation here).

I really should learn to read documentation more closely. It just hadn't occurred to me that so many things could have changed from the old loader.

That said, I was thinking of porting an application I'm writing at work over to ZF 1.8, but I never could get it working. The error I'm getting is:

Fatal error: Class 'Sidebar' not found in /path/to/SomeController.php on line 15

My directory structure is 100% correct, so I don't believe that I'm having the same problem at work that I was having at home, even though the error is the same.

Guess I'll just stick with the old loader for now. Deadlines approach, and I don't have time to dink around with this.

Edit: I should say that I'm not the only person having this problem. When hunting around on Google for a solution, I encountered about three posts complaining about the same error.

Edit #2: I think I figured it out. Won't know until I try it out tomorrow morning though. If I'm right, I'll feel pretty retarded.

Fedora: New Install ConnectionPool Errors

2009-06-23T09:50:00.000-07:00

It looks like the strange ConnectionPool errors are still alive and well in the recently released Fedora 3.2. I see these after the initial custom install of Fedora when using a non-bundled database, such as Postgresql, and an existing Tomcat 6 install.

An example start error looks like (as found in your catalina.out log):
----
java.lang.NoClassDefFoundError: org/apache/commons/pool/impl/GenericObjectPool at org.apache.commons.dbcp.BasicDataSourceFactory.createDataSource(BasicDataSourceFactory.java:152)
----

I guess what's happening is that Fedora is looking for a few commons*.jars in the wrong place... the result of some changes between Tomcat 5.5 and Tomcat 6.

The solution seems to be to copy the commons jars in your Fedora home directory into the Tomcat lib.

For example:
-cp /usr/local/fedora/client/lib/commons*.jar /usr/local/tomcat/lib/

I don't know if it's necessary to copy all the commons jars over, but I found that as I copied one over, it would just kick up another error asking for another... so I just decided to copy them all.

This problem has been around for a while. I'm a little surprised it wasn't taken care of in Fedora 3.2... but then, I figure the developers have bigger fish-to-fry (and there may be additional reasons not to make the change anyway).

More info: http://www.fedora-commons.org/confluence/display/DEV/mail/11502243

PGSQL: Moving Postgres Data in openSUSE

2009-06-23T07:34:00.000-07:00

It's pretty easy to move a Postgres data directory in openSUSE. These instructions assume you installed Postgres using Yast or Zypper, but some stuff would apply to other package managers or a manual install.

Note: You need to be su.

0) Stop postgres, if it's running:
-/etc/init.d/postgresql stop

1) Find the existing data directory. For me, it was under /var/lib/pgsql/data. If, for whatever reason, you don't find anything there, try this: find / -name pgsql | grep data

2) Copy the contents of that data directory into its new home. For example, let's say I have created a new data partition, which I've mounted as /postgdata.
-cp -R /var/lib/pgsql/data/* /postgdata

3) Change ownership and permissions of that directory:
-chown -R postgres /postgdata
-chgrp -R postgres /postgdata
-chmod -R 700 /postgdata

3) Open the postgres init file. This is generally under /etc/init.d/postgresql.
-vim /etc/init.d/postgresql

4) Look for a line beginning with "eval DATADIR=${POSTG..." and comment it out by prefixing it with a #. Just below that, add a new line reading:
-eval DATADIR=/postgdata

5) Save and start postgres.

ZF: GData Search through Zend Framework

2009-06-20T12:58:00.001-07:00

I don't know why I'm having such a tough time with this.

The Zend Framework has a handy wrapper for a Google book search, but I'll be damned if I can figure out how to do anything with any of the Dublin Core metadata that comes bundled with the Atom feed.

Here's the best chunk of code I can find on the ZF site:

$books = new Zend_Gdata_Books();
$query = $books->newVolumeQuery();

$query->setQuery('domino');
$query->setMinViewability('partial_view');

$feed = $books->getVolumeFeed($query);

foreach ($feed as $entry) {
echo $entry->getVolumeId();
echo $entry->getTitle();
echo $entry->getViewability();
}

All those 'getTitle', 'getVolumeId', etc are all fine and dandy, but the real meat is in the Dublin Core (author, ISBN, etc).

I'm working around this by simply ignoring the ZF wrapper and just using SimpleXML:

$url = "http://books.google.com/books/feeds/volumes?q=" . urlencode($isbn) . "&start-index=1&max-results=1";

$atom = simplexml_load_file($url);

foreach($atom->entry->children('http://purl.org/dc/terms') as $entry){
//Do Stuff
}

But it seems to me that there should be some way to do this through the ZF wrapper.

