 |
 |
Hi Robert I think there's certainly some merit to such an approach. Along with precision however, you'll need to take account of accuracy. I'm not sure that lower granularity would go hand in hand with less integrity though. Generally speaking, as precision increases it requires greater effort (and in terms of derived data, becomes impossible) to maintain the same level of accuracy (and as one generalizes so it becomes easier). A further axis to consider would be completeness - to what extent is the level of semantic information maintained - but this operates at different levels: a lacuna in the text may remove important information, and equally a gap in our knowledge may do the same (so a Linear A tablet can be physically complete yet inaccessible to us). Presumably there is a commensurate axis for (unidentified) inclusions as well but I don't know what the word for it would be :-) On Fri, Aug 27, 2010 at 9:37 PM, Robert Barron <[log in to unmask]> wrote: > The main idea of my paper, at the moment, is that the question itself is too > large and general to answer. > > I think that if one defines various "axes" on which to place projects, one > can decide what the impact/use of a specific project, at a specific time is. > > For example, on the axis of plain-text vs rich-text, Gutenberg and Perseus > are on opposite levels. > Another axis would be visualization, which would move from text to graphical > to GIS. > HESTIA would be further along than Perseus on this axis. > So I can say that Gutenberg is least quantitative and HESTIA is most > quantitative, based on those two parameters. > > Utility/Integrity and Granularity might be two more axes. > Perhaps a project which is low on granularity would need less integrity to > be as "qualitative" than a high granularity one. > > Robert Barron > Enterprise Management Specialist - IBM Israel > http://classicarete.blogspot.com > > > On Fri, Aug 27, 2010 at 9:09 PM, Martin Mueller > <[log in to unmask]> wrote: >> >> Perhaps it would be better to replace the opposition of 'quantity' and >> 'quality' >> with the degree of precision or granularity that is appropriate to >> different forms >> of inquiry. If you focus on what Mike Witmore at Wisconsin has called the >> "philologically exquisite" you may wonder whether the choice of >> capitalization in >> Herbert's poetry is the work of the poet or the printer. If you're editing >> Herbert >> that is an important question. On the other hand, if you are interested in >> lexical >> changes of a certain genre of poetry over two centuries, attention to that >> level of >> detail may not be necessary and may in fact get in the way. >> >> The larger the data set, the more coarse-grained the inquiry. I am not >> surprised >> that at the moment a majority of scholars think that coarse-grained >> inquiries across >> very large data sets have a greater pay-off than fine-grained of >> particular (and >> typically very well known) texts. >> >> But you also need different types of precision for different types of >> inquiry. I >> vividly remember a paper published a dozen years ago in which the authors >> used >> lexical analysis to claim that the Iliad and Odyssey were the work of >> different >> poets. One striking statistic in their analysis was the much greater >> frequency of >> the particle 'de' in the Iliad. That looked very persuasive until I >> divided the >> lines of the Iliad and Odyssey into lines spoken by the narrator and >> spoken by >> characters. The narrative:speech balance is 55:45 in the Iliad and 45:55 >> in the >> Odyssey. When you look at the distribution of 'de' by narrative or speech, >> it turns >> out that they are identical in the two epics, and the much greater Iliadic >> frequency >> of 'de' is simply a function of its having relatively more narrative -- >> not a very >> plausible piece of evidence for different authorship. For that type of >> inquiry, you >> wouldn't have to worry whether a given 'de' in your Teubner or OCT text is >> a 'ge' or >> 'te' in some manuscript: the degree of variance at that level is >> irrelevant. But you >> do have to distinguish sharply between narrative and speech. >> >> I happen to be 'chorizont', when it comes to the authorship of the Homeric >> poems, >> but the distribution of 'de' is not evidence for it. >> >> Aristotle in the opening chapter of the Nicomachean Ethics has a wonderful >> passage >> about the degree of precision that is appropriate to different forms of >> inquiry. >> Sometimes you want to look through a microscope and sometimes through a >> telescope. >> Digital technology will support both. >> >> >> > Hi Robert >> > >> > It's an interesting question and one that I've also been reflecting on >> > recently. Unfortunately I don't have any good articles to recommend >> > (and would be keen to hear of any) but a potentially interesting >> > statistic: >> > >> > I've just completed a survey of projects utilising