How to create a gazetteer - what I've learned

I recently completed the seventh year of continual work on the Mycenaean Atlas Project. In that time a great deal has been done but that's not what I want to write about now. I want to write about what I've learned during this work - and one thing in particular. 

 When creating a gazetteer or Atlas of the world of antiquity you must tell the user a story about each site. 

A gazetteer is not simply a matching list of feature names with latitude/longitude pairs. No usable gazetteer of antiquity can be written in this fashion.  Why not?  Because there are too many sites whose positions we do not know for certain.   So when choosing a lat/lon pair it almost always is a judgment - an approximation or a guess.   The user has to be told what went into that judgment.  In that case, it is crucial to give the user some hints about how to proceed in the face of uncertainty. At the very least the user must be given in full the references that you used (and with internet links if possible). 

Full disclosure is important when articles and/or books are not available to the gazetteer creator. For example, the ancient city of Iton in Phthiotis poses a number of geographical puzzles concerning not only Iton but the nearby sites of Zerelia, Karatsadagli, and Marmara. These sites form a little conceptual cluster which may or may not have something to do with a temple (the Barrington Atlas shows Iton and Marmara as cult sites). The gazetteer writer has to sort the sites out, locate them, evaluate their significance, and in particular, discuss the problem of whether there was or was not a cult of Athena Itonia in that area (as Strabo claims).  In my Commentary on Pleiades I went all the way back to the original investigators but I wasn't really able to seal the deal - to remove the last bit of ambiguity. The reason? I have no access to Lalonde's Athena Itonia: Geography and Meaning of an Ancient Greek War Goddess from just three years ago. Brill wants $195.00 for it and Amazon wants $143.00. Impractical either way. This isn't really a complaint. For one reason or another, this happens a lot and to just about everybody. How did I handle this? This is the message I put at the end of my comments for Pleiades 540935 which deals with Marmara. 

"Most of these questions should now be resolved in Lalonde [2019].  Regrettably, Lalonde's book was not available to me at the time of writing." 

Pleiades Commentary on Marmara (Pl. no. 540935)

 Now that the user is warned he or she might be able to get their hands on a copy of Lalonde and figure out the ambiguous parts. 

- Room has to be given to scholarly disagreements.  In the case of Iton which I mentioned above both Stählin [1924] and Philippson [1950] think that there was a temple there.  Roller [2018] thinks not.  The user needs to be told.

 - There are puzzles of transliteration to work through. The user has to be told what these are if they are material. 

 - Bibliography, bibliography, bibliography.  The user needs to know what research the position is based on.  Gazetteers are not just a circular citation game where gA cites gB and gB cites gC and gC cites gA. The user needs to be grounded at least as well as the gazetteer maker himself or herself was grounded. 

 - Misleading descriptions by the primary investigators. Investigators routinely get directions wrong, distances wrong, etc. Investigators are human. We get it. But when these problems are identified then the resolution (if there is one) must be explained to the user.   Here's an example: of a site in Euboea Sackett says 

" ..., Kherronisi, is a rocky headland immediately south of the island church Ayios Nikolaos. It is about 25 m. above the sea ... "[1] 

But Kherronisi isn't south of the Ayios Nikolaos - it's to the NE.  And the elevation of Kherronisi is not 25 m. - it is no more than 15 m.  I nearly made a serious error here but I finally convinced myself, in the teeth of Sackett's description, that Topostext was right in their placement and so I followed Brady Kiesling's solution.[2]

Pleiades Commentary on Helleniki (Pl. 540809)

Status of the commentary on Pleiades:  As of 12/11/22 I have completed some 400+ corrections/annotations to the Pleiades dataset.  These corrections complete the repair of the  rounded coordinates for mainland Greece as far north as Thessaly, the Cyclades, Crete, the Dodecanese, and some of the islands of the Aegean NE.  I have made the data table of these corrections and supplemental notes available to Pleiades. Perhaps they'll be interested.

