Semantic search
Maybe the hottest conference ever
The Wikipedia Hacking Days are over. We have been visiting Siggraph, we had a tour through the MIT Media Lab, some of the people around were Brion Vibber (Wikimedia's CTO), Ward Cunningham (the guy who invented wikis), Dan Bricklin (the guy who invented spreadsheets), Aaron Swartz (a web wunderkind, he wrote the RSS specs at 14), Jimbo Wales (the guy who made Wikipedia happen), and many other people. We have been working at the One Laptop per Child offices, the office to easily the coolest project of the world.
During our stay at the Hacking Days, we had the chance to meet up with the local IBM Semantic Web dev staff and Elias Torres, who showed us the fabulous work they are doing right now on the Semantic Web technology stack (never before rapid application deployment was so rapid). And we also met up with the Simile project people, where we talked about connecting their stuff like Longwell and Timeline to the Semantic MediaWiki. We actually tried Timeline out on the ISWC2006 conference page, and the RDF worked out of the box, giving us a timeline of the workshop deadlines. Yay!
Today started Wikimania2006 at the Harvard Law School. was not only a keynote by Lawrence Lessig, as great as expected, but also our panel on the Semantic Wikipedia. We had an unexpected guest (who didn't get introduced, so most people didn't even realize he was there), Tim Berners-Lee, probably still jetlagged from a trip to Malaysia. The session was received well, and Brion said, that he sees us on the way of getting the extension into Wikipedia proper. Way cool. And we got bug reports from Sir Timbl again.
And there are still two days to go. If you're around and like to meet, drop a note.
Trust me — it all sounds like a dream to me.
Meat Loaf
"But it was long ago
And it was far away
Oh God, it seemed so very far
And if life is just a highway
Then the soul is just a car
And objects in the rear view mirror may appear closer than they are."
Bat out of Hell II: Back into Hell was the first album I really listened to, over and over again. Where I translated the songs to better understand them. Paradise by the Dashboard Light is just a fun song. He was in cult classic movies such as The Rocky Horror Picture Show, Fight Club, and Wayne's World.
Many of the words we should remember him for are by Jim Steinman, who died last year and wrote many of the lyrics that became famous as Meat Loaf's songs. Some of Meat Loaf's own words better not be remembered.
Rock in Peace, Meat Loaf! You have arrived at your destination.
Meeting opportunities
I read in an interview in Focus (German) with Andreas Weigend, he says that publishing his travel arrangements in his blog helped him meet interesting people and allow for unexpected opportunities. I actually noticed the same thing when I wrote about coming to Wikimania this summer. And those were great meetings!
So, now, here are the places I will be in the next weeks.
- Oct 18-Oct 20, Madrid: SEKT meeting
- Oct 22-Oct 26, Milton Keynes (passing through London): Talk at KMi Podium, Open University, on Semantic MediaWiki. There's a webcast! Subscribe, if you like.
- Oct 30-Nov 3, Montpellier: ODBASE, and especially OntoContent. Having a talk there on Unit testing for ontologies.
- Nov 5-Nov 13, Athens, Georgia: ISWC and OWLED
- Nov 15-Nov 17, Ipswich: SEKT meeting
- Nov 27-Dec 1, Vienna: Keynote at Semantics on Semantic Wikipedia
- Dec 13-17, Ljubljana: SEKT meeting
- Dec 30-Jan 10, Mumbai and Pune: the travel is private, but this doesn't mean at all we may not meet for work if you're around that part of the world
Just mail me if you'd like to meet.
Milk consumption in China
Quiet disappointed by The Guardian. Here's a (rather) interesting article on the history of milk consumption in China. But the whole article is trying to paint how catastrophic this development might be: the Chinese are trying to triple their intake in milk! That means more cows! That's bad because cows fart us into a hot house!
The argumentation is solid - more cows are indeed problematic. But blaming it on milk consumption in China? Let's take a look at a few numbers omitted from the article, or stuffed into the very last paragraph.
- On average, a European consumes six times as much milk as a Chinese. So, even if China achieves its goal and triples average milk consumption, they will drink only half as much as a European.
- Europe has double the number of dairy cows than China has.
- China is planning to increase their milk output by 300% but only increase resources for that by 30% according to the article. I have no idea how that works, but sounds like a great deal to me.
- And why are we even talking about dairy cows? The number of beef cows in the US or in Europe each outnumber the dairy cows by a fair amount (unsurprisingly - a cow produces quite a lot of milk over a longer time, whereas its meat production is limited to a single event)
- There are about 13 million dairy cows in China. The US have more than 94 million cattle, Brazil has more than 211 million, world wide it's more than 1.4 billion - but hey, it's the Chinese milk cows that are the problem.
Maybe the problem can be located more firmly in the consumption habits of people in the US and in Europe than the "unquenchable thirst of China".
The article is still interesting for a number of other reasons.
MinCardinality
More on the Unique Name Assumption (UNA), because Andrew answered on it, with further arguments. He quotes Paul: " The initial problem was cardinality and OWL Flight attempts to solve the problem with cardinality. Paul put it succinctly: "So what is the point of statements with the owl:minCardinality predicate? They can't ever be false, so they don't tell you anything! It's kind of like a belt and braces when your belt is unbreakable." "
Again I disagree, this time to Paul: the minimal cardinality axiom does make sense. For what, they ask - well, for saying that there is a minimal cardinality on this relation. Yeah, you are right: this is an axiom which hardly can lead to an inconsisten ontology. But so what? You nevertheless can cut down the number of possible models with it and get more information out of the ontology.
"I would agree - this was my main problem - how do you explain to Joe (and Andrew) that all his CDs are the same rather than different."
That's turning around the argument. If the reasoner would claim that all of Joes CDs are the same, he would be doing a grave mistake. But so would he if he would claim that all are different: the point is, he just doesn't know. Without having someone to state sameness or difference explicitly, well, you can't know.
"I did comment that the resolution, local unique names using AllDifferent, didn't actually seem to solve the problem well enough (without consideration for scalability for example)."
I am not sure why that should be. It seems that Andrew would be happy if there was a file-wide switch claiming "If I use different URIs here I mean different objects. This file makes the UNA." These files would easily be translated to standard OWL files, but there would be less clutter inside (actually, everything that would need to be done is adding an axiom of allDifferent with all the names of the file).
"I have a feeling that context is a better solution to this problem (that might just be my golden hammer though)."
I don't understand this one, maybe Andrew will elaborate a bit on this.
If you imagine an environment with axioms floating around, from repository to repository, being crawled, collected, filtered, mapped and combined, you must not make the Unique Name Assumption. If you remain in your own personal knowledge base, you can embrace UNA. And everything you need between is one more axiom.
Is it that bad?
Molly Holzschlag (1963-2023)
May her memory be a blessing.
She taught the Web to many, and she fought for the Web of many.
More FOAF
Wow, I never can't get enough FOAF :) Besides my Nodix-FOAF file the AIFB Portal now also offers a FOAF export for all the people at the AIFB (using URIs for the people besides the mailbox SHA1-Sum as identifiers. Hah! FOAFers won't like that, but TimBL told us to do it this way in Galway a few weeks ago).
If you point your smushers at the FOAFs, I wonder if you can also compile the SWRC-output into it, as they use the same URI? And can you also, by adding my own FOAF from Nodix, that I am the same person? Anyone dare's to try? :)
It's a pity Morten's FOAF explorer is down, I'd really like to try it out and browse. Isn't there something similar out there?
A tidbit more on that is also posted on the AIFB blog, but from a different point of view.
Mother philosophy
I should start to write some content on this blog soon, but actually I am still impressed with this technology I am learning here every day...
When the FOIS2004 was approaching, an Italian newspapers published this under the heading "Philosophy - finally useful for something" (or so, my Italian is based on a autodidactic half day course). I found this funny, and totally untrue.
Philosophy always had the bad luck, that every time a certain aspect of it provoced wider attention, this aspect became a discipline of its own. Physics, geometry and mathematics are the classical examples, later on theology, linguistics, anthropology, and then, in the 20th century, logic went this way too. It's like philosophy being the big incubator for new disciplines (you can see that still in the anglo-american tradition of almost all doctors actually being Ph.D.s, philosophical doctors.
Thus this misconception becomes understandable. Now, let's look around - what's the next discipline being born from philosophy? Will it be business ethics? Will it be the philosophy of science, being renamed as scientific managment?
My guess is: due to the fast growing area of the Semantic Web, it will be ontology. Today, the Wikipedia already made two articles on it, ontologies in philosophy and ontologies in computer science. This trend will gain momentum, and even though applied ontology will always feed from the fundamental work done from Socrates until today, it will become a full-fledged discipline of its own.
Moving to Germany
We are moving to Germany. It was a long and difficult decision process.
Is it the right decision? Who knows. These kinds of decisions are rarely right or wrong, but just are.
What about your job? I am thankful to the Wikimedia Foundation for allowing me to move and keep my position. The work on Abstract Wikipedia and Wikifunctions is not done yet, and I will continue to lead the realization of this project.
Don’t we like it in California? We love so many things about California and the US, and the US has been really good to us. Both my wife and I grew here in our careers, we both learned valuable skills, and met interesting people, some of whom became friends, and who I hope to continue to keep in touch. Particularly my time at Google was also financially a boon. And it also gave me the freedom to prepare for the Abstract Wikipedia project, and to get to know so many experts in their field and work together with them, to have the project criticized and go through several iterations until nothing seems obviously wrong with it. There is no place like the Bay Area in the world of Tech. It was comparably easy to have meetings with folks at Google, Facebook, Wikimedia, LinkedIn, Amazon, Stanford, Berkeley, or to have one of the many startups reach out for a quick chat. It is, in many ways, a magical place, and no other place we may move to will come even close to it with regards to its proximity to tech.
And then there’s the wonderful weather in the Bay Area and the breathtaking nature of California. It never gets really hot, it never gets really cold. The sun is shining almost every day, rain is scarce (too scarce), and we never have to drive on icy streets or shovel snow. If we want snow, we can just drive up to the Sierras. If we want heat, drive inland. We can see the largest trees in the world, walk through the literal forests of Endor, we can hike hills and mountains, and we can walk miles and miles along the sand beaches of the Pacific Ocean. California is beautiful.
Oh, and the food and the produce! Don’t get me started on Berkeley Bowl and its selection of fruits and vegetables. Of the figs in their far too short season, of the dry-farmed Early Girl tomatoes and their explosion of taste, of the juicy and rich cherries we picked every year to carry pounds and pounds home, and to eat as many while picking, the huge diversity of restaurants in various states from authentic to fusion, but most of them with delicious options and more dishes to try than time to do it.
And not just the fruits and vegetables are locally sourced: be it computers from Apple, phones from Google, the social media from Facebook or Twitter, the wonderful platform enabling the Wikimedia communities, be it cars from Tesla, be it movies from Pixar, the startups, the clouds, the AIs: so. many. things. are local. And every concert tour will pass by in the Bay Area. In the last year we saw so many concerts here, it was amazing. That’s a place the tours don’t skip.
