Clozemaster is quite addictive

For the past few weeks I've been playing a lot of Clozemaster, an 8-bit style language learning game that is quite simple yet ends up being quite effective thanks to the new text input mode.

The way the site works is this: it takes all the sentences from Tatoeba.org, matches them up from language to language, and then turns it into a game where you fill in the missing word. Multiple choice is extremely easy and I only use that mode when using it on my phone. Otherwise the text input mode is what I go with all the time.

Now what is interesting about Closemaster is that because of its simplicity it works as good practice for synonyms. Sometimes you'll find yourself in a situation where you know the correct answer is let's say fertig, but the word doesn't quite fit. So then you have to start thinking about synonyms...try out bereit, and then it does fit. I can imagine that for some being marked wrong for entering a perfectly okay word would be irritating, but in practice I find it stretches the mind a bit more if you have to think not just about synonyms but also declensions. So sometimes maybe prêt doesn't fit but prêts or prêtes does. This roughness around the edges is part of what makes the game so appealing for me.

Read more...

How to avoid boredom when repeatedly listening to audio

Repetition is very important when learning a language. Repetition is very important. Repetition is very important.

The problem is that your mind doesn't always agree, especially when you are already somewhat proficient in a language. Once you've read and listened to a text once or twice that you completely or mostly understand you are now completely bored by it, and when listening to it again your mind has a strong tendency to wander.

To make things worse, even when repeatedly listening to a text you'll find yourself able to pay attention to the parts you understand, while the parts that are too complicated, too fast, too garbled to understand clearly just zoom by. So upon listening to a text five times for example your mind will have taken note five times of the parts you already understand well, and will more than likely have glossed over the more important parts.

My trick here is the following.

• I cut and paste a short section of text, about twenty or thirty words in length. Then I use Audacity to select the matching audio,
• a short section of text, about twenty or thirty words in length. Then I use Audacity to select the matching audio, and as I go along,
• about twenty or thirty words in length. Then I use Audacity to select the matching audio, and as I go along, I delete the preceding parts,
• words in length. Then I use Audacity to select the matching audio, and as I go along, I delete the preceding parts, and add a bit at a time.
• Then I use Audacity to select the matching audio, and as I go along, I delete the preceding parts, and add a bit at a time. This forces you
• use Audacity to select the matching audio, and as I go along, I delete the preceding parts, and add a bit at a time. This forces you to listen repeatedly,
• to select the matching audio, and as I go along, I delete the preceding parts, and add a bit at a time. This forces you to listen repeatedly, while keeping the content new.
• the matching audio, and as I go along, I delete the preceding parts, and add a bit at a time. This forces you to listen repeatedly, while keeping the content new. If necessary
• and as I go along, I delete the preceding parts, and add a bit at a time. This forces you to listen repeatedly, while keeping the content new. If necessary you can also use
• I delete the preceding parts, and add a bit at a time. This forces you to listen repeatedly, while keeping the content new. If necessary you can also use text-to-speech
• the preceding parts, and add a bit at a time. This forces you to listen repeatedly, while keeping the content new. If necessary you can also use text-to-speech such as
• and add a bit at a time. This forces you to listen repeatedly, while keeping the content new. If necessary you can also use text-to-speech such as in Google Translate,
• a bit at a time. This forces you to listen repeatedly, while keeping the content new. If necessary you can also use text-to-speech such as in Google Translate, if you can
• at a time. This forces you to listen repeatedly, while keeping the content new. If necessary you can also use text-to-speech such as in Google Translate, if you can stand
• This forces you to listen repeatedly, while keeping the content new. If necessary you can also use text-to-speech such as in Google Translate, if you can stand the artificial voice.

So in the example above you've encountered the term "matching audio" eight times before it gets deleted, but while doing so you're listening to something new at the end. When you end up deleting the term you encounter it one more time, and hopefully at that point you know whether you are familiar enough with it to delete it. If it still feels like you won't be able to recall it and use it on your own, then you may want to make a note of it or put it into your Anki deck.

For me this is actually the only way I can stand listening to something over and over again, unless it's content that well and truly interests me - the German audiobook of Demian, Rai's great biography of Wilhelm II, things like that. Otherwise I need to use this trick to keep my mind from wandering. If you're the same type of person then give this method a try.