Because there are about 5000 potentially correctible sites I won't ever be able to finish this.  But already Pleiades, for the Greek worlds and as seen through my Digital Atlas, has a much bouncier and usable feel.  It starts to be what it should be - a product that pays attention to correctness and usefulness and, as a result, gives the user confidence in the result.  And using the Digital Atlas you can also easily compare  Pleiades' solutions with the solutions of the Vici, DARE, and Trismegistos data sets as well as to the Topostext gazetteer and to de Graauw's Harbors.  

The Digital Atlas is now a working prototype of a full-up gazetteer of Antiquity.

And, as always, this is a useful reminder.

Footnotes

[1] Sackett et al. [1966] 42, #11.  'Elliniko'.

[2] Topostext is here, Pleiades is here, and my own solution is here.

Bibliography

Sackett et al. [1966] : Sackett, L.H. with V. Sankey, R.J. Howell, T.W. Jacobsen and M.R. Popham, 'Prehistoric Euboea: Contributions toward a Survey', Annual of the British School at Athens (61) 33-112.  1966.  Online here.

The Commentary on Pleiades

 The Mycenaean Atlas Project is more than just a Bronze Age gazetteer.  It is an accessible viewer for the entire ancient world - and at every period. It accomplishes this goal by integrating several very different data sets.  Here's an example:

The city of Aigion in Achaea as displayed in the Digital Atlas

One of those data sets is Pleiades.  But how do you handle or display a dataset with as many serious defects as Pleiades?  Its deficiencies are of three kinds.  Of Pleiades' approximately 39,000 records about 7,000 have no lat/lon coordinates at all.   As a result, these points cannot be displayed at all and  why they are in the dataset in the first place is a mystery.

Pleiades next problem is the 5,000 or so sites for which the lat/lon pairs are rounded to the nearest quarter degree.  This indefensible practice (and never corrected during the entire 20-year existence of this DB) puts these sites off by about 12 km. on average.  Even worse, it creates about 400 or so 'stacks' in each of which some 10 to 20 sites have exactly the same rounded coordinates and cannot be disambiguated.  (See the example of Boreion, below.)

The remainder 26,000 sites are simply inaccurate.  They are off by about 900 m. on average because this part of the DB was intended for the Barrington Atlas - a printed product in which accuracy, at the given scales, could never be better than a km. or so.  Accuracy is expensive and you never pay for more than you need.  This practice was fine for a printed product but a disaster when moved to the digital world in which representable locations are accurate to less than a meter.  The need to develop precise and accurate data for the digital world is far more urgent than it is for the world of the printed atlas.

When showing Pleiades data in a gazetteer viewer such as the Mycenaean Atlas Project the overwhelming temptation is to correct these numerous errors by adding additional information to the Pleiades data. Providing accurate lat/lon pairs is a chief priority.  The design problem is, somehow, to add that information in a way that new versions of the Pleiades database don't write over what people external to Pleiades, such as myself, have added.  How do we best add and store supplemental and updated information for Pleiades sites and how should this new data be displayed?

What is required is a Commentary on Pleiades.

      To support this Commentary I created a new DB table that would hold any additional information or corrected lat/lon pairs.  A table that is independent of the Pleiades database would not be overwritten by new versions of the Pleiades database.  It would exist alongside the Pleiades DB. And yet the new data in this table could still be connected to the old Pleiades database through the Pleiades id no.  That way if there was an entry in the new table for any specific Pleiades site record my browser would display THAT corrected/additional data instead of the incorrect Pleiades data.  This required that I accompany my new DB table with a new browser page in .html.  This will be the new Commentary on Pleiades page which stands between my viewer and Pleiades itself.  In addition to corrected information this new page maps the Pleiades places to their correct location on a new map instead of to the incorrect rounded quarter-degree locations where they are now displayed.

Design of old and new interaction with Pleiades.

The upper half of this diagram shows the current situation; the Mycenaean Atlas Project takes information from the unmodified local copy of the Pleiades database and displays it.  That display has a button that, when clicked, takes the user back to the original Pleiades page for more information.

The lower half of the diagram shows the new design.  In addition to the local copy of the Pleiades database as before (orange), there is a brand-new table that contains my own contributed material.  On the browser map, when the user clicks the Pleiades icon for more info they don't go to the Pleiades page as before but to the new Pleiades Commentary page that shows the corrected lat/lon pair along with more info and a browser map that shows the icon in its corrected position.  A link on that supplemental commentary page will send the user to the original Pleiades page (dashed line).