Anyway, I posed the question on the Zend forums:

Linky

OR09: A Closer Look at Fedora's Ingest Performance

2009-05-21T07:46:00.000-07:00

I'm glad I attended this session. They benchmarked Fedora ingest, and concluded that Fedora is quite robust and able to withstand a large amount of pummeling. One note is that they were able to get better performance out of MySQL using ISAM or tuned PostgreSQL.

http://fedora.fiz-karlsruhe.de/docs

Sun Microsystems is taking this data and continuing research on this topic.

OR09: Agile Fedora: AJAX, Low-cost Clustering, and Dynamic Metadata Forms for a Multicultural Website Project

2009-05-21T05:59:00.000-07:00

Here's another IR using PHP.

Between these guys, WGBH, and the Islandora project, me and my PHP are starting to feel a little less lonely.

And side note: they're also using Gentoo... my Linux distro of choice.

Their administration tool is crazy awesome... it's all Ajax based and looks very easy to use. However, as he stressed, it's more of a one-shot interface and not designed for multiple collections within a single environment.

The longer I'm here, the more I think it might be a really good idea to use the Islandora software. Even this presenter, as he was showing his own interface, suggested that future collections may use a pre-packaged interface.

Edit: Actually, I had a good talk with the WGBH guy, and I'm no longer convinced that using a pre-packaged interface is the best method for our own circumstances. We'll see.

OR09: Disseminating Broadcast Archives: Exposing WGBH Materials for Scholarly Use

2009-05-21T05:29:00.000-07:00

These guys seem to have faces similar challenges to what I anticipate we will experience with our own video collection. For example, they had to work with old formats and playback technology, copyright issues, etc.

They're doing some very cool things. Smaller tidbits I like are the Ajax citation generator.

And most exciting of all, they provided links to their in-progress Fedora-based video archive and a link to their Zend Framework code.

The video archive requires a username and password, which (of course) I won't post here.

OR09: Islandora: a Drupal/Fedora Repository System

2009-05-20T11:20:00.000-07:00

One small note on this.

I think it would be a mistake not to try this for at least the second or third build of the repository. It can work with any content model we develop, and I think it can work with the current infrastructure we have.

OR09: Fedora and Django for an image repository: a new front-end

2009-05-20T10:56:00.000-07:00

This developer initially used Muradora as a front end for his first IR but rejected it for his current IR for a series of reasons ranging from development lag to familiarity. For his current Fedora implementation, he used Django/Python.

Not sure what else to say. It was pretty cool.

OR09 Facilitating Wiki/Repository Communication with Metadata

2009-05-20T08:29:00.000-07:00

These guys are doing some really interesting stuff mixing wikis and Fedora, and they've developed Mediawiki plugins to list related objects from Fedora or to ingest wiki material into Fedora. One of the plugins is called Wiki2Fedora; I didn't catch the other plugin name.

Wiki integration is actually something I've been considering for a few days now. It's good to get a concrete example of how the NSDL Materials Digital Library Pathway executed this.

OR09: Technology Overview - New Releases and Roadmaps for DSpace and Fedora

2009-05-20T06:46:00.000-07:00

I'm not writing about any of the DSpace stuff here, even though some of it is interesting. In many ways, they're trying to make it more Fedora-like.
-----
Chris Wilper:
What's new in Fedora 3.2?
-Web client
-Will not be deprecating the Swing client for some time but it will happen eventually
-New Storage Abstraction: Akubra
-Store files by ID
-Pluggable
-Stackable
-Multiplexing (allow to store across multiple systems but Fedora treats it as a single system)
-Derby db support
-Run Fedora at arbitrary paths (to allow multiple Fedora instances)
-Proven SWORD 1.3 integration

Future and Continued Improvements:
-RELS-INT
-More controlled of what's indexed
-SPARQL
-Content Modeling
-Incremental improvements
-Enhanced Content Models
-Active Fedora
-Disseminators
-Improved ease of setup
-Support for PUT/POST/DELETE

Lightweight Interfaces
-AtomPUB
-ORE
-WebDAV
-FedoraFS (FUSE)

Security
-Fedora Enhanced Security

Improved Modularity
-Maven2
-Modern plugin architecture: OSGi/Spring DM
-----
I think that Akubra is something to take a serious look at. It may even be wise to implement it prior to going to production, even it it takes another month.

OR09: Strategic Overview Updates and Future Directions from the DSpace Foundation and Fedora Commons

2009-05-20T05:44:00.001-07:00

This is a discussion covering some of the more administrative details surrounding the DSpace/Fedora Commons merger.