Finally: in California, because so many people are not from here, we felt more like we belong just as well as everyone else, than anywhere else. Our family is quite a little mix, with passports from three continents. Our daughter has no simple roots. Being us is likely easier in the United States than in any of the European nation states with their millenia of identity. After a few years I felt like an American. In Germany, although it treated me well, after thirty years I still was an Ausländer.
As said, it is a unique place. I love it. It is a privilege and an amazing experience to have spent one decade of my life here.
Why are we moving? In short, guns and the inadequate social system.
In the last two years alone, we had four close-ish encounters with people wielding guns (not always around home). And we are not in a bad neighborhood, on the contrary. This is by all statistics one of the safest neighborhoods you will find in the East Bay or the City.
We are too worried to let the kid walk around by herself or even with friends. This is such a huge difference to how I grew up, and such a huge difference to when we spent the summer in Croatia, and she and other kids were off by themselves to explore and play. Here, there was not a single time she went to the playground or visited a friend by herself, or that one of her friends visited our house by themselves.
But even if she is not alone: going to the City with the kid? There are so many places there I want to avoid. Be it around the city hall, be it in the beautiful central library, be it on Market Street or even just on the subway or the subway stations: too often we have to be careful to avoid human excrement, too often we are confronted with people who are obviously in need of help, and too often I feel my fight or flight reflexes kicking in.
All of this is just the visible effect of a much larger problem, one that we in the Bay Area in particular, but as Americans in general should be ashamed of not improving: the huge disparity between rich and poor, the difficult conditions that many people live in. It is a shame that so many people who are in dire need of professional help live on the streets instead of receiving mental health care, that there are literal tent cities in the Bay Area, while the area is also the home of hundreds of thousands of millionaires and more than sixty billionaires - more than the UK, France, or Switzerland. It is a shame that so many people have to work two or more jobs in order to pay their rent and feed themselves and their children, while the median income exceeds $10,000 a month. It is a shame that this country, which calls itself the richest and most powerful and most advanced country in the world, will let its school children go hungry. Is “school lunch debt” a thing anywhere else in the world? Is “medical bankruptcy” a thing anywhere else in the world? Where else are college debts such a persistent social issue?
The combination of the easy availability of guns and the inadequate social system leads to a large amount of avoidable violence and to tens of thousands of seemingly avoidable deaths. And they lead to millions of people unnecessarily struggling and being denied a fair chance to fulfill their potential.
And the main problem, after a decade living here, is not where we are, but the trajectory of change we are seeing. I don’t have hope that there will be a major reduction in gun violence in the coming decade, on the contrary. I don’t have hope for any changes that will lead to the Bay Area and the US spreading the riches and gains it is amassing substantially more fairly amongst its population, on the contrary. Even the glacial development in self-driving cars seems breezy compared to the progress towards killing fewer of our children or sharing our profits a little bit more fairly.
After the 1996 Port Arthur shooting, Australia established restrictions on the use of automatic and semi-automatic weapons, created a gun buyback program that removed 650,000 guns from circulation, a national gun registry, and a waiting period for firearms sales. They chose so.
After the 2019 Christchurch shooting, New Zealand passed restrictions on semi-automatic weapons and a buyback program removed 50,000 guns. They chose so.
After the shootings earlier this year in Belgrade, Serbia introduced stricter laws and an amnesty for illegal weapons and ammunition if surrendered, leading to more than 75,000 guns being removed. They chose so.
I don’t want to list the events in the US. There are too many of them. And did any of them lead to changes? We choose not to.
We can easily afford to let basically everyone in the US live a decent life and help those that need it the most. We can easily afford to let no kid be hungry. We can easily afford to let every kid have a great education. We choose not to.
I don’t want my kid to grow up in a society where we make such choices.
I could go on and rant about the Republican party, about Trump possibly winning 2024, about our taxes supporting and financing wars in places where they shouldn’t, about xenophobia and racism, about reproductive rights, trans rights, and so much more. But unfortunately many of these topics are often not significantly better elsewhere either.
When are we moving? We plan to stay here until the school year is over, and aim to have moved before the next school year starts. So in the summer of ‘24.
Where are we moving? I am going back to my place of birth, Stuttgart. We considered a few options, and Stuttgart led overall due to the combination of proximity to family, school system compatibility for the kid, a time zone that works well for the Abstract Wikipedia team, language requirements, low legal hurdles of moving there, and the cost of living we expect. Like every place it also comes with challenges. Don’t get me started on the taste of tomatoes or peaches.
What other places did we consider? We considered many other places, and we traveled to quite a few of them to check them out. We loved each and every one of them. We particularly loved Auckland due to our family there and the weather, we loved the beautiful city of Barcelona for its food and culture, we loved Dublin, London, Zürich, Berlin, Vienna, Split. We started making a large spreadsheet with pros and contras in many categories, but in the end the decision was a gut decision. Thanks to everyone who talked with us and from whom we learned a lot about those places!
Being able to even consider moving to these places is a privilege. And we understand that and are thankful for having this privilege. Some of these places would have been harder to move for us due to immigration regulation, others are easy thanks to our background. But if you are thinking of moving, and are worried about certain aspects, feel free to reach out and discuss. I am happy to offer my experience and perspective.
Is there something you can help with? If you want to meet up with us while we are still in the US, it would be good to do so timely. We are expecting to sell the house quite a bit sooner, and then we won’t be able to host guests easily. I am also looking forward to reconnecting with people in Europe after the move. Finally, if you know someone who is interested in a well updated 3 bedroom house with a surprisingly large attic that can be used as a proper hobby space, and with a top walkability index in south Berkeley, point them our way.
Also, experiences and advice regarding moving from the US to Germany are welcome. Last time we moved the other way, and we didn’t have that much to move, and Google was generously organizing most of what needed to be done. This time it’s all on us. How to get a container and get it loaded? How to ship it to Germany? Where to store it while we are looking for a new home? How to move the cat? How to make sure all goes well with the new school? When to sell the house and where to live afterwards? How to find the right place in Germany? What are the legal hurdles to expect? How will taxes work? So many questions we will need to answer in the coming months. Wish us luck for 2024.
We also accept good wishes and encouraging words. And I am very much looking forward to seeing some of you again next year!
Mulan
I was surprised when Disney made the decision to sell Mulan on Disney+. So if you wanted to watch Mulan, you not only have to buy it, so far so good, but you have to join their subscription service first. The price for Mulan is $30 in the US, additionally to the monthly fee of streaming, $7. So the $30 don't buy you Mulan, but allow you to watch it if you keep up your subscription.
Additionally, on December 4 the movie becomes free for everyone with a Disney+ subscription.
I thought, that's a weird pricing model. Who'd pay that much money for streaming the movie a few weeks earlier? I know, it will be very long weeks due to the world being so 2020, but still. Money is tight for many people. Also, the movie had very mixed reviews and a number of controversies attached to it.
According to the linked report, Disney really knows what they're doing. 30% of subscribers bought the early streaming privilege! Disney made hundreds of millions in extra profit within three first few days (money they really will be thankful for right now given their business with the cruise ships and theme parks and movies this year).
The most interesting part is how this will affect the movie industry. Compare to Tenet - which was reviewed much better and which was the hope to revive the moribund US cinema industry, but made less than $30M - which also needs to be shared with the theaters and had much more distribution costs. Disney keeps a much larger share of the $30 for Mulan than Tenet makes for its production company.
The lesson from Mulan and Trolls 2, which also did much better than I would ever have predicted, for the production companies experimenting with novel pricing models, could be disastrous for theaters.
I think we're going to see even more experimentation with pricing models. If the new Bond movie and/or the new Marvel movie should be pulled from cinemas, this might also be the end of cinemas as we know them.
I don't know how the industry will change, but the swing is from AMC to Netflix, with the producers being caught in between. The pandemic massively accelerated this transition, as it did so many others.
My Erdös Number
After reading a post by Ora and one by Tim Finin, I tried to figure my own Erdös Number out. First, taking Ora's path, I came up with an Erdös of 7:
Paul Erdös - Stephan Hedeniemi - Robert Tarjan - David Karger - Lynn Stein - Jim Hendler - Steffen Staab - Denny Vrandečić
But then I looked more, and with Tim's path I could cut it down to 6:
Paul Erdös - Aviczir Fraenkl - Yaacov Yesha - Yelena Yesha - Tim Finin - Steffen Staab - Denny Vrandečić
The point that unnerved me most was that the data was actually there. Not only a subscription-only database for mathematical papers (why the heck is the metadata subscription only?), but there's DBLP, there's the list of Erdös 1 and 2 people on the Erdös Number project, there's Flink, and still, I couldn't mash up the data. This syntactic web sucks.
The only idea that brought me further - without spending even more time with that - was a Google search for "my erdös number" "semantic web", in the hope to find some collegues in my field that already have found and published their own Erdös number. And yep, this worked quite fine, and showed me two further, totally disjunctive paths to the one above:
Paul Erdös - Charles J. Coulborn - A. E. Brouwer - Peter van Emde Boas - Zhsisheng Huang - Peter Haase - Denny Vrandečić
and
Paul Erdös - Menachem Magidor - Karl Schlechta - Franz Baader - Ian Horrocks - Sean Bechhofer - Denny Vrandečić
So that's and Erdös of 6 on at least 3 totally different paths. Nice.
What surprises me - isn't this scenario obviously a great training project for the Semantic Web? Far easier than Flink, I suppose, and still interesting for a wider audience as well, like Mathematicians and Noble Laureates? (Oh, OK, not them, they get covered manually here).
Update
I wrote the post quite a time ago. A colleague of mine notified me in the meantime that I have a Erdös of only 4 by the following path:
Paul Erdös - E. Rodney Canfield - Guo-Quiang Zhang - Markus Krötzsch - Denny Vrandečić
Wow. It's the social web that gave the best answer.
2019 Update
Another update, after more than a dozen years: I was informed that I have now an Erdös number of 3 by the following path:
Paul Erdös - Anthony B. Evans - Pascal Hitzler - Denny Vrandečić
I would be very surprised if this post requires any further updates.
My horoscope for today
Here's my horoscope for today:
You may be overly concerned with how your current job prevents you from reaching your long-term goals. The Sun's entry into your 9th House of Big Ideas can work against your efficiency by distracting you with philosophical discussions about the purpose of life. These conversations may be fun, but they should be kept to a minimum until you finish your work.
How the heck did they know??
Need help with SQL
But I have no time to write it down... I will sketch my problem, and hopefully get an answer (I've started this blog in the morning, and now I need to close it, cause I have to leave).
Imagine I have following table (a triple store):
| Subject | Predicate | Object |
|---|---|---|
| Adam | yahoo im | adam@net |
| Adam | skye | adam |
| Berta | skype | berta |
How do I make a query that returns me the following table:
| Subject | o1 | o2 |
|---|---|---|
| Adam | adam@net | adam |
| Berta | - | berta |
The problems are the default value in the middle lower cell. The rest works (as maybe seen in the Semantic Mediawiki code, file includes/SMW_InlineQueries.php -- note that the CVS is not up to date for now, because SourceForge's CVS is down for days!)
