MAB Namespace

Da die DNB den Unterschied zwischen Namespace und Validation verwischt und dazu noch ihre Server umgezogen hat, ohne zumnidest die wichtigsten Validationsdateien auf die neuen Adressen umzuleiten, hier ein modifizierter (aber technisch gueltiger) Header einer MAB XML-Datei mit tatasaechlich am angegebenen Ort existierender Validationsdatei:

<?xml version="1.0" encoding="UTF-8"?>
<datei xmlns=""
xsi:schemaLocation="" >
<datensatz typ="h" status="n" mabVersion= "M2.0">
<feld ind=" " nr="001">BV037887150</feld>
<feld ind=" " nr="100">Gruner, Ludwig</feld>
<feld ind=" " nr="331">&lt;&lt;The&gt;&gt;
   Caryatides from the "Stanza dell'Eliodoro" in the Vatican,
   designed by Raffaelle d'Urbino</feld>
<feld ind=" " nr="410">London</feld>
<feld ind=" " nr="412">Gruner</feld>
<feld ind=" " nr="425">1852</feld>
<feld ind=" " nr="433">19 Bl.</feld>
<feld ind=" " nr="434">Ill.</feld>

Bei Bedarf – wenn etwa die DNB die Validationsdatei ( erneut unerreichbar verschiebt -, kann auch auf die hier abgelegte Datei zurueckgegriffen werden:

Naeheres zum MAB-XML Schema hier link.

Aber aufgepasst – die in den Beispielen und Schemapublikationen dort angegebene Validationsdatei ist ungueltig! An der dort angegebenen Adresse der physischen Validationsdatei xsi:schemaLocation=”” existiert nichts mehr. Ich habe deshalb eine Kopie auf gelegt und verlinke dorthin.

Zum vermurksten Namespace – der (unabsichtlich?) wie ein Dateiname aussieht und dazu auch noch wie der Name der nicht mehr existierenden Validationsdatei! – siehe auch die Diskussion hier.

Posted in Annotation | Leave a comment

Annotations and Notes (and Comments, too)

In the following text we want to present some of the different concepts for annotations, notes and comments which are present on the web.

We will briefly analyze the different setups, scopes and possibilties each solution offers.

We will also add some uses cases by the example of a fictional professor Brunelleschi who is using these tools at work in his institute and for himself.

But first of all, we need to set some things straight – first of all the terminology. In fact there are three fundamentally different ways to relate an information to a given resource available today:

Annotating which is adding a note to a line of text
Commenting which is adding a note to a website (mostly a post)
Note taking which is taking a note about a website

These meta-informations can have different scopes (or endpoints):

Site-Specific when notes, annotations and comments can relate to one specific web site only
Any Web Page, if Provided by Web Site Owner when notes, annotations and comments can relate to all websites where its owner has installed a specific script
Any Web Page when notes, annotations and comments can relate to any website

And finally, there are different places where these meta-informations are stored:

Server provided by Web Site when notes, annotations and comments reside on a server provided by the owner of the web site
Server provided by Third-Party when notes, annotations and comments reside on a third-party server, probably chosen by the user (free/open source or commercial offering)

We will now describe the most common cases in more detail.

1. Site-Specific Commenting

Our professor Brunelleschi browses through the websites searching for some place to spend the holidays – it’s August, after all  – when he comes upon a museum’s website where one of the artefacts is falsely presented as the work of one of his colleagues. And even as the work of one of his arch enemies!

Brunelleschi is really angry, but a bit undecided about what he should do: he could contact the museum staff, but right now, in mid August, there is zero probability that someone – and even less a knowledgable person – will answer his email. And he doesn’t want to end up dealing with disgruntled staff telling him he should try again later.

But then he sees that there’s a possibility to leave comments on the web sites: he decides to write a comment about the falsely attributed authorship. But wait – should he use his real name? Probably not, better avoid to give his enemies a pretext for denouncing him as petty and arrogant.

At the end Brunelleschi writes his comment using the pseudonym »Giovannino«.


Overview: Commenting is for leaving short messages regarding a blog post or web site. It might even be anonymous, even if you are forced to register with an email. The comment is not necessarily traceable to yourself, nor will it be part of your work material.

Target: The comment function works only on specially prepared web sites, like blogs.

Functions: Sharing the comment is only indirectly possible, by sending others a link to the whole website. Sometimes a function like “Inform me if someone replys to my post” is available. Replying to a comment is definitely possible. Tagging is mostly available.

