Salonanarchist | Leunstoelactivist

Has coverage of Amsterdam’s district politics declined?

The districts with their elected councils have always played an important role in Amsterdam politics, but since the elections of 19 March 2014 their powers have been curtailed (I’m simplifying; here’s an explanation in Dutch). One may ask what point there is in having districts if their role is largely reduced to implementing policies set by the city government. And would people still be interested in what they do?

To find out more, I looked up the names of almost 500 local politicians and counted how often they’re mentioned at the websites of the most important local media, newspaper het Parool and TV station AT5. The results are shown below:

Unsurprisingly, there were many mentions of local politicians during election month March 2014. Further, especially at AT5, there’s been a peak in mentions of district politicians in October 2010. This is probably related to their eager coverage of the so-called MuzyQ affair in the Oost district.

But the relevant question is of course: what happened after March 2014. At het Parool, mentions of district politicians declined. At AT5, the pattern is less clear-cut.

The chart below shows articles mentioning district politicians as a percentage of all articles mentioning local politicians. The graph shows the three-month moving average. This has the advantage of showing the trend more clearly, but the consequence is that there’s a lag before effects fully manifest themselves. This can be seen clearly for het Parool: in the graph, the effect of the elections extends until May, but in reality it wore off quicker (cf. the top chart).

As the chart shows, AT5’s coverage of district politics has been declining for years. It’s difficult to say whether the changes in the administrative system as of March 2014 have had a separate effect on AT5’s coverage.

At het Parool, there is a clear effect after March 2014. There are two possible interpretations:

  • One could argue that there’s been a decline in mentions of district politicians after the March 2010 elections as well; perhaps journalists need some time after an election to figure out which district politicians are newsworthy. In 2010, coverage of district politics recovered pretty fast as a consequence of the MuzyQ affair refered to above.
  • On the other hand, it’s been almost a year and a half now since the 2014 elections. One should expect any temporary effects to have worn off by now. All the same, mentions of district politicians have remained below twenty percent, whereas they were mostly above twenty percent before the elections.

I’m inclined to follow the second interpretation. It appears that district politics has become structurally less relevant - at least so in the eyes of the editors of het Parool.


Since I’ve been counting anyway, I can also report which politicians have been mentioned most often in het Parool. I restrict myself to council members.

2014 - now, city council
1. Jan Paternotte
2. Rutger Groot Wassink
3. Wil van Soest
4. Marjolein Moorman
5. Johnas van Lammeren

2014 - now, district councils
1. Hans Bremer
2. Pieter Rietman
3. Iwan Leeuwin
4. Reinout de Vries
5. Erna Berends

2010 –2014, city council
1. Laurens Ivens
2. Robert Flos
3. Jan Paternotte
4. Frank de Wolf
5. Maureen van der Pligt

2010 - 2014, district councils
1. Vera Bergkamp
2. Narish Parsan
3. Rutger Groot Wassink
3. Nelly Duijndam
4. Jan-Bert Vroege


There appears to be no data available on past roles of local politicians. Since I was looking for a project to learn PHP and MySQL, I decided to create one myself. Data is from a number of sources including my own News from Amsterdam website (no longer maintained). I used the search forms of the AT5 and Parool websites to look up articles containing names of local politicians (after removing accents, shortening double last names and setting to lowercase). I restricted the analysis to articles that appeared between March 2010 and July 2015 and included only those with a publication date within the period in which a politician was in office.

See the Dutch version of this article for a more elaborate discussion of how the data was collected and analysed.

On demolishing Amsterdam School housing blocks


Recently, some Amsterdam School housing blocks on the Tugelaweg in the Transvaal neighbourhood in Amsterdam have been demolished. I wasn’t happy about this: we should cherish the Amsterdam School heritage, even the buildings that are not highlights of the period. But fair’s fair: the new buildings that have replaced them look good and fit well with the blocks that have been spared.

At the opposite site of the neighbourhood, along the Transvaalkade, something similar has happened. There, the replacement buildings are even a striking imitation of the Amsterdam School.

As for the Tugela blocks: after some research I found out that the decision to demolish some of them hasn’t been taken lightly. Before the decision was made, one J. Schilt of the municipal Bureau of Monuments and Archeology had written a report on the matter. In dense prose and with an eye for interesting details, he described the history of the houses.

The Tugela blocks were built between 1915 and 1924 by the Handwerkers Vriendenkring (HWV), an institution rooted in the Jewish labour movement. In fact, one of the first Dutch trade unions, the General Diamond Workers’ Union, originated from the HWV. The HWV got to carry out part of the plans architect H.P. Berlage had made for the new Transvaal neighbourhood.

Meanwhile, the railroad along the Tugelaweg was still at surface level. The embankment, which gives the Tugelaweg a bit of a claustrophobic feel today, was added later.

During the German occupation, the Transvaal neighbourhood was initially an area where Jews from other neighbourhoods were forced to move to. Later, many residents were deported. Especially during the last winter of the war, wooden parts were looted from the empty houses and they ended up partly ruined. After the war, the houses were restored, but some of the top floors had to be demolished. Schilt commented:

The original architectural character has obviously suffered from the inevitable restaurations shortly after the liberation; however these construction traces also keep the memory alive of the war years and especially of the fate of the Jewish population.

Schilt concluded that some of the blocks should be spared and that any new construction should fit with the neighbourhood:

The result should in any case be a form of «Stadtreparatur»; that is, a form of careful urban renewal, and not some sort of conceited and contemporay vision of a star architect.

In any case, it’s good to know that the municipality apparently employs knowledgeable and motivated people to guard the city’s social history.

The picture shows the new blocks from the Retiefstraat; the Tugelaweg can be seen from the train between Amstel and Muiderpoort.


Making the pay gap transparent

As of 2017, US-listed companies will be required to report how the income of their CEO relates to the median pay of their workers. This was decided by the Securities and Exchange Commission (SEC) on Wednesday.

The decision is reminiscent of the so-called Tinbergen norm, which states that the highest income in a company shouldn’t be higher than five times the lowest income. Recently, Broer Akkerboom (formerly of Groningen University) has revealed that the norm was probably not formulated by the renowned Dutch economist Jan Tinbergen himself, even though everybody seems to think so.

Akkerboom quotes Nobel Prize winner Robert Shiller:

Jan Tinbergen had calculated that a CEO of a company should not earn more than five times the wage of a typical worker. In America, that has grown to 230 times the wage of a typical person! We kind of missed the whole Tinbergen norm.

Akkerboom notes that Shiller compares highest pay to typical pay, rather than to lowest pay. There’s something else about the quote: Shiller compares CEO pay not to other workers in the company, but to the wider population. I think that makes sense: why compare CEO pay to their own employees but not to the outsourced workers who clean their offices, serve their lunches and fix their computers? On the other hand, the new SEC norm includes at least some of the workers a company employs abroad.

Further, the Tinbergen-Shiller norm has the practical advantage that it can’t be tinkered with so easily (the NYT describes how the new SEC norm can be manipulated).

How would Dutch CEOs score on the Tinbergen-Shiller norm? The CEOs of the government-owned ABN AMRO bank and of government-owned Schiphol Airport get about 25 to 30 times the income of a typical Dutch person. Average CEO pay (excluding pension contributions) of companies included in the AEX is about 100 times that amount. So there’s some levelling to be done.


Can they track you by your smartphone battery status

Who knew. Apparently, websites can collect pretty detailed information about the battery of the laptop or smartphone you’re using. They can see if you’re currently charging the battery. If you are, they can see how long it’ll take before it’s fully charged. If you aren’t, they can see how long it’ll last.

I read about this in the Guardian, which has an article about a study that apparently found that the detailed information obtained through the HTML5 battery status API can in some cases be used to identify users, at least over a short period, even if they use a VPN or Chrome’s private browsing mode.

However, the API doesn’t work in all browsers. In fact, I googled around to find out how it works and it turns out you need different code to get it working in Firefox than in Chrome. And while I got the code working on my Macbook, it didn’t work on my iPhone - not even with Chrome.

So how about your device? You can check below if the code works with your combination of browser and device. Let me know!

Your battery status

Let me know what it shows on your device:


Shared space

Een stukje van de Amsterdamse De Ruijterkade tussen station en pont wordt shared space: voetgangers en fietsers moeten onderling maar uitmaken wie er voorrang heeft. De eerste reactie die in me opkwam: «kom op mensen, dit is een doorgaande fietsroute en geen speeltuin voor modieuze experimenten».

OK, dat was misschien wat kort door de bocht. Maar op Twitter blijkt dat mensen met verstand van zaken ook hun bedenkingen hebben:

  • Fietsprofessor Marco te Brömmelstroet en voormalig gemeenteraadslid Fjodor Molenaar willen het plan wel een kans geven maar alleen als het fietspad scootervrij wordt gemaakt.
  • Adviseur David Hembrow waarschuwt dat er onnodige ergernis zal ontstaan tussen fietsers en voetgangers. Ook al manoeuvreren fietsers voorzichtig tussen de voetgangers door, dan nog zullen voetgangers zich onveilig voelen en zich ergeren aan de fietsers.
  • Utrechts raadslid Bram Fokke ziet het ook niet zitten: «Doe het niet. Shared space werkt alleen bij overzichtelijke situaties en dan bij (zeer) spaarzaam gebruik».

De De Ruijterkade is niet zomaar een fietsroute. Het is een belangrijke verbinding tussen Oost en West en één van de drukste fietsroutes van Amsterdam. Op werkdagen rijden (pdf) er tijdens de avondspits meer dan 2.400 fietsers (dat zou overigens neerkomen op gemiddeld tenminste één fiets per 3 seconden; ik neem aan dat het er regelmatig een stuk meer zijn).

Niet voor niets is de De Ruijterkade tien jaar geleden al opgenomen in het Hoofdnet Fiets. Dat betekent dat er kwaliteitscriteria (pdf) gelden: je moet er bijvoorbeeld snel kunnen fietsen (nou ja, gemiddeld 12 tot 15 km/u) en er moet ruimte zijn om andere fietsers in te halen. Ik heb niet de indruk dat deze criteria leidend zijn geweest bij het nieuwe ontwerp.

Misschien valt het in de praktijk allemaal mee. Het gaat tenslotte maar om een korte onderbreking van het fietspad. Maar het valt te hopen dat de gemeente een vinger aan de pols houdt, ook als het gaat om de doorstroming van fietsers. En dat ze de boel weer terugdraaien als blijkt dat het gewoon niet werkt.

Update - Eric Plankeel wijst erop dat de situatie extra ingewikkeld wordt door de nieuwe tunnel onder het spoor voor fietsers en voetgangers, die uitkomt op de shared space en dit najaar wordt geopend.

Aangezien ik aanvankelijk dacht dat de opmerking over overstekende voetgangers ging, heb ik die cijfers ook nog even opgezocht. De GVB-ponten bij het station zetten op een werkdag gemiddeld zo’n 40.000 mensen over het IJ, waarvan 20 tot 40% voetgangers, met een flinke piek tijdens de ochtend- en avondspits. Dat worden er nog meer door woningbouw en groei van de werkgelegenheid in Noord.

Al met al kan je je inderdaad afvragen hoe realistisch de artist’s impression is.

Update 2 - Inmiddels is er ook een petitie voor een groene golf in plaats van shared space.

Shared space: het besluit

Via het raadsarchief valt te achterhalen dat het voorlopig ontwerp voor het Stationseiland in 2012 naar de raad is gestuurd (door wethouder Eric Wiebes?). In de bijlagen wordt de shared space aangekondigd (onder het motto de chaos als uitgangspunt nemen) en valt te lezen dat er ook nog plannen zijn om ribbels in het fietspad aan te brengen die zorgen voor een «rammelend effect» (heel fijn). Van verschillende kanten zijn destijds vraagtekens geplaatst.

Bij de bespreking in commissie en raad is de shared space niet aan de orde geweest. Misschien vonden raadsleden een prima idee, of misschien vonden ze andere aspecten (geen scooters in de tunnel onder het station) op dat moment belangrijker. Hoe dan ook, het voorlopig ontwerp is met algemene stemmen aangenomen.

Dat verklaart wellicht waarom raadsleden zich momenteel nauwelijks lijken te mengen in de discussie over het plan (met Jan-Bert Vroege als uitzondering): je kan moeilijk kritiek leveren op een plan waar je partij eerder mee heeft ingestemd. Mocht de nieuwe inrichting onverhoopt voor chaos zorgen of de doorstroming van fietsers frustreren, dan verandert de zaak natuurlijk.

Voor de volledigheid nog even de relevante passage uit het Ontwerpboek (een bijlage bij de voordracht die in 2012 naar de gemeenteraad is gestuurd):

Het gebied bij de aanlanding van de ponten is een druk knooppunt. Het grootste deel van de voetgangers en fietsers in dit gebied heeft de pont als bestemming. Voetgangers en fietsers komen hier in een soort golfbeweging in grote groepen van de ponten. Ze bewegen door elkaar en zullen elkaar hinderen op hun eigen toegewezen verkeersareaal.
Gekozen is om juist de chaos in het gebied als uitgangspunt te nemen en daarom te kiezen voor een inrichting conform de zogenaamde ‘shared space’ principes. Door zo min mogelijk het verkeer te regelen of te scheiden maar juist te mengen houden verkeersdeelnemers meer rekening met elkaar.
Het voorstel is om het gebied rond de pontaanlanding vorm te geven als een pleinachtige ruimte waar voetgangers en fietsers gezamenlijk gebruik van maken. De fietspaden houden dan ook voor dit plein op. Wel kan door middel van een lichte markering of aangepaste bestrating een route aangegeven worden voor de fietsers.
Om te voorkomen dat fietsers en brommers met teveel snelheid het gebied doorkruisen wordt gezocht naar een bestrating welke voor voetgangers goed te belopen is maar waarbij het voor fietsers en brommers vervelend is om snel over heen te gaan. Bijvoorbeeld door stroken afwijkende bestrating toe te passen die een ‘rammelend effect’ zullen hebben. Dit zal in de volgende fase van het project verder uitgewerkt worden.


Patterns in Tour de France excitement

I wrote two posts on patterns in Tour de France excitement, as measured by the ratio of #tdf tweets containing an exclamation mark. Since I published them, I’ve done some more data cleaning, most notably filtering out some marketing tweets. As a result, some of the weirder patterns have disappeared. Here’s a rewrite of the findings.

Hour of the day

Many stages show a peak in enthusiasm towards the end of the afternoon, when the riders finish. In some of the cases where such a peak is lacking, there appears to be a logical explanation. The most obvious examples are the rest days on 13 and 21 July. And the first stage was an individual time trial so there was not just one finish time.

In some cases, the pattern may be less pronounced because the winner had been riding ahead of the rest for some time before finishing (e.g. Simon Geschke in stage 17; Vincenzo Nibali in stage 19).


The chart shows how Tour de France excitement developed over time, by language. Overall, exitement was high at the start of the tour, but then it declined, reaching a low on the first rest day. After the rest day, excitement started high again and then declined somewhat. After the second rest day, it started high again, but then it didn’t decline - presumably because the Tour had gotten a bit more interesting (thank you Nairo Quintana and Vincenzo Nibali).

There’s often a peak in excitement in tweets in a particular language on days when the stage is won by a rider speaking that language. For example, there are peaks in German-language tweets on the days Tony Martin and Simon Geschke won the stage. As one might expect, the pattern is less pronounced for languages used by people from many different countries, such as English.

The pattern for French-language tweets is a bit intriguing. Starting around Quatorze Juillet there’s a substantial peak. It doesn’t seem to be associated with specific riders or incidents.

And for some reason, the stage wins by André Greipel don’t appear to have impressed the Germans much.


Using the Twitter API, I collected between 300,000 and 400,000 tweets containing the hashtag #tdf. Since the initial version of the analysis, I’ve made a few changes. For one thing, I excluded retweets, which seemed to yield somewhat more robust results. Also, I filtered out some silly marketing accounts (this and this). The latter adaptation didn’t make much difference for the total number of tweets included but did make a difference for the pattern on specific days.


Blijkt uit cijfers van de BOVAG dat de aanpak van snorscooters werkt?

De aanpak van snorfietsen werkt. Tenminste, dat concludeert de BOVAG. De Telegraaf schrijft erover:

De keiharde aanpak van snorscooters die veel te snel over het fietspad racen werkt. Er worden steeds minder snorscooters verkocht, terwijl bromfietsen na een jarenlange daling weer in populariteit stijgen.

De BOVAG is blij met dit soort nieuws. In mei wisten ze een vergelijkbaar bericht in het NOS-journaal te krijgen. Aan hun leden schreven ze:

De grote media-aandacht voor deze ontwikkelingen kan eraan bijdragen dat snorfietsers niet worden geconfronteerd met draconische maatregelen, zoals verplaatsing naar de rijbaan en een helmplicht.

Maar klopt de interpretatie van de BOVAG? Zeker, in mei werden er 15% minder nieuwe snorfietsen geregistreerd dan in dezelfde maand vorig jaar. Maar dat is niet het hele verhaal.

In de eerste plaats gaat het vooralsnog om een tijdelijk effect: in juni lag het aantal registraties alweer op het niveau van de voorgaande jaren. Het is een beetje hachelijk om je beleid te baseren op een effect dat alweer uitgewerkt lijkt te zijn.

Bovendien betekent een daling van het aantal registraties niet dat er minder snorfietsen zijn; hoogstens dat de groei even iets minder hard gaat. Maar de groei gaat gewoon door, zoals het aantal snorfietsen al jaren groeit. Landelijk van 292 duizend in 2007 naar 605 duizend in 2015; in Amsterdam van 8 duizend naar 32 duizend (CBS).

Update - BOVAG meldt dat ik «vergeten» ben rekening te houden met het feit dat er in de cijfers voor de eerste zes maanden van 2015 1.781 zogenaamde speed pedelecs zitten (iets waar hun eigen overzicht overigens geen melding van maakt). Maakt dat wat uit?

In de eerste plaats verandert dit natuurlijk niets aan het feit dat er maandelijks duizenden snorfietsen bijkomen (ik zag dat ik niet de eerste ben die daar op wijst).

BOVAG verschaft geen gegevens over het aantal speed pedelecs per maand. Ook is niet bekend hoeveel er vorig jaar in de cijfers zaten, al zullen het er toen minder zijn geweest. Kortom, het is gissen wat de ontwikkeling van het aantal snorfietsen exclusief speed pedelecs zou zijn geweest. Wat ik wel weet is dat het bijna uitgesloten is dat je de stijging van vorige maand (juni) hiermee weg zou kunnen verklaren.

Al met al verandert er niet zoveel aan de conclusie. Het zou goed kunnen dat strengere controles hebben gezorgd voor een tijdelijk dipje in de scooterverkoop, maar op basis van de beschikbare gegevens kan je niet concluderen dat er sprake is van een duurzaam effect.

Update 2 - Bij BOVAG gaan ze de discussie in ieder geval niet uit de weg. In een vervolgtweet suggereren ze dat ik het citaat van hun website uit zijn verband heb gerukt door de slotzin weg te laten: «Draag deze boodschap uit richting uw klanten, wijs hen op de risico’s en op hun eigen verantwoordelijkheid, en bied aan om de snorfiets technisch in orde te maken».

Ik heb het BOVAG-artikel geciteerd om te laten zien waarom BOVAG blij is met berichtgeving over dalende registratiecijfers voor snorfietsen. Dat ze hun achterban voorlichten is mooi, maar staat een beetje los van het onderwerp van mijn artikel (het gebruik en de interpretatie van de cijfers). Overigens heb ik wel gelinkt naar het BOVAG-artikel zodat iedereen kan opzoeken wat de context van het citaat is.


Apparently, it’s still possible to fool Google

Researchers have found that men are almost six times more likely than women to be shown ads on news websites for a career coaching service for $200k+ executive positions. The findings suggest some of the algorithms involved in tracking internet users have discriminatory outcomes. They might lead to «deeper investigations by either the companies themselves or by regulatory bodies», the authors add (via WP).

Not just the findings are interesting, but so is the research method. The researchers created AdFisher, basically a smart web scraper built with Python. AdFisher can create large numbers of «agents», have them visit certain websites or alter their profile via the Google Ad Settings, and then see what ads it gets shown on websites like the Times of India or the Guardian. Further, it will organise these activities in such a way that experimental and control conditions can be compared, and it will even analyse the results, using machine learning to figure out what may have triggered differences in what ads are shown.

Somehow this reminded me of the patent Apple (!) obtained for a cloning service to fool the companies that are tracking you. The service would mimick some of your normal online behaviour, but also do other stuff, such as faking an interest in basket weaving. This way it would contaminate the profile these companies keep of you, perhaps to the point of making it useless.

So would you be able to get away with that? If you open a bunch of browser windows with Google searches, Google will ask you to fill out a captcha to make sure you’re human («Our systems have detected unusual traffic from your computer network. This page checks to see if it’s really you sending the requests, and not a robot»). This is a very simple example, but given the fast-developing ability to analyse patterns in online behaviour, you’d expect that companies like Google and Facebook would have become eerily accurate at identifying (real) internet users and telling them from bots.

Against that background, it’s somehow reassuring that it’s apparently still possible to fool Google by creating a fake profile.

P.s. I’ve never been shown ads for a career coaching service for $200k+ executive positions, but if they do turn up I’ll just tell Google I’m a woman.


Tweeting #oxi

The responses of European leaders to the outcome of last Sunday’s referendum in Greece were pretty unanimous. Germany’s vice-chancellor Sigmar Gabriel (a social-democrat) said Tsipras had torn down the bridges between Greece and the rest of Europe. Spanish PM Mariano Rajoy said Greece must follow Europe’s rules. And Dutch PM Mark Rutte somewhat pedantically said he was «really angry» about the referendum and that the Greeks better not come up with a «lame story» (flutverhaal).

For a different perspective, I turned to Twitter. The hashtag #oxi, associated with a ‘no’ vote in the referendum, has become a bit of a symbol of opposition to EU-imposed austerity. I collected some 110,000 tweets containing #oxi (and not #nai) from around last Sunday. The #oxi tweets that are geotagged are shown on the map. It appears that quite a few tweets came from Spain and Italy, but also from the UK and Ireland, and - who’d have thought - the Netherlands (one wonders if bar de Druif in Amsterdam is an #oxi stronghold). In Spain, #oxi territory seems to overlap with areas where progressive party Podemos won in the mayoral elections earlier this year.

Note that only a small proportion of tweets are geotagged, so one shouldn’t rush to conclusions based just on the map. An alternative approach is to look at the language of tweets.

To interpret these findings properly one should take various factors into account, including the number of people who speak a language and how many are on Twitter. But whichever way you look at it, the number of Spanish-language #oxi tweets is impressive. There may well be a connection with the popularity of Podemos.

To get an idea of the contents of the #oxi-tweets I looked up the most-favourited tweets in some of the key languages. A few examples:

The joy of losing fear. Long live Greece! (es)

Today I’m going to eat a Greek tortilla. And what’s that? The same as the Spanish one, but with more huevos [eggs / balls]. (es)

Threats. Blackmail. Fear. Propaganda. The courageous Greek people defied it all. But now they desperately need our help. (en)

Tonight I feel truly European. As if Greece had voted for me against the technocrats and austerity. (fr)

A small, proud nation can change Europe. We should help them (it)

What if we take #oxi as an opportunity to rigorously curtail the world of banks, speculators and finance across the EU? (de)


I searched the Twitter api for tweets using the search terms #oxi and #nai. I analysed tweets containing either #oxi or #nai (not both). Some have argued that ochi would be more appropriate than oxi; in French sometimes oki is used and of course the Greeks have their own alphabet. That said, #oxi appears to be a pretty universal symbol for a no vote in the Greek referendum and for opposition to austerity.

The number of #nai tweets was very small (less than two thousand). Locations of tweets were derived from the location data provided by the Twitter api. As indicated, only a small number of tweets contain this information; further, there may be cultural differences in the extend to which people allow their device to send location data with their tweets. Twitter also provides language data which appears to be pretty accurate (although they occasionally mistake Catalan for French). Note that language data cannot be simply linked to countries: for example, quite a few tweets in Dutch will be from Belgium while on the other hand, Dutch twitterers frequently tweet in English.

I used Python to collect and process the data, R for analysis and d3.js and Leaflet for visualisation.

Rijwiel, fiets or machine: Dutch words for bicycle

A new book on the early history of Dutch bicycles briefly discusses the words that were used to refer to them:

The government spoke of a rijwiel [riding wheel], manufacturers and salespersons of «machines» and «safeties» and among users, the typical Dutch word fiets (or sometimes viets) gained popularity at some point.

I especially like the term machine, which has a nice futuristic ring to it (I wanted to use this image to illustrate my point, but it’s copyrighted until 2027). The term machine also links to the emancipatory role of the bicycle, which was described as the «freedom machine» by an American feminist.

So when did the Dutch call bicycles machines? In his book Fiets!, Ewoud Sanders indicates the word machine had been popular at least since 1880 (he adds that people sometimes spoke of toer- and renmachines, which makes the term machine sound even cooler). He quotes a linguist who wrote in 1911:

Anyone above age 35, has said or heard being said: vélocipède, vélo, safety, bicycle, rijwiel, kar, machine, fiets, and knows more or less the context in which these words are or were used. But who could even roughly date them, without much reading up and asking around?

To find out more, I turned to the Delpher historical archive of Dutch newspapers. I first looked for occurences of velocipede, rijwiel and fiets in newspaper articles (this method obviously has limitations which I discuss in the Method section below).

The term velocipede was used in the second half of the 19th century. Both fiets and rijwiel became more popular after the 1880s. The term rijwiel fell into disuse after the war, whereas use of the term fiets kept growing.

I also looked into occurences of these terms in ads, which resulted in a slightly different pattern.

The term rijwiel was used more often and stayed in use longer in ads than in articles. This may have something to do with the fact that many bicycle manufacturers had the term rijwiel in their brand name. Second, there’s an intriguing spike in the use of the terms rijwiel and especially fiets in the 1940s. Something to do with the war?

A closer look reveals that the spike occured during the first years of the German occupation. A possible explanation can be found in Pete Jordan’s book on Dutch cycling history. He explains that the German occupiers immediately plundered petrol supplies. The result: «Within days after the capitulation, motorists switched to the bicycle in such high numbers that bicycle shops saw their sales more than double, and orders for bicycle manufacturers rose to ‘fantastic heights’».

But let’s return to the term machine. Obviously, there’s no point in simply counting occurances of this term in articles and ads, for in most cases it will have nothing to do with bicycles. To solve this problem, I looked at ads and articles containing the term machine in combination with either rijwiel or fiets. Not a perfect solution, but let’s see what happens.

The results suggest that use of the term machine to refer to a bicycle peaked in the 1890s in articles and in the 1900s for ads (I suspect the slight rise in the 1960s is due to noise). An early example is an ad from 1885:

Offered for sale, at a moderate price, one well-maintained English Vélocipède (Royal Bicycle), used for only 5 months, front wheel 1.24 Meter. Owner has acquired a lower Machine.

Today, the term machine is often associated with devices that run on fuel or electricity (in fact, the Italian woord macchina simply means car). But there’s no reason why this should remain so. The Dutch Wikipedia defines a machine as «a device consisting of a frame, a drive mechanism and other specific parts. It’s a mechanism that can convert a form of movement or energy into another form of movement or energy».

That sounds pretty much like a description of a bicycle. Cyclists, let’s reclaim the term machine. Not instead of fiets, of course, but as an addition.


I used the Delpher website to search the database of Dutch newspapers which is claimed to contain about 10% of publications between 1618 and 1995. Selection criteria for the database may affect the results. In addition, scanned newspapers have been OCR’ed and this is not entirely error-free. I tried to identify publications which use the term machine to refer to bicycles by searching for co-occurances of machine and either fiets or rijwiel. Obviously this isn’t error-free either; for example, a shop may advertise both bicycles and sewing machines. Initially there was a strange peak in combinations of rijwiel and machine in the 1960s which turned out to be due to unrelated mentions of the terms in a section called Beurs van Amsterdam so I filtered those out.

The books referred to are:
Kaspar Hanenbergh en Michiel Röben (2015), Ons stalen ros: Nederland wordt een land van fietsers. 1820 tot 1920. Uitgeverij Ons stalen ros.
Pete Jordan (2013), De fietsrepubiek. Podium.
Ewoud Sanders (1997), Fiets! De geschiedenis van een vulgair jongenswoord. Sdu / Standaard Uitgeverij.

Update - Reinier Asscheman asked about the term tweewieler (two-wheeler). This showed a small peak in the 1880s (ads) and 1890s (articles) and started to rise again in the second half of the 20th century. At its highest level, there was still only one mention for every seven mentions of fiets in ads and one in thirty for articles.