Catholic Version of the Church Fathers

Ken McGuire

https://community.logos.com/discussion/comment/789262#Comment_789262

Fr Devin Roza said:

Finally, may I point out that all of this might cost radically less than what Faithlife initially estimated... at least if Faithlife is able to partner with the University of Leipzig. They have decided to digitalize Migne themselves and offer it as open source to the world for free.  A good chunk of Patrologia Latina is done in rough draft form. And they are working on PG as well. CSEL is also included in their list of works to tackle (I have close to 8000 references to CSEL in my library). This is the same type of project that brought us the Perseus collection (in fact, they have officially joined forces with Perseus). 

Wow! All of this is quite impressive. I have to book-mark this - and pray for it to come like Perseus....

EastTN

https://community.logos.com/discussion/comment/789262#Comment_789262

Fr Devin Roza said:

Mark Barnes said:

What I'm saying is that it's almost impossible to to make links from references like the one in my example above unless we have author/work datatypes in addition to Schaff and Migne datatypes. Creating a Migne datatype on its own will only help for reference that explicitly refer to Migne, and that solves only a fraction of the problem.

Thanks, Mark. We actually agree about this - my suggestion of mapping implies what you state here, as well as my suggestion that the author/work datatypes (which I was simply referring to as "other" datatypes of fathers) be created as need be when other resources exist of those works. 

I think there's a persistent misunderstanding about what MJ is proposing.

MJ. Smith said:

... I'm not proposing logosref:Migne_XX_33_4, I'm proposing (insisting even) that the rules to generate Augustine.De_Civ._Dei  have to cover all documents in Migne and need to be "implemented" for all of Migne so that new resources are not released without datatypes because of the work required to create new datatypes. ...

If I'm correctly understanding what she's proposing, the data types would not be limited to links that explicitly reference Migne. The primary use her proposal makes of Migne is simply to define the universe of patristic works for which data types should be created.

Louis St. Hilaire

https://community.logos.com/discussion/comment/789262#Comment_789262

I don't know if I have a whole lot to add to this discussion, but maybe I can help clarify a couple of things from an insider perspective.

First of all, data type creation isn't a trivial process. Our data types and tagging are smart because we encode a lot of information in them, and this is what makes them expensive to implement. Each data type requires research, coding, testing, and documentation. Each data type adds at least a little bit to the complexity of the software and the tagging process (and, thus, the overall expense of new resources).

Secondly, we always reference Migne when we create new patristic book/chapter/paragraph data types because we want to make sure we can use the data type to tag references everywhere, and Migne is usually the standard to be referred to, but the existence of Migne as a standard doesn't really make the process for creating these data types less complex

For the purposes of citations and data types, Migne really isn't "one thing" (unless you're talking about column numbers, which we're not). The work/book/chapter/paragraph citations aren't really a single unified citation scheme but hundreds of unique, independent, schemes. The work/book/chapter/paragraph divisions don't usually originate with Migne but come from some combination of the authors' original book divisions, the capitula of medieval copyists, and the numbering of previous editions. Each has its own history, its own quirks, and its own variants that have to be understood and encoded if we're going to do it right.

The fact that they've all been collected into a giant set doesn't really make them any less complex as individuals or provide any advantage other than a common location to refer to. Consequently, I don't see an easy, top-down solution that avoids the hard work of dealing with each author and work individually. And the risk of a top-down solution based on superficial analysis is that you'll end up with something like a Bible data type that really only supports the KJV--or worse, because Migne, as a corpus, is orders of magnitude larger and more complex than the Bible.

To support book/chapter/paragraph numbers for every text in Migne would require hundreds of new data types. We would have to ramp up data type creation significantly. And, to do it well, we would want to put people on it who are familiar with the literature, have some facility with the ancient languages, and know our data type system and codebase. Those people are high demand for other projects.

I think we can exercise more foresight in the creation of data types (and Kyle is shifting data type development to do this--e.g., if we're already creating a data type for so-and-so's works, and we've only got 6 of his 10 commonly cited works, we'll go ahead and support all 10 in the data type at the outset), but given an effectively infinite number of things we could add data type support for, we have to make some kind of judgment about how to allocate our finite resources.

Currently, we do so based primarily on what will deliver the most functionality soonest. This is why we tend to focus on things that are available in the system and commonly cited. Unless we're going to devote significantly more resources to creating new data types, it doesn't make a lot of sense to spend a lot of time on things that we don't have a source or many citations for. (I should also mention that without a source in Logos, it's quite a bit more difficult to analyze a text, test a data type, and QA the references we tag, so we increase the risk of having to re-do things when we acquire a source anyway.)

Fr Devin Roza said:

Finally, may I point out that all of this might cost radically less than what Faithlife initially estimated... at least if Faithlife is able to partner with the University of Leipzig. They have decided to digitalize Migne themselves and offer it as open source to the world for free.  A good chunk of Patrologia Latina is done in rough draft form. And they are working on PG as well. CSEL is also included in their list of works to tackle (I have close to 8000 references to CSEL in my library). This is the same type of project that brought us the Perseus collection (in fact, they have officially joined forces with Perseus). 

Let's just say that we're already watching these projects very closely.

MJ. Smith

https://community.logos.com/discussion/comment/789297#Comment_789297

Thanks for the perspective, Louis.

SineNomine

https://community.logos.com/discussion/comment/789297#Comment_789297

Louis St. Hilaire said:

Fr Devin Roza said:

Finally, may I point out that all of this might cost radically less than what Faithlife initially estimated... at least if Faithlife is able to partner with the University of Leipzig. They have decided to digitalize Migne themselves and offer it as open source to the world for free.  A good chunk of Patrologia Latina is done in rough draft form. And they are working on PG as well. CSEL is also included in their list of works to tackle (I have close to 8000 references to CSEL in my library). This is the same type of project that brought us the Perseus collection (in fact, they have officially joined forces with Perseus). 

Let's just say that we're already watching these projects very closely.

I shall just say thank you.