Goals of the organization:
-Support: Provide leadership, support, and advocacy for our existing platforms and communities
-Innovate: Develop new services and offerings tha extend beyond our existing software to serve both new and existing communities
-Sustain: Develop a sustainable non-profit organization by bringing in revenue and funding from multiple souces

Combined, this organization has the largest market share of open repositories. They cite the figure of 700 instances... I'm actually surprised it isn't more.

One thing that Fedora Commons have recently opened up are the Solution Communities. Some of these might be worth exploring:
http://fedora-commons.org/confluence/display/FCCWG/Home

OR09: From the Desktop to the Cloud: Leveraging Hybrid Storage Architectures in Your Repository

2009-05-19T10:35:00.000-07:00

I like what these guys are doing. They've developed a storage controller plugin for EPrints to replicate objects in the cloud, which they configure using XML.

Most of his examples use Amazon as a cloud service. The cool of thing about Amazon's service is that your object gets distributed across three continents. When a user accesses an object from your repository through the Amazon service, Amazon detects which is the closest copy and provides the user that copy. This saves on bandwidth and provides for a faster download for the user.

OR09: DuraCloud

2009-05-19T10:04:00.000-07:00

DuraCloud is more of a service than an actual application. The idea is they will serve as a sort of proxy between the user and any number of cloud providers. The advantage of this is that users requiring large amounts of storage capacity but without the budget to purchase large amounts of storage can inject their data into the cloud. It could also be used for redundancy and distributing data.

DuraCloud would support Fedora, DSpace, EPrints, and Zenity.

It is not-for-profit with the potential for corporate sponsorship... likely from Microsoft, Amazon, and Sun/Oracle.

I'm mostly concerned about how the cloud can handle security. For public collections, this seems great; I'm not so sure about private or restricted collections.

OR09: Adding OAI-ORE Support to Repository Platforms

2009-05-19T07:59:00.000-07:00

This presentation focused on methods and challenges for harvesting metadata using OAI-ORE from a DSpace repository.

Most of the presentation dealt with DSpace-specific issues, but the topic itself does apply to other repositories.

OR09: An Iterative Approach to Building Sustainable Repository Services on Fedora

2009-05-19T07:26:00.000-07:00

I sometimes feel overwhelmed after seeing some of the cool stuff institutions are doing with their Fedora collections. This presentation provided one of those moments.

Sadly, I really don't think we have enough time to explore some of these ideas.

OR09: Reusing Open-Access Content Using Authoring Tools

2009-05-19T07:05:00.000-07:00

This is pretty cool.

They've created a plugin for MS Office products to search and add citations, images, etc from a repository without needing to ever migrate away from Office. It is then possible to use the plugin to add the new document to a DSpace repository.

He went over several usability studies focusing on retrieving data from the repository and dropping it into Word and Powerpoint. Results were positive.

Looks like a great tool.

OR09: Policy-Based Distributed Data Management Systems

2009-05-19T05:32:00.000-07:00

The iRODS software allows developers a method for setting policies to track and manage data. I don't know how much this software would apply to anything I am currently working on. After all, Fedora itself does perform some (though not all) of the tasks provided by iRODS. Presently, rule-writing is done in C and Python but PHP and Java are coming soon.

A rule can call another rule or be recursive. Some can be quite complex. Most are put in place immediately but some may be deferred.

It is developed to be used in a distributed environment.

One note is that JPL uses this and NASA National Center for Computational Sciences is evaluating it.

iRods does include a search function.

One thing iRODS does that could be very useful is provide a periodic data integrity check and automate replacement of corrupted data.

Actually, you know what? I think I really might want to evaluate this software. It does provide a lot of important functionality that Fedora doesn't provide. Hopefully their PHP and Java components come soon.

That said, their focus is more in distributed environments, which usually means really big collections. Fedora's model may be better for medium collections dependent on relationships.

http://www.irods.org
http://www.dice.unc.edu
http://www.diceresearch.org
http://www.irods.org/index.php/Fedora

OR09: Microsoft Research

2009-05-18T12:46:00.000-07:00

Microsoft research is more of a pure research side of the company.

Goal: Optimize and extend Microsoft software to meet the specific needs of the academic community.

I thought this segment might be more about an actual repository product that Microsoft was putting together, but it a appears that it has more to do with Microsoft figuring out ways to get Microsoft products to fit a particular need.

Edit: I was wrong. Here's Microsoft's very new repository project:

http://research.microsoft.com/en-us/projects/zentity/
-They will be releasing this open source, but he stresses that it will be based on proprietary Microsoft technologies.

The Microsoft SWORD plugin for MS Office seems to be a part of this.

-Interoperability is essential
-Leverage existing community protocols, practices, guidelines, etc.
-Optimize for data-driven research
-Data preservation

OR09: DSpace Foundation and Fedora Commons

2009-05-18T11:52:00.000-07:00

Here's a session I'm interested in. This focuses on the new Duraspace foundation: the merger of the DSpace Foundation and the Fedora Commons. The gentleman introducing the session referred to it as a "creepy monopoly" in which Microsoft is the underdog (Microsoft Research is also present on the stage).

Duraspace strategy:
-Independent, not-for-profit organization
-Run it like a startup... entrepreneurial
-Mission-driven... software cannot be the cart leading horse (need the mission driving the software rather than vice-versa)
-Diversified portfolio... technology and business

Fedora Commons and DSpace Foundation realized that they had a lot of commonalities in their organizational goals. The concluded that if they work together, they can accomplish more than alone.

Goals of Joining:
Support: Provide leadership, support, and advocacy for our existing software and related FOSS communities
Innovate: Develop new services and offerings that extend beyond existing platforms
Sustain: Develop a sustainable non-profit organization by bringing in revenue and funding from multiple sources

OR09: NSF DataNet: Curating Scientific Data

2009-05-18T10:34:00.000-07:00

I don't really have much to say about this. You may be better off just reading the synopsis:

https://or09.library.gatech.edu/general101.php

One protocol he mentioned using is something called THUMP. It may be worth taking a look at this, though I don't really think it fits anything I'm currently working on.

I was also interested in the graph he presented comparing the various data transfer protocols. Apparently they're using wget to archive web sites, which they've found to be especially fast, but on their graph, they had rsync listed as being just about the slowest thing available. I'm a little surprised by this. I've always had good luck with rsync. Of course, the advantage to wget is that there is really very little configuration, which may also explain why they're using it.

This session was largely about coming up with requirements and methods for implementation for a repository housing scientific datasets.

OR09: Connecting Authors and Repositories through SWORD

2009-05-18T07:57:00.000-07:00

SWORD:
-Lightweight protocol for deposit into repositories

There are SWORD clients using web, MS Office, etc...

This project is based on the Atom publishing protocol (AtomPub).

They demonstrate a web client and a Facebook client. I was hoping he would demonstrate the MS Office client, but alas....

http://www.ukoln.ac.uk/repositories/digirep/index/SWORD

This could be a great tool to use for ingesting objects into a repository and is definitely worth exploring.

Edit: actually a Microsoft rep is here to discuss the SWORD Office add-on. He did provide a demonstration. It looks exceptionally easy to use.

OR09: LORE: A Compound Object Authoring and Publishing Tool for Literary Scholars based on the FRBR

2009-05-18T07:30:00.000-07:00

LORE is pretty cool. It's a Firefox plugin that allows authors, etc to create compound objects, which can then be imported into Fedora. (I'm oversimplifying this.)

It is not currently available to the public, but they will be releasing it under GPLv3 in the next few months.

OR09: Many Lightweight Views into Complex Repository Content: Enabling Rapid Application Development for Fedora Repositories

2009-05-18T06:35:00.000-07:00

These guys have developed an interesting method to quickly develop Fedora-based applications using Rails.

He demonstrates an application to generate metadata from Flickr tags, an application for Jewish Women, and something else.

His focus seems to be on ways to develop faster, more agile software rather than longer application plans often based on RPFs, etc.

He notes that many developers are using JSON rather than XML.

XML, he believes, is useful but not all-useful.

(Maybe I should learn a bit more about using JSON.)

I like his ideas a lot, but of course, there are situations when agile development isn't that much of a possibility.

OR09: Author Identifiers in Scholarly Repositories

2009-05-18T06:13:00.000-07:00

I'm going to do some blogging during the Open Repositories 09 conference. Here's the first entry.

Here are some interesting ideas by the developers during the Author Identifiers presentation:

1) Facebook application to allow authors to drop their publications list into their profiles
2) Javascript widget to drop publications list into their own home pages

I'm not too sure about the Facebook application idea... that would be a whole new can of worms. However, I think the Javascript widget could be quite promising. It may also get authors a bit more interested in the repository and making sure that it's kept up-to-date.

Industry: Oracle + Sun... I'm troubled

2009-04-21T08:07:00.001-07:00

I've been thinking of the proposed Sun/Oracle merger quite a lot since the powers-that-be announced it yesterday.