Requirements: On the user side, mostly some kind of login is required to avoid automaized spam. Web site owners instead need to setup their site to allow commenting (and sometimes need to approve comments).

Storage: Comments are stored at the specific web site.There is no export functionality.

Reuse: Reuse is not foreseen.

Available services: “Comment” functions on web sites and most blogs, like WordPress, TypePad, Blogsmith, MovableType &c. Normally the comment relates to the whole post only, but there is also an exception: the plugin for WordPress, which lets you comment on single paragraphs.

End User Perspective: Very convenient to leave a comment or start a discussion. On longer threads, the missing export function becomes a nuisance: I’ve seen more serious threads copied into personal blogs by their authors just to preserve them.

2. General Web Annotations

But the same museum’s website has also some interesting stuff Brunelleschi has never seen before: artefacts which show stylistic and technical properties which can only be ascribed to foreign artisans, most probably coming from the Orient.

Obviously, the museum staff does not recognize the importance of these artefacts and even describe them as early modern provincial works.

No need to tell them, though! Brunelleschi opens his electronic notebook – he paid for it by credit card, but that was worth it – and starts to take notes.

Most of the time he »clips« the relevant webpage – that is, he grabs the text and the relevant images only without the usual garbage on the header and footer of the page – and into his electronic notebook and then adds, on a new line, his personal remarks.

This is quite fast and very handy, and much better than mere bookmarking because when he looks up his notes  – or makes a search  – he quickly finds all the texts he has clipped inside his notebook.

He knows exactly about the importance of his electronic notebook, and therefore he’s happy he’s found a commercial provider with a good reputation – not one who might be gone tomorrow – and with the possibility to export and archive all his notes.

Brunelleschi’s notebook is even securely protected with a seemingly good encryption, to make sure nobody of his colleagues can snoop in. There is only one exception: to a certian young researcher who’s dealing with oriental artists he’s given acces to all the notes relating to oriental stuff: it’s enough to »tag« (another one of these catch words! his notes with a relevant tag like »Oriental« and both of them have access to the same notes. By sharing their notes both are profiting from the other’s expertise.


Overview: Note taking allows for rapidly collecting ideas from different web sites and store them into an electronic notebook. Depending on the software you can also copy text fragments from the web page into you notes or »clip« the whole story.

Target: All web sites

Functions: Some note taking apps allow sharing notes (via URL), and sometimes even following, where you automatically see the other’s notes. Tagging is mostly available, but I haven’t yet heard of a reply function.

Requirements: On the user side, a subscription to one of the note taking services is necessary. The service can then be used for all websites. Web site owners instead need not do anything.

Storage: Notes are stored at the service provider, who might give you the possibility to export them into compatible formats (mostly XML).

Reuse: Reuse of the collected material is one of the main selling points of these systems, so tools are provided on-site or even as separate applications for your desktop.

Available services: Elektronische Notizblöcke wie Evernote, Simple Note, Springpad etc.

End user perspective: Very much common among junior researchers for its ease of use and ubiquitous availability, including mobile devices. Highly regarded for collecting and ordering (using »tags«) research material.

3. Site-Specific Annotations

The real day-time work of Brunelleschi though is the administration of a study group for his research institute, which has grown to 5 full time and 2 part time workers. It takes a lot of time and energy to guide his people … and to avoid they chase phantoms!

Fortunately, Brunelleschi has taken care to install a commenting system which allows him to look at all the writings his workers are preparing and to leave a note: most of the time it’s a simple advice, a positive note, sometimes he takes care to have a longer discussion.

It’s also kind of an incubator, because all discussions – his notes and his worker’s replies – are behind a firewall and cannot be seen from outside. But for the study group, these notes and replies which always bear the author’s name and a time stamp, are permanently visible and, in fact, end up being part of the project.

At the end of the project, all texts – and all the notes – will be permanently archived in the institute’s long term archival system, where future researchers will find them.


Overview: A side-wide annotation systems allow to annotate large texts, or group of texts, within a controlled environment. Its users are mostly employees who automatically have access to the system. The system might mostly be used for revisioning documents (e.g. manuals).

Target: Unfortunately, the system works only on specially prepared websites.

Functions: Sharing annotations is automatically part of the process as a whole group of people will be assigned to the task. But you will be able to share annotations only withing the work group. Replying, and maybe even tagging will certainly be possible, but a “following” function is rare.