This whole process is transparent to the user who will only see the new commentary page when he or she clicks on a Pleiades place for which I have created commentary. 

The new Pleiades Commentary page will look like this (annotations in red):

The new Commentary on Pleiades page (here for Belminatis, Pleiades 570152).

Now I'd like to show a situation in which this commentary is invaluable.

Mt. Boreion in Arcadia is Pleiades 570158.  Pleiades gives its position rounded to the nearest quarter degree like this: 37.25, 22.25.  This is exactly the same position as seventeen other Pleiades sites.  I list them here:

Pleiades grouping 18 sites at the very same location: 37.25 N, 22.25 E.

This sort of practice is common in Pleiades and it must be corrected.  Even a casual user does not deserve to have all these sites shown in the identical location.

I corrected the rounded position of Boreion (37.25 N, 22.25 E) to  37.429300°,  22.342900°.  On the next map I compare my position, Topostext's position, and  Pleiades position:

Placement of Mount Boreion (Attica) in Pleiades, Topostext, and Mycenaean Atlas Project.

I have not adopted Topostext's position here but the rounded Pleiades position (at the bottom of the map) is 21 km. from my proposed location and almost 19 km. distant from Topostext's. 

My Commentary to Pleiades will change all this. If you search for 570158 or 'Boreion' and display it on the digital atlas you will be shown this next page (Be sure to select 'Pleiades' layer from the (not shown) map layer selector on the top right):

Mt. Boreion (Pleiades 570158) displayed in correct position.

The top link in the info box ("Pleiades (S) 570158") will not bring the user to the main Pleiades page for 'Boreion'.  Instead the new Pleiades Commentary screen will be displayed:

Commentary on Pleiades - supplemental information page

My goal for the future is to correct all the locations whose lat/lon coordinates have been rounded.  For the Peloponnese alone this amounts to about 160 such rounded markers.  Presently there are about 150 corrected Pleiades sites (primarily for the Peloponnese) and the work of adding corrections to Pleiades will be ongoing.

Pleiades' pages and how to fix them

 How Pleiades pages look:

If ever there was a mutual appreciation society that society is Pleiades.  Look at all the bloviating thank-yous and credits-given on the left side of this window.  No user cares about that and all this extraneous info should be relegated to a secondary page that users can look at if and when they want to.

The map is way too small.  I mean, the map is what it's about, right?  Unless naming the several  Pleiades honchos over and over is the actual purpose of Pleiades ... that couldn't be the real purpose could it?

With respect to the map panel (which is way too tiny to be useful) we're zoomed in so far that context is lost. What can P. do about restoring that context?   Even for a specialist it's easy to forget exactly where we are.  And don't get me started about the annoying zoom-in-from-outer-space (ZIFOS, tm) which is cool and kicky the first time you see it but is actually a time-waster and the fiftieth time makes me -like Oedipus - want to scratch out my eyeballs

How Pleiades pages should look:

In the revised version we've done away with the annoying zoom-in-from-outer-space and settled on a nice large map.   Context which is lost by being zoomed-in is restored with the new spotter map which you see on the upper right.  The spotter map is zoomed out (level 4,5?) so that you can tell at a glance where you are in the world.

All the ridiculous back-scratching, log-rolling, and mutual thank-yous are moved off to one of the tabs on the top.  Users can click on that tab if they want to see whose bright idea this site was but - spoiler alert - they won't.  I left the bibliography off of this new map but there is room for one or a button that will pop the bibliography up on a new window.  New windows for subsidiary purposes are easy to write and not enough advantage is taken of this idea.

And, in fact, now that I look at it, we should lose the right-hand side margin which is crapped up with stuff that could easily be buttons.  How about this:

The new and improved, user-forward, Pleiades page!

Lean and mean.  No wasted space.  User's needs are foremost.  The search field has been moved to where it belongs and is enlarged so as to be actually visible.  Do you see the new Credits button?  No?  Good.

Instead we have something actually useful which is a new bibliography button.  When you click on it you get a simple bibliography page (not shown).  Now if  Pleiades would just fix all the broken bibliography links and then link to something actually useful instead of a Zotero page.