I'm troubled.

It would be one thing if Sun were just a hardware vendor. I see nothing wrong with Oracle peddling its own brand of overpriced servers. I can even kinda-sorta stomach giving them control of Java, though it would not surprise me if Oracle decided to release future versions under some sort of proprietary license.

But once we start talking about MySQL, I really begin to cringe.

When MySQL agreed to sell to Sun last year, there were a lot of red flags that went up from all around the industry and in open source community. When MySQL's founders left the company soon after the sale took place, more red flags went up.

That said, I think that Sun's handling of the software was OK. They kept the community edition Free and Open Source and looked to continue doing so.

Enter Oracle.

Here's a company that has been gobbling up database software for years now. They've tried picking up, unsuccessfully, MySQL before. If this merger goes through, they'll own their biggest rival for peanuts. The only widely-used database remaining will be Microsoft SQL. PostgreSQL is also still on the scene, but it's not widely used enough to be considered competition.

Keep in mind that nobody uses MySQL for big, enterprise-level databases. Oracle owns that market. However, MySQL dominates the web application market. This merger will give Oracle a sweeping stranglehold on the database market.

Oracle may even decide to keep MySQL completely Free and Open Source, but does it really matter when they've attained a monopoly? Give them five years... when people have forgotten that MySQL was once an autonomous company... and watch Oracle surreptitiously pull MySQL behind a proprietary license and eventually swallow it entirely. Or maybe they'll re-brand it as Oracle: Community Edition.

I'm of the opinion that congress should put a stop to this merger. It will harm competition in the database market. At the very least, congress should require that Oracle sell off the MySQL division. I can see a company like Novell being a more fitting suitor.

This merger could harm other industries as well. Oracle is heavily invested in Linux. Would they start promoting Solaris instead? Even giving them control over Java could be problematic. But neither of these reasons are problematic enough to put a stop to the merger.

But a database monopoly? Yeah, that's a problem.

Fedora: Open Repositories Conference

2009-04-20T12:15:00.000-07:00

I'll be attending the Open Repositories Conference in May. It looks to be a good show.

Here the event schedule.

Hopefully I'll have time to blog about a few of the sessions.

MMORPG: Open Sourcing Myst Online

2009-04-17T05:53:00.000-07:00

I've been a Myst fan for years, and I played Uru briefly back when it came out (before the project died the first time).

And now, Uru: Myst Online has gone open source.

Time permitting, this is a project I wouldn't mind getting involved with. I'd be especially interested in helping to develop a Linux client.

Kindle Alternatives: AT&T and Verizon Want In!

2009-04-02T07:02:00.000-07:00

I'm not really sure what this means yet, except to suggest that Sprint is making a pretty penny off its partnership with Amazon.

Maybe we'll see Sony's reader include cellular connectivity in the not-too-distant future.

I will say this: in terms of range, the Sprint network is relatively small in comparison to Verizon and AT&T. If another ebook reader worked with one of the larger networks, it would open ebook technology to millions of new people.

Link: Engadget

This American Life: Recession 101

2009-04-02T05:38:00.001-07:00

With the G-20 summit going on right now, it may be worth revisiting the events that got us into this economic situation in the first place.

This American Life is one of the best sources I've found for taking all this recession stuff and making it understandable for those of us without an economics degree.

Everyone could benefit from listening to these episodes.

Begin with these first three. They go into the specifics of what leaders have done and what they're currently doing. The episode Bad Bank has even been praised by Treasury Secretary Tim Geithner and Senate Finance Committee Chair Max Baucus.

The Giant Pool of Money

A special program about the housing crisis produced in a special collaboration with NPR News. We explain it all to you. What does the housing crisis have to do with the turmoil on Wall Street? Why did banks make half-million dollar loans to people without jobs or income? And why is everyone talking so much about the 1930s? It all comes back to the Giant Pool of Money.

Another Frightening Show about the Economy

Alex Blumberg and NPR's Adam Davidson—the two guys who reported our Giant Pool of Money episode—are back, in collaboration with the Planet Money podcast. They'll explain what happened this week, including what regulators could've done to prevent this financial crisis from happening in the first place. You can learn more about the daily ins and outs and join the discussion on the Planet Money blog.

Bad Bank

The collapse of the banking system explained, in just 59 minutes. Our crack economics team—the guys who explained the mortgage crisis, Alex Blumberg and NPR’s Adam Davidson—are back to help all of us understand the news. For instance, when we talk about an insolvent bank, what does it actually mean, and why are we giving hundreds of billions of dollars to rich bankers who screwed up their own businesses? Also, two guys go to New Jersey to look at a toxic asset.

