Notable people born before 1700 tended to be born in years that are multiples of 10. I wonder if a lot of dates are less accurate than they're thought to be.
"When exact birth year was not tagged, the
individual was prorated over the appropriate time period."
Which makes sense as you still get a relatively accurate portrait of the scheme. I wonder how large a roll the loss of records have played in erasing a large chunk of our history.
"Judging from the spikes every decade prior to about 1800, it seems that Wikipedia generously applies birth year categories when birth decades would be more appropriate. Nonetheless, the birth categories turns out to be far better curated than the machine-readable PersonData field."
I'd expect the prorating to spread the dates evenly across the decade precisely to avoid this kind of periodicity. It seems to suggest dates have been inaccurately recorded with year precision as, say 1690 when some author along the line didn't know if it was 1686 or 1694.
"I have no good explanation for the trough centered around 1700..."
Never studied history, but some googling suggests that the Early Modern Period concluded just after that trough and the Age of Revolutions kicked off. French Revolution started in 1789. Industrial revolution in the UK started around that same time.
Looking for active or dull periods by event would have to be adjusted by 30-50 years to allow for those born to achieve notoriety, right? 1678 seems to be the deepest part of the trough suggesting that the early 1700s were dull?
Wikipedia makes any century seem particularly exciting...
Part of my intuition wants to buy into the idea that the age of exploration, and the sudden European expansion into the colonial/imperial territories of that period profoundly redirected the attention of people away from their peers, and instead toward the nature of the world at large (and maybe even humanity's place in the universe), and reduced the intensity of the run-away social chain reaction that produces interesting personalities that steal attention from each other.
If this analysis was only performed on English wikipedia, then there might be a degree of bias regarding European history. In any event, one would never expect the graph to bottom out, but possibly the reason the trough doesn't dip any lower than it does, could be that the rest of world history remains static, and picks up the slack, continuing to generate notable personalities in areas not disrupted by exploration.
Similarly, my intuition leans toward thinking that the long, steady, even duration of the trough might be an artifact of the nature of communication during the period, which also likely introduces a degree of delay between cause and effect. Long slow letter writing suddenly has to traverse an ocean or two, to maintain communication across populations spanning multiple continents.
I had seen that, but it also seemed to blend into a range of others including the earlier and more devastating Black Death. The Great Plague of London killed 100k, Black Death 75-200 million.
When I point with a cursor on a graph line, there are numbers with a caption "notable births". And then "notable births fraction". So, I suppose, first one is count...
This is interesting to me mostly because I a Wikipedia admin since 2003 and am also writing an ancient history book since 2010 wherein my scope begins to cease exactly where Wikipedia picks up.
I haven't understood why Wikipedia insists on notability requirements? Why not let every human have their own Wikipedia page? After all we are no longer limited by number of pages that can be printed.
The notability requirement is not there to keep out the plebs, it's there to keep the information on Wikipedia accurate. In effet, it sets a lower bound on the reliability of primary sources, improving the quality of the project as a whole.
I think that's right, but in another way Wikipedia does keep the plebes out:
When you defend a reverted change on the talk page all too often you get something along the lines "I don't care about this discussion, it was an IP anyway. Stay away!".
Wikipedians often see "IPs" as vandals and don't even pretend to judge on the merits.
Together with some "ownership" feeling that many Wikipedia users have regarding "their" pages, it makes for a very unproductive and frustrating experience for a casual would-be supporter.
Unless you're willing to really commit to Wikipedia, learn all the jargon and be a part of the power games, Wikipedia is effectively read-only.
As an "IP" you're mostly restricted to fixing typos.
That's perfectly fine, but a stark contrast to how the project is painting itself.
Most IRL people still don't know that wikipedia is editable (starting with my flatmate, art school age 24). Granted: They just miss the point entirely and assume it's another Linux-or-Facebook free thing. So IP is very efficient for demoing: it makes sparkles in their eyes. Then the change is reverted, serious participation requires an account.
When I'm working from a corporate computer, I keep my IP and don't log in. That way, it leaves the trail open if anyone wants to trace a bias back to my company.
In my opinion, the verifiability requirement has mostly (though not 100%) superseded the notability requirement on Wikipedia over the past ~5 years. Nowadays it's close to necessary and sufficient for something to have good third-party sources: if something has good sources it's usually presumed notable because it clearly has been noted by others (e.g. by scholars, if the sources are scholarly sources), while if something doesn't have good sources, then whether it's notable or not is moot because there's no basis for writing a Wikipedia article on it anyway.
The trough and then peak between 1970-1990 is interesting. Any insights? Are people more likely to have an article written about them in their 20s than 30s?
An interesting intersection of those is that the bar for notability is really low for ancient history, if any information is available. Minor people about whom some documentation has survived are interesting to historians, simply because the number of ancient Greeks who we know by name and about whom any information has survived is small, so essentially all of them are of interest to historians. Same with texts: every ancient surviving poem is at least somewhat notable and has some study of it (even surviving poem fragments), which is definitely not the case with every surviving poem from 1995.
True that, but if you were writing poems in say, ancient Greece, you likely weren't Little Billy writing love ballads to Susan from across the street. That someone was writings poems itself makes them notable than their compadres.
There's this woman, Allia Potestas [1], who is notable just because we happen to have her epitaph, and the epitaph happens to mention details of some sort of polyandrous relationship configuration that she had. I doubt she would be notable if she lived today.
Additionally, I'm pretty sure that poetry in Ancient Greece was if anything more commonplace than it is today, and not necessarily some high action of the literati. I think we're used to a world where the marginal cost of reproducing creative works is very low, so everyone can hear the same songs and read the same books as everyone else - before the era of easy reproduction, I think you had a lot more "local talent" generating their own little poems, songs, plays, etc.