Next, On to Peripleo!  If it ever works.

Digital Antiquities Databases; Their Visualization and The Mycenaean Atlas Project Digital Atlas

Several of the common digital antiquities databases have very poor or non-existent visualization tools.  Their focus is showing you individual sites on a map and with precious little other support.  The least satisfactory in this regard is Arthur deGraauw's Harbors project.  Dr. deGraauw is an outstanding expert in the field of harbor research; he freely and generously makes his data available to everyone.  This data is, however, in the form of a spreadsheet (computer types refer to this as a 'flat file') and no map visualization is possible on his site.  The remainder among these various DBs provide a dedicated page for each site along with minimal, mostly useless information about it, but without any larger visualization.  The best user interface among these databases is Vici which provides a visualization map for each site and shows you the context.  I do notice, though, that when I use Vici that there is often no site description or indication of the site's significance.   Here's an example: this shows Neftina on Lemnos which appears to be a quarry although we are not told this:

Vici is actually pretty good and contains nearly everything you want in map visualization.  Vici also  provides context in the form of other near-by sites.  The zoom-level is user controlled and jumping to another site does not require you to go back into outer space before zooming in again.  

Multiple Databases

When research involves the several popular databases dealing with antiquity it necessitates running back and forth from one window to the next and trying to put all this data into context.

   

There must be a better way.

What do we require when visualizing geographic data from antiquity?  Ideally we should be able to see all the data simultaneously on one map so the databases are inter-comparable.  Is the site coverage around Peristeria in Messenia significantly different between Pleiades and Vici?   To answer this question you need these two DBs on the same map so that they can be compared.  And each site on this combo-map should link back to the dedicated page which contains more info about that one site.  The map itself should be interactive. You should be able to click/tap on it and have it redraw centered on where you've clicked/tapped; restoring the data context at the same time.

 We're talking about a lot of data here.  The M.A.P. Digital Atlas can display abut 100,000 data points.  So when the user looks for one site (Chania on Crete, say) how many surrounding DB sites should be drawn automatically? 

Dumping the entire DB is too much.   It's time consuming.  Most of the DB would not be of concern to the user.  And the program response would take a significant hit.

Showing just the individual site is too little.  (Pleiades is worst in this regard.  P. annoyingly zooms in from outer space for every site search and it never shows any context.) 

Pleiades: Searching for Kydonia.
Phase 1: Outer Space!

Topostext shows the context around a sought-for site but it also forces the user back into outer space in most instances.)

The Mycenaean Atlas Project Digital Atlas chooses a middle way.  What is needed is enough context so that the user can work for an extended period on that specific level (Zoom levels 12-18 in most instances).  The Digital Atlas creates and  populates a frame which is about 1/5 of a degree in lat and lon.  In practice this seems to afford adequate space for investigating the immediate surroundings with little or no detectable drawing time.  When the user clicks/taps on the map it is redrawn with the new frame centered at the click.  The new frame is filled in from whatever data sets the user has chosen.

Part of Lemnos in the Mycenaean Atlas Project Digital Atlas
Features (red F paddles), M.A.P. sites (Bronze Age) in blue and Trismegistos sites are displayed

In any solution all the data sets should be visible.  In practice, even when limited by a frame, this can produce a great deal of confusing data overlay.  The user should, therefore, be able to  pick and choose among the various sets - showing all, or none, or some combination.  One wants to be able to see what Pleiades has covered as opposed to what Vici has covered, for example.  The M.A.P. Digital Atlas features a pull-down menu which allows the user to select as many/few database layers as desired.

The Lemnos map (detail)
Layer Menu visible along with 'spotter' map

In sum, then, the Digital Atlas from the Mycenaean Atlas Project adheres to the following design features:

1. Supports visualization for the following DBs: DARE, deGraauw's Harbors Project, Mycenaean Atlas Project (with features), Pleiades, Topostext, Trismegistos, VICI
2. Displays these databases in a frame that is 0.18° 'square'.
3. Differential selection among the seven data sets
4. Persistent zoom level (user-controlled zoom) (The user is never catapulted into outer space against his/her will)
5. User-controlled redraw of the frame by clicking anywhere on the map
6. Links from each DB icon to the relevant details page (Pleiades, Topostext etc., even deGraauw's Harbors).
7. User-selectable map sheet backgrounds

The result is a quick and easy way to examine coverage for a place (e.g. Siteia on Crete) among the several databases.