Read more...

My new favourite Chrome extension: Youtube Playback Speed Control

It's a simple one you can see here, which slows down or speeds up playback on YouTube.

Why is that useful, given that YouTube already has the option to slow down or speed up playback? Well, it's because of this.



The first reason is that the only slower option below standard is 0.5, which is awkwardly slow at times. 0.5 can be good when necessary, but when you understand most of a language and just want to slow things down a little bit it's overkill. This Chrome extension gives the option of 0.75, which is perfect.

The other reason is that speed can be controlled with a single keystroke, instead of having to click on the menu. That lets you slow things down when necessary, and even speed the video up to 2 or 3 times normal speed when going through a section without dialogue.

Read more...

New Readlang.com is even better than before

Readlang.com just got a lot better, with one notable feature in particular: the option to show the translation of a word above it, instead of replacing the word. That makes it look like this:

Having the translation on a different line makes it look a little like a certain book or two of mine.

He also asked whether the font used for the line above is a bit too 'cute', and it would be if that were the only font (I couldn't imagine reading a whole book in that) but as a word here and there it stands out nicely and almost looks like somebody has scribbled in the meaning above on a notebook or something, giving it a very natural look.

Read more...

Armenian Wikipedia, Rosetta, Mormon language learning, Gaza tunnel diggers

I haven't been writing here much recently, simply because I've been so occupied with reading. Maybe two months ago I came across this study showing the effects of massive amounts of reading (1+ million words) and how it compared to time spent abroad, and apparently the general rule is 1 million words equalled about the same improvement as a year overseas. This one here is interesting too, especially the chart a few pages down. That chart shows that the volume of text needed to be exposed to 10,000 words at least once is 600,000+ words. To see them three times each: 3+ million.

So I thought I would read The Wheel of Time (4+ million words) in two languages and see what a difference it makes. And also to truly get a feel for what a million words is like. A million is a lot of words. I'm currently halfway through book 5. I read up to book 8 in English before back in the 1990s, and decided that I would stop reading until the series was completed, which it was two years ago.

I don't like to treat Page F30 as a polyglot blog, given the limitations it places on what one can write about, but this spreadsheet might be interesting to readers.

Words means the number of words I've deleted that day (I use Readlang.com for this), words learning is the number of words registered there, and it goes up as new words are added and others I don't need to review anymore are deleted. Total is the total number of new words I assume have been put in my head since starting this, 1400 is just a number to make the graph look nice, pages is the number of pages read that day, words read is pages multiplied by 350. Then divide that by new words, which gives words per new word. That last column has turned out to be less meaningful than I thought because it depends on my mood whether I click on a word or not. Sometimes I feel like adding a word I kind of know but want to see a few more times anyway, and other times I don't.

On days off I can read 20,000+ words, on work days I can get 10,000 words in, sometimes fewer.

Date	Words	Running total	Words learning	Total		Pages	Words read	New words	Words per new word
July 7	22	201	720	921	1400	61	21350	45	474,44
July 8	25	226	751	977	1400	56	19600	56	350,00
July 9	25	251	792	1043	1400	68	23800	66	360,61
July 10	45	296	772	1068	1400	56	19600	25	784,00
July 11	137	433	652	1085	1400	30	10500	17	617,65
July 12	33	466	664	1130	1400	63	22050	45	490,00
July 13	16	482	670	1152	1400	31	10850	22	493,18
July 14	8	490	693	1183	1400	45	15750	31	508,06
July 15	2	492	715	1207	1400	39, on page 685	13650	24	568,75
July 16	13	505	734	1239	1400	102, on page 39	35700	32	1 115,63
July 17	2	507	737	1244	1400	29, on page 68	10150	5	2 030,00
July 18	110	617	637	1254	1400	21, on page 89	7350	10	735,00
July 19	120	737	535	1272	1400	64, on page 153	22400	18	1 244,44
July 20	55	792	490	1282	1400	41, on page 194	14350	10	1 435,00
July 21	41	833	456	1289	1400	19, on page 213	5795	7	827,86
July 22	39	872	431	1303	1400	17, on page 230	5950	14	425,00
July 23	18	890	418	1308	1400	13, on page 243	4550	5	910,00
July 24	36	926	423	1349	1400	34, on page 277	11900	41	290,24
July 25	35	961	415	1376	1400	29, on page 306	10150	27	375,93
July 26	18	979	450	1429	1400	54, on page 360	18900	53	356,60
July 27	24	1003	449	1452	1400	40, on page 400	14000	23	608,70
July 28	24	1027	448	1475	1400	40, on page 440	14000	23	608,70
July 29	8	1035	449	1484	1400	25, on page 465	8750	9	972,22
July 30	36	1071	453	1524	1400	47, on page 512	16450	40	411,25
July 31	52	1123	453	1576	1400	82, on page 594	28700	52	551,92