And here are two episodes that provide more anecdotal evidence of our economic situation. If you want the nitty gritty, you won't find them here. But you will hear stories about regular folk experiencing life during tough times.

Scenes from a Mall

This American Life spends several days in a mall in suburban Tennessee, not only to find out what people are buying in this grim financial year, but to document daily life in the mall during the run-up to Christmas. Also, a rift in a national association of professional Santas—the Amalgamated Order of Real Bearded Santas (yes, there is such a group).

Scenes from a Recession

The economy works in mysterious ways. This week, we highlight the unusual circumstances our economic drought has left us in, and the newly hatched plans being made to survive it: including a partially-renovated condo building in Chicago, whose developers have abandoned it—though they didn’t bother to tell the 19 unit owners who still live there, paying their mortgages. And a story which tracks the FDIC during its most covert operation: taking over an unsuspecting bank.

Daily Linux: Moving and Renaming Files

2009-03-30T13:14:00.000-07:00

More and more, librarians are expected to understand Unix shell commands. In fact, the first mandatory class I took back in library school was centered around this very thing.

This is the first post in my ongoing series of common Unix commands and tips. Most of these commands will work in any Unix environment (including OSX), but my interest has always been Linux, so I'll call this series the "Daily Linux".

That said, even though I'm calling this "Daily Linux", don't expect me to post something every day. This will probably be a once or twice a week thing.

Without further ado:

It may seem odd but you use the same command for both moving and renaming files: mv.

The basic syntax to move a single file is:

mv file1 somedirectory/file1

If you are worried about overwriting an existing file, use the -i (interactive) option.

mv -i file1 somedirectory/file1

This will give you the option to cancel the operation prior to overwriting the file. If you do not include the -i option, the shell will overwrite the file without asking.

The command can also be used to move entire directories.

mv directory1/ somedirectory/directory1/

Side Tip: Keep in mind that unlike many other commands in Linux, mv does not require a recursive option. In other words, handling directories is nearly identical to handling individual files.

If you would like to see the files as they're being copied, use the -v (verbose) option.

mv -v directory1/ somedirectory/directory1/

One thing to keep in mind is that this:

mv directory1/*

Is different than this:

mv directory/

In the first case, you are moving the contents of the directory. In the second case, you're moving the directory itself plus the contents.

Renaming files operates in a similar fashion to moving.

To rename file1 into file2:

mv file1 file2

To make it interactive and prevent yourself from overwriting an existing file:

mv -i file1 file2

Moving and renaming can be performed in the same operation:

mv file1 somedirectory/file2

To move and rename a directory:

mv directory1/ somedirectory/directory2/

There are additional options I didn't go through here. To learn more, take a look at mv's manual page:

man mv

Kindle Alternatives: eSlick Reader

2009-03-30T06:40:00.000-07:00

Sometimes I feel like I'm just using this blog to announce new ebook readers, but there are just so many coming out, and I feel like I should note them all.

Anyway, here we have Foxit's eSlick reader, which supports PDF, TXT, and MP3 (in case you want to listen to music while you read).

Like the Papyrus - and unlike the Kindle - the eSlick excludes any sort of wifi or cellular, which means that you'll either need to copy your books and music using USB or direct to SD.

I could see this device being good for libraries for the same reasons that the Papyrus could be good.

And one more point about the eSlick: it's the least expensive ebook reader on the market right now.

It is currently available for order, but they are out of stock, so your package may take a while to arrive.

Image taken from Engadget.

This American Life: Image Makers

2009-03-27T09:32:00.000-07:00

Another This American Life from the archives.

Act One follows the band The High Strung on their library-hopping tour across the US. Their plan, in part, is to change the image of libraries from a dusty building of books and a few sour looking faces into a place of epic awesome.

Anyway, it's a great episode.

Link: This American life

Kindle Alternatives: Samsung Papyrus

2009-03-27T05:45:00.000-07:00

Here's a nice looking eBook reader from Samsung. Unlike the Kindle, this device is sports a touchscreen, and I also like the aluminum casing.

The downside? Yes, there's always a downside... no wifi or cellular connectivity. I guess that means you'll be dropping all your books onto the device using USB.

The upside to that? I can see this device being good for libraries interested in building an arsenal of eBook readers. The lack of connectivity translates into a bit more security, and it would also prevent readers from downloading books directly from any sort of Kindle store.

Link: Engadget

Project Update

2009-03-26T15:37:00.000-07:00

I usually have at least one Android-related project going on at any given time.