I have said that each icon has links to a site details page.  Pleiades icon to Pleiades site, Topostext icon to Topostext site, etc.  The only exception is Dr. deGraauw's Harbors project.  There are no detail pages on his site for individual harbors and so I created a harbor details page for him.  Those pages faithfully reflect HIS data.  Here's an example:

Sample detail page for Arthur deGraauw's Harbor data.
This is Neftina on Lemnos

Looking at this I see that this page should have a 'spotter' map like the Digital Atlas.  Always something more to do.

Databases aren't just created for random and unstatable purposes.  They are created for reasons.  These reasons take their form in the software that is written to execute over them.  One thing that is absolutely essential is careful thinking about the user interface and visualization.  With awkward or non-existent visualization tools the database is incapable of effectively displaying itself.

 

Do Not Use the Pleiades Data Set

 

The river Quartaccio is a brook (described as 'fossa') just near Marina di San Nicola on the Lazio coast of Italy. This brook diverges from the Statua (another brook) at 41.9298 N, 12.135 E and flows N towards the town of Quartaccio and Valcanneto. Along this brook at about 41.9356 N, 12.1381 E a modern road crosses it. This is approximately the position of a Roman bridge which is given no name. This bridge is marked at this place in the Barrington Atlas (Map 43, A2, no. 2).   Here's what it looks like in Google Earth:

All in all it took me about half an hour to trace down the right Quartaccio (there are several) and take a stab at the position of the bridge based on the Barrington Atlas.

In Pleiades this bridge is no. 426578 and its position is given by Pleiades as 41.875 N, 12.125 E which is in the Tyrrhenian some 3.69 km. from the nearest land and 6.83 km from the putative bridge.  See it at the bottom of the next photo:

Now these gross errors of placement in the Pleiades data are nothing new. I looked at five examples last time and I attributed them to careless digitization and the complete lack of any quality control.   I did notice that Pleiades puts their accuracy estimate for the Quartaccio Bridge as 'Rough'. I thought, 'Aha!',  if I can see what 'Rough' means to the Pleiades group then maybe I can see  how far the  problem goes.   I got the 'Rough' points out of the database that I made from Pleiades' data and plotted that on Google Earth. There are about 5000 points labelled 'Rough' in the Pleiades DB. I used a select to retrieve the first 3000 of these (about 1/10 of the entire Pleiades DB). Here's the result of plotting these 3000:

This is shocking.  

Even if the location of these sites was not known (most of them are known as I will show) Pleiades has made no effort to place them even approximately.  These points are simply dumped off at the nearest half-degree or quarter-degree vertex.  Nothing else could make such a pattern.

And most of these points are not singles.  They are multiples of 5 to 10 distinct locations which are simply dumped on top of each other. All these sites have simply been rounded to the nearest degree or half-degree. And most of these points are easily locatable. When I was criticizing P. for errors I truly believed that the number was, perhaps, about 100 points. Now I see that it is thousands of points that could be known but which are simply dumped off at the nearest vertex.

Greece looks like this:

I've said that many of these points are just carelessly dumped in a pile even though the locations they denote are easily locatable.  Here's an example.  In the Corinthian Gulf Pleiades has overlain six points at 38.25 N, 22.25 E.

I've drawn red lines from the vertex where they were dumped to their real locations.  The line drawn to Phocis is drawn to the nearest point of land in Phocis.  The lines from the  rivers are drawn from the mouths of the respective rivers.  The Styx is the modern Mavronero above Solos.  The village of Solos is marked.  Every point was easily locatable and the average error is 13.9 km.  

In Pleiades-world four rivers, an entire province, and a gulf are all in exactly the same location.   All of their 'rough' points are exactly the same.

I can think of no toponymical practice that can explain this pattern.  What's absolutely clear is that the Pleiades data set has no conceivable use for any scholarly purpose.   

