Re: Question about islandora！

[Melissa Anez]

Hello Tianye,

We have a pretty good collection of user documentation that should help to get you started with most of these questions: http://wiki.duraspace.org/display/ISLANDORA6122/Islandora

Specifically:

1. what is the data model ?

http://wiki.duraspace.org/display/ISLANDORA6122/Chapter+3+-+Getting+Started+with+Islandora 

2.What is the metadata architecture?

http://wiki.duraspace.org/display/ISLANDORA6122/Case+Study+-+Thinking+Through+a+Data+Problem

3.What are the core functions and services?

http://wiki.duraspace.org/display/ISLANDORA6122/Chapter+1+-+Introducing+Islandora

For questions 4 & 5, we may need to know a little more about what kind of data you are looking to manage, such as the file format and the types of metadata you would want to store.

- Melissa

On Sunday, January 27, 2013 1:01:04 PM UTC-4, tianye wrote:

Hello everyone，I am from Beijing, China, is a librarian, hope to use islandora platform to establish institutional repositories, but my native language is not English, so there are a lot of problems i do not understand。I hope you can help me. My question is as follows:

1. what is the data model ?

2.What is the metadata architecture?

3.What are the core functions and services?

4.How to support the ability of the type of non-text objects? Especially scientific data,

5.Ability to support complex/compound object?

 

thank you

[tianye]

thank you very much! For questions 4 & 5,for example,scientific data,image,video and so on,How to support these objects?

and compare to ProjectHydra, what are the advantages?

My English is not very good, I am sorry!

[Melissa Anez]

Your English is fine, no need for apologies :)

Different types of objects can be supported using different Solution packs. These are Drupal modules that contain their own content models, forms, and viewers for different kinds of files. For instance, to support images you could use:

Basic Image Solution Pack: https://wiki.duraspace.org/display/ISLANDORA6122/Basic+Image+Solution+Pack
OR
Large Image Solution Pack (for really big TIFF files): https://wiki.duraspace.org/display/ISLANDORA6122/Large+Image+Solution+Pack

For Videos:
Video Solution Pack: https://wiki.duraspace.org/display/ISLANDORA6122/Video+Solution+Pack

We don't have a specific Solution Pack designed to support scientific data. An existing one may still work - it could depend on the exact file type. We will be developing a Generic Research Solution Pack to deal with that kind of object, but I'm afraid it is not yet available.

I don't know very much about Hydra myself, so I will have to leave that question for a more knowledgeable community member.

You can try out Islandora with some of your data to see how it works by using our "sandbox": http://sandbox.islandora.ca/

This is a fully functional Islandora site that you can try as an administrator to test how it works.

- Melissa

[tianye]

thank you！！！！！！

[Aaron Coburn]

and compare to ProjectHydra, what are the advantages?

I am much more familiar with Islandora than with Hydra, but here is a brief comparison according to what I know. You may wish to pose the same question to the Hydra group.

1. Hydra and Islandora both are layers that sit on top of Fedora and Solr, so in some ways, they are very similar.

2. Hydra is written in Ruby, while Islandora uses PHP inside of a Drupal framework

--this is not necessarily a pro or a con, and it may be entirely irrelevant if you don't plan to customize the code

3. Islandora uses GSearch (a java servlet) to connect Fedora with Solr, while Hydra uses a Ruby gem (Fedora-Solrizer)

--personally, I count this as a disadvantage for Islandora, because anytime the XML transformation rules (XSLT) are changed, the entire tomcat server (with fedora along with it) must be restarted. I actually use some custom code to connect Fedora to Solr in our system so that I don't have to restart tomcat -- and all of fedora -- when I adjust the XSLT. (GSearch must -- AFAIK -- run on the same host as Fedora, while other types of connectors can be more easily distributed across multiple machines)

4. The Islandora code is very closely tied to Drupal, so all interface (HTML/CSS/JavaScript) code will need to be customized within the context of a Drupal theme. Hydra, on the other hand, exposes a REST API to the underlying repository (and search engine). Hydra comes with a UI (hydra-head), but it is decoupled from the rest of the architecture. That makes Hydra slightly more of a framework than Islandora.

--in my opinion, if you plan to make *substantial* changes to the UI or the editorial workflow, there may be some advantages to using Hydra, because it directly exposes a REST API.

5. Access controls are handled quite differently -- with Islandora, access controls are managed by Fedora using XACML; with Hydra, the access controls are handled within the rails portion of the hydra stack.

--I would count this as a significant advantage for Islandora, especially if there are other (non-islandora) applications that will be accessing the fedora repository directly. If you use Hydra, then all application access to the repository must pass through the Hydra API.

I think much of this comes down to personal preference vis-à-vis architecture, programming language, and the degree to which you plan to customize the system. 