Requirements: On the user side, a subscription to service is necessary – if not automatically provided company-wide. On the provider side, these systems are mostly deployed by larger corporations, research institutions, and organizations like Europeana which can justify the complex and costly installations.

Storage: Comments are stored at the company or institute only, which might – or might not – give you the possibility to export them.

Reuse: Most systems are closed in the sense that they don’t allow for (or foresee) output and reuse outside the system itself.

Available services: The only working(!) system I know of is Highlighter (link) which is a commercial service mostly used in educational context.

End user perspective: The end user will see the system mostly as employee of a company or institute – adding notes to the document pool will be part of his normal work, but the result will mostly not be his own “intellectual property”.

4. General Web Annotations

For his personal studies about Dante’s »Divina Commedia« and its influence on early Renaissance artists our professor Brunelleschi is using yet another instrument, which allows him line-for-line commenting on texts, but is adapt at being used on all web pages, not only texts which reside on his institute’s server.

In this case it is necessary to annotate texts residing foreign servers, mostly in the US but also in Italy and some texts in Germany, and to add annotations without these servers even knowing about it.

At the same time, Brunelleschi can virtually work together with his colleague in L’Aquila, Italy, because the both see the same texts and there is a mechanism to reply to the other’s annotations. On a normal day, the time it takes his colleague in Italy to respond to one of his annotations is only a couple of minutes!

Brunelleschi can display the texts, the layered annotations and even a special page where all text fragments and annotations are collected, which serves him well when he has to write down an article.


Overview: Annotating texts line-for-line is a very demanding task for any application especially if it’s supposed to work on any website. Although extremely useful for intensive scientific research, until spring 2012 there was no working(!) application available for the common user.

Target: All web sites

Functions: Sharing annotations, following other users, tagging and replying are available.

Requirements: On the user side, a subscription to (the only) one of the commercial annotating services is necessary. The service can then be used for all websites. Web site owners instead need not do anything.

Storage: Annotations are stored at the commercial service provider, who gives you the possibility to export them into compatible formats (mostly XML).

Reuse: Reuse is at the forefront of all services, with possibilities to re-arrange material, and re-tag it, up to export functionality as MSWord.

Available services: As of now, I only know of Annotary link. A kind of similar service is Scrible which copies (clips) a website and then lets you annotate it (the clipped website, not the original one!).

End user perspective: Great for scientific work. Very young service provider, so no long time experience available. Export functions – as of now – are only print/pdf, but more (MSWord) is promised.

5. General Web / Public Commenting

At the end of the day, and before leaving for home, Brunelleschi finds some time to dedicate to his hobby: in fact, few know that he is a dedicated fan of a special kind of small hand-built boats named »gozzi« which are typical of the Amalfi and Sorrento peninsula. In the last days thee was some talk about a new boat by famous builder Aprea, and Brunelleschi immerges himself into the discussion.

Finally he finds the first images of the new boat – an open 7.50m boat – and puts a comment on the web page. At the same time, he shares the comment with his friends by using »Circles« function, which allows him to fine-tune who exactly will see his post: only his close friends, only the Aprea aficionados or the general public. His choices depend on the subject matter, but for today he shares his comment with everyone. Soon he sees the first responses, and some further links to other images.

The comment and response system is quite rudimentary, nothing scientific here, it’s more like a marketplace where everyone shares his comments, replies to the others’ comments and then moves on. The exact membership of each circle is hard to know, because people come and go, or might only be silent onlookers.


Scope: Public commenting systems are mostly used for rapidly sharing news to a larger, sometimes undefined audience. The content of the comment is mostly not much more than a simple “Hey, look here!” but it’s ease of use makes up for its shortcomings.

Target: Specially prepared websites

Functions: Sharing, following, tagging and replying are all implemented.

Requirements: On the user side, a subscription to one of the social services is necessary. The service can then be used for all participating websites. Web site owners instead mostly need to install only a small script on their site.

Storage: Comments are stored at the social service provider, who rarely (as of now, I only know of Google’s Takeout service) might give you the possibility to export them.

Reuse: Only few services allow export of social comments, and thus a possible reuse – although such a reuse would be quite meaningless without the accompanying platform (and the other users’ comments).

Available services: Social comment systems like Google+, FB Comments &c.

End user perspective: Useful for quickly and easily spreading news.


Add: RE-USE (e.g. Word.Doc export for own work, etc.)

Add: screenshots!