At present, I'm working on putting an aBook together. This will be a new work of fiction, as I'm no longer putting out classics.

I hope to have it out by next weekend. I'll announce its release here.

Android: Archos MID

2009-03-26T09:49:00.000-07:00

One thing I'm interested in seeing is better electronic book support on phones. I've worked on an ebook packager application for Android and thrown out a bunch of examples of books into the market, but the G1's screen may be just a little too small for really comfortable reading.

This new tablet by Archos may provide a better alternative for folks who are interested in using Android for electronic books. The screen is nice and wide, meaning that you could increase text size to a more comfortable level, but the device still looks relatively portable.

Link: Engadget

Pwn2own: Mobile

2009-03-25T11:30:00.000-07:00

Here's a small follow-up to my previous post about the Pwn2own browser competition.

Apparently all mobile devices present and hackable left the contest unscathed. Now we can all breathe a collective sigh of relief.

I was a little worried that my beloved Android was going to end up violated (not because I have some reason to distrust it but for reasons of pride), but it, too, survived.

Link: Gizmodo

This American Life: Mapping

2009-03-25T08:00:00.000-07:00

I think this story on maps could appeal to some librarians.

We love managing information, and here we have a story about managing the environment, so to speak. One theme is how each map has a specific focus. These cartographers aren't just mapping roads, they're figuring out interesting correlations between objects (and sometimes sounds and smells), and they're figuring out ways to manage those correlations. Sound familiar catalogers?

The story may also strike a cord with those interested in information visualization.

Link: This American Life

Kindle Alternatives: Astak Mentor

2009-03-24T07:59:00.000-07:00

I'm not too sure about this one. It's nice and small, and might, actually, fit in a pocket. I still think you might be better off with just getting an e-book application for your phone if size is your main concern.

Link: Engadget

Fedora: Search Tidbit

2009-03-23T07:34:00.000-07:00

One Fedora point which I think deserves a special note is the search mechanism.

After installing the application, you will find a default search page at http://yourhostname:port/fedora/search. I believe this search uses either the API-A-Lite or API-A 'findObjects' method, but I've come to the conclusion that this method is not really that appropriate for a production environment.

First of all, using findObjects will only permit you to search the Dublin Core fields as well as a few administrative fields. This means that if you decide to create your own metadata fields, you will have no way to search through it.

Another issue I've encountered is the inability to turn off case sensitive searching. Maybe there is a way, but if so, I haven't found it.

So the solution? Unless you want to index Fedora and build your own search, it looks like the tool to use is the Fedora Generic Search (GSearch). This application sits alongside Fedora and uses Lucene/Solr/Zebra to index the FOXML records.

I'm still learning how to set up GSearch, and I'm finding it a little confusing. Once I understand it a bit better, I'll probably post some details here.

One side note: I recently sat in on a webinar with Thorny Staples, a Fedora bigwig, who pretty much said exactly what I had already suspected about the inadequacies of findObjects in a production environment.

Pwn2own: How fast can you hack a browser?

2009-03-19T07:20:00.000-07:00

If you know what you're doing, it doesn't take long at all.

So the gist is that Safari, Internet Explorer, and Firefox were all hacked on the first day of the annual hacker's competition, Pwn2own.

Google Chrome survives.

Now this doesn't mean that libraries should go out and rush to replace Firefox with Chrome. Keep in mind that Chrome hasn't even been out long, and it also doesn't have near the userbase of Safari, IE, or Firefox. That translates into fewer hackers testing the code.

Still, I am impressed that Chrome managed to survive so long.

Why Opera wasn't included in the Pwn2own competition is a bit of a mystery.

NASA: Spacebat

2009-03-18T12:10:00.000-07:00

"A bat that was clinging to space shuttle Discovery’s external fuel tank during the countdown to launch the STS-119 mission remained with the spacecraft as it cleared the tower, analysts at NASA’s Kennedy Space Center concluded."

Link to NASA report: The Tale of Spacebat

Someone really needs to write a children's book about Spacebat. It would be a touching adventure about a poor, crippled bat with dreams of becoming an astronaut, and through hard work and determination, he achieves his dream.

Huh... sounds like Gattica.

Of course, the author might need to glorify the ending a bit. The last thing we want is kids to read a story about a poor crippled bat who burns up on his way into orbit.

NASA Librarian

2009-03-18T08:07:00.000-07:00

I've migrated my posts from the NASA Librarian blog (nasalibrarian.blogspot.com) over here and deleted the original blog.