It would be interesting to plot the PageRank score of each person's page as well, showing their significance. I would expect a long tailed graph (as we filter only the very significant persons the farther back we go in time; less notable ones gradually pass into oblivion).
I think even better metric would be number of translations to other languages for each person's page (maybe in combination with PageRank), I noticed it correlates with "notability" (not just for people) pretty good and I even have been meaning to do it myself for some time :)
The trend isn't surprising. 15 minutes of fame is easier to achieve than 300 years. It's also less work to research Justin Timberlake than Justinian because one need not even open a book.
I am surprised; I would have expected the over-weighting of recent births to be far larger. The graph shows 1920s births and 1970s births as about equally likely to be notable. That doesn't fit with my experience that every minor 30-something writer or actor has their own page.
I guess the birth date of people on Wikipedia normalized in order to be indexable?
I have a (stupid) theory that I would like to test but I don't know how to go about it. The theory is that many (most) people who go on to achieve great things are orphaned (or estranged) from their father at a young age (before they're 6).
A first step to test it would be to calculate the correlation between the age of the person when their father died, and some measure of their "greatness" (length of article? number of sources?)
But I don't think the information about the year of death of each parent is normalized; any idea in how to go about this?
But you're going to have a serious problem with missing data. Date of parent's death will only be there (in structured form) if the parent is themselves notable enough to have a wikipedia article. Otherwise, it'll only be in the text.
So I think you'll only get a meaningful result if you manually clean the data. Choose a 'neutral' source of notable people (e.g. Who's Who). Hire somebody via odesk or mechanical turk to find and record age of father's death. Compare with averages across the entire population.
Basic manual research is now cheap; take advantage of it!
45 comments
[ 3.4 ms ] story [ 123 ms ] threadOf Internet , i cannot disprove either.
Looks like it's a Wikipedia problem.
Never studied history, but some googling suggests that the Early Modern Period concluded just after that trough and the Age of Revolutions kicked off. French Revolution started in 1789. Industrial revolution in the UK started around that same time.
http://en.wikipedia.org/wiki/Early_modern_period
Looking for active or dull periods by event would have to be adjusted by 30-50 years to allow for those born to achieve notoriety, right? 1678 seems to be the deepest part of the trough suggesting that the early 1700s were dull?
Wikipedia makes any century seem particularly exciting...
http://en.wikipedia.org/wiki/18th_century
Anyone more familiar with that period?
If this analysis was only performed on English wikipedia, then there might be a degree of bias regarding European history. In any event, one would never expect the graph to bottom out, but possibly the reason the trough doesn't dip any lower than it does, could be that the rest of world history remains static, and picks up the slack, continuing to generate notable personalities in areas not disrupted by exploration.
Similarly, my intuition leans toward thinking that the long, steady, even duration of the trough might be an artifact of the nature of communication during the period, which also likely introduces a degree of delay between cause and effect. Long slow letter writing suddenly has to traverse an ocean or two, to maintain communication across populations spanning multiple continents.
https://en.wikipedia.org/wiki/Age_of_Exploration
http://en.wikipedia.org/wiki/User:Mike_Linksvayer/Article_of...
When you defend a reverted change on the talk page all too often you get something along the lines "I don't care about this discussion, it was an IP anyway. Stay away!".
Wikipedians often see "IPs" as vandals and don't even pretend to judge on the merits.
Together with some "ownership" feeling that many Wikipedia users have regarding "their" pages, it makes for a very unproductive and frustrating experience for a casual would-be supporter.
Unless you're willing to really commit to Wikipedia, learn all the jargon and be a part of the power games, Wikipedia is effectively read-only.
As an "IP" you're mostly restricted to fixing typos.
That's perfectly fine, but a stark contrast to how the project is painting itself.
When I'm working from a corporate computer, I keep my IP and don't log in. That way, it leaves the trail open if anyone wants to trace a bias back to my company.
Then it should be a verifiablity requirement.
I wrote a longer-winded version of this argument a while ago: http://www.kmjn.org/notes/wikipedia_notability_verifiability...
Everyone can have a page on Facebook or some other site with different goals to wikipedia.
http://en.wikipedia.org/wiki/World_population
Additionally, I'm pretty sure that poetry in Ancient Greece was if anything more commonplace than it is today, and not necessarily some high action of the literati. I think we're used to a world where the marginal cost of reproducing creative works is very low, so everyone can hear the same songs and read the same books as everyone else - before the era of easy reproduction, I think you had a lot more "local talent" generating their own little poems, songs, plays, etc.
1. https://en.wikipedia.org/wiki/Allia_Potestas
I have a (stupid) theory that I would like to test but I don't know how to go about it. The theory is that many (most) people who go on to achieve great things are orphaned (or estranged) from their father at a young age (before they're 6).
A first step to test it would be to calculate the correlation between the age of the person when their father died, and some measure of their "greatness" (length of article? number of sources?)
But I don't think the information about the year of death of each parent is normalized; any idea in how to go about this?
But you're going to have a serious problem with missing data. Date of parent's death will only be there (in structured form) if the parent is themselves notable enough to have a wikipedia article. Otherwise, it'll only be in the text.
So I think you'll only get a meaningful result if you manually clean the data. Choose a 'neutral' source of notable people (e.g. Who's Who). Hire somebody via odesk or mechanical turk to find and record age of father's death. Compare with averages across the entire population.
Basic manual research is now cheap; take advantage of it!
May 2, 2014 English Wikipedia Archive http://dumps.wikimedia.org/enwiki/20140502/enwiki-20140502-p...