It should work in general, with as many columns as I like on the answer table (based on the predicates in the first one).
Oh, and if you solved this -- or have an idea -- it would be nice if it worked with MySQL 4.0, i.e. without Subqueries.
Any ideas?
New OWL tools
The KAON2 OWL Tools get more diverse and interesting. Besides the simple axiom and entity counter and dumper, the not so simple dlpconverter, and the syntactic transformer from XML/RDF to OWL/XML and back, you now also have a filter (want to extract only the subClassOf-Relations out of your ontology? Take filter), diff and merge (for some basic syntactic work with ontologies), satisfiable (which checks if the ontology can have a satisfying model), deo (turning SHOIN-ontologies in SHIN-ontologies by weakening, should be sound, but naturally not complete) and ded (removes some domain-related axioms, but it seems this one is still buggy).
I certainly hope this toolbox will still grow a bit. If you have any suggestions or ideas, feel free to mail me or comment here.
New home in Emeryville
Our new (temporary home) is the City of Emeryville. Emeryville has a population of almost 13,000 people. The apartment complex we live in has about 400 units, and I estimate that they have about 2 people on average in each. Assuming that about 90% of the apartments are occupied, this single apartment complex would constitute between 5 and 10% of the population of the whole city.
New people at Yahoo and Google
Vint Cerf starts working at Google, Dave Becket moves to Yahoo. Both like the Semantic Web (Vint said so in a German interview with c't, and I probably don't have to remind you about Daves accomplishments).
I'm sure, Yahoo got Dave because of his knowledge about the Semantic Web. And I wonder if Google got Vint out of the same reason? Somehow, I doubt it.
New tagline is my New Year's resolution
I just changed the tagline of this blog. The old one was rather, hmm, boring:
"Discovering the Semantic Web, Ontology Engineering and related technologies, and trying to understand these amazing ideas - and maybe sharing an idea or two... "
The new one is at the same time my new year's resolution for 2006.
"Kicking the Semantic Web's butt to reality"
'nuff said, got work to do.
New versions: owlrdf2owlxml, dlpconvert
New versions of owlrdf2owlxml and dlpconvert are out.
owlrdf2owlxml got renamed, as it was formerly known as rdf2owlxml. But as a colleague pointed out, this name can easily be misunderstood, meaning to transform arbitrarily RDF to OWL. It doesn't do that, it only transforms OWL to OWL, from RDF/XML-serialisation to XML Presentation Syntax. And it seems to work quite stable, it can even transform the famous wine ontology. Version 0.4 out now.
dlpconvert lost a lot of its bugs. And as most of you were feeding RDF/XML to it, well, now you can do it officially (listen to the users), too. It reads both syntaxes, and creates a Prolog program out of your ontology. Version 0.7 is out.
They are both based on KAON2, the Karlsruhe Ontology Infrastructure module, written by Boris Motik. My little tools are just wrapped around KAON2 and using its functionality. To be honest, I'm thinking of writing quite a number of little tools like this, who offer different functionality, thus providing you with a nice toolkit to handle ontologies efficiently. I don't lack ideas right now, it's just I' m not sure that there's interest in this.
Well, maybe I should just start and we'll see...
By the way, both tools are not only available as web services, but you may also download them as command line tools from their respective websites and play around it on your PC. That's a bit more comfortable than using a browser as your operating system.
No April Fool's day
This year, I am going to skip April Fool's day.
I am not being glib, but quite literal.
We are taking flight NZ7 starting on the evening of March 31 in San Francisco, flying over the Pacific Ocean, and will arrive on April 2 in the early morning in Auckland, New Zealand.
Even if one actually follows the flight route and overlays it over the timezone map, it looks very much like we are not going to spend more than a few dozen minutes, or at most a few hours, in April 1, if all goes according to plan.
Looking forward to it!
Here's the flight data of a previous NZ7 flight, from Sunday: https://flightaware.com/live/flight/ANZ7/history/20230327/0410Z/KSFO/NZAA/tracklog
Here are the timezones (but it's Northern winter time). Would be nice to overlay the two maps: 
Where's Wikifunctions when it's needed?
The question seems to be twofold: how often do we cross the dateline, and how close are we to local time midnight while crossing the dateline. For a perfect date miss one would need to cross the dateline exactly once, at a 24 hour difference, as close as possible to local midnight.
Normbrunnenflasche
It's a pity there's no English Wikipedia article about this marvellous thing that exemplifies Germany so beautifully and quintessentially: the Normbrunnenflasche.
I was wondering the other day why in Germany sparkling water is being sold in 0.7l bottles and not in 1l or 2l or whatever, like in the US (when it's sold here at all, but that's another story).
Germany had a lot of small local producers and companies. To counter the advantages of the Coca Cola Company pressing in the German market, in 1969 a conference of representatives of the local companies decided to introduce a bottle design they all would use. This decision followed a half year competition and discussion on what this bottle should look like.
Every company would use the same bottle for sparkling water and other carbonated drinks, and so no matter which one you bought, the empty bottle would afterwards be routed to the closest participating company, not back home, therefore reducing transport costs and increasing competitiveness against Coca Cola.
The bottle is full of smart features. The 0.7l were chosen to ensure that the drink remained carbonated until the last sip, because larger bottles would last longer and thus gradually loose carbonization.
The form and the little pearls outside were chosen for improved grip, but also to symbolize the sparkles of the carbonization.
The metal screw cap was the real innovation there, useful for drinks that could increase pressure due to the carbonization.
And finally two slightly thicker bands along the lower half of the bottle that would, while being rerouted for another usage, slowly get more opaque due to mechanical pressure, thus indicating how well used the individual bottle was, so they could be taken out of service in time before breaking at the customer.
The bottles were reused an average of fifty times, their boxes an average of hundred times. More than five billion of them have been brought into circulation in the fifty years since their adoption, for an estimated quarter of a trillion fillings.
Northern Arizona
Last week we had a wonderful trip through Northern Arizona.
Itinerary: starting in Phoenix going Northeast through Tonto National Forest towards Winslow. In Tonto, we met our first surprise, which would become a recurring pattern: whereas we expected Arizona in April to be hot, and we were prepared for hot, it had some really cold spells, and we were not prepared for cold. We started in the Sonoran Desert, surrounded by cacti and sun, but one and a half hours later in Tonto, we were driving through a veritable snow storm, but fortunately, just as it was getting worrisome, we crossed the ridge and started descending towards Winslow to the North.
The Colorado Plateau on the other side of the ridge was then pleasant and warm, and the next days we traveled through and visited the Petrified Forest, Monument Valley, Horseshoe Bend, Antelope Canyon, and more.
After that we headed for the Grand Canyon, but temperatures dropped so low, and we didn't have the right outfit for that, we stayed less than a day there, most of it huddled in the hotel room. Still, the views we got were just amazing, and throwing snowballs was an unexpected fun exercise.
Our last stop took us to Sedona, where we were again welcomed with amazing views. The rocks and formations all had in common that they dramatically changed with the movement of the sun, or with us moving around, and the views were always fresh.
Numbers: Our trip took us about 950 miles / 1500 kilometeres of driving, and I was happy that it was a good Jeep for this trip. The difference in altitude went from 1000 feet / 330 meters in Phoenix up to 8000 feet / 2400 meters driving through Coconino. Temperatures ranged from 86° F / 30° C to 20° F / -7° C.
What I learned again is how big this country is. And how beautiful.
Surprises: One thing that surprised me was how hidden the Canyons can be. Well, you can't hide Grand Canyon, but it is easy to pass by Antelope Canyon and not realizing it is there. Because it is just a cut in the plateau.
I also was surprised about how flat and wide the land is. I have mostly lived in areas where you had mountains or at least hills nearby, but the Colorado Plateau has large wide swaths of flat land. "Once the land was as plane as a pancake".
I mentioned the biggest surprise already, which was how cold it got.
Towns: it was astonishing to see the difference between, on the one side, a town such as Page or Sedona and on the other side Winslow. All three have a similar population, but Page and Sedona felt vigorous, lively, clean, whereas Winslow felt as if it was on the decline, deserted, struggling.
The hotel we stayed in in Winslow, La Posada, was a beautiful, weird, unique jewel that I hesitate to flat-out recommend, it is too unusual for that, but that I still enjoyed experiencing. It is clearly very different from any other hotel I ever stayed in, full of history, and embracing themes of both suicide and hope, respectfully trying to grow with the native population, and aiming to revive the city's old town, and it is difficult to really capture the vibe it was sending out.
For pictures, I am afraid I am pointing to my Facebook posts, which should be visible without login:
- Scottsdale
- Palo Verde tree
- Desert Botanical Garden, Phoenix
- Desert to snowstorm
- Winslow
- La Posada hotel
- Petrified forest
- Monument valley
- Horseshoe bend
- Antelope canyon
- Grand canyon
- Sedona
OK
I often hear "don't go for the mediocre, go for the best!", or "I am the best, * the rest" and similar slogans. But striving for the best, for perfection, for excellence, is tiring in the best of times, never mind, forgive the cliché, in these unprecedented times.
Our brains are not wired for the best, we are not optimisers. We are naturally 'satisficers', we have evolved for the good-enough. For this insight, Herbert Simon received a Nobel prize, the only Turing Award winner to ever get one.
And yes, there are exceptional situations where only the best is good enough. But if good enough was good enough for a Turing-Award winning Nobel laureate, it is probably for most of us too.
It is OK to strive for OK. OK can sometimes be hard enough, to be honest.
May is mental health awareness month. Be kind to each other. And, I know it is even harder, be kind to yourself.
Here is OK in different ways. I hope it is OK.
Oké ఓకే ਓਕੇ オーケー ओके 👌 ওকে או. קיי. Окей أوكي Օքեյ O.K.
OWL 2.0
I posted this to the public OWL dev mailing list as a response to a question posed by Jim Hendler quite some while ago. I publish it here for easier reference.
Quite some while ago the question of OWL 2.0 was rised here, and I wrote already two long replies with a wishlist - but both were never sent and got lost in digital nirvana, one due to a hardware, the second due to a software failure. Well, let's hope this one passes finally through. That's why this answer is so late.
Sorry for the lengthy post. But I tried to structure it a bit and make it readable, so I hope you find some interesting stuff here. So, here is my wishlist.
- I would like yet another OWL language, call it OWL RDF or OWL Superlite, or whatever. This is like the subset of OWL Lite and RDFS. For this the difference between of owl:Class and rdf:Class needs to be somehow standardly solved. Why is this good? It makes moving from RDF to OWL easier, as it forces you to keep Individuals, Classes and Relations in different worlds, and forgets about some of the more sophisticated constructs of RDF(S) like lists, bags and such. This is a real beginners language, really easy to learn and implement.
- Defined Semantics for OWL FUll. It is unclear -- at least to me -- what some combinations of RDF(S)-Constructs and OWL DL-constructs are meant to mean.
- Add easy reification to OWL. I know, I know, making statements about statements is meant to be the root of all evil, but I find it pretty useful. If you like, just add another group of elements to OWL, statements, that are mutually disjoint from classes, instances and relations in OWL DL, but there's a sublanguage that enables us to speak about statements. Or else OWL will suck a lot in comparison to RDF(S) and RDF(S) + Rules will win, because you can't do a lot of the stuff you need to do, like saying what the source of a certain statement is, how reliable this source is, etc. Trust anyone? This is also needed to extend ontologies toward probabilistic, fuzzy or confidence-carrying models.
- I would love to be able to define syntactic sugar, like partitionOf (I think, this is from Asun's Book on Ontology Engineering). ((A, B, C) partitionOf D) means that every D is either an A or a B or a C, that every A, B or C is a D, and that A, B and C are mutually disjunct. So you can say this already, but it needs a lot of footwork. It would be nice to be able to define such shotcuts that lever upon the semantics of existing constructors.
- That said, another form of syntactic sugar - because again you can use existing OWL constructs to reach the same goal, but it is very strenuous to do so - would be to define UNA locally. Like either to say "all individuals in this ontology are mutually different" or "all individuals with this namespace are mutually different". I think, due to XML constraints the first one would be the weapon of choice.
- I would like to be able to have more ontologies in the same file. So you can use ontologies to group a number of axioms, and you also could use the name of this group to refer to it. Oh well, using the name of an ontology as an individual, what does this mean? Does it imply any further semantics? I would like to see this clarified. Is this like named graphs?
- The DOM has quite nicely partitioned itself in levels and modules. Why not OWL itself? So you could have like a level 2 ontology of mereological questions, and such stuff, all with well defined semantics, for the generic questions. I am not sure there are too many generic questions, but taxonomy is (already covered), mereology would be, and spatiotemporal and dynamic issues would be as well. Mind you, not everyone must use them, but many will need them. It would be fine to find stan dard answers to such generic questions.
- Procedural attachments would be a nice thing. Like have a a standardized possibilities to add pieces of code and have them executed by an appropriate execution environment on certain events or requests by the reasoner. Yes, I am totally aware of the implications on reasoning and decidability, but hey, you asked what people need, and did not ask for theoretical issues. Those you understand better.
- There are some ideas of others (which doesn't mean that the rest is necessarily original mine) I would like to see integrated, like a well-defined epistemic operator or streamlining the concrete domains to be more consistent with abstract domains, or to define domain and range _constraints_ on relations, and much more. Much of this stuff could be added optional in the sense of point 7.
- And not to forget that we have to integrate with rules later, and to finally have an OWL DL query language. One goal is to make it clear what OWL offers over simply adding rules atop of RDF and ignoring the ontology layer completely.
So, you see, this is quite a list, and it sure is not complete. Even if only two or three points were finally picked up I would be very happy :)
OWL luna, nicer latex, OWL/XML to Abstract Syntax, and more
After a long hiatus, due to some technical problems, finally I could create a new version of the owl tools. So, version 0.27 of the owl tools is now released. It works with the new version of KAON2, and includes six months of bug fixing, but also a number of new features that have introduced a whole new world of new, exciting bugs as well.
The owl latex support was improved greatly. Translation of an owl ontology is done now more careful, and the user can specify much more of the result than before.
A new tool is owl luna - luna like local unique name assumption. It adds an axiom to your ontology that states that all individuals are different from each other. Due to most ontology editors not allowing to do this automatically, here you find a nice maintenance tool to make your ontology much less ambiguous.
The translations of OWL/RDF to OWL/XML and back have been joined in one new tool, called owl syntax, that allows you to translate owl ontologies also to OWL Abstract Syntax, a much nicer syntax for owl ontologies.
owl dlpconvert has been extended as it now also serialzes it results as RuleML if you like. So you can just pipe your ontologies to RuleML.
So, the tools have both become sharper and more numerous, making your toolbelt to work with owl in daily life more usable. Get the new owl tools now. And if you peek into the source code, the code has underwent a major clean up, and you will also see the new features I am working on, that have to do with ontology evaluation and more.
Have fun with the tools! And send me your comments, wishes, critiques!
On the competence of conspiracists
“Look, I’ll be honest, if living in the US for the last five years has taught me anything is that any government assemblage large enough to try to control a big chunk of the human population would in no way be consistently competent enough to actually cover it up. Like, we would have found out in three months and it wouldn’t even have been because of some investigative reporter, it would have been because one of the lizards forgot to put on their human suit on day and accidentally went out to shop for a pint of milk and like, got caught in a tik-tok video.” -- Os Keyes, WikidataCon, Keynote "Questioning Wikidata"
One world. One web.
I am in Beijing at the opening of the WWW2008 conference. Like all WWWs I was before, it is amazing. The opening ceremony was preceded by a beautiful dance, combining tons of symbols. First a woman in a traditional Chinese dress, then eight dancers in astronaut uniforms, a big red flag with "Welcome to Beijing" on it (but not on the other side, when he came back), and then all of them together... beautiful.
Boris Motik's paper is a best paper candidate! Yay! Congratulations.
I rather listen to the keynote now :)
Blogging from my XO One.
Oscar winning families
Yesterday, when Jamie Lee Curtis won her Academy Award, I learned that both her parents were also nominated for Academy Awards. Which lead to the question: who else?
I asked Wikidata, which lists four others:
- Laura Dern
- Liza Minnelli
- Nora Ephron
- Sean Astin
Only one of them belongs to the even more exclusive club of people who won an Academy Award, and where both parents also did: Liza Minnelli, daughter of Vincente Minelli and Judy Garland.
Also interesting: List of Academy Award-winning families
Our four freedoms for our technology
(This is a draft. Comments are welcome. This is not meant as an attack on any person or company individually, but at certain practises that are becoming increasingly prevalent)
We are not allowed to use the devices we paid for in the ways we want. We are not allowed to use our own data in the way we want. We are only allowed to use them in the way the companies who created the devices and services allow us.
Sometimes these companies are nice and give us a lot of freedom in how to use the devices and data. But often they don’t. They close them down for all kinds of reasons. They may say it is for your protection and safety. They might admit it is for profit. They may say it is for legal reasons. But in the end, you are buying a device, or you are creating some data, and you are not allowed to use that device and that data in the way you want to, you are not allowed to be creative.
The companies don’t want you to think of the devices that you bought and the data that you created as your devices and your data. They want you to think of them as black boxes that offer you services they create for you. They don’t want you to think of a Ring doorbell as a camera, a microphone, a speaker, and a button, but they want you to think of it as providing safety. They don’t want you to think of the garage door opener as a motor and a bluetooth module and a wifi module, but as a garage door opening service, and the company wants to control how you are allowed to use that service. Companies like Chamberlain and SkyLink and Genie don’t allow you to write a tool to check on your garage door, and to close or open it, but they make deals with Google and Amazon and Apple in order to integrate these services into their digital assistants, so that you can use it in the way these companies have agreed on together, through the few paths these digital assistants are available. The digital assistant that you buy is not a microphone and a speaker and maybe a camera and maybe a screen that you buy and use as you want, but you buy a service that happens to have some technical ingredients. But you cannot use that screen to display what you want. Whether you can watch your Amazon Prime show on the screen of a Google Nest Hub depends on whether Amazon and Google have an agreement with each other, not on whether you have paid for access to Amazon Prime and you have paid for a Google Nest Hub. You cannot use that camera to take a picture. You cannot use that speaker to make it say something you want it to say. You cannot use the rich plethora of services on the Web, and you cannot use the many interesting services these digital assistants rely on, in novel and creative combinations.
These companies don’t want you to think of the data that you have created and that they have about you as your data. They don’t want you to think about this data at all. They just want you to use their services in the way they want you to use their services. On the devices they approve. They don’t want you to create other surfaces that are suited to the way you use your data. They don’t want you to decide on what you want to see in your feed. They don’t want you to be able to take a list of your friends and do something with it. They will say it is to protect privacy. They will say that it is for safety. That is why you cannot use the data you and your friends have created. They want to exactly control what you can and cannot do with the data you and your friends have created. They want to control how many ads you must see in order to be allowed to see your friends’ posts. They don't want anyone else to have the ability to provide you creative new interfaces to your feed. They don’t want you yourself the ability to look at your feed and do whatever you want with it.
Those are devices you paid for.
These are data you and your friends have created.
And more and more we are losing our freedom of using our devices and our data as we like.
It would be impossible to invent email today. It would be impossible to invent the telephone today. Both are protocols that allow everyone to communicate with anyone no matter what their email provider or their phone is. Try reading your friend’s Facebook feed on Instagram, or send a direct message from your Twitter account to someone on WhatsApp, or call your Skype contact on Facetime.
It would be impossible to launch the Web today - many companies don’t want you browsing the Web. They want you to be inside of your Facebook feed and consume your content there. They want you to be on your Twitter feed. They don’t want you to go to the Website of the New York Times and read an article there, they don’t want you to visit the Website of your friend and read their blog there. They want you to stay on their apps. Per default, they open Websites inside their app, and not in your browser, so you are always within their app. They don’t want you to experience the Web. The Web is dwindling and all the good things on it are being recut and rebundled within the apps and services of tech companies.
Increasingly, we are seeing more and more walls in the world. Already, it is becoming impossible to pay and watch certain movies and shows without buying into a full subscription in a service. We will likely see the day where you will need a specific device to watch a specific movie. Where the only way to watch a Disney+ exclusive movie is on a Disney+ tablet. You don’t think so? Think about how easy it is to get your Kindle books onto another Ebook reader. How do you enable a skill or capability available in Alexa on your Nest smart speaker? How can you search through the books that you bought and are in your digital library, besides by using a service provided by the company that allows you to search your digital library? When you buy a movie today on YouTube or on iMovies, what do you own? What are you left with when the companies behind these services close that service, or go out of business altogether?
Devices and content we pay for, data we and our friends create, should be ours to use in empowering and creative ways. Services and content should not be locked in with a certain device or subscription service. The bundling of services, content, devices, and locking up user data creates monopolies that stifle innovation and creativity. I am not asking to give away services or content or devices for free, I am asking to be allowed to pay for them and then use them as I see fit.
What can we do?
As far as I can tell, the solution, unfortunately, seems to be to ask for regulation. The market won’t solve it. The market doesn’t solve monopolies and oligopolies.
But don’t ask to regulate the tech giants individually. We don’t need a law that regulates Google and a law that regulates Apple and a law that regulates Amazon and a law to regulate Microsoft. We need laws to regulate devices, laws to regulate services, laws to regulate content, laws that regulate AI.
Don’t ask for Facebook to be broken up because you think Mark Zuckerberg is too rich and powerful. Breaking up Facebook, creating Baby Books, will ultimately make him and other Facebook shareholders richer than ever before. But breaking up Facebook will require the successor companies to work together on a protocol to collaborate. To share data. To be able to move from one service to another.
We need laws that require that every device we buy can be made fully ours. Yes, sure, Apple must still be allowed to provide us with the wonderful smooth User Experience we value Apple for. But we must also be able to access and share the data from the sensors in our devices that we have bought from them. We must be able to install and run software we have written or bought on the devices we paid for.
We need laws that require that our data is ours. We should be able to download our data from a service provider and use it as we like. We must be allowed to share with a friend the parts of our data we want to share with that friend. In real time, not in a dump download hours later. We must be able to take our social graph from one social service and move to a new service. The data must be sufficiently complete to allow for such a transfer, and not crippled.
We need laws that require that published content can be bought and used by us as we like. We should be able to store content on our hard disks. To lend it to a friend. To sell it. Anything I can legally do with a book I bought I must be able to legally do with a movie or piece of music I bought online. Just as with a book you are not allowed to give away the copies if the work you bought still enjoys copyright.
We need laws that require that services and capabilities are unbundled and made available to everyone. Particularly as technological progress with regards to AI, Quantum computing, and providing large amounts of compute becomes increasingly an exclusive domain for trillion dollar companies, we must enable other organizations and people to access these capabilities, or run the risk that sooner or later all and any innovation will be happening only in these few trillion dollar companies. Just because a company is really good at providing a specific service cheaply, it should not be allowed to unfairly gain advantage in all related areas and products and stifle competition and innovation. This company should still be allowed to use these capabilities in their products and services, but so should anyone else, fairly prized and accessible by everyone.
We want to unleash creativity and innovation. In our lifetimes we have seen the creation of technologies that would have been considered miracles and impossible just decades ago. These must belong to everybody. These must be available to everyone. There cannot be equity if all of these marvellous technologies can be only wielded by a few companies on the West coast of the United States. We must make them available to all the people of the world: the people of the Indian subcontinent, the people of Subsaharan Africa,the people of Latin America, and everyone else. They all should own the devices they paid for, the data they created, the content they paid for. They all should have access to the same digital services and capabilities that are available to the engineers at Amazon or Google or Microsoft. The universities and research centers of the world should be able to access the same devices and services and extend them with their novel and creative ideas. The scrappy engineers in Eastern Europe and India and Nigeria and Central Asia should be able to call the AI models trained by Google and Microsoft and use them in novel ways to run their devices and chip-powered cars and agricultural machines. We want a world of freedom, tinkering, where creativity and innovation are unleashed, and where everyone can contribute their ideas, their creativity, and where everyone can build their fortune.
Papaphobia
In a little knowledge engineering exercise, I was trying to add the causes of a phobia to the respective Wikidata items. There are currently about 160 phobias in Wikidata, and only a few listed in a structured way what they are afraid of. So I was going through them, trying to capture it in s a structured way. Here's a list of the current state:
Now, one of those phobias was the Papaphobia - the fear of the pope. Now, is that really a thing? I don't know. CDC does not seem to have an entry on it. On the Web, in the meantime, some pages have obviously taken to mining lists of phobias and creating advertising pages that "help" you with Papaphobia - such as this one:
This page is likely entirely auto-generated. I doubt it that they have "clients for papaphobia in 70+ countries", whom they helped "in complete discretion" within a single day! "People with severe fears and phobias like papaphobia (which is in fact the formal diagnostic term for papaphobia) are held prisoners by their phobias."
This site offers more, uhm, useful information.
"Group psychotherapy can also help where individuals share their experience and, in the process, understand and recover from their phobia." Really? There are enough cases that we can even set up a group therapy?
Now, maybe I am entirely off here - maybe, papaphobia is really a thing. With search in Scholar I couldn't find any medical sources (the term is mentioned in a number of sociological and historical works, to express general sentiments in a population or government against the authority of the pope, but I could not find any mentions of it in actual medical literature).
Now could those pages up there be benign cases of jokes? Or are they trying to scam people with promises to heal their actual fears, and they just didn't curate the list of fears sufficiently, because, really, you wouldn't find this page unless you actually search for this term?
And now what? Now what if we know these pages are made by scammers? Do we report them to the police? Do we send a tip to journalists? Or should we just do nothing, allowing them to scam people with actual fears? Well, by publishing this text, maybe I'll get a few people warned, but it won't reach the people it has to reach at the right time, unfortunately.
Also, was it always so hard to figure out what is real and what is not? Does papaphobia exist? Such a simple question. How should we deal with it on Wikidata? How many cases are there, if it exists? Did it get worse for people with papaphobia now that we have two people living who have been made pope?
My assumption now is that someone was basically working on a corpus, looking for words ending in -phobia, in order to generate a list of phobias. And then the term papaphobia from sociological and historical literature popped up, and it landed in some list, and was repeated in other places, etc., also because it is kind of a funny idea, and so a mixture of bad research and joking bubbled through, and rolled around on the Web for so long that it looks like it is actually a thing, to the point that there are now organizations who will gladly take your money (CTRN is not the only one) to treat you for papaphobia.
The world is weird.
Partial copyright for an AI generated work
Interesting development in US cases around copyright and AI: author Elisa Shupe asked for copyright registration on a book that was created with the help of generative AI. Shupe stated that not giving her registration would be disabilities discrimination, since she would not have been able to create her work otherwise. On appeal, her work was partially granted protection for the “selection, coordination, and arrangement of text generated by artificial intelligence”, without referral to the disability argument.
Pastir Loda
Vladimir Nazor is likely the most famous author from the island of Brač, the island my parents are from. His most acclaimed book seems to be Pastir Loda, Loda the Shepherd. It tells the story of a satyr that, through accidents and storms, was stranded on the island of Brač, and how he lived on Brač for the next almost two thousand years.
It is difficult to find many of his works, they are often out of print. And there isn't much available online, either. Since Nazor died in 1949, his works are in the public domain. I acquired a copy of Pastir Loda from an antique book shop in Zagreb, which I then forwarded to a friend in Denmark who has a book scanner, and who scanned the book so I can make the PDF available now.
The book is written in Croatian. There is a German translation, but that won't get into the public domain until 2043 (the translator lived until 1972), and according to WorldCat there is a Czech translation, and according to Wikipedia a Hungarian translation. For both I don't know who the translator is, and so I don't know the copyright status of these translations. I also don't know if the book has ever been translated to other languages.
I wish to find the time to upload and transcribe the content on Wikisource, and then maybe even do a translation of it into English. For now I upload the book to archive.org, and I also make it available on my own Website. I want to upload it to Wikimedia Commons, but I immediately stumbled upon the first issue, that it seems that to upload it to Commons the book needs to be published before 1928 and the author has to be dead for more than 70 years (I think that should be an or). I am checking on Commons if I can upload it or not.
Until then, here's the Download:
Philosophische Grundlagen
I had a talk on Philosophical Foundations of Ontologies last week at the AIFB. I prepared it in German (and thus, all the slides were in German) and just before I started I got asked if I may give the talk in English.
Having never heard a single lesson in philosophy in English and having read English philosophy only on Wikipedia before, I said yes. Nevertheless, the talk was very well perceived, and so I decided to upload it. It's pure evil PowerPoint, no semantic slides format, and I didn't yet manage to translate it to English. If anyone can offer me some help with that - I simply don't know many of the technical terms, and I don't have ready access to the sources - I would be very happy! Just drop me a note, please.
Philosophische Grundlagen der Ontologie (PowerPoint, ca. 4,5 MB)
Broken link
Playing around with Aliquot
- Warning! Very technical, not particularly insightful, and overly long post about hacking, discrete mathematics, and rabbit holes. I don't think that anything here is novel, others have done more comprehensive work, and found out more interesting stuff. This is not research, I am just playing around.
Ernest Davis is a NYU professor, author of many publications (including “Rebooting AI” with Gary Marcus) and a friend on Facebook. Two days ago he posted about Aliquot sequences, and that it is yet unknown how they behave.
What is an Aliquot sequence? Given a number, take all its proper divisors, and add them up. That’s the next number in the sequence.
It seems that most of these either lead to 0 or end in a repeating pattern. But it may also be that they just keep on going indefinitely. We don’t know if there is any starting number for which that is the case. But the smallest candidate for that is 276.
So far, we know the first 2,145 steps for the Aliquot sequence starting at 276. That results in a number with 214 digits. We don’t know the successor of that number.
I was curious. I know that I wouldn’t be able to figure this out quickly, or, probably ever, because I simply don’t have enough skills in discrete mathematics (it’s not my field), but I wanted to figure out how much progress I can make with little effort. So I coded something up.
Anyone who really wanted to make a crack on this problem would probably choose C. Or, on the other side of the spectrum, Mathematica, but I don’t have a license and I am lazy. So I chose JavaScript. There were two hunches for going with JavaScript instead of my usual first language, Python, which would pay off later, but I will reveal them later in this post.
So, my first implementation was very naïve (source code). The function that calculates the next step in the Aliquot sequence is usually called s in the literature, so I kept that name:
const divisors = (integer) => {
const result = []
for(let i = BigInt(1); i < integer; i++) {
if(integer % i == 0) result.push(i)
}
return result
}
const sum = x => x.reduce( (partialSum, i) => partialSum + i, BigInt(0) )
const s = (integer) => sum(divisors(integer))
I went for BigInt, not integer, because Ernest said that the 2,145th step had 214 digits, and the standard integer numbers in JavaScript stop being exact before we reach 16 digits (at 9,007,199,254,740,991, to be exact), so I chose BigInt, which supports arbitrary long integer numbers.
The first 30 steps ran each under a second on my one decade old 4 core Mac occupying one of the cores, reaching 8 digits, but then already the 36th step took longer than a minute - and we only had 10 digits so far. Worrying about the limits of integers turned out to be a bit preliminary: With this approach I would probably not reach that limit in a lifetime.
I dropped BigInt and just used the normal integers (source code). That gave me 10x-40x speedup! Now the first 33 steps were faster than a second, reaching 9 digits, and it took until the 45th step with 10 digits to be the first one to take longer than a minute. Unsurprisingly, a constant factor speedup wouldn’t do the trick here, we’re fighting against an exponential problem after all.
It was time to make the code less naïve (source code), and the first idea was to not check every number smaller than the target integer whether it divides (line 3 above), but only up to half of the target integer.
const divisors = (integer) => {
const result = []
const half = integer / 2
for(let i = 1; i <= half; i++) {
if(integer % i == 0) result.push(i)
}
return result
}
Tiny change. And exactly the expected impact: it ran double as fast. Now the first 34 steps ran under one second each (9 digits), and the first one to take longer than a minute was the 48th step (11 digits).
Checking until half of the target seemed still excessive. After all, for factorization we only need to check until the square root. That should be a more than constant speedup. And once we have all the factors, we should be able to quickly reconstruct all the divisors. Now this is the point where I have to admit that I have a cold or something, and the code for creating the divisors from the factors is probably overly convoluted and slow, but you know what? It doesn’t matter. The only thing that matters will be the speed of the factorization.