The Pleiades data set is approaching twenty years of existence.  In that time the P. team had plenty of time to fix these gross errors.  I concluded long ago that Pleiades never had any interest in creating a scholarly (or even a reliable) map of the ancient world.  Soon I will dedicate one of these posts to discussing what I think their real interest is.

The Pleiades data is unsuitable as a reference.  

It is unsuitable for study of the Ancient World.  

It is unusable as a basis for developing further geographic tools.

Pleiades owes the scholarly community an explanation of why their data so bad and what they intend to do to rectify this very poor database.

Correspondent Comments on the Suitability of Pleiades Data for Scholars

A friend of mine replied to my post on the inaccuracy and unsuitability of Pleiades data for scholarly work.  (http://mycenaeanatlasproject.blogspot.com/2018/02/pleiades-data-does-crowd-sourcing-for.html)

I reproduce his letter here:

"Pleiades isn't structured to provide a single, accurate set of coordinates, though I think it hopes to evolve in that direction. Its most useful role currently is as a set of identifiers that allow links to superior gazetteers. For example, the huge error for ancient Messene is the result of displaying a calculated representative point that includes one spurious DARMC location (from the modern village of Messene), in addition to a mildly inaccurate DARMC location plus a very accurate DARE location. (DARE has assimilated a bunch of Google Earth-validated ToposText points for Greece, and from other sources as well, but uses Pleiades IDs as an easy pivot to other resources). Some of the tools Pleiades funding has produced for the purpose of improving its data are not being used very much -- one problem being a technological gap between laborious on-the-ground collectors ... and people who automate things."

Now I look at it piece by piece (original letter in red, my replies in black)

"Pleiades isn't structured to provide a single, accurate set of coordinates,"

So then where do we go from here?

" ..., though I think it hopes to evolve in that direction."

Spoiler alert: they're not going to. This would involve an enormous amount of work - actual scholarship. They're not going to commit to this because they think that this can be done on the cheap - through copying other data sets or through crowd sourcing. That's not the way that any of this works. My experience with them is that they will correct an error if you bring it forcefully to their attention but not otherwise.

"For example, the huge error for ancient Messene is the result of displaying a calculated representative point that includes one spurious DARMC location (from the modern village of Messene), in addition to a mildly inaccurate DARMC location plus a very accurate DARE location. (DARE has assimilated a bunch of Google Earth-validated ToposText points for Greece, and from other sources as well, but uses Pleiades IDs as an easy pivot to other resources). "

You've explained Messene but what about all the other errors? Nor have you questioned my estimate that approx. 1/3 of Pleiades has serious errors. What you're describing sounds like a real incestuous tangle. I don't even want to get into unpacking this beyond saying that topographical accuracy does not come from copying other data sets. It's like the old saw about buying a used car: you're just buying someone else's problems.

" Some of the tools Pleiades funding has produced for the purpose of improving its data are not being used very much -- one problem being a technological gap between laborious on-the-ground collectors (like me) and people who automate things. "

Sounds like you're describing Recogito. Is that what you mean? Are there other tools that they support? I tried out their conversion tool Geocollider. It failed miserably.

Everything about Pleiades/Pelagios/Peripleo is sham.  The Barrington Atlas data was useful for its printed purpose but now they're trying to roll that data over into the digital world where its approximative nature makes it unfit for use. And they've wrapped the whole thing up with bad and outmoded ideas - not from scholarly practice, from anthropology or toponymy or history or classical studies or any other relevant discipline - but from computer science. None of what they're doing (crowd sourcing and linked data) has anything to do with any scholarly practice or purpose but this is what they're selling and they're getting pots of money for it. In the end actual scholars wind up exactly where they started - having to do the topography of the Mediterranean from scratch. I actually know a fellow (from a very prestigious school) who's preparing a study of Mediterranean habitations. I was shown one of his spreadsheets and it was stuffed with errors since he had relied on Pleiades. In fact that's where my blog post came from.

I've asked myself what their game is. I suspect that what they want is to license their data (or perhaps their follow-on project Pelagios/Peripleo) to schools for so much per seat and deny access to non-customers. That's a time-honored approach in Computer Science. First get a lot of contributors to fork over their work for nothing under the name of something noble-sounding like 'Open Data' or the 'Semantic Web'. Second, license the whole to third parties and keep all the money.