Add: Annotator & Yuma

Posted in Annotation | Comments Off on Annotations and Notes (and Comments, too)



The Documents

The project comprises 3 different types of documents:

  1. completely modern texts like Mark Smith’s edition of Alhacen’s De Optica, and Klaus Bergdolt’s edition of Ghiberti’s Commentary
  2. classical texts like the well-known Friedrich Risner’s edition of Alhacen and Witelo,
    and lastly:
  3. image scans of manuscripts like the already mentioned manuscript in the Casanatense library. Hopefully we will be able to show the manuscript in the Vatican library, too.

Work on these documents is still ongoing, with some documents lacking only some text figures, others, most notably the Ghiberti, parts of the text due to the complexity of transcription. The finished project will be hosted on its own platform on a dedicated server. So, please don’t copy the present URL because right now what you see is the development version.


First some words the modern texts. First of all, Marks Smith’s reconstruction of the Latin translation of Alhacen:

As you see, we present the Latin text side-by-side with Mark Smith’s English translation.

We also present Mark Smith’s critical annotations via pop-ups:


I’m not too happy about these pop-ups. First, because there are not visible all the time but only if your mouse hovers on top of the relevant symbol.

Second and most important, because it is hard or impossible to create a scientific annotation for a pop-up: I will come to this part later on. For now I just want to say that I will try to find another solution: maybe footnotes under each paragraph or in the page margins.

Another challenge will be to add the necessary text figures alongside the explanations provided by Mark Smith. Most probably I will present these right inside the existing text documents.


Let me also show you our approach regarding Klaus Bergdolt’s editon of Ghiberti’s Commentary, the third book.

What I did here was something very similar to Mark Smith’s edition, with the rendering of the original Italian text alongside the German translation. Unfortunately, there far more footnotes and they are of two separate types: notes regarding the Italian text, like in a text critical edition, and notes regarding its interpretation.

I therefore decided to keep the original distinction by letters or numbers for each note, even more so as they are cited that way inside the text.

Since the original layout is quite complex, I also provide a link to the scanned page:

A full rendering of Bergdolt’s text will take some time, but I hope we can provide at least the Italian original and the German translation, adding the text critical notes sometime later.


I will only briefly mention the manuscript in the Casanatense library, which I show here:

The single scans can be chosen from the middle strip and are then displayed in the main viewport, as it is more or less usual these days. We also also have an autoscroll feature which guides you through the document.

We took special care to display the single scans always with maximum screen size, that is the page scan automatically adapts to various screen dimensions.

As I mentioned earlier, we are also in contact with the Vatican library for a presentation of their manuscript inside the Perspectiva+ platform. Let’s hope we can get their permission!


I will now speak of Friederich Risner’s edition of Alhacen and Witelo, which was printed in 1572.

The scans for the print edition have been provided by the Documentation Service Strasbourg and  the transcription was done by the ECHO project of the Max Planck Institute for the History of Science.

As of today, the text presents all the original type forms, abbreviations and ligatures. Here is a sample:

Obviously, a normalized transcription, like the one displayed by the ECHO project, would be much nicer. But before I set out to modernize the transcription the underlying text first has to be checked for typos (of which there are many).

Let me now turn to the task of turning the digitized text into modern edition.

Text presentation always depends on its audience, its medium, and its time and place.

For the Perspectiva+ project I tried to create a modern Risner which is highly legible and reusable for research.  Therefore I did not want to maintain all aspects of the original print edition.

This means that the layout is adapted to modern medium and a modern audience by re-ordering the text flow, reconnecting line breaks, suppressing page breaks, and so on. I always opted for legibility and re-usability.

First, the whole text of each of his books has been put together in one web page, allowing for easy scrolling and searching:

A hierarchy of first, second and third level headers for books, chapters and sections is presented in the toolbar on the left.

For better readability, page break markers are put inside the running text, instead of breaking the text flow:

In order to give readers the chance to check the original page I’ve made the small page break markers clickable:

I further introduced a »liquid text flow« with 1, 2 or 3 columns which allow for optimum readability on different screens. Mobile browsers will render the text on 1 column:

Most older screens will end up using 2 columns:

But modern wide screens will render the same text using 3 columns:

It’s also working on mobile devices. Here’s a screenshot from a current production smartphone in landscape mode:

Since text flow is different in our digital version, figures will also be positioned differently than in Risner’s edition – although we managed to keep them still within the original paragraph. Here’s an example from Risner (p.6):

And here’s our digital rendition in a 2-column layout:

Unfortunately, the resolution of the scans available to us was too low to allow for further enlargements.