So let's let to the other items in the title.

Armenian Wikipedia: I remember being disappointed at its tiny size a year or two ago, and apparently it has exploded since then due to a national campaign to have everyone write one article each. Here's one video encouraging Armenians to contribute:





Rosetta: the mission, not the language learning software. I an extremely pleased with them releasing daily images of the comet as Rosetta approaches. Dawn had better do something similar when it approaches Ceres in the next few months, as the lack of images during the Vesta approach was simply unacceptable.

• CometWatch 30 July
• CometWatch 29 July
• CometWatch 28 July
• CometWatch 25-27 July
• CometWatch - 24 July

Mormons: NPR has an article on the techniques they use to learn languages. I've always been impressed by the way they learn languages, since the first time I ever spoke with a non-Japanese person only in Japanese with a Mormon from Brazil in Nagoya.

Gaza tunnel diggers: This documentary with French subtitles is one of the more interesting documentaries I've seen in a while. It is a few years old, and is just 50 minutes of following a few of these tunnel diggers around as they do their thing. It has music but no extra commentary, which is great. A documentary without the typical dozens and dozens of quips from specialists speaking on the matter in their office is a rare treat.

Read more...

Germany takes a 2-day rest, simple yet interesting tool for forcing repetition

Germany in 1912 is taking a rest for two days. I'm about a third of the way through the images I have.

Why Germany in 1912? Well, the book where the images come from was published exactly then (next volume was 1913), and there's something about books published just before the war that draw me. They are a final snapshot of a Europe that turned into something completely different in the space of a few years, and the last few years hold something akin to nostalgia even though I was never there.

And now to the "tool for forcing repetition". I'm not sure how else to put it. The tool is here:

http://ikindalikelanguages.com/re/

It's a very simple idea I had that I proposed to our friend at ikindalikelanguages and interlinearbooks.com who then made it happen (may still be a few bugs). I was listening to something the other day that mentioned a study claiming that one needs to encounter a new word thirty times in order to truly learn it. I haven't been able to find the study but it probably refers to encountering a word thirty times in differing situations. However, the second-best thing may be to force oneself to encounter it thirty times however one can.

So how does one do that? Well, you could find a book and read it thirty times. Or you could read one sentences thirty times and go on to the next one. The former is a Herculean task, the latter is probably not possible due to the utter tedium of it all, the inability to stop one's mind from wandering while doing it.

So perhaps if one wants to truly repeat something thirty (or ten, or five, or what have you) times, the best way to do it is the following. Let's take the first paragraph from this article for example and pretend we want to read each sentence a number of times. For the sake of brevity we will do one paragraph five times per sentence instead of the whole article, thirty times per sentence.

Well, first we start with the first sentence.

Poverty Point ist ein archäologischer Fundort im Nordosten des US-Bundesstaates Louisiana.

Then the first and the second.

Poverty Point ist ein archäologischer Fundort im Nordosten des US-Bundesstaates Louisiana.
Das Gelände auf einer Hangkante über der Talebene des Mississippis birgt in Größe und Komplexität einzigartige Erdwerke einer präkolumbischen Indianerkultur, die auf das 18. bis 10. Jahrhundert v. Chr. datiert werden.

Then the first three.

Poverty Point ist ein archäologischer Fundort im Nordosten des US-Bundesstaates Louisiana.
Das Gelände auf einer Hangkante über der Talebene des Mississippis birgt in Größe und Komplexität einzigartige Erdwerke einer präkolumbischen Indianerkultur, die auf das 18. bis 10. Jahrhundert v. Chr. datiert werden. Ihre Erbauer gehörten zu einer Jäger-, Sammler- und Fischer-Kultur, die bereits einfache Keramiken herstellte und das Material für ihre steinernen Werkzeuge über Entfernungen von zum Teil über 2000 Kilometern bezog.

Then the first four.

Poverty Point ist ein archäologischer Fundort im Nordosten des US-Bundesstaates Louisiana.
Das Gelände auf einer Hangkante über der Talebene des Mississippis birgt in Größe und Komplexität einzigartige Erdwerke einer präkolumbischen Indianerkultur, die auf das 18. bis 10. Jahrhundert v. Chr. datiert werden. Ihre Erbauer gehörten zu einer Jäger-, Sammler- und Fischer-Kultur, die bereits einfache Keramiken herstellte und das Material für ihre steinernen Werkzeuge über Entfernungen von zum Teil über 2000 Kilometern bezog. Bei der 38. Tagung des UNESCO-Welterbekomitees vom 15. bis 25. Juni 2014 soll Poverty Point als Welterbestätte ausgewiesen werden.

Then the first five.

Poverty Point ist ein archäologischer Fundort im Nordosten des US-Bundesstaates Louisiana.
Das Gelände auf einer Hangkante über der Talebene des Mississippis birgt in Größe und Komplexität einzigartige Erdwerke einer präkolumbischen Indianerkultur, die auf das 18. bis 10. Jahrhundert v. Chr. datiert werden. Ihre Erbauer gehörten zu einer Jäger-, Sammler- und Fischer-Kultur, die bereits einfache Keramiken herstellte und das Material für ihre steinernen Werkzeuge über Entfernungen von zum Teil über 2000 Kilometern bezog. Bei der 38. Tagung des UNESCO-Welterbekomitees vom 15. bis 25. Juni 2014 soll Poverty Point als Welterbestätte ausgewiesen werden. Die Begründung lautet, dass Poverty Point ein herausragendes Bauwerk einer Jäger- und Sammler-Kultur sei, die größte Siedlung im Nordamerika seiner Zeit und möglicherweise „die größte Siedlung von Jägern und Sammlern aller Zeiten“.

Now we've seen the first sentence five times so it drops away. Next would be sentences two to six (then three to seven, then four to eight, etc.) but the text is only five sentences long so the next one is two to five.

Das Gelände auf einer Hangkante über der Talebene des Mississippis birgt in Größe und Komplexität einzigartige Erdwerke einer präkolumbischen Indianerkultur, die auf das 18. bis 10. Jahrhundert v. Chr. datiert werden. Ihre Erbauer gehörten zu einer Jäger-, Sammler- und Fischer-Kultur, die bereits einfache Keramiken herstellte und das Material für ihre steinernen Werkzeuge über Entfernungen von zum Teil über 2000 Kilometern bezog. Bei der 38. Tagung des UNESCO-Welterbekomitees vom 15. bis 25. Juni 2014 soll Poverty Point als Welterbestätte ausgewiesen werden. Die Begründung lautet, dass Poverty Point ein herausragendes Bauwerk einer Jäger- und Sammler-Kultur sei, die größte Siedlung im Nordamerika seiner Zeit und möglicherweise „die größte Siedlung von Jägern und Sammlern aller Zeiten“.

Then three to five:

Ihre Erbauer gehörten zu einer Jäger-, Sammler- und Fischer-Kultur, die bereits einfache Keramiken herstellte und das Material für ihre steinernen Werkzeuge über Entfernungen von zum Teil über 2000 Kilometern bezog. Bei der 38. Tagung des UNESCO-Welterbekomitees vom 15. bis 25. Juni 2014 soll Poverty Point als Welterbestätte ausgewiesen werden. Die Begründung lautet, dass Poverty Point ein herausragendes Bauwerk einer Jäger- und Sammler-Kultur sei, die größte Siedlung im Nordamerika seiner Zeit und möglicherweise „die größte Siedlung von Jägern und Sammlern aller Zeiten“.

Then four to five:

Bei der 38. Tagung des UNESCO-Welterbekomitees vom 15. bis 25. Juni 2014 soll Poverty Point als Welterbestätte ausgewiesen werden. Die Begründung lautet, dass Poverty Point ein herausragendes Bauwerk einer Jäger- und Sammler-Kultur sei, die größte Siedlung im Nordamerika seiner Zeit und möglicherweise „die größte Siedlung von Jägern und Sammlern aller Zeiten“.

Then just five:

Die Begründung lautet, dass Poverty Point ein herausragendes Bauwerk einer Jäger- und Sammler-Kultur sei, die größte Siedlung im Nordamerika seiner Zeit und möglicherweise „die größte Siedlung von Jägern und Sammlern aller Zeiten“.

And that's what the tool does. You decide how many times you want to repeat the text, paste the text in, then click and read the first sentence. Then click and read it again. Then click and read it again. It's a nice way to force yourself to repeat what you are reading, while still retaining that sense of accomplishment in getting through a book bit by bit.

Read more...

Readlang now available in Estonian

I've been meaning to mention this for about two weeks now - Readlang.com now has Estonian as one of its options (upon my request). I haven't been able to use it yet for Estonian, but will when I get the time.

More interesting than that though is this post by Steve Ridout, the creator of the site, on exactly how it was created, how many people have subscribed, number of visitors, amount of revenue raised, everything. It's a very long read.

One other thing I noticed a few days ago is that vocabulary is now sorted by source, so if you want to go over all the vocab learned in Book A and not Book B, then you can do that now. I think I will start naming the text I paste in now, because at the moment I usually just hit a bunch of letters for the title, read the text, click on words I don't know, and delete it after I'm done and the words I want to see later are in the system. Now of course I don't remember what bbbbb, eee, and eeeeeeeeeeee were actually about.

Read more...

How much Chinese a beginner can learn in 4 months

I came across a great forum post today by someone who spent 4 months in Taiwan studying Chinese very instensively, about 70-80 hours a week. He (I will assume it's a he) started this experience as a beginner so it shows what can realistically be accomplished in such a short period of time, and the post is extremely detailed, including living and tutoring expenses, so it may help for others thinking of planning the same thing. Personally I do not like immersing myself in a language at this stage and also don't get the strong urge to do so until later on, but precisely this makes for an interesting comparison with how I learned Korean. One big difference right from the start: I never had to hire a tutor for conversation since making friends accomplished that, and being at an intermediate level upon arrival makes it that much easier to make such friends. At the lowest levels you need either an incredibly patient friend, or will have to pay a tutor / arrange a language exchange.

Immersing oneself in a language early on vs. later may seem to be irrelevant at first sight since both involve improving in a language, but it fact it is not: in most cases the student will end up returning to his home country, and after that "real life" will often set in. Back to school, back to work, out of money, etc. etc. If you, for example, have about $3000 in living expenses saved up but can make the trip now vs. later, later may be the better option if three or four months of exposure at a more advanced level can make all the difference between coming back an advanced beginner or coming back mostly fluent.

If you have a few hundred thousand saved up for language learning, however, then it won't matter a great deal. Spend four months abroad, come back, use lots of Italki or pay people to speak to you in person, and you will be able to continue the immersive experience wherever you happen to be.

Anyhow, read the post if at all curious what this person did in four months in Taiwan. Highly recommended.

Read more...

Was ist mit mir geschehen? Interlinearbooks is up to four books now

Interlinearbooks.com, the site originally inspired by this and this (well, just the first one), is now up to four books, each in a different language. Last time I mentioned it the site had two books and two languages: Swedish and Lithuanian. The fourth I won't mention here so go there to see it, but for the third I will steal an image from the site for readers to guess. Though I doubt it will be all that hard for anyone familiar with German literature.

Read more...

The ideal dictionary: L2 -- L2 --|-- L1

A few days ago I ordered a book from Wook.pt, a very highly unrecommended site for e-books. Portuguese is one of my many languages somewhere in between Conversation Ridge and Mount Fluency, so many of the books I would like to read in English I try to buy in Portuguese whenever I can. Anything written in the non-Brazilian variant is especially hard to find, so when I find something written in that type of Portuguese I usually snap it up. Which book I bought this time I won't say because then someone will ruin the ending for me.

Unfortunately the reader on that site is about the worst I've ever used, and after a week of trying on various computers and browsers I was finally able to get it working on an Android tablet. At least I can read it. At least I was able to practice the fine art of sending angry customer emails during that time (which were answered by automated messages). Pelo menos posso ler o livro que comprei.

One positive thing I can say about the online reader though is that it comes with a dictionary, a dictionary for Portuguese readers of course. Click on a word for about a second and one can see the definition, and with the explanations being in Portuguese it gives more of an immersive experience.

However, there are times when one reads and is able to understand the definition of a word...but still doesn't know exactly what word it is. This is particularly the case with things like types of clothing, trees, and the like. Let's say for example I came across the word abeto. What's an abeto? It's this:

(Botânica) árvore conífera da família das pináceas (Pinaceae) que se encontra na Europa, Ásia e América do Norte

Árvore da família das pináceas; pinheiro-alvar; madeira de abeto.

Okay, it's a kind of coniferous tree. How about a freixo?

designação comum a várias espécies de árvores da família das Oleáceas.

Árvore oleácea, tipo da tribo das fraxíneas.

Okay, it's another tree. And even with a more complete explanation containing leaf size, tree height and anything else one might want to know, I've never learned these trees as an L1 speaker along with this information so I still wouldn't know.

So this is one of the situations where one requires quick access to one's L1 to check the meaning of the word. Teachers in schools that are forbidden to use anything but the L2 will remember a lot of frustration here, where a student asks for the meaning of a word and the word just happens to be one of these. How does one explain a word like this without using the student's native language? Acting it out, drawing it on the board, looking up an image online...but even with an image of a tree, it's still better to just say fir, or ash. Abeto is fir, freixo is ash.

So in short, the best dictionary for an intermediate language learner is the following: one with the definition of the L2 word given entirely in the L2, with a tiny button somewhere that can be clicked on if necessary to show the L1 definition. Ideally this would only be used once every dozen or so words, but always available as a last resort. Without it, one must leave the book for a time to go off to Wiktionary or wherever to find what the word actually is, in spite of having a general concept of what it means.

Read more...

How to think in Spanish

I love these textbooks on Archive.org. One of them comes complete with movements to carry out while learning phrases.

Read more...

Kinderseele has arrived in the mail

This book is adorable! Here it is compared to an early proof of Demian (where the title is spaced a tad too high and to the left).

Lulu shows the exact dimensions of one book compared to the other.

Demian - 244 pages, 6 x 9, 0.93 lbs.
Kinderseele - 165 pages, 4.25 x 6.88, 0.4 lbs.


For any future longer projects I'll probably go with the size between these two, but for a book of its size (one quarter the length of Demian) the pocketbook size is perfect.

This weekend I will give it a look over in paper form to see if any small changes need to be made in the translation itself, so expect a post on that around Sunday or a bit after. But the design (as before, conceived by me, perfected by Lance Kraemer) is perfect.

Read more...

Readlang now has Afrikaans, plus some new features

I haven't written about Readlang.com in a while, and during that time it has had some interesting features added.

One change is that Afrikaans has been added - I thought it was already there but it was not, and Steve added it fairly soon after I requested it. Don't forget you can get anything else supported by Google Translate added too...unless it doesn't have spaces between the words.

I forget if I've mentioned this one, but the web reader is for me the most useful change in the past few months. Now the Chrome extension will keep the format of a page you are reading when you click on the green R, in effect putting an invisible Readlang screen on top of the page instead of importing the document and taking you to Readlang to read it and learn the content. This is much better than the previous system, where importing a text often resulted in a few pages of unrelated content (categories, other headlines, email addresses, physical addresses, etc.) before you got to the article. If you look at the front page of Welt.de for example the first text you will encounter is Abo, then Shop, then TV-Programm, Wetter, Anmelden, Registrieren, Home, Politik, Wirtschaft, and so on.

This change is interesting - a Reddit-like voting system for documents readers have shared. I won't be sharing any of mine as I prefer to delete anything I've read through once, but looking at what others are sharing can give one some good ideas for where to get content.

Finally the most recent one - the ability to type the answers to vocabulary cards instead of just clicking. Makes the process a bit more active.

Read more...

My second book is done. Kinderseele by Hermann Hesse

Readers of Page F30 will know that I created an interlinear translation of Hermann Hesse's Demian about two years ago, and published it on Lulu about half a year later. The motive behind an interlinear translation is to provide a translation, as literal as possible, that is lined up directly below the original text. Such a translation is exactly what I would have liked to buy a few years ago when I first started learning German, when I found the following:

1) the original German text was what I most wanted to read but could not understand it,
2) the English translation simply didn't feel like the same work (and in many cases was inaccurate), and
3) switching back and forth between the two was somewhat hard on the eyes and the brain, plus a non-literal translation did not always help in showing the meaning of the original text when the translator chose a different way to express an idea than Hesse did.

Because I so enjoyed making the translation the next question was what to translate next. I began reading Hesse's other works and ended up with a few candidates, but some of them were quite long, while none of them I enjoyed as much as I did Demian. Steppenwolf for example I enjoyed for the first half, the second...not so much. Eventually I came across a book called Klingsors letzter Sommer that I found on Gutenberg.org, and got ready for a book about a painter's last summer, as the synopsis said.

The first page of this book began with the word Kinderseele, which I first thought was the first chapter of Klingsors letzter Sommer. Because just about nothing I had ever come across in Hesse's works has ever truly resembled Demian, I was taken aback by this work which, from the first page onward, almost seemed like a rewriting of it. There again was the mention of unser Vaterhaus, "unsere" Welt, the light and comforting world of Sinclair's parents ("hinten führte die Treppe aus der dunklen Kühle empor und zu Licht und hellem Behagen"), then a comrade at school that had a certain aspect that Sinclair was drawn to (different from Demian here, but still), and so on. Eventually I realized this was a different book from the other two in Klingsors letzter Sommer, and read it through.

Kinderseele is not officially a prelude to Demian, but truly resembles one. Though it is a story that takes place when Sinclair (Hesse) is eleven years old, and Demian begins when Sinclair is ten, besides the discrepancy in numbers it reads like one that takes place one year before Demian. Take for example this part from Demian where he first steals from his own money box:

Es ging sehr leicht, sie aufzubrechen, es war nur ein dünnes Blechgitter zu durchreißen; aber der Riß tat weh, erst damit hatte ich Diebstahl begangen. Bis dahin hatte ich nur genascht, Zuckerstücke und Obst.
Dies nun war gestohlen, obwohl es mein eigenes Geld war.

The part in bold means "until then I had only filched (small things) like candy and fruit"...and in Kinderseele we see him stealing exactly that.

After that we have the relationship between him and his father, events that get out of control after Sinclair lies to a fellow student, and many more. Suffice to say, if it is not officially a prologue to Demian, everything in it reads as one.

So naturally I began the translation right away, finished it relatively quick due to its smaller length (about a quarter the size of Demian), and would have published it a few months ago were it not for a weird font issue with Libreoffice that caused half the text to disappear when I tried to make the document into a pdf. A new edition of Libreoffice has fixed this, and I was able to finish the project on Lulu.com this morning.

Here is the link to the book on lulu.com.

Cover design improvement is by Lance Kraemer, as before, as well as the proofreading which was done by Olivier Simon. This time I have made the book a pocket size edition, so it is 165 pages at 4.25 x 6.88, compared to Demian's 244 pages at 6 x 9. Price is also much less: $9.50 instead of $17.50, partly because of the smaller size but also because of opting for a smaller profit margin this time because Kinderseele is a much less well known work, and the more that can read it the better.

The cover looks like this.

Now that it is done I have ordered myself a copy, and will receive it in about a week. Though I have proofread it a number of times already, reading a paper edition is always a different experience than a file on a screen, and I could see myself making a change or two here or there, though I don't expect any major changes from the current edition. I will probably post about it again in two weeks' time, after I have received the first paper edition and have finished going through it.

One final note about the translation: the translation this time is still literal, but I have made it somewhat less literal than the one used in Demian due to the rarity of the book. It is not easy to simply pick up another translation of Kinderseele for comparison, and so I wanted to make it easier for the reader to understand. The smaller size of the book is another reason for this, because not as many words are able to fit on a single line. So for the most part it comes across just like the translation of Demian, just slightly more readable.

Read more...

Interlinearbooks.com has just launched

A short while after publishing my first interlinear book: Demian - An Interlinear Translation, a friend of mine contacted me about how I had arranged the translation below the original text. The book looks like this:

which makes it look like I used some sort of software to automatically arrange one line below the other. Unfortunately that was not the case. I spent a while searching for something that could do the trick, found nothing, and eventually just went with massive hitting of the spacebar below each and every line. I would not recommend doing that. Fortunately my love of the book itself made this less than a dreary task and I was able to spacebar this book into existence.

A while later this friend began developing his own software to automate the task, started a test site that he (and I, a bit) used for a few months on two more open-source books, and yesterday he has given me permission to announce the website to the world.

So here it is: Interlinearbooks.com.

I am pleased to see that he has explained the methodology as well:

just in case anyone comes along and asks why the translation below is so awkward at times. This is exactly the point - such a translation is supposed to keep nudging the reader toward the original text and away from the translation. However, it should never be so unclear that the reader is forced to go and pick up a dictionary to confirm what a sentence is supposed to mean.

Read more...

Half review, half request for review - Sayafter.me

About a week ago I learned about a website called Sayafter.me, a site for learning languages that is based on comparing one's pronunciation with that of a native speaker. At the moment it's only available to practice one's English, though the site owner says he will be adding other languages later. Because English is my first language I was able to verify at least that it correctly recognizes my pronunciation as correct (even though it was using RP and my pronunciation is Canadian), so far so good. I also had a Korean I know that doesn't know much English try it out and she got about 30 - 50% every time, also so far so good. But I'm curious to see what others at varying skill levels get.

If English is your L2 give this site a try and see what results you get, and let me know in the comments below. I'm curious about Olivier in particular.

Once other languages are added I'm going to try them out, and will be sure to not post about the results if my pronunciation is deemed to be bad.

(Just kidding)

Read more...

Bilinguis shows promise

Not as good as an interlinear text, but this site displaying Librivox texts side by side with audio is not bad at all (albeit still lacking in content).

Read more...

Now trying out Bliu Bliu

Bliu Bliu is pretty fun so far - it's a website for learning languages that seems somewhat similar to others like Lingq or Reading With Texts except that BB has an algorithm that guesses for you which words it thinks you know (based on which words you've already told it you know and don't know), and then goes around the web picking texts for you that seem suited to your level. Meanwhile, the selected texts can be recorded by native speakers if they so choose, meaning about one in twenty or so will come with audio.

The guessing algorithm is fun, resulting in a much more effortless experience than something like Lingq where you will say for example that you know the word chose but when the word choses comes up it will assume you don't know. On the other hand it also means that you will spend as much time marking words that you don't know (that BB assumes you did) as vice versa, so you have to be sure that you are scanning each text slowly and checking whether you actually know words that aren't highlighted.

That's all I have to write about it after a day of trying it out, may have more to say on the subject in a week or so.

Read more...

European radio stations all in one spot

My new favourite go to site is this one, a collection of radio stations in Europe by country, even sub-countries like the Faroe Islands and tiny ones like San Marino. Because they are grouped by country and not language that means that the one for Spain has Basque, Catalan and Galician along with the expected Spanish, France has Occitan, and Slovenia has Hungarian.

Galician (at least on that one station) comes across as more of a Portuguese-influenced Spanish than a Spanish-influenced Portuguese, at least phonetically. When written it comes across the other way.

Read more...

Sentence breaks now available on Readlang

Readlang.com now has a new feature, beyond a doubt the most exciting new feature the site has ever added, because I was the one that suggested it and now it's a reality. It's a simple one, which turns this:

into this:

Much easier to read. I came up with this idea when I was doing it myself to compare readability, and the thought struck me that it would be fairly easy to implement for a programmer of Steve Ridout's talent.

For those already using Readland: activate it by clicking on the settings tool, then AA, then change no sentence breaks to sentence breaks.

Read more...