Although I'm not sure that they can really carry this out successfully because it's the Underpants Gnomes business model.


Bibliography

DARE: The Digital Atlas of the Roman Empire.  https://dare.ht.lu.se/

DARMC: The Digital Atlas of Roman and Medieval Civilizations. https://darmc.harvard.edu/

Geocollider: https://pleiades.stoa.org/news/blog/introducing-geocollider

Recogito:  https://recogito.pelagios.org/

Underpants Gnomes: https://vimeo.com/79954057

Pleiades Data - Does Crowd Sourcing for Toponymy Actually Work?

In a previous post I discussed the magnitude of the errors we could expect in a positioning system limited to two-place decimal fractions.  I suggested that the average error that we could expect in such a system at latitudes approximately 35° N would be about 400 m.  I thought that this imprecision would bar such a system from serious work in toponymy and I asked what product would use a system so limited in mathematical range and so obviously unsuitable for the purpose.

The Pleiades Project is an attempt to adapt crowd-sourcing to the field of toponymy.  They have received generous financial support from the National Endowment for the Humanities to the amount of $1,140,780.  You can read about their grants here, here, here, and here.

What has the world of scholarship received for all this money?

Recently I was checking the positions for a random sample of geographic locations of classical and hellenistic locations in Greece.  The locations were derived from Pleiades.  This list was not selected by me but by a colleague.  Out of 45 locations 17 (36%) had significant errors.  

The smallest error was 357 m. and the largest was 3775 m.  The average error or arithmetic mean (in the erroneous part of the sample) was 1,514.7 m.   The median error was 1200 m.   I present no standard deviation because the errors are not normally distributed. In fact it appears as though the error distribution in Pleiades data might be bimodal. This suggests that there is more than one underlying cause for the Pleiades system’s inaccuracies.   

From my error worksheet.  Y-dimension is error in m.

The 13 less erroneous locations may derive from crowd contributions.  The uppermost 4 positions  (Passaron, Skotoussa, Messene, Antigonia) may be remnants of the original digitization of the Barrington Atlas data – however that was accomplished.  In other words I am suggesting that it appears as though crowd-sourcing tends to smooth out but not eliminate the original digitizing errors.  I emphasize that these are suppositions on my part.  But, clearly, the complicated history of Pleiades' data generation has left a signature in the error results.

Here is a link to my worksheet.  Occasional references in that worksheet to sites as 'Fnnn' or 'Cnnn' may be resolved at the site http://helladic.info/.

If these results are upheld by others then I would suggest that Pleiades is not an appropriate component of any scholarly work.  If a system with a two-place fractional component has an average error of nearly 400 m. then the average error of Pleiades data of more than 1500 m. suggests that Pleiades data - at some point in its generation - never had an accuracy better than 0.5 to 1.5 fractional places (10^-0.5 to 10^-1.5).

I estimate that it takes at least 2 hours of research to reliably establish a location from Bronze Age or later sites in Greece.  I do not know how many data points Pleiades claims but if it is, for example, 10000 points then it would require an effort of about 20000 man hours to complete a reliability review for Pleiades.  At 2200 man hours in a man year that would require about 9 man years to complete.  This is an order of magnitude estimate.

Crowd-sourcing in toponymy studies does not appear to work.

This defective data of Pleiades casts a shadow downstream - for example in such derivative products as Pelagios/Peripleo.

If Pleiades cannot undertake a good faith reliability study it should be rejected by the scholarly community.

Fact Computing

‘In this life, we want nothing but Facts, sir; nothing but Facts!’

Mr. Gradgrind in Hard Times

Dickens

Recently a friend suggested to me that Helladic.info was a natural candidate for Pelagios Commons.

Sites of classical learning being assimilated

 into Pelagios Commons/Peripleo

‘Absolutely not’, I replied.

And then found myself tongue-tied because I couldn’t adequately express why not.  So that’s what this blog post is about. 

