BHL future plans

31 views
Skip to first unread message

Roderic D. M. Page

unread,
Apr 16, 2026, 7:39:05 AMApr 16
to Taxacom
Hi all,

In a little over a week BHL will gather in London and there will be a "stakeholders" meeting where people (incl. me) sit in a room and say what they think BHL should do next.

Obviously most users won't be in that room. If you have thoughts on what BHL could be doing better, I'd be curious to hear them.

Regards,

Rod

David Redei

unread,
Apr 16, 2026, 8:22:54 AMApr 16
to Taxacom
First of all: currently many of the journals are only uploaded up to 1921 or 1922. Upload more recent issues up to the limit allowed by copyright (right now 1936).

David Redei

--
You received this message because you are subscribed to the Google Groups "Taxacom" group.
To unsubscribe from this group and stop receiving emails from it, send an email to taxacom+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/taxacom/28a74e53-aba4-4387-b823-11e08b4ecf8en%40googlegroups.com.

Francisco Welter-Schultes

unread,
Apr 16, 2026, 9:36:03 AMApr 16
to tax...@googlegroups.com
Leave the user interface as is, do not change the deep link URLs. Keep
everything stable. The form BHL shows its digitized content is very
useful, do not change it.

Since a while ago links from AnimalBase to BHL content result in an
error. Clicking on such a BHL link leads to a security check page where
one has to click a checkbox to confirm being a human. This looks fine,
but clicking this checkbox results in this same page being just reloaded
and the same question appears again. This goes endless. So the page does
apparently not understand.

The error source is that AnimalBase uses http links to BHL content, not
https links - we do not have a programmer any more to change that and to
replace http by https in our links. So it would be good if it is
possible that the BHL response not only understands that someone coming
in with a https hyperlink is a humen being, but that also someone coming
in from a http hyperlink is a human being.

Have a successful meeting
Francisco

Roderic D. M. Page

unread,
Apr 16, 2026, 10:06:51 AMApr 16
to Taxacom
Good point, so you'd like BHL to keep up with the moving wall of public domain content, which in the US is 95 years after publication(!). 

Roderic D. M. Page

unread,
Apr 16, 2026, 10:32:31 AMApr 16
to Taxacom
Hi Francisco,

I suspect the interface will change at some point (it doesn't work on mobile devices for example). But this need not affect direct links to pages  (assuming that you mean links of the form https://www.biodiversitylibrary.org/page/58147827 ).

The security check is a pain, but seems to be the only way BHL can withstand the bots at the moment.

I'm surprised that HTTP links are a problem, for me the get automatically switched to HTTPS, for example http://www.biodiversitylibrary.org/page/58147827.

I tried to explore this further but I couldn't connect to http://www.animalbase.uni-goettingen.de

Regards,

Rod

Rafaël Govaerts

unread,
Apr 16, 2026, 10:35:28 AMApr 16
to Roderic D. M. Page, Taxacom

Hi,

I would suggest 3 points that would help us at IPNI.

 

  1. That a big effort was done to find missing volumes, some journals have 1-2 volumes missing, others most volumes. We sometimes find them in obscure places in cellars, so it seems important to digitize those before they disappear.
  2. Add ephemeral publications, especially old nursery catalogues, there are already a lot but again often very incomplete and often few remaining.
  3. This may be duplication but rather than have links to bibdigital, have the data in BHL in some way so names can link to protologues, currently that is not possible for those.

 

Best wishes,

Rafaël

 

 

 

From: tax...@googlegroups.com <tax...@googlegroups.com> On Behalf Of Roderic D. M. Page
Sent: 16 April 2026 13:39
To: Taxacom <tax...@googlegroups.com>
Subject: [TAXACOM] BHL future plans

 

 

CAUTION: This email originated from outside the organisation. Do not click links or open attachments unless you recognise the sender and know the content is safe.

Suspicious Email Reporting Form

Do not directly forward any suspicious emails to IT Support; please use the linked form to report it.