All images – the single scan pages and the text figures – have been artificially enhanced or corrected as to:

  • white balance,
  • tonal range,
  • micro contrast,
  • sharpness.

These interventions might be regarded as heretic, but please keep in mind that our goals are legibility and usability – not the presentation of a specific historical document in its most authentic state.  Anyone interested in the actual state of the specimen has to consult the uncorrected RAW files or, better yet, the document itself.

As to fonts, for the main text we use a combination of Enriqueta and, for a few special characters only, Junicode. Both fonts’ X-height has been been aligned in order to make them nearly indistinguishable. Here’s an example of both fonts used together:

All fonts are automatically downloaded and used inside the browser — the user won’t even notice it.


Let’s now see what we did to make the texts accessible and reusable for research.

Most important for scientific use are stable citation links. We therefore implemented not only links to whole books, but also to fragments like chapters, sections and single paragraphs.  These links are represented by blue circles:

They can easily be copied:

Links are also important for creating annotations. In fact annotations were a high priority for us.

At the start of the project we defined the basic functionality an annotation system should provide:

  • Text fragments: Can the user make annotations not only to sections but also to single words?
  • Images:  Again, can the user make annotations not only to whole images but to details inside the image?
  • Sharing:  Can the user create annotations for his scientific research and share them with other users?
  • Replying:  Can a user see another user’s annotations and make a direct reply?
  • Exporting:  Can a user export his annotations and keep them for himself?
  • Interchange: Can annotations be interchanged between different annotations systems?
  • Archiving:  Can the annotations be preserve for the future?

And, the most basic questions:

  • Scope:  Can the annotation system be used for all web sites and documents across the internet or is it a closed environment?

Some systems do not allow their users to make annotations to any website or text on the internet, but only to a given website. Kind of user lock-in where the annotation tools can only be applied on the documents hosted on one specific web site.

As of today, no available system fulfills all requirements. We decided therefore to … give our users a choice of some of the best existing tools. We aim to deploy 3 annotation tools:

First, Annotator: a free/open source annotation tool which can be used for the Perspectiva material but also for other websites. It’s only down side as of today is its lack of backend functionality.

Second, Annotary: a relatively young proprietary annotation system which also works anywhere in the web with a nice graphical backend,

Third, Highlighter: another commercial system with a really great backend but which is restricted to the Perspectiva website.

Our users can test these tools and choose the annotation system they prefer. A short help section for how to use the different systems is displayed on every page.

For sake of brevity, I’ll now show only the Annotator system, mainly because it’s the most technically interesting.

Annotator is one of the projects hosted by the Open Knowledge Foundation (link). Its development (link) is being done alongside the specifications to an annotation framework defined by the Open Annotation Collaboration project (link). Therefore there is little chance  that the service could one day get shut down. From an end user perspective, this means vendor-neutral annotating, and even a possibility to be compatible with future annotation services.

Here are some screenshots of the Annotator tool in use on the Perspectiva+ site.

We highlight some text and click on the symbol for annotations …


The annotation window pops up  …

We start to write our annotation … in this example we want to refer to a similar introduction by Risner made in part II of his volume on optics, by Vitelo.

Instead of a mere text we want to give a direct link. So first we create the title, using square brackets, …


Then we prepare the actual URL link, using parentheses. Now we have to switch to the text we want to link to and copy the citation link!

Here we are in the Vitelo part of Risner’s publication – the blue buttton becomes visible on hovering and shows the citation link.

Here we copy the link … a right-mouse-click will do …


And here we paste the link in our annotation:

We save the annotation …

And we’re done.  The ugly code has become invisible leaving only a nicely formatted link:


The use of cross-links inside documents or from one document to another (even on another website!)  allows scientific reuse. Different users can annotate the documents, re-edit them and, if they want so, even share their annotations.

What’s still lacking in the Annotator system is a print functionality, so users can print their annotations alongside the original documents, and annotations for images. We hope to be able to provide at least the latter inside the project time frame.

Talk held at the conference “Perspective as Practice: On the Circulation of Optical Knowledge In and Outside the Workshop”, Berlin 12-13 October 2012. The relevant PDF is PerspectivaSlides, the original presentation file is PerspectivaSlides-ODP.

Posted in Annotation | Leave a comment