Monday, March 11, 2013

WebWise - Picking a Platform

This is an overview of the second session I attended at the WebWise Conference in Baltimore on Wednesday, March 6, 2013.  The presenter for Picking a Platform was Tom Scheinfeldt of the Roy Rosenzweig Center for History and New Media at George Mason University.  He blogs at foundhistory.org.  I attended this session in the context of Whilbr.  We've been evaluating new platforms for Whilbr, and our recent focus has been on OCLC's CONTENTdm.

When starting out a web based project, consider:

  • Aim: What is the website trying to accomplish?  What is the point?  Closely tied to who is the audience?
  • Audience:  General public" isn't well enough defined.  School children?  Scholars?  New members?  People internal to your institution?  Even if these are all your people, list them out.
  • Collection (type, size, permissions)
  • Metadata: We have a tendency to work so hard at developing and presenting the metadata.  One option among many is to present no metadata - just keep that in mind.  The decision about how much metadata to display will influence your platform.  Sometimes a title and date may be sufficient, depending on your audience needs.
  • Resources (technical, financial, human)
  • Timetable: Be flexible and consider if it will take way too long to implement the project based on using a given technology.
  • Branding: Decide how closely you want this associated with your institution. This is not a secondary concern.  It affects the range of choices you have for hosting your platform.  There are some platforms that will not allow you to make the project look like the rest of your website.
Don't let the technology choice drive the project.  This limits your options and you can end up hurting the user experience.  It could tie your vision to a set of tools that may not accommodate your vision.  You should lobby for choice.
 
Some things to consider: Self hosted or cloud?  These days this is a significant choice.

First, these are some self-hosted options.  You can go big and heavy or small and light.
 
Islandora - Building a rich digital repository ecosystem.  Islandora is a platform for implementing Fedora - a Drupal front-end - this is a big and heavy option.  Allows for a robust collections back-end and an attractive pretty front end that the rest of the world can use.  It can do a lot - extremely flexible.  Can design back end and metadata however you want it.  But with flexibility comes lots of responsibility.  You need someone who knows how to implement them.  Requires someone on your staff who can do the hard work of implementing - working through all the choices.  Need expertise.  Islandora is open source - free in that regard, but requires substantial investment of expertise.  Discovery Garden is a vendor that helps with implementation.
 
OCLC's CONTENTdm - Can be hosted locally or by OCLC.  Like Islandora, but more restrictions on customization.  Constraints can be good in some cases.  Allows for relative ease of implementation.  Using this, it's going to be easier to get going than Islandora.

VirtualExhibit for PastPerfect 4.0 - Is as easy to use as a Windows program.  Getting farther down the road.  Although it's very friendly and easy to use, it's not very customizable.
 
Greenstone is an open source option.
 
Omeka is a good self-hosting optoin for collections driven web sites.  You need a linux server and ability to maintain.  For Omeka and most other open source solutions there are vendor options for support. teachinghistory.org was developed using a drupal front-end, which allows the addition of new features. Though this requires a staff member in an office working on this almost all the time. For complicated community driven web sites they use Drupal.  Omeka is used for Lincoln at 200. It's a collections driven web site with a narrative framework. See newberry.org. It provides some of the skin-ability of Drupal, but with the ability to include metadata the way drupal doesn't just out of the box. If you want to do the kind of customization of the Lincoln site, you need someone experienced in php.

Wordpress.  If you want a text-driven web site, Wordpress may be sufficient.  In a sense Wordpress does the least. It has its limitations in that it can't take a lot of data types.  It's a blog with text and some images.  On the other hand, there's something very flexible about that in that you don't need a full-time developer. There are all kinds of Wordpress plugins that are relatively easy to use without a great deal of expertise.  The Omeka homepage happens to be in Wordpress.

Toward the very light end there is Mediawiki.  It's relatively easy to install.  Not a lot of options to make it look pretty, but relatively transparent in how to make it work.  You can get a lot of content for little investment.

There are plenty of other platforms out there that would require you to host on your own server - such as Joomla.    Most of these require you have a linux server and someone who can maintain it.

Next, Tom covered some of the cloud options that wouldn't involve the library providing the hosting. Many of these options don't require any local IT or experience.  They give a lot of freedom to work at your own pace.  But, you may be limited in the features you want to use.  You may not be able to brand it and integrate it with your main web site.  The URL may be something other than your library's web site.

Omeka offers a cloud version that runs $50 to $100 a year.  You will be disappointed in how it works.  The half a dozen themes they offer are pretty limited.

Viewshare:Interfaces to our Heritage is a free product from the Library of Congress.  You can export an excel spreadsheet into Viewshare and it will plot on a map. It allows you to imbed views from a chart into your blog or web site.  All of this is in the cloud.  Views are created on server side.  You are imbedding these views into your blog, but everything is hosted on LC server.  You can do an awful lot without technical support, but options are limited.

Historypin: Pin your history to the world.  It's a powerful piece of software to aggregate content from users all over the world.  You're adding your content to a collective system.  This is your stuff living alongside everyone else's stuff.  You lose ownership and branding, but you gain a huge audience.

You could add your materials to the Wikimedia Commons - Wikipedia to get your stuff seen by lots of people.  But of course you loose control - copyright, branding, association with your institution.  If your object is to get stuff into the hands of the most people possible, this may work for you.

Another option is social media sites, such as Pinterest.  Same issues/opportunities as wikipedia.  Facebook another option.  It's a tradeoff.  You gain a lot of audience and ease, but lose control.

At Tom's project they are maintaining legacy servers but are moving more and more content to the cloud, specifically to Amazon web services.  There are tradeoffs.  When Amazon goes down your site goes down too. But then you need to assume that all servers will go down at some point, so maybe it's better to have the Amazon technician fixing it than you.
 
We need to be comfortable with the idea of using multiple platforms simultaneously.  All of your online content doesn't need to be in the same platform.  After all, we use different tools for different things.

-Joe

No comments: