Tuesday, May 11, 2021
BrainFuck program for stage0 seeds
In the past week, I created a BrainFuck generator, which generates BrainFuck from a relatively simple programming language, for the purpose of writing a BrainFuck program that could replace the seeds for the oriansj/stage0 github repository. This BrainFuck basically implements the command line:sed 's/[;#].*$//g' $input_file | xxd -r -p > $output_filementioned in the README.md. I extended my BrainFuck interpreter / code generator with the -eof0 command line flag for returning 0 when end of the input has been reached. I used the following two command to test it:
./bf -eof0 hex0.bf <~/git/bootstrap-seeds/POSIX/x86/hex0_x86.hex0 >hex0-seed cmp -l hex0-seed ~/git/bootstrap-seeds/POSIX/x86/hex0-seedWhere hex0.bf contains the BrainFuck program generated by the BrainFuck generator page. I know that it does really solve the basic problem of generating a binary without using a binary that one cannot trust 100%, but it does give an alternative for the sed and xxd commands or using. Even using a BrainFuck interpreter written in BrainFuck of a BrainFuck to x86 executable compiler written in BrainFuck cannot solve the fundamental trust problem.
Fixing carriage-return line-feed
This evening, I noticed that some of my files of my website contained other line ending than the windows carriage-return line-feed combination. I wrote a small program to fix these. With no command line switches it just reports the number of errors in each file that contains incorrect line endings. The command line switch -l will also report the lines and the combination with which that line ends. The command line switch -f will fix all files mentioned on the command line. This does require a directory org to be present, to which the original files will be copied.Monday, May 10, 2021
Link
Sunday, May 9, 2021
Suddenly warm
Yesterday morning, the temperature dropped as low as -1.3° Celsius (and -5.7° Celsius at ground) level at Twenthe Airport, while today it went up to 27.0° in the afternoon and remained above the 20° for the rest of the day, while in the past weeks the temperature but barely reachted 15°. This might also explain why today, there were still some flowers on our magnolia. In the afternoon, Conny and I went walking in a forrest just North of the area where we walked last Sunday. On the way there, we stopped near triangulation point 340331 and around 12:52 we succeeded in locating it just below the surface. After having it photographed we covered it up again. When we rested on a bench during our walk, we say a bee hover over the ground and than started to dig at some spot. After about a minute it had disappeared under the ground not leaving any visible hole. At home we created a small kitchen garden from the germinating seeds that Conny got from the supermarket Albert Heijn as MoestuinMaatjes.Link
Tuesday, May 4, 2021
Using Brainfuck for bootstrap
Yesterday, I came along a mentioning of live-bootstrap on Hacker News, which also lead me to the Bootstrappable builds website. I also found the Bootstrappable builds article on LWN.net. This morning, I spend some time digging into it, trying to understand the live-bootstrap GitHub reposity. I actually ended up looking at the oriansj/stage0 GitHub reposity in particular the hex0_x86.hex0. I am a little surprised that it is opening files as specified on the command line, instead of simply reading from the standard input and writing to standard output. According to bootstrap-seeds README.md file, one way to process the hex0_x86.hex0 file is using the sed UNIX utility to remove the comments and the xxd UNIX utility to change the hexadecimal pairs to raw bytes. However, the README.md also states (in capitals): "NEVER TRUST ANYTHING IN HERE". Because, yes, you have to trust the implementation of sed and xxd. I wondered whether Brainfuck could be of any help in this area or as a basis for a bootstrap environment, because it is a very simple language for which a large number of interpreters in many different languages have been implemented. I even spend some time working on a C program to generate Brainfuck code as I did before for the Brainfuck interpreter written in Brainfuck.Addition May 6: In a comment on Hacker News, Kragen Javier Sitaker writes that he did consider Brainfuck for a Universal virtual computer (UVC) and bootstrapping a compiler from a tiny compiler using Brainfuck.
Sunday, May 2, 2021
Koffiegoot
Conny and I went walking this afternoon. But before we started on our walking trip, we first visited a marke stone and looked for a triangulation point. At 13:36, we visited the marke stone Steenen paal bij de Nijkerksdijk. It actually is not a marke stone, because the border of the two marke Haaksbergen en Honesch and Langelo runs to the East of it, but a stone that marked the border between the border of the provinces Overijssel and Gelderland as it was established in 1775. At 13:42, we searched for triangulation point 340331. On the side of the road we found a sign to locate the triangulation point stating the number 340 (on top) and 331 (on the bottom) with the number 2,7 in the middle. The 2,7 seems to indicate the distance to the actual triangulation point. It was not clear in which direction we should search. At home, I concluded from some pictures, which I only noticed then, that the point was close to the road and just below the surface. We probably have stood on top of it. The first part of our walking trip was along a brook called Koffiegoot (coffee gutter), which probably was called like this because of its brown colour. At about 14:10, we found the marke stone Hellekampspaal, which is rather large, about 170 cm tall and measuring 48 by 48 cm wide, and also served to mark the border between the two provinces and was placed there after a long running border conflict between 1611 and 1775.Friday, April 30, 2021
Link
Wednesday, April 28, 2021
End of curfew
Today, the curfew ends together with some other restrictions, one of them being that you no longer have to make reservation to visit a non-essential shop. On my way home, I visited charity shop Het Goed. There were some long queues before some large retail shops this afternoon. It is a little strange that some restrictions are lifted, because there is no real sign that the number of infected people is dropping. The number of people at the ICU of hospitals in the Netherlands remains high. We are still at the top op the fourth wave were the numbers are higher than during the second and third wave. Yesterday was Koningsdag (King's Day) and many people went to the city centers to celebrate it. It seems that people no longer view COVID-19 as a serious problem and are acting like it.Saturday, April 24, 2021
Witte Veen
This morning Conny and I went walking in Het Witte Veen (The White Peat), a peat and forest area near the German border. At, 9:01, we visited the triangulation point 340204, which is on a stone. At 10:18, we visited triangulation point 340329, which is a bolt on a small building. At 10:35, we visited triangulation point 359306. Not sure if we actually saw it, because some note seems to suggest it is located underground.Thursday, April 22, 2021
GOGBOT café, яєαℓιту ιѕ ѕуηтнєтι¢
I watched this GOGBOT cafeé on YouTube. It has talks by:- Sander Duivestein about deepfales and synthetic media.
- Iris Withuis about Real Fake / The influence of virtual influencers.
- Askold Kurov about Welcome to Chechnya.
Tuesday, April 20, 2021
Groot Peter
Conny and I walked to the shopping center called Winkelcentrum Zuid to visit some shop and way home we went to see the marke stone Steen bij Groot Peter (33.34-05). This a replica that is placed on September 16, 2016 in the North East corner of the intersection of Wesselerbrinklaan (runing North-South), Het Bijvank (on the West). and Het lang (on the East). The original marke stone was in the middle of Wesselerbrinklaan just South of the intersection . Behind the stone (as seen from the center of intersection) there are a number of oak trees that belonged to a farm called Groot Peter. The kadaster of 1832 shows a farm called Peters at that location. Often when a son of farmer would start a new farm near the farm of his father, the names of the farms were prefixed with the words Groot (big) for the original farm and Klein (small) for the new farm. This is probably how the name Groot Peter came into existence. The adjective Klein and Groot are also found in surnames from people coming from this region, because for a long time people were named after the farm where they came from instead of with a surname or patronym.Links
- Kyteman - Root Canal Therapy
- Colin Benders - Reunion - #Eurorack Jamsession
- Colin Benders X Boris Acket | Residency | Tivoli Vredenburg
- Mysteryland 2020 | Colin Benders Live Modular Techno Set | Modular Mayhem
Monday, April 19, 2021
Hexagon puzzel
I gave Conny a small wooden puzzle as a present. In the Netherlands it is sold under the name Hexagon puzzel. It is produced in China, Yunhe County, Lishui City. The product model is XC-808 with the name Interesting Changeful Puzzle. She wanted to know how many solutions it had. After some wrong tries (including not realizing that the pieces could not be used upside down), I arrived at the conclusion that there are 4702 solutions excluding solutions that are the same with respect to rotation of the whole solution. I took the btc24_to_ec.cpp program, which I developed for the Beat the Computer No. 24 puzzle, and made it into a more generic program for puzzles on a hexagon grid (with either hexagons or triangles), calling it hexagon_to_ec.cpp. The command line arguments 1 --flat need to be used for the Hexagon puzzel. The output is fed into an exact cover solver and postprocessed with a small program that makes a letter string of each solution. Below all the solutions are shown in a random manner or can be traversed using the buttons. (The 'a' piece is always placed in the same orientation.)
|
|
Sunday, April 18, 2021
In full bloom
Today, finally, all the flowers of our magnolia opened. It is almost three weeks ago that the first flowers opened. The temperature has been quite low for the time of the year in recent weeks, even dropping below zero degrees Celsius, causing some of the flowers to turn brown. But this afternoon all the flowers had opened and you could smell the nice scent of the magnolia in the garden.Saturday, April 17, 2021
Snippert
Conny and I went walking through the adjecent forests Snippert and Het Haagse Bos. At one point we came to a location where the path was submerged in water and we wondered whether there was a way to get to the other side when there appeared a woman on the other side, who wanted to get to our side. After some searching, we found signs of a small path leading around the area, which seemed to have been used for some time, because at one place there were three wooden trunks laid side by side over a water filled ditch. During our trip we found two two marke stones:- At 14:43, Steen tegen het ossenschot (20.32-01).
- At 15:25, Steen aan 't Haagerbrook (20.32-02).
Thursday, April 15, 2021
Genealogy
I have been thinking about genealogy the last days. Yesterday evening, I spend some time adding some data to WikiTree based on some data from the Dutch resource site: WieWasWie. First of all, I realized that the WikiTree is American centered because, for example, the concept of a middle name is not universal. Here in the Netherlands, we do not have the concept of a middle name. Someone can have one or more firstnames, where having four firstnames is not rare. (Not to talk about a generation name in some Asian countries.) The second thing I realized is that although it is possible to add sources, it is not a source based approach. During the night, I thought about an algorithmic approach to genealogy, where the 'family tree' is constructed based on a collection of records. This morning, I spend some time transcribing the marriage certificate of Leonardus van Bragt and Margaretha Dorothea Faase. The result of this is:-
Heden den Zevenden Mei ACHTTIENHONDERD
NEGENTIG, verschenen voor ons, Ambtenaar van den burgerlijken stand der gemeente
Schoten in het huis derzelve gemeente,
ten einde een huwelijk aan te gaan:
Leonardus van Bragt
oud drie en twintig jaren, van beroep bloemkweeker
geboren te Velsen en wonende te Velsen,
meerderjarig zoon van Willem van Bragt van beroep
bloemkweeker en van Maria Elisabeth Alders, zonder
beroep, beiden wonende te Velsen, ter Eene
en Margaretha Dorothea Faase
oud vier en twintig jaren, van bezoep zonder,
geboren te Bloemendaal en wondende te Schoten,
meerderjarige dochter van Jacob Faase, van beroep
bloemkweeker, wonende te Schoten en van Petronella
Catharina van der Veld, overleden, ter Andere Zijde,
van welk huwelijk de vereischte afkondiging zonder stuiting hebben plaats gehad op
Zondagen, den Twingtigsten en Zeven en twintigsten April deze jaar
binnen deze gemeente en binnen de gemeente Velsen
Door de aanstaande echtgenooten zijn ons overgelegd:
1º Het bewijs, dat de huwelijksafkondigingen zonder Sluiting
zijn afgeloopen te Velsen, 2º hunne geboorte-akten,
3º de akte van overlijden der moeder van de bruid, 4º het
bewijs dat de bruidegom heeft voldaan aan de wet op
de Nationale Militie.
Voorts verklaarden de ouders van de bruidegom en de
vader van de bruid, die mede voor ons verschenen, hunne
toestemming tot het aangaan van dit huwelijk te geven.
Waarna wij hun in het openbaar hebben afgevraagd, of zij elkander aannemen tot
echtgenooten en getrouw de plichten zullen vervullen, welke door de Wet aan de huwelijken
staat verbonden zijn; hetwelke door hen toestemmend beantwoord zijnde, hebben wij in naam
de Wet verklaard, dat zij door het huwelijk zijn vereenigd. In tegenwoordigheid van:
Franciscus Faase, oud zestig jaren, van beroep bloemkweeker,
wonende te Bloemendaal, oom-; Johannes Franciscus
Faase, oud een en zestig jaren, wonende te Schoten, van be„
roep bloemkweeker, oom-; Cornelis Hendrik van Looij, oud
acht en veertig jaren, van beroep winkelier, wondende te Bloe„
mendaal, behuwdoom van de echtgenoot en Johannes
Alders, oud een en vijftig jaren, van beroep uurwekma„
ker, wonende te Schoten, oom van den echtgenoot.
En is hiervan door ons opgemaakt deze akte, welke na voorlezing, door ons, de com„
paranten en de getuigen is ondertekend.
Where I have marked all the handwritten parts in italics. The ordinal indicator in the certificate has two dashes instead of one in the above transcription. I have used the Double Low-9 Quotation Mark as the two commas that indicate a continuation with the next line. Something that nowadays is usually done with a dash. This already shows some of the problems with transcribing documents. And, so far, no one has verified the transcription. In a distributed information system, you would at least have some mechanism to sign a transcription to indicate who did the transcription and who reviewed it with digital signatures. And there is also the problem that there could be multiple interpretations of the document and that one would like to record variants.
Next, I spend some time thinking about a way to represent all the data from the above marriage certificate in some form of JSON. Most of the information is already included in the page, but, for example, the information about the witnesses is not included. The extra information could be used to calculate the probability that two people mentioned in different records are indeed one and the same person. The information is also not in a computer readable form, although, I expect that it is retrieved from some kind of database with a welldefined structure. It would be nice if there is some kind of REST API to query this database. Below an example of how the information in the marriage certificate could be represented in a way that could be processed by a means of an application that constructs a family tree.
{ "source":{
"Document type":"BS Huwelijk",
"Institution name":"Noord-Hollands Archief",
"Institution place":"Haarlem",
"Collection region":"Noord-Holland",
"Sourcenumber":"3",
"Registration date":"07-05-1890",
"Certificate place":"Schoten"},
"data":{
"marriage":{
"date":"18900507",
"place":"Schoten",
"groom":{
"name":"Leonardus van Bragt",
"profession":"bloemkweeker",
"birthplace":"Velsen",
"livingplace":"Velsen",
"age":"23",
"father":{
"name":"Willem van Bragt",
"profession":"bloemkweeker",
"livingplace":"Velsen"},
"mother":{
"name":"Maria Elisabeth Alders",
"profession":"-",
"livingplace":"Velsen"}},
"bride": {
"name":"Margaretha Dorothea Faase",
"profession":"-",
"birthplace":"Bloemendaal",
"livingplace":"Schoten",
"age":"24",
"father":{
"name":"Jacob Faase",
"profession":"bloemkweeker",
"livingplace":"Schoten"},
"mother":{
"name":"Petronella Catharina van der Veld",
"profession":"-",
"deceased";""}},
"witnesses":[
{
"name":"Franciscus Faase",
"age":"60",
"profession":"bloemkweeker",
"livingplace":"Bloemendaal",
"relationship":{"type":"uncle", "of":"bride"}},
{
"name":"Johannes Franciscus Faase",
"age":"60",
"livingplace":"Schoten",
"profession":"bloemkweeker",
"relationship":{"type":"uncle", "of":"bride"}},
{
"name":"Cornelis Hendrik van Looij",
"age":"48",
"profession":"winkelier",
"livingplace":"Bloemendaal"
"relation":{"type":"uncle through marriage", "of":"groom"}},
{
"name":"Johannes Alders",
"age":"51",
"profession":"uurwekmaker",
"livingplace":"Schoten",
"relation":{"type":"uncle", "of":"groom"}}]}}
"signed":[
{"person":"Franciscus Johannes Faase", "signature":"bd078fe...."}]}
As is often the case with a more formal approach to solving a problem is that it requires a lot of more effort (and a different kind of mindset) than a more informal approach. Also it requires a greater attention to details of the people involved. And we should not forget that reality is always more complex than one thinks at first and that it is thus very difficult to develop a formalism which covers all cases.
Wednesday, April 14, 2021
Links
- Maslow's pyramid of code review
- Dependency Graph of Propositions in Euclid's Elements
- Euclid's Elements
Tuesday, April 13, 2021
Link
Sunday, April 11, 2021
Het Haagse Bos
Because last week, the German governments demands all negative test for COVID-19 for people crossing the border from the Netherlands, Conny and I decided to halt our search for border poles and instead search for some marke stones and go for a walk through the forests known as Het Haagse Bos where a marke stone can be found. We found the following marke stones:- At 12:34, Steen aan de kamp in het veld (19.33-02).
- At 12:44, Steen op het Wachtveld (19.32-06).
- At 12:47, Steen bij Hendrik Beumer (19.32-01).
- At 13:24, Snippertsteen (19.20.32)
Wednesday, April 8, 2021
A096004
After I came home from work, I worked on a program (at first based on hexagonnum.cpp) to calculate the number of convex polygons that can be constructed from a given number of equal sized, equilateral triangles. When it did produce a sequence that looked correct, I entered it in On-Line Encyclopedia of Integer Sequences. To my surprise, it did not return any results. Instead of suggesting it as a new sequence, I took the advice to first search the encyclopedia. I came along sequence A096004: Number of convex triangular polyominoes containing n cells, which looked like describing what I had calculated. When I compared it with my own sequence, I discovered that the value for fifteen triangles/cells was different. After some debugging, I discovered a bug in my program and after fixing it, it returned the same sequence.Wednesday, April 7, 2021
Snow
There was about 1cm of snow on my car this morning. The roofs of houses, plants and some place on the ground were also covered with snow. In the past days, there were a lot of winter showers with snow and hail alternating with sunshine, several times even at the same time. On Monday evening and Tuesday morning, there was also some snow on roofs and some other areas. Todaym we also had some rain showers, but later in the afternoon, again some hail.Monday, April 5, 2021
Voorthuizen
During the Easter weekend, Conny and I went to Voorthuizen where we rented hotel room at holiday part De Boshoek. We arrived on Friday after we had searched for 'border stones' of the city Apeldoorn. In 1750 the border of the municipality of Apeldoorn was established with wooden poles. Around 1794 the 26 wooden poles were replaced with stone poles. Most of these stone poles have since been removed. Along the entry roads, stone name signs of the city have been placed. It is these name signs, on the East side of the city, we have searched. Conny used the report (in Dutch) Speurtoch naar grenspalen and Google streetview to record the location of these in Google maps. Later, I discovered that some of these poles and also some others are indicated on various topographic maps.On Saturday, we visited the Green Cathedral, which I visited before on August 25, 2016. We also walken around castle Geerestein just North of Woudenberg.
Yesterday, we continued our search for the name signs. This time on the West of the city. We also entered the crown domain Het Loo where we found another name sign just besides one of the original border stones with the letter A and E on opposing sides. We continued walking through the woods, which we felt were rather boring. Today, we decided to go home early and not visit any places on the way home, because rain, wet snow, hail, thunder and strong winds were predicted. We did see some wet snow on the way home.
Thursday, April 1, 2021
De Poppe
Conny and I continued our search for border poles. We walked in an area called De Poppe. We encountered the following border stones:- At 14:04, pole 17.
- At 14:07, pole 17-I.
- At 14:59, pole 17-II.
- At 15:04, pole 18.
- At 15:20, pole 20.
- At 15:41, pole 20-II.
- At 15:46, pole 20-III.
Wednesday, March 31, 2021
24.4° Celsius
The temperature at Twenthe Airport went up to 24.4° Celsius, which breaks the previous record of 23.7° for the temperature on this date in 2017. The predicted temperature according to the prediction at the start of the day, was 21.8° Celsius.Tuesday, March 30, 2021
Link
Monday, March 29, 2021
Idoneal numbers
On February 5, 2019, I wrote about hexagon numbers. Today, I discovered that the numbers I found, are sequence A229757 in the On-Line Encyclopedia of Integer Sequences. With the sequence, there are references to two publications: Reguläre Dreieckpflasterungkonvexer Polygone.and Tiling Convex Polygons with Congruent Equilateral Triangles by Eike Hertel and Christian Richter. The second paper is an extension of the first one with respect to proofs. The numbers for pentagons the numbers are related to the Idoneal numbers. It is not known if there are more idoneal numbers than the one that are known, but there are no more if generalized Riemann hypothesis is true. As a pentagon can often be changed into a hexagon by removing, one would expect that both sequence match with an offset of one. That is indeed the case for many numbers, but not all.First two flowers
The first two flowers of our magnolia have opened today.Sunday, March 28, 2021
Revisiting Coevorden
Conny and I revisited Coevorden to look for border poles we had missed on previous visits after I created a GPX file with waypoints using the ArcGis viewer and downloaded this (through Google Drive) to OsmAnd on my Galaxy Tab 3 tablet. The weather was a bit rainy for most of the time. We had brought some devices to remove weeds and brambles. We encountered the following border poles:- At 13:05, pole 133-I again.
- At 13:10, pole 133-II. Almost submerged.
- At 13:19, pole 134.
- At 13:25, pole 134-I again.
- At 13:30, pole 134-II again.
- At 13:35, pole 135 again.
- At 13:36, pole 135-I. (Plate says 135 with carved I.)
- At 13:39, pole 135-II.
- At 13:44, pole 135-III.
- At 13:45, pole 136.
- At 13:52, poles 137 and 137-I again.
- At 15:15, pole 147-I again.
- At 15:17, poles 147-II (N) and 147-II (D) again.
- At 15:21, pole 148 (D) again.
- At 15:22, pole 148 (N) laying in a ditch.
- At 16:25, pole 152-I (D). Almost submerged.
Saturday, March 27. 2021
Link
Wednesday, March 24. 2021
Link
Tuesday, March 23, 2021
Border changes around Losser
I continued working on a KML file (to be viewed in Google Earth) about the border changes just North of Losser. I spend a long time determining the position of the border poles 8 to 11 using the available data. The positions I have chosen, are inferred from cadaster data from now and 1832, do not perfectly match with the positions I calculated from the positions of the assurance stones and the map by Otto Koolmann. They are about seven meter off. My GPS measurements of the locations are probably not very accurate. The possibility that the border poles have been moved between the two dates, should not be excluded, because it is known that the poles were washed away several times by the meandering river. I do not know if it is possible to get more accurate GPS locations (without having to spend hunderds of Euro's). I found the ArcGis viewer where you can get GPS coordinates on a recent topographic map. It did show a dot at the location of one of the assurance stones. The GPS locations where close to the one that I measured myself. I used the viewer to adjust the GPS coordinates of some of the border stones along the current border. The KML file contains the following folders:- Losserbrugge: Some border poles based on a map just North of border pole 6 where there was a neutral area.
- Koolmann: With the position of the assurances poles and the positions of the border poles based on the map of Otto Koolmann around the start of the twentied century.
- Before 1949: With border poles on the border before April 23.1949.
- 1949 - 1963: With (wooden) border poles for the border from April 23, 1949 till August 1, 21963.
- After 1963: With some border poles placed after August 1, 1963.
- From 1988: With some border poles that were only shown on a map after 1988.
Sunday, March 21, 2021
Verzekeringsstenen
This morning, I was looking through the book Grensgang: een historische reis langs de randen van Overijssel (in Dutch) written by Jan ten Hove (starting from the back). It is about the history of the borders of Overijssel described as a trip along the border. On page 149, I found a reproduction of a drawing by Otto Koolmann about the 'Verzekeringsstenen', which could be translated (from Dutch) to English as 'assurance stones'. They are at a place where a turn of the border with Germany was established in the middle of the river Dinkel. To define it, on each side of the river two poles were placed, where the crossing point of those poles defined the exact position of the turn in the border. However, the river, as most rivers, was meandering, and over time some of the stone poles, (mostly on the German side) were washed away. To overcome this problem, four new stones were placed at a larger distance of the river as kind of assurance. Hence the name of these poles. We realized that the two unmarked stones we found last Thursday might have been two of the four verzekeringsstenen. We also read some account of some unmarked stones on the other side of the river. This afternoon, Conny and I decided to have a closer look and try to determine the location of the stones with GPS. We encountered the following stones and border poles:- At 13:17, unmarked stone on the former Dutch side at approximately 52°17'00.5"N 7°01'30.6"E.
- At 13:40, unmarked stone on the former Germany side at approximately 52;deg;17'01.0"N 7°01'40.2"E.
- At 13:42, unmarked stone on the former Germany side at approximately 52°17'03.3"N 7°01'39.5"E.
- At 13:46, pole 11-II.
- At 13:52, pole 11-III.
- At 13:56, pole 12.
- At 14:25, unmarked stone on the former Dutch side at approximately 52°17'01.7"N 7°01'30.4"E about 15 to 20 meters to the North of the other.
Saturday, March 20, 2021
Parsing workshop
This afternoon, at 14:00 (CET), I gave the workshop A modern approach to parsing programs as part of the CyberSaturdays of TkkrLab. For some unknown reason, the mircophone of my headset was not working, and I had to use a MacBook of someone else. My intention of the workshop was that everyone, no matter their level, would get something out of the workshop. But afterwards, I felt that I had failed this objective. I realized that in the past weeks, I spend a lot of time working on the materials and the programs, instead of thinking about the workshop itself. I think that my previous experience would be enough to just improvise during the workshop, but I am afraid I was blinded by my pride to realize that giving a workshop online is a quite different. I think, I should have taken some time at the start to see the attendees and ask them about their experience level. I also think, I should have started with an example of parsing an expression and evaluating its value and use this as a starting point to explain everything that is needed to parse it. Something, like I start doing at 22:22 of the presentation. Taking the normal priority rules used in numerical expression, I could have explained the use of priorities and how you could use a formal grammar to describe it.And then only explain something about the history of parsing and how I started developing IParse and that I took a very different approach from the traditional way of developing parsers. I remembered, for example, that IParse cannot parse all Context-Free Grammars (CFG), but that this has never been a problem for all the practical grammars that I have encountered in the past 20 years. Apparently, grammars used for programming languages are more limited than the general class of CFG, probably due to what can be easily comprehended by us humans.
Thursday, March 18, 2021
Graafschapspad
Conny and I spend some time walking near Losser, searching for border poles. We ended up walking a part along a walking path called Graafschapspad which was flooded at certain locations probably because of the high water level in the Dinkel river. We encountered the following border poles:- At 11:39, pole 4.
- At 12:25, pole 5-II, which is almost submerged.
- At 12:29, poles 5-III and 6.
- At 13:22, pole 11.
- At 13:30, unmarked border stone.
- At 13:32, unmarked border stone.
- At 14:25, pole 6-I.
- At 14:31, pole 5-I, which is almost submerged.