So my next step (source code) was a combination of a still quite naïve approach to factorization, with another function that recreates all the divisors.
const factorize = (integer) => {
const result = [ 1 ]
let i = 2
let product = 1
let rest = integer
let limit = Math.ceil(Math.sqrt(integer))
while (i <= limit) {
if (rest % i == 0) {
result.push(i)
product *= i
rest = integer / product
limit = Math.ceil(Math.sqrt(rest))
} else {
i++
}
}
result.push(rest)
return result
}
const divisors = (integer) => {
const result = [ 1 ]
const inner = (integer, list) => {
result.push(integer)
if (list.length === 0) {
return [ integer ]
}
const in_results = [ integer ]
const in_factors = inner(list[0], list.slice(1))
for (const f of in_factors) {
result.push(integer*f)
result.push(f)
in_results.push(integer*f)
in_results.push(f)
}
return in_results
}
const list = factorize(integer)
inner(list[0], list.slice(1))
const im = [...new Set(result)].sort((a, b) => a - b)
return im.slice(0, im.length-1)
}
That made a big difference! The first 102 steps all were faster than a second, reaching 20 digits! That’s more than 100x speedup! And then, after step 116, the thing crashed.
Remember, integer only does well until 16 digits. The numbers were just too big for the standard integer type. So, back to BigInt. The availability of BigInt in JavaScript was my first hunch for choosing JavaScript (although that would have worked just fine in Python 3 as well). And that led to two surprises.
First, sorting arrays with BigInts is different from sorting arrays with integers. Well, I find already sorting arrays of integers a bit weird in JavaScript. If you don’t specify otherwise it sorts numbers lexicographically, instead of by value:
[ 10, 2, 1 ].sort() == [ 1, 10, 2 ]
You need to provide a custom sorting function in order to sort the numbers by value, e.g.
[ 10, 2, 1 ].sort((a, b) => a-b) == [ 1, 2, 10 ]
The same custom sorting functions won’t work for BigInt, though. The custom function for sorting requires an integer result, not a BigInt. We can write something like this:
.sort((a, b) => (a < b)?-1:((a > b)?1:0))
The second surprise was that BigInt doesn’t have a square root function in the standard library. So I need to write one. Well, Newton is quickly implemented, or copy and pasted (source code).
Now, with switching to BigInt, we get the expected slowdown. The first 92 steps run faster than a second, reaching 19 digits, and then the first step to take longer than a minute is step 119, with 22 digits.
Also, the actual sequences started indeed diverging, due to the inaccuracy of large JavaScript integer: step 83 resulted in 23,762,659,088,671,304 using integers, but 23,762,659,088,671,300 using BigInt. And whereas that looks like a tiny difference on only the last digit, the number for the 84th step showed already what a big difference that makes: 20,792,326,702,587,410 with integers, and 35,168,735,451,235,260 with BigInt. The two sequences went entirely off.
What was also very evident is that at this point some numbers took a long time, and others were very quick. This is what you would expect from a more sophisticated approach to factorization, that it depends on the number and size of the factors. For example, calculating step 126 required to factorize the number 169,306,878,754,562,576,009,556, leading to 282,178,131,257,604,293,349,484, and that took more than 2 hours with that script on my machine. But then in Step 128 the result 552,686,316,482,422,494,409,324 was calculated from 346,582,424,991,772,739,637,140 in less than a second.
At that point I also started taking some of the numbers, and googled them, surfacing a Japanese blog post from ten years ago that posted the first 492 numbers, and also confirming that the 450th of these numbers corresponds to a published source. I compared the list with my numbers and was happy to see they corresponded so far. But I guesstimated I would not in a lifetime reach 492 steps, never mind the actual 2,145.
But that’s OK. Some things just need to be let rest.
That’s also when I realized that I was overly optimistic because I simply misread the number of steps that have been already calculated when reading the original post: I thought it was about 200-something steps, not 2,000-something steps. Or else I would have never started this. But now, thanks to the sunk cost fallacy, I wanted to push it just a bit further.
I took the biggest number that was posted on that blog post, 111,953,269,160,850,453,359,599,437,882,515,033,017,844,320,410,912, and let the algorithm work on that. No point in calculating steps 129, which is how far I have come, through step 492, if someone else already did that work.
While the computer was computing, I leisurely looked for libraries for fast factorization, but I told myself that no way am I going to install some scientific C library for this. And indeed, I found a few, such as the msieve project. Unsurprising, it was in C. But I also found a Website, CrypTool-Online with msieve on the Web (that’s pretty much one of the use cases I hope Wikifunctions will also support rather soonish). And there, I could not only run the numbers I already calculated locally, getting results in subsecond speed which took minutes and hours on my machine, but also the largest number from the Japanese website was factorized in seconds.
That just shows how much better a good library is than my naïve approach. I was slightly annoyed and challenged by the fact how much faster it is. Probably also runs on some fast machine in the cloud. Pretty brave to put a site like that up, and potentially have other people profit from the factorization of large numbers for, e.g. Bitcoin mining on your hardware.
The site is fortunately Open Source, and when I checked the source code I was surprised, delighted, and I understood why they would make it available through a Website: they don’t factorize the number in the cloud, but on the edge, in your browser! If someone uses a lot of resources, they don’t mind: it’s their own resources!
They took the msieve C library and compiled it to WebAssembly. And now I was intrigued. That’s something that’s useful for my work too, to better understand WebAssembly, as we use that in Wikifunctions too, although for now server side. So I rationalized, why not see if I can get that run on Node.
It was a bit of guessing and hacking. The JavaScript binding was written to be used in the browser, and the binary was supposed to be loaded through fetch. I guessed a few modifications, replacing fetch with Node’s FS, and managed to run it in Node. The hacks are terrible, but again, it doesn’t really matter as long as the factorization would speed up.
And after a bit of trying and experimenting, I got it running (source code). And that was my second hunch for choosing JavaScript: it had a great integration for using WebAssembly, and I figured it might come in handy to replace the JavaScript based solution. And now indeed, the factorization was happening in WebAssembly. I didn’t need to install any C libraries, no special environments, no nothing. I just could run Node, and it worked. I am absolutely positive there are is cleaner code out there, and I am sure I mismanaged that code terribly, but I got it to run. At the first run I found that it added an overhead of 4-5 milliseconds on each step, making the first few steps much slower than with pure JavaScript. That was a moment of disappointment.
But then: the first 129 steps, which I was waiting hours and hours to run, zoomed by before I could even look. Less than a minute, and all the 492 steps published on the Japanese blog post were done, allowing me to use them for reference and compare for correctness so far. The overhead was irrelevant, even across the 2,000 steps the overhead wouldn’t amount to more than 10 seconds.
The first step that took longer than a second was step 596, working on a 58 digit number. All the first 595 steps took less than a second each. The first step that took more than a minute, was step 751, a 76 digit number, taking 62 seconds, factorizing 3,846,326,269,123,604,249,534,537,245,589,642,779,527,836,356,985,238,147,044,691,944,551,978,095,188. The next step was done in 33 milliseconds.
The first 822 steps took an hour. Step 856 was reached after two hours, so that’s another 34 steps in the second hour. Unexpectedly, things slowed down again. Using faster machines, modern architectures, GPUs, TPUs, potentially better algorithms, such as CADO-NFS or GGNFS, all of that could speed that up by a lot, but I am happy how far I’ve gotten with, uhm, little effort. After 10 hours, we had 943 steps and a 92 digit number, 20 hours to get to step 978 and 95 digits. My goal was to reach step 1,000, and then publish this post and call it a day. By then, a number of steps already took more than an hour to compute. I hit step 1000 after about 28 and a half hours, a 96 digit number: 162,153,732,827,197,136,033,622,501,266,547,937,307,383,348,339,794,415,105,550,785,151,682,352,044,744,095,241,669,373,141,578.
I rationalize this all through “I had fun” and “I learned something about JavaScript, Node, and WebAssembly that will be useful for my job too”. But really, it was just one of these rabbit holes that I love to dive in. And if you read this post so far, you seem to be similarly inclined. Thanks for reading, and I hope I didn’t steal too much of your time.
I also posted a list of all the numbers I calculated so far, because I couldn’t find that list on the Web, and I found the list I found helpful. Maybe it will be useful for something. I doubt it. (P.S.: the list was already on the Web, I just wasn't looking careful enough. Both, OEIS and FactorDB have the full list.)
I don’t think any of the lessons here are surprising:
- for many problems a naïve algorithm will take you to a good start, and might be sufficient
- but never fight against an exponential algorithm that gets big enough, with constant speedups such as faster hardware. It’s a losing proposition
- But hey, first wait if it gets big enough! Many problems with exponential complexity are perfectly solvable with naïve approaches if the problem stays small enough
- better algorithms really make a difference
- use existing libraries!
- advancing research is hard
- WebAssembly is really cool and you should learn more about it
- the state of WebAssembly out there can still be wacky, and sometimes you need to hack around to get things to work
In the meantime I played a bit with other numbers (thanks to having a 4 core computer!), and I will wager one ambitious hypothesis: if 276 diverges, so does 276,276, 276,276,276, 276,276,276,276, etc., all the way to 276,276,276,276,276,276,276,276,276, i.e. 9 times "276".
I know that 10 times "276" converges after 300 steps, but I have tested all the other "lexical multiples" of 276, and reached 70 digit numbers after hundreds of steps. For 276,276 I pushed it further, reaching 96 digits at step 640. (And for at least 276,276 we should already know the current best answer, because the Wikipedia article states that all the numbers under one million have been calculated, and only about 9,000 have an unclear fate. We should be able to check if 276,276 is one of those, which I didn't come around to yet).
Again, this is not research, but just fooling around. If you want actual systematic research, the Wikipedia article has a number of great links.
Thank you for reading so far!
P.S.: Now that I played around, I started following some of the links, and wow, there's a whole community with great tools and infrastructure having fun about Aliquot sequences and prime numbers doing this for decades. This is extremely fascinating.