I've been worried for a while now that including NASA in my blog title and URL might make it sound too official, so I finally just axed it. Better to be safe than sorry... usually.

Playstation 3 as movie viewing station in libraries?

2009-03-18T05:54:00.000-07:00

Ok, so now I'm brainstorming.

The PS3 has:

DVD playback

Bluray playback

A video download service

WMV, AVI, MPEG support

So could a library buy a handful of PS3s and set them up in movie viewing booths? Most libraries already have a wide collection of DVDs, and the PS3 would give them the option to expand to Bluray if they so desired. If a patron wants to watch a film the library doesn't have, it may be possible to find it using the video download service. If there are films that are in high demand (let's say it's required viewing for a university class), the library could prepare by dumping a digital version of the film on each PS3 hard drive.

Ok, brainstorming session over.

Docstoc

2009-03-17T08:35:00.000-07:00

I have no idea how this happened, but one of my academic papers from graduate school has mystically appeared on the internet through a service called Docstoc.

Here's the link

The paper itself examines the idealism behind the GPL, specifically related to the publication the Linux Journal. Some librarians may find it interesting.

I'm not troubled by its sudden appearance on the internet, but I am a little curious as to how it got there.

DSpace vs. EPrints vs. Fedora (in brief)

2009-03-14T08:36:00.000-07:00

It seems there are three big-wigs in open source institutional repositories these days. As far as sheer girth of active installations around the world, I don't think anything compares to DSpace, but thanks to a strong showing in Europe, my impression is that EPrints probably isn't that far behind. In both cases, a single install will provide you with most of the tools that you need to build a repository, assuming that you don't care to make too many customizations to either the interface or the metadata. Their large userbases also makes for active development and relatively frequent software updates. In the case of EPrints, I also know that there's paid support available, which can be nice for a smaller library without a paid technical staff. I'm not aware of any direct support from the DSpace team, though this may warrant a little more research.

Fedora is a different animal. Rather than provide you with a pre-packaged product, which you could begin using right out of the box, you get something more akin to a framework. The closest thing you get to an interface is this Fedora administration client, a java application which must be installed on each workstation you intend to use for ingesting, content model development, viewing objects, etc. There is no web-based administration tool, though I've heard this is in the works.

Instead Fedora provides a series of APIs for viewing and modification objects. These APIs make it possible for programmers to build user interfaces in the programming language of their choice for searching and administrating the repository.

I am partial to Fedora because of its supreme flexibility, though I am still learning the ins-and-outs of how it works underneath. Still though, Fedora isn't for most libraries, especially if your library doesn't have a dedicated technical staff. When I say technical, I mean to say that you likely need at least a programmer with knowledge of SOAP and/or REST, a XML expert, and an interface designer.

Another point to consider when examining both DSpace and Fedora is systems administration. Both applications are Java based, which means they will need someone who can administrate something like Tomcat or JBoss. I'll do a post on installing Fedora sometime in the near future.

Eprints is Perl based, so you will probably need someone comfortable with Apache and CPAN. One nice feature about EPrints though is its official Debian repository (if you're using Debian or Ubuntu), which can pull down the required Perl modules along with EPrints itself. You will still need to take care of your Apache configuration and what have you, but the EPrints wiki is a big help for that. Once suggestion I would have for the EPrints developers is creating YaST and YUM repositories as well, for Red Hat, Fedora, and SuSE users. Even though the EPrints software is relatively easy to use, installing it without a repository is often substantially more difficult than installing DSpace or Fedora.

Hope this helps.

OLE Project

2009-03-13T11:07:00.000-07:00

I just learned about the OLE Project.

The 'OLE' stands for, of course:

Open
Library
Environment

Now this acronym is exceptionally enticing, especially for someone with my interests.

Open: I think of open source... Linux... GNU... all that jazz.
Library: I think of libraries.
Environment: I think of green technologies... the next-gen power grid... the Chevy Volt.

But according to the project overview:

The goal of the Open Library Environment Project is to define a next-generation technology environment based on a thoroughly re-examined model of library operations and connected to other enterprise technology systems.

You get the idea.

But in any case, it looks like they're examining some interesting topics. I'm sure I'll learn more during their webcast at the end of this month.

Let's try this again...

2009-03-12T15:15:00.000-07:00

Starting blogs is an annual affair for me.

I put one up; post for a few months. And then, almost all at once, I fade away.

This time I plan to make it work. In fact, let's make that an oath: I hearby swear to post at least two updates every week, and if I don't, may Zeus strike me down with all his electric wrath.

Ok, there.