Forgot password? | Forgot username? | Register

Development of an improved taxonomy module

Development of an improved taxonomy module

I have asked KE when implementation of the proposal (defined in this forum) as the default module might begin. The response I received was as follows:

KE perceives that there is not enough buy-in from the major museums for them to make it happen this year unless there is financial underwriting from all of us. They estimate a significant cost to develop the module, plus a lesser, but still significant cost for each museum to migrate existing data.

We'd like some clarification from KE regarding if/when they expect to be able to implement this, or what more buy-in they need. A significant amount of time was spent by the participants over a number of years to develop this model that reflects all of their needs and views. We continued to proceed, given numerous logistical challenges and sparse funding because KE said they would appreciate our input and would implement a consensus-based result. To have come this far and to get such a lack luster response from them is disappointing.

Given the amount of time spent on developing this from FMNH, AMNH, PMY, NYBG and others and that there are similar projects happening right now (Bibliography), it seems fair to ask for some (re)assurances that there will be pay-off for us at the end before we go forward with any other EMu user-defined improvement projects.

Please feel free to add your comments and questions to this forum. Also, any messages supporting the implementation of this proposal as the default can be iterated/reiterated here.

Joanna at FMNH

Edited by: - 01-Jan-70 09:00:00

Joanna McCaffrey
Biodiversity Informatics Mgr.
useravatar
Offline
39 Posts
Female  Birthday  Website 
Administrator has disabled public posting. Please login or register in order to proceed.

Re: Development of an improved taxonomy module

Hi Joanna,

I'm sorry my previous response endorsing your final draft and congratulating you on what you and others on the Taxonomy working group didn't get up (not sure why). We would certainly like to see this module implemented as the default Taxonomy module. And - we certainly appreciate the amount of time you have all spent on its development. You've raised an important issue regarding the significant contribution made by users into module development and/or improvement when there's no guarantee that the improved models will be adopted by KE in the short term.

Di Bray at Museum Victoria (Melbourne, Australia)

Administrator has disabled public posting. Please login or register in order to proceed.

Re: Development of an improved taxonomy module

Joanna and fellow taxonomy users,

Our apologies for the slow response. Your email arrived during our Easter holiday period and several of us have been away for a few days.

In regard to your proposal for a new Taxonomy module, firstly we'd like to sincerely thank you for the effort that you and many others have put into this. As you are all aware, we are very supportive of any efforts to improve our taxonomic support and in particular efforts that result in a more standardized system and we hope that this proposal takes us a significant step towards that end.

In your email, you've identified three areas of concern - buy-in, development costs and rollout costs. These are somewhat intertwined but nonetheless I'd like to address each separately.

DEVELOPMENT COSTS:
This is a substantial development project. It is incompatible with the existing Taxonomy module and so cannot be implemented as a simple upgrade. Rather, we must create a new module. Besides, given that at least one very large museum has indicated it will not be moving from the existing module any time soon, we will be forced to maintain the existing module into the future. We anticipate implementing a new module and then allowing clients to choose which version - the old or the new - should be included in their EMu implementation.

We haven't costed the development fully but it is certainly in the tens of thousands of dollars. This in itself is not a problem. We have our own budget set aside for ongoing improvement of EMu and we would be happy to use some of that to fund a revision of the Taxonomy module.

Naturally our budget for ongoing EMu improvement is limited. The only criterion we apply to use of that budget is that the development be for the good of as broad a user base as possible.

ROLLOUT COSTS
Rollout of the new Taxonomy module to an existing site could be achieved as part of a normal upgrade. However, this proposal takes the Taxonomy module in a direction which is incompatible with the existing module. So while it is easy to put the new module onto a client's machine, their existing taxonomic data is held in the old module in an incompatible format. So at every site, there is a need for the existing data to be migrated to the new module. This migration will include some simple mapping of old fields to new fields but it will also involve splitting data, creating multiple records from single records, establishing links between records, populating attributes by inference from existing fields. In other words, it is not simply adding new fields to an existing model; it changes the structure of the existing data.

There will also be impacts flowing through to the catalogue. For example, most natural history catalogues draw some attributes from Taxonomy to aid in more efficient searching. These catalogues will have to be modified to draw from the new module and all catalogue records updated.

Additionally there are likely to be changes required to reports and to existing web interfaces.

We would anticipate that the core of the migration (i.e. the basic fields) would be relatively simple and would be common across all sites. However we expect that at many sites, if not all, there will be many local migration hurdles which have to be overcome plus changes to the catalogue, reports and the web. Thus there will almost certainly be some site-specific effort required for each museum. Each museum will also be required to extensively test the migration before going live to ensure that it has accurately mapped the data.

In addition to this, many clients - eleven in all - currently use sub-classed versions of taxonomy. As their existing Taxonomy modules are different from the standard module, the migration process to the new module will have to be customized individually for them. Additionally, those clients must decide if they want to have their existing customizations re-applied to the new Taxonomy module.