--

You received this message because you are subscribed to the Google Groups "Taxacom" group.
To unsubscribe from this group and stop receiving emails from it, send an email to taxacom+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/taxacom/28a74e53-aba4-4387-b823-11e08b4ecf8en%40googlegroups.com.



Disclaimer

The information contained in this communication from the sender is confidential. It is intended solely for use by the recipient and others authorized to receive it. If you are not the recipient, you are hereby notified that any disclosure, copying, distribution or taking action in relation of the contents of this information is strictly prohibited and may be unlawful.

This email has been scanned for viruses and malware, and may have been automatically archived by Mimecast Ltd, an innovator in Software as a Service (SaaS) for business. Providing a safer and more useful place for your human generated data. Specializing in; Security, archiving and compliance. To find out more Click Here.

Roderic D. M. Page

unread,
Apr 16, 2026, 11:28:48 AMApr 16
to Taxacom
Hi Rafaël,

Just to be clear:

1. So you would like BHL to check whether its journal runs are complete, or you would like a way to bring those gaps to BHL's attention? If you find a missing volume in a cellar :O would it help if there was a way to get that digitised (e.g., at Kew)

2. BHL has a LOT of nursey catalogues, but I assume most are American via the USDA. Are there other sources of nursey catalogues BHL could pursue.

3. I suspect there is politics behind this, but I've always thought that external links were a pain compared to having the content in BHL.

Regards,

Rod

Rafaël Govaerts

unread,
Apr 16, 2026, 3:52:35 PMApr 16
to Roderic D. M. Page, Taxacom
  1. So you would like BHL to check whether its journal runs are complete, or you would like a way to bring those gaps to BHL's attention? If you find a missing volume in a cellar :O would it help if there was a way to get that digitised (e.g., at Kew)

“BHL to check whether its journal runs are complete”, yes, though this may be done in some automated way (dear I say AI?).

 

  1. BHL has a LOT of nursey catalogues, but I assume most are American via the USDA. Are there other sources of nursey catalogues BHL could pursue.

One of the largest collections I know is at Wageningen university library in the special collections. But I suspect many libraries have some (like Kew).

 

  1. I suspect there is politics behind this, but I've always thought that external links were a pain compared to having the content in BHL.

I’m sure there will be politics, but politics is about compromise as well.

Best wishes,

Rafaël

 

 

 

Regards,

 

Rod

 

 

On Thursday, 16 April 2026 at 15:35:28 UTC+1 Rafaël Govaerts wrote:

Hi,

I would suggest 3 points that would help us at IPNI.

 

  1. That a big effort was done to find missing volumes, some journals have 1-2 volumes missing, others most volumes. We sometimes find them in obscure places in cellars, so it seems important to digitize those before they disappear.
  2. Add ephemeral publications, especially old nursery catalogues, there are already a lot but again often very incomplete and often few remaining.
  3. This may be duplication but rather than have links to bibdigital, have the data in BHL in some way so names can link to protologues, currently that is not possible for those.

 

Best wishes,

Rafaël

 

--

You received this message because you are subscribed to the Google Groups "Taxacom" group.
To unsubscribe from this group and stop receiving emails from it, send an email to taxacom+u...@googlegroups.com.

David Campbell

unread,
Apr 16, 2026, 7:26:53 PMApr 16
to Taxacom
Developing a way to let users flag inaccurate information could be useful feedback; of course, that would also require some way of assessing the feedback and making corrections if needed.  

The optical character recognition is not good at all in many older publications.  Some plates have especially ornate and faint script that doesn't even register, but even pages of text may have few to no correctly identified characters.  

Being able to search within one set of search results might be helpful, or equivalently having more capacity to use AND, NOT, etc. in entering a single search.  

The automatic finding of scientific names often misses all the real names while providing many imaginary ones.  One thing that might help is teaching the algorithm that taxa names do not appear in publications significantly older than the date of the publication of the names.  Possibly it would also help to identify the taxonomic focus of a publication in same way as a guide to which names are more likely to appear.  The list of possible names that the algorithm searches for is not very strong in molluscan paleontology; I would guess that there are other taxa also not as well-represented in their source list (I think it's CoL, if I remember correctly).  