Popculture in logics
- You ⊑ ∀need.Love (Lennon, 1967)
- ⊥ ≣ compare.You (Nelson, 1985)
- Cowboy ⊑ ∃sing.SadSadSong (Michaels, 1988)
- ∀t : I ⊑ love.You (Parton, 1973)
- ∄better.Time ⊓ ∄better⁻¹.Time (Dickens, 1859)
- {god} ⊑ Human ⊨ ? (Bazilian, 1995)
- Bad(X)? (Jackson, 1987)
- ⃟(You ⊑ save.I) (Gallagher, 1995)
- Dreamer(i). ∃x : Dreamer(x) ∧ (x ≠ i). ⃟ ∃t: Dreamer(you). (Lennon, 1971)
- Spoon ⊑ ⊥ (Wachowski, 1999)
- ¬Cry ← ¬Woman (Marley, 1974)
- ∄t (Poe, 1845)
Solutions: Turn around your monitor to read them.
sǝlʇɐǝq ǝɥʇ 'ǝʌol sı pǝǝu noʎ llɐ ˙ǝuo
ǝɔuıɹd ʎq ʎllɐuıƃıɹo sɐʍ ʇı 'ƃuos ǝɥʇ pǝɹǝʌoɔ ʇsnɾ pɐǝuıs ˙noʎ oʇ sǝɹɐdɯoɔ ƃuıɥʇou ˙oʍʇ
˙uosıod ʎq uɹoɥʇ sʇı sɐɥ ǝsoɹ ʎɹǝʌǝ ɯoɹɟ '"ƃuos pɐs pɐs ɐ sƃuıs ʎoqʍoɔ ʎɹǝʌǝ" ˙ǝǝɹɥʇ
ʞɔɐɹʇpunos ǝıʌoɯ pɹɐnƃʎpoq ǝɥʇ ɹoɟ uoʇsnoɥ ʎǝuʇıɥʍ ʎq ɹɐlndod ǝpɐɯ ʇnq uoʇɹɐd ʎllop ʎq ʎllɐuıƃıɹo 'noʎ ǝʌol sʎɐʍlɐ llıʍ ı 'ɹo - ",noʎ, ɟo ǝɔuɐʇsuı uɐ ɥʇıʍ pǝllıɟ ,ǝʌol, ʎʇɹǝdoɹd ɐ ƃuıʌɐɥ" uoıʇdıɹɔsǝp ǝɥʇ ʎq pǝɯnsqns ɯɐ ı 'ʇ sǝɯıʇ llɐ ɹoɟ ˙ɹnoɟ
suǝʞɔıp sǝlɹɐɥɔ ʎq sǝıʇıɔ oʍʇ ɟo ǝlɐʇ ɯoɹɟ sǝɔuǝʇuǝs ƃuıuǝdo ǝɥʇ sı sıɥʇ ˙(ʎʇɹǝdoɹd ǝɥʇ ɟo ǝsɹǝʌuı suɐǝɯ 1- ɟo ɹǝʍod" ǝɥʇ) ǝɯıʇ ɟo ʇsɹoʍ ǝɥʇ sɐʍ ʇı ˙sǝɯıʇ ɟo ʇsǝq ǝɥʇ sɐʍ ʇı ˙ǝʌıɟ
(poƃ)ɟoǝuo ƃuıɯnsqns sn ʎq pǝlıɐʇuǝ sı ʇɐɥʍ sʞsɐ ʇı ʎllɐɔısɐq ˙ʇıɥ ɹǝpuoʍ ʇıɥ ǝuo 5991 ǝɥʇ 'sn ɟo ǝuo sɐʍ poƃ ɟı ʇɐɥʍ ˙xıs
pɐq ǝlƃuıs ʇıɥ ǝɥʇ uı "pɐq s,oɥʍ" ƃuıʞsɐ 'uosʞɔɐɾ lǝɐɥɔıɯ ˙uǝʌǝs
ɔıƃol lɐpoɯ ɯoɹɟ ɹoʇɐɹǝdo ʎılıqıssod ǝɥʇ sı puoɯɐıp ǝɥʇ ˙"ǝɯ ǝʌɐs oʇ ǝuo ǝɥʇ ǝɹ,noʎ ǝqʎɐɯ" ǝuıl ǝɥʇ sɐɥ ʇı ˙sısɐo ʎq 'llɐʍɹǝpuoʍ ˙ʇɥƃıǝ
˙ooʇ ǝuo ǝɹɐ noʎ ǝɹǝɥʍ ǝɯıʇ ɐ sı ǝɹǝɥʇ ǝqʎɐɯ puɐ ˙(ǝɯ ʇou ǝɹɐ sɹǝɥʇo ǝsoɥʇ puɐ 'sɹǝɯɐǝɹp ɹǝɥʇo ǝɹɐ ǝɹǝɥʇ) ǝuo ʎluo ǝɥʇ ʇou ɯɐ ı ʇnq ˙ɹǝɯɐǝɹp ɐ ɯɐ ı" ˙ǝuıƃɐɯı 'uıɐƃɐ uouuǝl uɥoɾ ˙ǝuıu
(ǝlɔɐɹo ǝɥʇ sʇǝǝɯ ǝɥ ǝɹoɟǝq ʇsnɾ oǝu oʇ ƃuıʞɐǝds pıʞ oɥɔʎsd ǝɥʇ) xıɹʇɐɯ ǝıʌoɯ ǝɥʇ ɯoɹɟ ǝʇonb ssɐlɔ ˙uoods ou sı ǝɹǝɥʇ ˙uǝʇ
ʎǝuoɯ ǝɯos sʇǝƃ puǝıɹɟ sıɥ os ƃuıʎl ʎlqɐqoɹd sɐʍ ǝɥ ʇnq 'puǝıɹɟ ɐ oʇ sɔıɹʎl ǝɥʇ pǝʇnqıɹʇʇɐ ʎǝlɹɐɯ ˙"ʎɹɔ ʇou" sʍolloɟ "uɐɯoʍ ʇou" ɯoɹɟ ˙uǝʌǝlǝ
ǝod uɐllɐ ɹɐƃpǝ ʎq '"uǝʌɐɹ ǝɥʇ" ɯoɹɟ ɥʇonb ˙ǝɹoɯɹǝʌǝu :ɹo ˙ǝɯıʇ ou sı ǝɹǝɥʇ ˙ǝʌlǝʍʇ
Power in California
It is wonderful to live in the Bay Area, where the future is being invented.
Sure, we might not have a reliable power supply, but hey, we have an app that connects people with dogs who don't want to pick up their poop with people who are desperate enough to do this shit.
Another example how the capitalism that we currently live failed massively: last year, PG&E was found responsible for killing people and destroying a whole city. Now they really want to play it safe, and switch off the power for millions of people. And they say this will go on for a decade. So in 2029 when we're supposed to have AIs, self-driving cars, and self-tieing Nikes, there will be cities in California that will get their power shut off for days when there is a hot wind for an afternoon.
Why? Because the money that should have gone into, that was already earmarked for, making the power infrastructure more resilient and safe went into bonus payments for executives (that sounds so cliché!). They tried to externalize the cost of an aging power infrastructure - the cost being literally the life and homes of people. And when told not to, they put millions of people in the dark.
This is so awfully on the nose that there is no need for metaphors.
San Francisco offered to buy the local power grid, to put it into public hands. But PG&E refused that offer of several billion dollars.
So if you live in an area that has a well working power infrastructure, appreciate it.
Prediction coming true
I saw my post from seven years ago, where I said that I really like Facebook and Google+, but I want a space where I have more control about my content so it doesn't just disappear. "Facebook and Google+ -- maybe they won't disappear in a year. But what about ten?"
And there we go, Google+ is closing in a few days.
I downloaded my data from there (as well as my data from Facebook and Twitter), to see if there is anything to be salvaged from that, but I doubt it.
Productivity pro tip
- make a list of all things you need to do
- keep that list roughly in order of priority, particularly on the first 3-5 items (lower on the list it doesn't matter that much)
- procrastinate the whole day from doing the number 1 item by doing the number 2 to 5 items
Progress in lexicographic data in Wikidata 2023
Here are some highlights of the progress in lexicographic data in Wikidata in 2023
- Greek jumped from 0% to 45% right away
- Panjabi jumped right away from 0% to 71% (but on an admittedly small corpus)
- Italian made a huge jump from 52% to 82% by increasing the number of forms from 9,000 to 286,000
- Turkish jumped from 0.9% to 22%
- Sindhi climbed from 15% to 25%
- Farsi climbed from 15% to 24% increasing the number of forms from 4,000 to 33,000
- Western Panjabi climbed from 36.9% to 47.9%
- Hindi climbed from 49.9% to 65.9%
- Breton increased from 56% to 67%
- Croatian increased from 40% to 45%
- Dutch went from 20% to 29%
- French from 82.9% to 86.9%, mostly by dealing better with apostrophes in the analysis
- Nynorsk pushed from 75% to 80% by increasing the number of forms from 18,000 to 68,000
- Danish from 83.9% to 86.9% by increasing the number of forms in Wikidata from 65,000 to 170,000
- German from 76% to 79% by increasing the number of forms in Wikidata from 90,000 to 200,000
- Spanish pushed from 88% to 91% by increasing the number of forms from 280,000 to 430,000
What does the coverage mean? Given a text (usually Wikipedia in that language, but in some cases a corpus from the Leipzig Corpora Collection), how many of the occurrences in that text are already represented as forms in Wikidata's lexicographic data. Note that every percent more gets much more difficult than the previous one: an increase from 1% to 2% usually needs much much less work than from 91% to 92%.
RDF in Mozilla 2
I read last week Brendan Eich's post on Mozilla 2, where he said that with Moz2 they hope that they can "get rid of RDF, which seems to be the main source of "Mozilla ugliness"". Danny Ayers commented on this, saying that RDF "should be nurtured not ditched."
Well, RDF in Mozilla always was crappy, and it was based on a pre-1999-standard RDF. No one ever took up the task -- remind you, it's open source -- to dive into the RDF inside Mozilla and repair and polish it, make it a) compatible to the 2004 RDF standard, (why didn't RDF get version numbers, by the way?) and b) clean up the code and make it faster.
My very first encounter with RDF was through Mozilla. I was diving into Mozilla as a platform for application development, which seemed like a very cool idea back then (maybe it still is, but Mozilla isn't really moving into this direction). RDF was prominent there: as an internal data structure for the developed applications. Let's repeat that: RDF, which was developed to allow the exchange of data on the web, was used within Mozilla as an internal data structure.
Surprisingly, they had performance issues. And the code was cluttered with URIs.
I still think it is an interesting idea. In last year's Scripting for the Semantic Web workshop I presented an idea on integrating semantic technologies tightly into your programming. Marian Babik and Ladislav Hluchy picked that idea up and expanded and implemented the work much better than I ever could.
It would be very interesting how this combination could really be put to work. An internal knowledge base for your app that is queried via SPARQL. Doesn't sound like the most performant idea. But then -- imagine not having one tool accessing that knowledge base. But rather a system architecture with a number of tools accessing that knowledge base. Adding data to existing data. Imagine just firing up a newly installed Adressbook, and -- shazam! -- all your data is available. You don't like it anymore? Switch back. No changes lost. Everything is just views.
But then again, didn't the DB guys try to do the same? Yes, maybe. But with semantic web technology, shouldn't it be easier to do it, because it is meant to integrate?
Just thinking. Would love to see a set of tools based on a plugabble knowledge base. I'm afraid, performance would suck. I hope we will see.
RDF is not just for dreamers
Sometimes I stumble upon posts that leave me wonder, what actually do people think about the whole Semantic Web idea, and about standards like RDF, OWL and the like. Do you think academia people went out and purposefully made them complicated? That they don't want them to get used?