Unfortunately all of this involves time and effort - from both KE and each museum. Again, we haven't accurately costed this - in fact, it must be costed on an individual site basis - but if I had to guess, I would expect it to be around $5,000 at "simple" sites and perhaps $20,000 to $30,000 at the more complex sites (e.g. with lots of environments, plus a customized taxonomy module to begin with).

These are real costs and local to individual sites and so unfortunately we would have to pass these on to our clients. We would of course do our best to keep these rollout costs to a minimum. But the fact that there will be some rollout costs may have an impact on whether a museum adopts the new module or not.

BUY-IN
Which brings me to the issue of, as Joanna put it, "buy-in".

As I said above, it is very much in KE's interest that the Taxonomy module be improved and that we achieve more standardization of Taxonomy across the entire client base. It is equally important that we invest our development efforts on areas of benefit to as many users as possible (even though this proposal is of interest only to natural history museums who make up only a small proportion of the total EMu client base, they tend to be the largest clients).

The discussion on the forum has been excellent with some very good contributions and healthy debate. Clearly there are some clients very interested in the proposal. But comment has come from a small proportion of Taxonomy users and so it is difficult for us to assess how widely it is supported and indeed whether this proposal achieves the goals of wide applicability and increasing standardization.

It certainly appears that FMNH, AMNH, YPMNH, NYBG and MV fully support this proposal. We have already heard that the two largest museums - NMNH and NHM - have no plans to move to the new model.

But our records show that some 30 EMu clients have the Taxonomy module (we can't tell how extensively they are using it) and we would really like to hear from as many of those as possible. Does this proposal represent an improvement to our taxonomy support and do you intend to adopt it bearing in mind that there will be some rollout costs?

We look forward to hearing from all natural history museums.

Kind regards,
John Doolan

Administrator has disabled public posting. Please login or register in order to proceed.

Re: Development of an improved taxonomy module

John

Although there are some features of the proposed alternative taxonomy module that really excite us, such as the hierarchy tab, in general the current taxonomy module meets our needs and the costs of shifting to the alternative taxonomy module would outweigh the benefits for us. Given our other development priorities, we do not intend to adopt the alternative taxonomy module in the near future.

Kind regards
Philip

Museum of New Zealand Te Papa Tongarewa

Administrator has disabled public posting. Please login or register in order to proceed.

Re: Development of an improved taxonomy module

Here at NSW (in Sydney) we have been watching the development of an 'improved' taxonomy module with interest. Our basic philosophy towards the function of our "NSW Collections" database (EMu) has moved more to wanting EMu to link to other web-based sites that hold information that we require (with respect to the Taxonomy and Bibliography modules - e.g. IPNI in an important source of nomenclatural and bibliographical data). There appears to be a tendency to want to store every thing within a database. Contrary to this, we would like to minimise the duplication of effort by having direct links to these external sources of information. However, even though that is (perhaps) our preferred position, it is clear that there appears to still be a need to hold datasets that are being managed and developed by others.

With respect to the proposed Taxonomy module changes: The proposal approximates to how we are using or, at least, trying to use this module.

"Taxonomy (Botany) screen 1": Yes, we like the modifications. We assume that the "Used?" field = accepted name or not. The two "Publ. date" fields do not appear to be of much value for botany - we would be using "Typification (shated) screen 3" for our date information.

"Classification(Botany) screen 2": Yes

"Hierarchy (shared) screen 2": Yes - although we tend not to store very much of the higher level classification levels in our "NSW Collections" database, as we prefer to link to various web-sites for this information.

"Citations (shared)screen ?": Yes - the most important feature that we like is the ability to refer to a single page within the protologue for a new taxon - this has been difficult to do within the current Bibliography module without causing considerable duplication of referencing.

"Hybrids/Cultivars (Botany)" screen ?": Yes - looks good.

"Typification (shared) screen 3": Yes - good.

"Validity (Botany) screen 1": Yes - good, but we have found that we do not have the resources to maintain this type of data for all the taxa that we are the custodians for providing information. It would be a nice to have but "watch this space" because many records would have empty fields on this tab!

>>KE perceives that there is not enough buy-in from the major museums for them to make it happen this year unless there is financial underwriting from all of us. They estimate a significant cost to develop the module, plus a lesser, but still significant cost for each museum to migrate existing data.<<

NSW's position, like many other institutions, is that our resources are 'significantly' limited, but I doubt that this equates to the 'significant' cost to develop and migrate, as mentioned by KE. We would like to be involved, but it will depend on how significant the 'significant' actually is! However, we like the improvements that have been proposed for this module.

Administrator has disabled public posting. Please login or register in order to proceed.
There are 0 guests and 0 other users also viewing this topic