A significant challenge is that different publications may format names many different ways.  The genus may be listed on one page, followed by a series of species names with just an abbreviation or no genus name, for example.  

Being able to make the publication full screen, without the additional headers, menu, etc. might be useful at times.  

--
You received this message because you are subscribed to the Google Groups "Taxacom" group.
To unsubscribe from this group and stop receiving emails from it, send an email to taxacom+u...@googlegroups.com.


--
Dr. David Campbell
Professor, Geology
Department of Natural Sciences
110 S Main St, Campus Box #7270
Gardner-Webb University
Boiling Springs NC 28017

Martin Kalfatovic

unread,
Apr 16, 2026, 10:23:01 PMApr 16
to Taxacom
Rod, will you go to the business meetings too? Will the stakeholder meeting cover the hard topics of governance, funding, sustainability, data security and durability, and a new iteration that maintains content/data/access if the current one fails? 

I think it's all well and good for a stakeholders meeting to discuss what "BHL should look like", but if there's not meaningful contributions from external stakeholders on the above topics, it could be an exercise in window dressing.

Martin 

Roderic D. M. Page

unread,
Apr 17, 2026, 4:02:07 AMApr 17
to Martin Kalfatovic, Taxacom
Hi Martin,

I think so (my BHL status is not entirely clear).

The stakeholders are people/organisations involved in either adding data to BHL, consuming data, or both. “Regular users” aren’t represented, hence my post here. Governance and money is on everyone’s mind, and some of the stakeholders will have definite views on that.

As to a “new iteration” there are noises about that. For the moment BHL runs on the existing platform, but images and text are also on Amazon’s AWS so if, say the Internet Archive goes down, BHL can switch to AWS for images.

Interesting times.

Regards,

Rod

Sent from my iPad

On 17 Apr 2026, at 03:23, Martin Kalfatovic <martin.k...@gmail.com> wrote:

Rod, will you go to the business meetings too? Will the stakeholder meeting cover the hard topics of governance, funding, sustainability, data security and durability, and a new iteration that maintains content/data/access if the current one fails? 
--
You received this message because you are subscribed to the Google Groups "Taxacom" group.
To unsubscribe from this group and stop receiving emails from it, send an email to taxacom+u...@googlegroups.com.

Roderic D. M. Page

unread,
Apr 17, 2026, 8:25:28 AMApr 17
to David Campbell, Taxacom
Hi David,

Finding names is both BHL’s “killer feature” and one of its frustrations. Doug Yanega has similar comments.

I think OCR will improve over time, there have been big advances in AI tools for OCR. 

The name finding algorithms are essentially looking for strings that “look like” taxonomic names, so they make mistakes. There are other approaches that some people are playing with, part of the challenge is scaling to cover all of BHL.

Linking names and subsequent abbreviations of the name is a challenge. Some of the older name finding tools handled this quite well, the current ones maybe not so much.

Regards,

Rod

Andy Mabbett

unread,
Apr 17, 2026, 8:48:35 AMApr 17
to Taxacom
On Fri, 17 Apr 2026 at 00:27, David Campbell <pleur...@gmail.com> wrote:


> Being able to make the publication full screen, without the
> additional headers, menu, etc. might be useful at times.

Much more than "might be" - this is definitely an issue when using
smaller screens, such as on a notebook computer.

Roderic D. M. Page

unread,
Apr 17, 2026, 10:33:41 AMApr 17
to Taxacom
Thanks everyone who responded, I took the comments, threw them into ChatGPT, tweaked the outcome, and posted it on the BHL forum https://forum.biodiversitylibrary.org/t/what-the-taxacom-community-wants-from-bhl/91?u=rdmpage 

Hopefully this captures the things discussed.

Regards,

Rod
Reply all
Reply to author
Forward
0 new messages