Pelagios Commons is an effort to bring together a large number of websites (themselves all concerned with various aspects of Classical learning) under a single umbrella and, in some sense, merge them into a common resource.  Pelagios is, in computer parlance, a 'front end' or 'concentrator'.  They perform no original scholarship; they take the results and research of various other sites, smash them into homogeneous factlets, and then spew them out again through their results engine which is called Peripleo.

Since they contribute nothing to scholarship itself, what is it that Pelagios Commons is really trying to accomplish?  They are trying to demonstrate that Classical Studies, in all its forms, is suitable for, and can be represented by, something called 'Open Linked Data'.

What is ‘Open Linked Data’?

Wikipedia defines ‘Linked Data’ as:

 ‘In computing, linked data ... is a method of publishing structured data so that it can be interlinked and become more useful through semantic queries.’[2]

(emphasis is mine)

(‘Open Data’ is merely linked data which has no copyright or other usage restrictions.)

‘Open Linked Data’ might be visualized as an enormous network of little fact nuggets.  The ‘interlinking’ mechanism is, of course, the internet but the underlying goal is given in the definition – it is to facilitate ‘Semantic Queries’ and to make the entire congeries of facts ‘more useful’.

It’s the ‘Semantic Queries’ part which gives it away – Open Linked Data is, conceptually, part of the Semantic Web of Tim Berners-Lee.  Now I’m opposed to the idea of any scholar's wasting their time on Semantic Web efforts and have blogged about it here.  I strongly urge my readers to go back and read that entire post but I reproduce the conclusion here:

“And no automation can replace scholarship.  By scholarship I mean the several activities of gathering evidence, organizing, patient collation, reflection, judgment and the expression of these activities in the form of essays, books, diagrams and, yes, even in the form of web sites or blogs.  There is no grand slam against reality; no Tower to the Heavens that we can build that will let us storm the citadel of knowledge.   We have to patiently scrape away at the matrix of the Unknown with our small intellects in order to see it more plainly.”

I oppose the Semantic Web because it is a totalizing (and trivializing) view of knowledge that is inappropriate for fields in the Social Sciences such as Classical Studies.

For example, as the proprietor of Helladic.info I take my field to be the locations of Bronze Age sites.  So far, there are about 2400 such sites in my database and it seems as though site locations would be prima facie appropriate for an Open Linked Data approach.  But it turns out that there is ambiguity about nearly every one of those sites and many of them - as they stand in the DB right now - are likely to be wrong.  My database is not a database of sites so much as it is a database of scholarly arguments about the location and meanings of Bronze Age sites.  In other words, a database of ambiguities, many of which can never be resolved.   And if this is true of relatively straight-forward things like lat/lon pairs how much more true must it be of other Social Science topics in Classical Studies?  Where will we find the semantic web approach to Slavery in ancient Greece and Rome?  Where will we find the Open Linked Data representation of the efflorescence in Fifth-century Athens?  Where will we find the automated web-linked explanation of the collapse at the end of LH III?  Oh, but I forget.  There are available simple fact-based answers, suitable for the Semantic Web, on all these topics.  For example:

Slavery was a bad but necessary thing for Greece and Rome in an age with no petroleum, electricity, or engines.  Classical Athens experienced an efflorescence in the Fifth-century because of the indomitable Will of its people and their love of Freedom.  Mycenaean civilization collapsed at the end of LH III because some Invaders from the Sea destroyed everything.

But I exalt myself onto a plane where I do not belong.  Tolstoy was here long before me.  The famous chapter 1 in Epilogue 2 of War and Peace is perfectly ready for semantic net representation. [2]

Open Linked Data is a trivializing approach to knowledge.

Ready for your close-up, Mr. Gradgrind?

Notes
[1] It's pointless to footnote Wikipedia but the above quote was taken from the article 'Linked Data' on February 9, 2018.

[2] "Louis XIV was a very proud and self-confident man; he had such and such mistresses and such and such ministers and he ruled France badly. His descendants were weak men and they too ruled France badly. And they had such and such favorites and such and such mistresses. Moreover, certain men wrote some books at that time. At the end of the eighteenth century there were a couple of dozen men in Paris who began to talk about all men being free and equal. This caused people all over France to begin to slash at and drown one another. ..."