Semantic >> > technologies in archaeology and cultural heritage, which include a >> > handful of Classics projects as well. The survey covers about 50 >> > projects (the majority of those undertaken in the field to date). One >> > of the questions was whether the projects were deemed to emphasise >> > 'utility' vs 'integrity' of data. In other words, was a certain degree >> > of data corruption permissible if the overall dataset is easier to >> > make use of? >> > >> > I had presumed that integrity would be foremost, leading to problems >> > in adopting many of the automated 'semantifying' services developed >> > for other sectors. It turns out however that 2/3 of respondents see >> > utility as more desirable (for their project) than data integrity. >> > Obviously there are many caveats that need to be taken into account >> > (the archaeological emphasis, the requirements of semantic >> > technologies, etc.) but the results certainly suggest that a lot of >> > researchers are willing to permit some wooliness and uncertainty when >> > trying to ascertain the Big Picture. It's also worth bearing in mind >> > that our sources are _inherently_ corrupt (and largely arbitrary) so >> > perhaps too great an emphasis on 'quality' would be meaningless in any >> > case. >> > >> > A final thought is that these two trends play off each other. The >> > HESTIA project required us to do a lot of cleaning by hand, but this >> > was possible largely because the bulk of the work had already been >> > done automatically by Perseus. We intend to feed our results back to >> > Perseus in turn. Likewise, the hard work done by Pleiades in order to >> > provide a high quality and technically sophisticated gazetteer will >> > enable us to produce better (but still imperfect) results applying NLP >> > to the Google Corpus for our Google Ancient Places project. Thus, >> > while it remains meaningful to ask this question for a given project, >> > perhaps it is less so across the discipline as whole. >> > >> > Best >> > >> > Leif >> > >> > >> > >> > On Thu, Aug 26, 2010 at 6:47 PM, Robert Barron <[log in to unmask]> >> > wrote: >> >> Hi, >> >> >> >> Seeing as the list has erupted in a flurry of activity, I thought I'd >> >> ask a question :) >> >> >> >> I'm looking for material on the qualitative vs quantitative difference >> >> of the use of digital research in the classics. >> >> I'll explain with an example: >> >> Reading ancients texts on-line in, say, project Gutenberg, is a a >> >> difference of quantity compared to reading it in a Loeb book. >> >> You're seeing the same material, it's just faster/cheaper/easier to >> >> access. >> >> >> >> On the other hand, reading it on Perseus is a real difference - you >> >> are getting more context. If you add something like project Hestia for >> >> Herodotus then the difference is even more marked. >> >> >> >> I'll quote Wikipedia on "Digital Humanities" : Most researchers across >> >> the disciplines agree with Fr Roberto Busa's argument that the primary >> >> effect of computing is not to accelerate the pace of humanities >> >> research, but rather to provide new ways of approach and new paradigms >> >> for the enduring problems in the study of human cultural artifacts. >> >> >> >> I've found plenty of implicit material, but I wondered if there are >> >> any explicit articles on this subject (besides my future seminar >> >> paper, of course :) Â ) >> >> >> >> Robert Barron >> >> Enterprise Management Specialist - IBM Israel >> >> http://classicarete.blogspot.com >> >> >> >> >> >> >> >> On Wed, Aug 25, 2010 at 9:45 PM, Matteo Romanello >> >> <[log in to unmask]> wrote: >> >>> Dear members of the list, >> >>> As part of my PhD I'm working on a bibliography about "Classics and >> >>> the >> >>> Computers". >> >>> I'm looking specifically at general surveys, studies and discussions >> >>> about >> >>> the relationship between classics and the computers, also known as >> >>> Digital >> >>> Classics. >> >>> I drafted a first list that I'd be happy to share with other Digital >> >>> Classicists or anyone else having an interest in this. >> >>> I'd also welcome additions to my initial list: I'm thinking in >> >>> particular of >> >>> publications that I have unintentionally neglected and/or publications >> >>> in >> >>> other languages that I was not aware of. Therefore I'd like to share >> >>> it >> >>> using a tool that allows others to easily augment it. >> >>> Do you think that for this purpose it'd be better a group on Zotero or >> >>> a >> >>> page on the DigitalClassicist wiki, or what else? >> >>> Best, >> >>> Matteo >> >>> ______________ >> >>> Matteo Romanello >> >>> PhD candidate >> >>> Centre for Computing in the Humanities (CCH) >> >>> King's College, London >> >>> http://wiki.digitalclassicist.org/User:MatteoRomanello >> >>> http://kcl.academia.edu/MatteoRomanello >> >>> http://www.linkedin.com/in/matteoromanello >> >> >> > >> >> >> -- > >