Christopher St. John wrote down some nice experience with using RDF for logging. And he was pretty astonished that "RDF can actually be a pretty nifty tool if you don't let it go to your head. And it worked great."
And then: "Using RDF doesn't actually add anything I couldn't have done before, but it does mean I get to take advantage of tons of existing tools and expertise." Well, that's pretty much the point of standards. And the point of the whole Semantic Web idea. There won't be anything you will be able to do later, that you're not able to do today! You know, assembler was pretty turing-complete already. But having your data in standard formats helps you. "You can buy a book on RDF, but you're never going to buy a book on whatever internal debug format you come up with"
Stop being surprised that some things on the Semantic Web work. And don't expect miracles either.
RIP Christopher Alexander
RIP Christopher Alexander, the probably most widely read actual architect in all of computer science. His work, particularly his book "A Pattern Language" was popularized, among others, by the Gang of Four and Design Pattern work, and is frequently read and cited in Future of Programming and UX circles for the idea that everyone should be able to create, but in order to enable them, they need patterns that make creation possible. His work inspired Ward Cunningham when developing wikis and Will Wright when developing that most ungamelike of games, Sim City. Alexander died on March 17, 2022 at the age of 85.
RIP Niklaus Wirth
RIP Niklaus Wirth;
BEGIN
I don't think there's a person who created more programming languages that I used than Wirth: Pascal, Modula, and Oberon; maybe Guy Steele, depending on what you count;
Wirth is also famous for Wirth's law: software becomes slower more rapidly than hardware becomes faster;
He received the 1984 Turing Award, and had an asteroid named after him in 1999; Wirth died at the age of 89;
END.
RIP Steve Wilhite
RIP Steve Wilhite, who worked on CompuServe chat for decades and was the lead of the CompuServe team that developed the GIF format, which is still widely used, and which made the World Wide Web a much more colorful and dynamic place by having a format that allowed for animations. Wilhite incorrectly insisted on GIF being pronounced Jif. Wilhite died on March 14, 2022 at the age of 74.
Rainbows end
Rainbows end.
The book, written in 2006, was set in 2025 in San Diego. Its author, Vernor Vinge, died yesterday, March 20, 2024, in nearby La Jolla, at the age of 79. He is probably best known for popularizing the concept of the Technological Singularity, but I found many other of his ideas far more fascinating.
Rainbows end explores themes such as shared realities, digital surveillance, and the digitisation of the world, years before Marc Andreessen proclaimed that "software is eating the word", describing it much more colorfully and rich than Andreessen ever did.
His other work that I enjoyed is True Names, discussing anonymity and pseudonymity on the Web. A version of the book was published with essays by Marvin Minsky, Danny Hillis, and others. who were inspired by True Names.
His Science Fiction was in a rare genre, which I love to read more about: mostly non-dystopian, in the nearby future, hard sci-fi, and yet, imaginative, exploring novel technologies and their implications on individuals and society.
Rainbows end.
Released dlpconvert
There's so much to do right now here, and even much, much more I'd like to do - and thus I'm getting a bit late on my announcements. Finally, here we go: dlpconvert is released in a 0.5 version.
You probably think: what is dlpconvert? dlpconvert is able to convert ontologies, that are within the dlp-fragment, from one syntax - namely the OWL XML Presentation Syntax- into another, here Datalog. Thus you can just take your ontology, convert it and then use the result as your program in your Prolog-engine. Isn't that cool?
Well, it would be much cooler if it were stronger tested, and if it could read some more common syntaxes like RDF/XML-Serialisation of OWL ontologies, but both is on the way. As for testing I would hope that you may test a bit - as for the serialisation, it should be available pretty soon.
dlpconvert is based totally on KAON2 for the reduction of the ontology.
I will write more as soon as I have more time.
Research visit
So, I have arrived for my very first longer research visit. I am staying at the Laboratory of Applied Ontologies in Rome. I have never been to Rome before, and all my type of non-work experience I'll chat about in my regular blog, in German. Here I'll stick to Semantic Stuff and such.
So, if you are nearby and would like to meet -- give me a note! I am staying in Rome up to the WWW, i.e. up to May 20th. The plan is to work on my dissertation topic, Ontology Evaluation, especially in the wake of the EON2006 workshop, but that's not all it seems. People are interested in and knowledge about Semantic Wikis as well. So there will be quite a lot stuff happening in the next few weeks -- I'm excited about it all.
Who knows what will happen? If my plan works out, at the end of the stay we will have a common framework for Ontology Evaluation. And I am not talking about this paper frameworks -- you know, that are presented in papers with titles starting "Towards a..." or "A framework for...". No, but real software, stuff you can download, and play with.
Restarting, 2019 edition
I had neglected Simia for too long - there were five entries in the last decade. A combination of events lead me to pour some effort back into it - and so I want to use this chance to restart it, once again.
Until this weekend, Simia was still running on a 2007 version of Semantic MediaWiki and MediaWiki - which probably helped with Simia getting hacked a year or two ago. Now it is up to date with a current version, and I am trying to consolidate what is already there with some content I had created in other places.
Also, life has been happening. If you have been following me on Facebook (that link only works if you're logged in), you have probably seen some of that. I married, got a child, changed jobs, and moved. I will certainly catch up on this too, but there is no point in doing that all in one huge mega-post. Given that I am thinking about starting a new project just these days, this might be the perfect outlet to accompany that.
I make no promises with regards to the quality of Simia, or the frequency of entries. What I would really love recreate would be a space that is as interesting and fun for as my Facebook wall was, before I stopped engaging there earlier this year - but since you cannot even create comments here, I have to figure out how to make this even remotely possible. For now, suggestions on Twitter or Facebook are welcome. And no, moving to WordPress or another platform is hardly an option, as I really want to stay with Semantic MediaWiki - but pointers to interesting MediaWiki extensions are definitely welcome!
Running out of text
Many of the available text corpora have by now been used for training language models. One untapped corpus so far have been our private messages and emails.
How fortunate that none of the companies that train large language models have access to humongous logs of private chats and emails, often larger than any other corpus for many languages.
How fortunate that those who do have well working ethic boards established, who would make sure that such requests are evaluated.
How fortunate that we have laws in place to protect our privacy.
How fortunate that when new models are published also the corpora are being published on which the models are being trained.
What? Your telling me, "Open"AI is keeping the training corpus for GPT-4 secret? The company closely associated with Microsoft, who own Skype, Office, Hotmail? The same Microsoft who just fired an ethics team? Why would all that be worrisome?
P.S.: To make it clear: I don't think that OpenAI has used private chat logs and emails as training data for GPT-4. But by not disclosing their corpora, they might be checking if they can get away with not being transparent, so that maybe next time they might do it. No one would know, right? And no one would stop them. And hey, if it improves the metrics...
SSSW Day 1
Today's invited speaker was Frank von Harmelen, co-editor of the OWL standard and author of the Semantic Web Primer. His talk was on fundamental research challenges generated by the Semantic Web (or: two dozen Ph.D. topics in a single talk). He had the idea after he was asked one day in the cafeteria "Hey Frank, whazzup in the Semantic Web?"
In the tradition of Immanuel Kant's four famous questions on philosophy, Frank posed the four big research challenges:
- Where does the metadata come from?
- Where do the ontologies come form?
- What to do with the many different ontologies?
- Where's the Web in the Semantic Web?
He derived many research questions that arise when you bring results from other fields (like databases, natural language, machine learning, information retrieval or knowledge engineering) to the Semantic Web and not just change the buzzwords, but take the implications that come along with the Semantic Web seriously.
Some more notes:
- What is the semantic equivalent to a 404? How should a reasoner handle the lack of referential integrity?
- Inference can be cheaper than lookup on the web.
- Today OWL lite would probably have become more like OWL DLP, but they didn't know better than
The other talks were given by Asun Gómez-Pérez on Ontological Engineering, and Sean Bechhofer on Knowledge Representation Languages for the SemWeb, pretty good stuff by the people who wrote the book. I just wonder if it was too fast for the people who didn't know about it already, and too repeting for the others, but well, that's always the problem with these kind of things.
The hands-on session later was interesting: we had to understand several OWL ontologies and explain certain inferences, and Natasha Noy helped us with the new Protégé 3.1. It was harder than I thought quite some times. And finally Aldo Gangemi was giving us some exercises with knowledge representation design patterns, based on DOLCE. This was hard stuff...
Wow, this was a lot of namedropping. The social programme (we were hiking today) around the summer school, and the talks with the peers are sometimes even more interesting than the actual summer school programme itself, but this probably won't be too interesting for most of you, and it's getting late as well, so I just call it a day.
SSSW Day 2
Natasha Noy gave the first talk today, providing a general overview on Mapping and Alignment algorithms and tools. Even though I was not too interested in the topic, she really caught my interest with a good and clean and structured talk. Thank for that! After, Steffen Staab continued, elaborating on the QOM approach to ontology mapping, having some really funny slides, but, as this work was mostly developed in Karlsruhe I already knew it. I liked his appeal for more tools that are just downloadable and usable, without having to fight for hours or days just to create the right environment for them. I totally agree on that!
The last talk of the day was from Aldo Gangemi on Ontology Evaluation. As I consider making this the theme of my PhD-thesis - well, I am almost decided on that - I was really looking forward to his talk. Although it was partially hard to follow, because he covered quite a broad approach to this topic, there have been numerous interesting ideas and a nice bibliography. Much to work on. I especially didn't yet see the structural measures he presented applied to the Semantic Web. Not knowing any literature on them, I am still afraid, that they actually fail [SSSW Day 1|Frank's requirements from yesterday]]: not just to be taken from graph theory, but rather to have the full implications of the Semantic Web paradigm been applied to them and thought through. Well, if no one did that yet, there's some obvious work left for me ;)
The hands-on-sessions today were quite stressy, but nevertheless interesting. First, we had to powerconstruct ontologies about different domains of traveling: little groups of four persons working on a flight agency ontology, a car rental service ontology and a hotel ontology. Afterwards, we had to integrate them. Each exercise had to be done in half a hour. We pretty much failed miserably in both, but we surely encountered many problems - which was the actual goal: in OWL DL you can't even concatenate strings. How much data intefration can you do then?
The second hands-on-session was on evaluationg three ontologies. It was quite interesting, although I really think that many of these things can happen automatically (I will work on this in the next two weeks, I hope). But the discussion afterwards was quite revealing, as it showed how differently people think about some quite fundamental issues, the importance they give to structural measures compared to the functional ones. Or, differently said: the question is, is a crappy ontology on a given domain better than a good ontology that doesn't cover your domain of interest? (The question sounds strange to you? To me as well, but well...)
Pitily I had to miss today's social special event, a football match between the students of the Summer School. Instead I had a very interesting chat with a colleague from the UPM, who came here for a talk, and who also wants to make her PhD in Ontology Evaluation, Mari Carmen Suárez de Figueroa. Interesting times are lying ahead.