My sense (which my be unfounded) is that Hydra allows or expects that the system administrator is going to do much more direct content modeling within fedora, while Islandora has wrapped many standard content types into its solution packs (image, book, etc). That is to say -- and others may disagree here -- Islandora seems to be more of a turn-key solution, while Hydra is more of a framework that can be leveraged to build your own application.

Also, just to be clear, I have never used Hydra (I have read through significant portions of their code); our system is based on Islandora, but it has been significantly modified from the standard Islandora release.

I hope that helps!

Aaron

On Monday, January 28, 2013 1:11:34 AM UTC+8, Melissa Anez wrote:

Hello Tianye,

We have a pretty good collection of user documentation that should help to get you started with most of these questions: http://wiki.duraspace.org/display/ISLANDORA6122/Islandora

Specifically:

1. what is the data model ?

http://wiki.duraspace.org/display/ISLANDORA6122/Chapter+3+-+Getting+Started+with+Islandora 

2.What is the metadata architecture?

http://wiki.duraspace.org/display/ISLANDORA6122/Case+Study+-+Thinking+Through+a+Data+Problem

3.What are the core functions and services?

http://wiki.duraspace.org/display/ISLANDORA6122/Chapter+1+-+Introducing+Islandora

For questions 4 & 5, we may need to know a little more about what kind of data you are looking to manage, such as the file format and the types of metadata you would want to store.

- Melissa

On Sunday, January 27, 2013 1:01:04 PM UTC-4, tianye wrote:

Hello everyone，I am from Beijing, China, is a librarian, hope to use islandora platform to establish institutional repositories, but my native language is not English, so there are a lot of problems i do not understand。I hope you can help me. My question is as follows:

1. what is the data model ?

2.What is the metadata architecture?

3.What are the core functions and services?

4.How to support the ability of the type of non-text objects? Especially scientific data,

5.Ability to support complex/compound object?

 

thank you

--
 
 

[tianye]

HI，Aaron

Thank you very much for your answer, your answer is very helpful for me.

The Library where I was is a research-oriented libraries，At present, Proceed with the research project is "The ability to support scientific data of 

Islandora and  Hydra“， That is，We are ready to select  Islandora or Hydra，to support the storage of scientific data in our research institutions，so， we are studying these two institution repository which is more suitable.

I think you are very proficiently in lslandora and the Hydra,so I hope you can give me some advices, I am very appreciate it!

Best wishes！

[tianye]

Look forward to your reply, thank you

On Monday, January 28, 2013 11:33:12 PM UTC+8, Aaron Coburn wrote:

[Mark Leggott]

Aaron has provided a great summary of the key difference between the 2 frameworks, so thanks for that. A few additional comments/highlights:

- Islandora does try to be more of a turnkey solution as Aaron suggests, especially with the deployment of the Solution Packs. The latest releases for Drupal 6 and 7 will be our later this month and include the same set of Solution Packs, with changes in the 7 release for new functions. Hydra sites tend to be more unique and built on top of more localized content models and approaches. 

- Islandora is more closely aligned with the core Fedora architecture, for example in our use of GSearch, XACML and the CMA.

- Islandora is no less customizable than a Hydra implementation, in fact I would argue that it is more customizable with less effort, given the opportunity to leverage Drupal and Fedora. UPEI runs dozens of Islandora sites, some primarily for research data, and we extend the Solution Packs or create custom content models as needed. Some custom research data sites include chemistry molecule files, gene sequence data, climate change data, etc.

- The Islandora community is larger than Hydra in terms of the number of institutions that have deployed the stack (http://islandora.ca/community). Islandora institutions are wide ranging, from small to large, while Hydra sites tend to be larger with dedicated dev teams. Islandora also has commercial support options, including hosting, custom development and annual support via discoverygarden.ca.

The Hydra and Islandora teams are still committed to building integration between the 2 Fedora stacks, and to that end we have demonstrated an Islandora stack accessing a Hydra-created Fedora collection with minor tweaks. There are additional levels we have defined, including the ability to run both stacks on a single Fedora using all CRUD functions from either on the same collections. The roadmap we have developed for this integration is not well defined, but the new efforts of the Fedora Futures project (https://wiki.duraspace.org/display/FF/Fedora+Futures+Home) have integration between Hydra and Fedora front and centre.

Mark

--
You received this message because you are subscribed to the Google Groups "islandora" group.
To unsubscribe from this group and stop receiving emails from it, send an email to islandora+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

Mark Leggott, Islandora Lead

Robertson Library

University of Prince Edward Island
550 University Ave. Charlottetown, PE C1A 4P3
Office - 902-566-0460  Cell - 902-314-7507

islandora.ca   mleg...@islandora.ca  Skype: markleggott